,coral ,reefs1 ,volcanic ,islands1 ,south ,american ,geology by ,charles ,darwin ,,editorial ,,note ,although in some respects more technical in their subjects and style than ,darwin's 8,journal10 the books here reprinted will never lose their value and interest for the originality of the observations they contain4 ,many parts of them are admirably adapted for giving an insight into problems regarding the structure and changes of the earth's surface1 and in fact they form a charming introduction to physical geology and physiography in their application to special domains4 ,the books themselves cannot be obtained for many times the price of the present volume1 and both the general reader1 who desires to know more of ,darwin's work1 and the student of geology1 who naturally wishes to know how a master mind reasoned on most important geological subjects1 will be glad of the opportunity of possessing them in a convenient and cheap form4 ,the three introductions1 which my friend ,professor ,judd has kindly furnished1 give critical and historical information which makes this edition of special value4 ,g4,t4,b4 ,,contents ,,the ,,structure ,,and ,,distribution ,,of ,,coral ,,reefs4 ,,critical ,,introduction ,,introduction ,chapter ,i--,,atolls ,,or ,,lagoon-,,islands4 ,section ,i--,,description ,,of ,,keeling ,,atoll4 ,corals on the outer margin4--,zone of ,nullipor4--,exterior reef4--,islets4--,coral-conglomerate4--, lagoon4--,calcareous sediment4--,scari and ,holuthuri subsisting on corals4--,changes in the condition of the reefs and islets4--,probable subsidence of the atoll4--,future state of the lagoon4 ,section ,,ii--,,general ,,description ,,of ,,atolls4 ,general form and size of atolls1 their reefs and islets4--,external slope4--,zone of ,nullipor4--,conglomerate4--,depth of lagoons4--,sediment4--,reefs submerged wholly or in part4--,breaches in the reef4--,ledge-formed shores round certain lagoons4--,conversion of lagoons into land4 ,section ,,iii--,,atolls ,,of ,,the ,,maldiva ,,archipelago--,,great ,,chagos ,,bank4 ,maldiva ,archipelago4--,ring-formed reefs1 marginal and central4--,great depths in the lagoons of the southern atolls4--,reefs in the lagoons all rising to the surface4--,position of islets and breaches in the reefs1 with respect to the prevalent winds and action of the waves4--,destruction of islets4--,connection in the position and submarine foundation of distinct atolls4--,the apparent disseverment of large atolls4--,the ,great ,chagos ,bank4--,its submerged condition and extraordinary structure4 ,chapter ,,ii--,,barrier ,,reefs4 ,closely resemble in general form and structure atoll-reefs4--,width and depth of the lagoon-channels4--,breaches through the reef in front of valleys1 and generally on the leeward side4--,checks to the filling up of the lagoon-channels4--,size and constitution of the encircled islands4--,number of islands within the same reef4--,barrier-reefs of ,new ,caledonia and ,australia4--,position of the reef relative to the slope of the adjoining land4--,probable great thickness of barrier-reefs4 ,chapter ,,iii--,,fringing ,,or ,,shore-,,reefs4 ,reefs of ,mauritius4--,shallow channel within the reef4--,its slow filling up4--,currents of water formed within it4--,upraised reefs4--,narrow fringing-reefs in deep seas4--,reefs on the coast of ,e4 ,africa and of ,brazil4--,fringing-reefs in very shallow seas1 round banks of sediment and on worn-down islands4--,fringing-reefs affected by currents of the sea4--,coral coating the bottom of the sea1 but not forming reefs4 ,chapter ,,iv--,,on ,,the ,,distribution ,,and ,,growth ,,of ,,coral-,,reefs4 ,section ,i--,,on ,,the ,,distribution ,,of ,,coral-,,reefs1 ,,and ,,on ,,the ,,conditions ,,favourable ,,to ,,their ,,increase4 ,section ,,ii--,,on ,,the ,,rate ,,of ,,growth ,,of ,,coral-,,reefs4 ,section ,,iii--,,on ,,the ,,depths ,,at ,,which ,,reef-,,building ,,polypifers ,,can ,,live4 ,chapter ,v--,,theory ,,of ,,the ,,formation ,,of ,,the ,,different ,,classes ,,of ,,coral-,,reefs4 ,the atolls of the larger archipelagoes are not formed on submerged craters1 or on banks of sediment4--,immense areas interspersed with atolls4--,recent changes in their state4--,the origin of barrier-reefs and of atolls4--,their relative forms4--,the step-formed ledges and walls round the shores of some lagoons4--,the ring-formed reefs of the ,maldiva atolls4--,the submerged condition of parts or of the whole of some annular reefs4--,the disseverment of large atolls4--,the union of atolls by linear reefs4--,the ,great ,chagos ,bank4--,objections1 from the area and amount of subsidence required by the theory1 considered4--,the probable composition of the lower parts of atolls4 ,chapter ,,vi--,,on ,,the ,,distribution ,,of ,,coral-,,reefs ,,with ,,reference ,,to ,,the ,,theory ,,of ,,their ,,formation4 ,description of the coloured map4--,proximity of atolls and barrier- reefs4--,relation in form and position of atolls with ordinary islands4--,direct evidence of subsidence difficult to be detected4--,proofs of recent elevation where fringing-reefs occur4--,oscillations of level4--,absence of active volcanoes in the areas of subsidence4--,immensity of the areas which have been elevated and have subsided4--,their relation to the present distribution of the land4--,areas of subsidence elongated1 their intersection and alternation with those of elevation4--,amount and slow rate of the subsidence4--,recapitulation4 ,appendix ,containing a detailed description of the reefs and islands in ,plate ,,iii4 ,,geological ,,observations ,,on ,,volcanic ,,islands4 ,,critical ,,introduction ,chapter ,i--,,st4 ,,jago1 ,,in ,,the ,,cape ,,de ,,verde ,,archipelago4 ,rocks of the lowest series4--,a calcareous sedimentary deposit1 with recent shells1 altered by the contact of superincumbent lava1 its horizontality and extent4--,subsequent volcanic eruptions1 associated with calcareous matter in an earthy and fibrous form1 and often enclosed within the separate cells of the scori4--,ancient and obliterated orifices of eruption of small size4--,difficulty of tracing over a bare plain recent streams of lava4--,inland hills of more ancient volcanic rock4--,decomposed olivine in large masses4--,feldspathic rocks beneath the upper crystalline basaltic strata4--,uniform structure and form of the more ancient volcanic hills4--,form of the valleys near the coast4--,conglomerate now forming on the sea beach4 ,chapter ,,ii--,,fernando ,,noronha2 ,,terceira2 ,,tahiti1 ,,etc4 ,,fernando ,,noronha4--,precipitous hill of phonolite4 ,,terceira4--,trachytic rocks3 their singular decomposition by steam of high temperature4 ,,tahiti4--,passage from wacke into trap2 singular volcanic rock with the vesicles half-filled with mesotype4 ,,mauritius4--,proofs of its recent elevation4--,structure of its more ancient mountains2 similarity with ,st4 ,jago4 ,,st4 ,,paul',s ,,rocks4--,not of volcanic origin4--,their singular mineralogical composition4 ,chapter ,,iii--,,ascension4 ,basaltic lavas4--,numerous craters truncated on the same side4--,singular structure of volcanic bombs4--,aeriform explosions4--,ejected granite fragments4--,trachytic rocks4--,singular veins4--,jasper1 its manner of formation4--,concretions in pumiceous tuff4--,calcareous deposits and frondescent incrustations on the coast4--,remarkable laminated beds1 alternating with1 and passing into obsidian4--,origin of obsidian4--,lamination of volcanic rocks4 ,chapter ,,iv--,,st4 ,,helena4 ,lavas of the feldspathic1 basaltic1 and submarine series4--,section of ,flagstaff ,hill and of the ,barn4--,dikes4--,turk's ,cap and ,prosperous ,bays4--,basaltic ring4--,central crateriform ridge1 with an internal ledge and a parapet4--,cones of phonolite4--,superficial beds of calcareous sandstone4--,extinct land-shells4--,beds of detritus4--,elevation of the land4--,denudation4--,craters of elevation4 ,chapter ,v--,,galapagos ,,archipelago4 ,chatham ,island4--,craters composed of a peculiar kind of tuff4--,small basaltic craters1 with hollows at their bases4--,albemarle ,island2 fluid lavas1 their composition4--,craters of tuff2 inclination of their exterior diverging strata1 and structure of their interior converging strata4--,james ,island1 segment of a small basaltic crater2 fluidity and composition of its lava-streams1 and of its ejected fragments4--,concluding remarks on the craters of tuff1 and on the breached condition of their southern sides4--,mineralogical composition of the rocks of the archipelago4--,elevation of the land4--,direction of the fissures of eruption4 ,chapter ,,vi--,,trachyte ,,and ,,basalt4--,,distribution ,,of ,,volcanic ,,isles4 ,the sinking of crystals in fluid lava4--,specific gravity of the constituent parts of trachyte and of basalt1 and their consequent separation4--,obsidian4--,apparent non-separation of the elements of plutonic rocks4--,origin of trap-dikes in the plutonic series4--,distribution of volcanic islands2 their prevalence in the great oceans4--,they are generally arranged in lines4--,the central volcanoes of ,von ,buch doubtful4--,volcanic islands bordering continents4--,antiquity of volcanic islands1 and their elevation in mass4--,eruptions on parallel lines of fissure within the same geological period4 ,chapter ,,vii--,,australia2 ,,new ,,zealand2 ,,cape ,,of ,,good ,,hope4 ,new ,south ,wales4--,sandstone formation4--,embedded pseudo-fragments of shale4--,stratification4--,current-cleav age4--,great valleys4--,van ,diemen's ,land4--,palozoic formation4--,newer formation with volcanic rocks4--,travertin with leaves of extinct plants4--,elevation of the land4--,new ,zealand4--,king ,george's ,sound4--,superficial ferruginous beds4--,superficial calcareous deposits1 with casts of branches2 its origin from drifted particles of shells and corals4--,their extent4--,cape of ,good ,hope4--,junction of the granite and clay-slate4--,sandstone formation4 ,,geological ,,observations ,,on ,,south ,,america4 ,,critical ,,introduction ,chapter ,i--,,on ,,the ,,elevation ,,of ,,the ,,eastern ,,coast ,,of ,,south ,,america4 ,upraised shells of ,la ,plata4--,bahia ,blanca1 ,sand-dunes and ,pumice-pebbles4--,step-formed plains of ,patagonia1 with upraised shells4--,terrace-bounded valley of ,santa ,cruz1 formerly a sea-strait4--,upraised shells of ,tierra del ,fuego4--,length and breadth of the elevated area4--,equability of the movements1 as shown by the similar heights of the plains4--,slowness of the elevatory process4--,mode of formation of the step-formed plains4--,summary4--,great shingle formation of ,patagonia2 its extent1 origin1 and distribution4--,formation of sea-cliffs4 ,chapter ,,ii--,,on ,,the ,,elevation ,,of ,,the ,,western ,,coast ,,of ,,south ,,america4 ,chonos ,archipelago4--,chiloe1 recent and gradual elevation of1 traditions of the inhabitants on this subject4--,concepcion1 earthquake and elevation of4--,,valparaiso1 great elevation of1 upraised shells1 earth or marine origin1 gradual rise of the land within the historical period4--,,coquimbo1 elevation of1 in recent times2 terraces of marine origin1 their inclination1 their escarpments not horizontal4--,guasco1 gravel terraces of4--,copiapo4--,,peru4--,upraised shells of ,cobija1 ,iquique1 and ,arica4--,lima1 shell-beds and sea-beach on ,san ,lorenzo4--,human remains1 fossil earthenware1 earthquake debacle1 recent subsidence4--,on the decay of upraised shells4--,general summary4 ,chapter ,,iii--,,on ,,the ,,plains ,,and ,,valleys ,,of ,,chile3--,,saliferous ,,superficial ,,deposits4 ,basin-like plains of ,chile2 their drainage1 their marine origin4--,marks of sea-action on the eastern flanks of the ,cordillera4--,sloping terrace-like fringes of stratified shingle within the valleys of the ,cordillera2 their marine origin4--,boulders in the valley of ,cachapual4--,horizontal elevation of the ,cordillera4--,formation of valleys4--,boulders moved by earthquake-waves4--,saline superficial deposits4--,bed of nitrate of soda at ,iquique4--,saline incrustations4--,salt-lakes of ,la ,plata and ,patagonia2 purity of the salt2 its origin4 ,chapter ,,iv--,,on ,,the ,,formations ,,of ,,the ,,pampas4 ,mineralogical constitution4--,microscopical structure4--,buenos ,ayres1 shells embedded in tosca-rock4--,buenos ,ayres to the ,colorado4--,s4 ,ventana4--,bahia ,blanca2 ,m4 ,hermoso1 bones and infusoria of2 ,p4 ,alta1 shells1 bones1 and infusoria of2 co-existence of the recent shells and extinct mammifers4--,buenos ,ayres to ,st4 ,fe4--,skeletons of ,mastodon4--,infusoria4--,inferior marine tertiary strata1 their age4--,horse's tooth4 ,,banda ,,oriental4--,superficial ,pampean formation4--,inferior tertiary strata1 variation of1 connected with volcanic action2 ,macrauchenia ,patachonica at ,s4 ,julian in ,patagonia1 age of1 subsequent to living mollusca and to the erratic block period4 ,,summary4--,area of ,pampean formation4--,theories of origin4--,source of sediment4--,estuary origin4--,contemporaneous with existing mollusca4--,relations to underlying tertiary strata4 ,ancient deposit of estuary origin4--,elevation and successive deposition of the ,pampean formation4--,number and state of the remains of mammifers2 their habitation1 food1 extinction1 and range4--,conclusion4--,supplement on the thickness of the ,pampean formation4--,localities in ,pampas at which mammiferous remains have been found4 ,chapter ,v--,,on ,,the ,,older ,,tertiary ,,formations ,,of ,,patagonia ,,and ,,chile4 ,rio ,negro4--,s4 ,josef4--,port ,desire1 white pumiceous mudstone with infusoria4--,port ,s4 ,julian4--,santa ,cruz1 basaltic lava of4--,p4 ,gallegos4--,eastern ,tierra del ,fuego2 leaves of extinct beech-trees4--,summary on the ,patagonian tertiary formations4--,tertiary formations of the ,western ,coast4--,chonos and ,chiloe groups1 volcanic rocks of4--,concepcion4--,navidad4--,coquimbo4 --,summary4--,age of the tertiary formations4--,lines of elevation4--,silicified wood4--,comparative ranges of the extinct and living mollusca on the ,west ,coast of ,s4 ,america4--,climate of the tertiary period4--,on the causes of the absence of recent conchiferous deposits on the coasts of ,south ,america4--,on the contemporaneous deposition and preservation of sedimentary formations4 ,chapter ,,vi--,,plutonic ,,and ,,metamorphic ,,rocks3--,,cleavage ,,and ,,foliation4 ,brazil1 ,bahia1 gneiss with disjointed metamorphosed dikes4--,strike of foliation4--,rio de ,janeiro1 gneiss-granite1 embedded fragment in1 decomposition of4--,la ,plata1 metamorphic and old volcanic rocks of4--,s4 ,ventana4--,claystone porphyry formation of ,patagonia2 singular metamorphic rocks2 pseudo-dikes4--,falkland ,islands1 palozoic fossils of4--,tierra del ,fuego1 clay-slate formation1 cretaceous fossils of2 cleavage and foliation2 form of land4--,chonos ,archipelago1 mica-schists1 foliation disturbed by granitic axis2 dikes4--,chiloe4--,concepcion1 dikes1 successive formation of4--,central and ,northern ,chile4--,concluding remarks on cleavage and foliation4--,their close analogy and similar origin4--,stratification of metamorphic schists4--,foliation of intrusive rocks4--,relation of cleavage and foliation to the lines of tension during metamorphosis4 ,chapter ,,vii--,,central ,,chile3--,,structure ,,of ,,the ,,cordillera4 ,central ,chile4--,basal formations of the ,cordillera4--,origin of the porphyritic clay-stone conglomerate4--,andesite4--,volcanic rocks4--,section of the ,cordillera by the ,peuquenes or ,portillo ,pass4--,great gypseous formation4--,peuquenes line2 thickness of strata1 fossils of4--,portillo line4--,conglomerate1 orthitic granite1 mica-schist1 volcanic rocks of4--,concluding remarks on the denudation and elevation of the ,portillo line4--,section by the ,cumbre1 or ,uspallata ,pass4--,porphyries4--,gypseous strata4--,section near the ,puente del ,inca2 fossils of4--,great subsidence4--,intrusive porphyries4--,plain of ,uspallata4--,section of the ,uspallata chain4--,structure and nature of the strata4--,silicified vertical trees4--,great subsidence4--,granitic rocks of axis4--,concluding remarks on the ,uspallata range2 origin subsequent to that of the main ,cordillera2 two periods of subsidence2 comparison with the ,portillo chain4 ,chapter ,,viii--,,northern ,,chile4--,,conclusion4 ,section from ,illapel to ,combarbala2 gypseous formation with silicified wood4--,panuncillo4--,coquimbo2 mines of ,arqueros2 section up valley2 fossils4--,guasco1 fossils of4--,copiapo1 section up valley2 ,las ,amolanas1 silicified wood4--,conglomerates1 nature of former land1 fossils1 thickness of strata1 great subsidence4--,valley of ,despoblado1 fossils1 tufaceous deposit1 complicated dislocations of4--,relations between ancient orifices of eruption and subsequent axes of injection4--,iquique1 ,peru1 fossils of1 salt-deposits4--,metalliferous veins4--,summary on the porphyritic conglomerate and gypseous formations4--,great subsidence with partial elevations during the cretaceo-oolitic period4--,on the elevation and structure of the ,cordillera4--,recapitulation on the tertiary series4--,relation between movements of subsidence and volcanic action4--,pampean formation4--,recent elevatory movements4--,long-continued volcanic action in the ,cordillera4--,conclusion4 ,index to 8,coral-,reefs0 ,index to 8,volcanic ,islands0 ,index to 8,south ,american ,observations0 ,,the ,,structure ,,and ,,distribution ,,of ,,coral ,,reefs4 ,,critical ,,introduction ,a scientific discovery is the outcome of an interesting process of evolution in the mind of its author4 ,when we are able to detect the germs of thought in which such a discovery has originated1 and to trace the successive stages of the reasoning by which the crude idea has developed into an epoch-making book1 we have the materials for reconstructing an important chapter of scientific history4 ,such a contribution to the story of the 8making of science0 may be furnished in respect to ,darwin's famous theory of coral-reefs1 and the clearly reasoned treatise in which it was first fully set forth4 ,the subject of corals and coral-reefs is one concerning which much popular misconception has always prevailed4 ,the misleading comparison of coral-rock with the combs of bees and the nests of wasps is perhaps responsible for much of this misunderstanding2 one writer has indeed described a coral-reef as being 8built by fishes by means of their teeth40 ,scarcely less misleading1 however1 are the references we so frequently meet with1 both in prose and verse1 to the 8skill10 8industry10 and 8perseverance0 of the 8coral-insect0 in 8building0 his 8home40 ,as well might we praise men for their cleverness in making their own skeletons1 and laud their assiduity in filling churchyards with the same4 ,the polyps and other organisms1 whose remains accumulate to form a coral-reef1 simply live and perform their natural functions1 and then die1 leaving behind them1 in the natural course of events1 the hard calcareous portions of their structures to add to the growing reef4 ,while the forms of coral-reefs and coral-islands are sometimes very remarkable and worthy of attentive study1 there is no ground1 it need scarcely be added1 for the suggestion that they afford proofs of design on the part of the living builders1 or that1 in the words of ,flinders1 they constitute breastworks1 defending the workshops from whence 8infant colonies might be safely sent forth40 ,it was not till the beginning of the present century that travellers like ,beechey1 ,chamisso1 ,quoy and ,gaimard1 ,moresby1 ,nelson1 and others1 began to collect accurate details concerning the forms and structure of coral-masses1 and to make such observations on the habits of reef-forming polyps1 as might serve as a basis for safe reasoning concerning the origin of coral-reefs and islands4 ,in the second volume of ,lyell's 8,principles of ,geology10 published in #ahcb1 the final chapter gives an admirable summary of all that was then known on the subject4 ,at that time1 the ring-form of the atolls was almost universally regarded as a proof that they had grown up on submerged volcanic craters2 and ,lyell gave his powerful support to that theory4 ,charles ,darwin was never tired of acknowledging his indebtedness to ,lyell4 ,in dedicating to his friend the second edition of his 8,naturalist's ,voyage ,round the ,world10 ,darwin writes that he does so 8with grateful pleasure1 as an acknowledgment that the chief part of whatever scientific merit this journal and the other works of the author may possess1 has been derived from studying the well-known and admirable #8,principles of ,geology4'0 ,the second volume of ,lyell's 8,principles0 appeared after ,darwin had left ,england2 but it was doubtless sent on to him without delay by his faithful friend and correspondent1 ,professor ,henslow4 ,it appears to have reached ,darwin at a most opportune moment1 while1 in fact1 he was studying the striking evidences of slow and long-continued1 but often interrupted movement on the west coast of ,south ,america4 ,darwin's acute mind could not fail to detect the weakness of the then prevalent theory concerning the origin of the ring-shaped atolls--and the difficulty which he found in accepting the volcanic theory1 as an explanation of the phenomena of coral-reefs1 is well set forth in his book4 ,in an interesting fragment of autobiography1 ,darwin has given us a very clear account of the way in which the leading idea of the theory of coral-reefs originated in his mind2 he writes1 8,no other work of mine was begun in so deductive a spirit as this1 for the whole theory was thought out on the west coast of ,south ,america1 before ,i had seen a true coral-reef4 ,i had therefore only to verify and extend my views by a careful examination of living reefs4 ,but it should be observed that ,i had during the two previous years been incessantly attending to the effects on the shores of ,south ,america of the intermittent elevation of the land1 together with the denudation and deposition of sediment4 ,this necessarily led me to reflect much on the effects of subsidence1 and it was easy to replace in imagination the continued deposition of sediment by the upward growth of corals4 ,to do this was to form my theory of the formation of barrier-reefs and atolls40 ,on her homeward voyage1 the ,beagle visited ,tahiti1 ,australia1 and some of the coral-islands in the ,indian ,ocean1 and ,darwin had an opportunity of testing and verifying the conclusion at which he had arrived by studying the statements of other observers4 ,i well recollect a remarkable conversation ,i had with ,darwin1 shortly after the death of ,lyell4 ,with characteristic modesty1 he told me that he never fully realised the importance of his theory of coral-reefs till he had an opportunity of discussing it with ,lyell1 shortly after the return of the ,beagle4 ,lyell1 on receiving from the lips of its author a sketch of the new theory1 was so overcome with delight that he danced about and threw himself into the wildest contortions1 as was his manner when excessively pleased4 ,he wrote shortly afterwards to ,darwin as follows3--8,i could think of nothing for days after your lesson on coral-reefs1 but of the tops of submerged continents4 ,it is all true1 but do not flatter yourself that you will be believed till you are growing bald like me1 with hard work and vexation at the incredulity of the world40 ,on ,may #bdth1 #ahcg1 ,lyell wrote to ,sir ,john ,herschel as follows3--8,i am very full of ,darwin's new theory of coral-islands1 and have urged ,whewell to make him read it at our next meeting4 ,i must give up my volcanic crater forever1 though it cost me a pang at first1 for it accounted for so much40 ,dr4 ,whewell was president of the ,geological ,society at the time1 and on ,may #cast1 #ahcg1 ,darwin read a paper entitled 8,on ,certain ,areas of ,elevation and ,subsidence in the ,pacific and ,indian oceans1 as deduced from the ,study of ,coral ,formations10 an abstract of which appeared in the second volume of the ,society's proceedings4 ,it was about this time that ,darwin1 having settled himself in lodgings at ,great ,marlborough ,street1 commenced the writing of his book on 8,coral-,reefs40 ,many delays from ill-health and the interruption of other work1 caused the progress to be slow1 and his journal speaks of 8recommencing0 the subject in ,february #ahci1 shortly after his marriage1 and again in ,october of the same year4 ,in ,july #ahda1 he states that he began once more 8after more than thirteen month's interval10 and the last proof-sheet of the book was not corrected till ,may #fth1 #ahdb4 ,darwin writes in his autobiography1 8,this book1 though a small one1 cost me twenty months of hard work1 as ,i had to read every work on the islands of the ,pacific1 and to consult many charts40 ,the task of elaborating and writing out his books was1 with ,darwin1 always a very slow and laborious one2 but it is clear that in accomplishing the work now under consideration1 there was a long and constant struggle with the lethargy and weakness resulting from the sad condition of his health at that time4 ,lyell's anticipation that the theory of coral-reefs would be slow in meeting with general acceptance was certainly not justified by the actual facts4 ,on the contrary the new book was at once received with general assent among both geologists and zoologists1 and even attracted a considerable amount of attention from the general public4 ,it was not long before the coral-reef theory of ,darwin found an able exponent and sturdy champion in the person of the great ,american naturalist1 ,professor ,james ,d4 ,dana4 ,two years after the return of the ,beagle to ,england1 the ships of the ,united ,states ,exploring ,expedition set sail upon their four years' cruise1 under the command of ,captain ,wilkes1 and ,dana was a member of the scientific staff4 ,when1 in #ahci1 the expedition arrived at ,sydney1 a newspaper paragraph was found which gave the ,american naturalist the first intimation of ,darwin's new theory of the origin of atolls and barrier-reefs4 ,writing in #ahgb1 ,dana describes the effect produced on his mind by reading this passage3--8,the paragraph threw a flood of light over the subject1 and called forth feelings of peculiar satisfaction1 and of gratefulness to ,mr4 ,darwin1 which still come up afresh whenever the subject of coral islands is mentioned4 ,the ,gambier ,islands in the ,paumotus1 which gave him the key to the theory1 ,i had not seen2 but on reaching the ,feejees1 six months later1 in #ahdj1 ,i found there similar facts on a still grander scale and of a more diversified character1 so that ,i was afterward enabled to speak of his theory as established with more positiveness than he himself1 in his philosophic caution1 had been ready to adopt4 ,his work on coral-reefs appeared in #ahdb1 when my report on the subject was already in manuscript4 ,it showed that the conclusions on other points1 which we had independently reached1 were for the most part the same4 ,the principal points of difference relate to the reason for the absence of corals from some coasts1 and the evidence therefrom as to changes of level1 and the distribution of the oceanic regions of elevation and subsidence--topics which a wide range of travel over the ,pacific brought directly and constantly to my attention40 ,among the ,reports of the ,united ,states ,exploring ,expedition1 two important works from the pen of ,professor ,dana made their appearance2--one on 8,zoophytes10 which treats at length on 8,corals and ,coral-,animals10 and the other on 8,coral-,reefs and ,islands40 ,in #ahgb1 ,dana prepared a work of a more popular character in which some of the chief results of his studies are described2 it bore the title of 8,corals and ,coral-,islands40 ,of this work1 new and enlarged editions appeared in #ahgd and #ahij in ,america1 while two editions were published in this country in #ahgb and #ahge4 ,in all these works their author1 while maintaining an independent judgment on certain matters of detail1 warmly defends the views of ,darwin on all points essential to the theory4 ,another able exponent and illustrator of the theory of coral-reefs was found in ,professor ,j4 ,b4 ,jukes1 who accompanied ,h4,m4,s4 ,fly1 as naturalist1 during the survey of the ,great ,barrier-,reef--in the years #ahdb to #ahdf4 ,jukes1 who was a man of great acuteness as well as independence of mind1 concludes his account of the great ,australian reefs with the following words3--8,after seeing much of the ,great ,barrier-,reefs1 and reflecting much upon them1 and trying if it were possible by any means to evade the conclusions to which ,mr4 ,darwin has come1 ,i cannot help adding that his hypothesis is perfectly satisfactory to my mind1 and rises beyond a mere hypothesis into the true theory of coral-reefs40 ,as the result of the clear exposition of the subject by ,darwin1 ,lyell1 ,dana1 and ,jukes1 the theory of coral-reefs had1 by the middle of the present century1 commanded the almost universal assent of both biologists and geologists4 ,in #ahei ,baron von ,richthofen brought forward new facts in its support1 by showing that the existence of the thick masses of dolomitic limestone in the ,tyrol could be best accounted for if they were regarded as of coralline origin and as being formed during a period of long continued subsidence4 ,the same views were maintained by ,professor ,mojsisovics in his 8,dolomit-riffe von ,sdtirol und ,venetien10 which appeared in #ahgi4 ,the first serious note of dissent to the generally accepted theory was heard in #ahfc1 when a distinguished ,german naturalist1 ,dr4 ,karl ,semper1 declared that his study of the ,pelew ,islands showed that uninterrupted subsidence could not have been going on in that region4 ,dr4 ,semper's objections were very carefully considered by ,mr4 ,darwin1 and a reply to them appeared in the second and revised edition of his 8,coral-,reefs10 which was published in #ahgd4 ,with characteristic frankness and freedom from prejudice1 ,darwin admitted that the facts brought forward by ,dr4 ,semper proved that in certain specified cases1 subsidence could not have played the chief part in originating the peculiar forms of the coral-islands4 ,but while making this admission1 he firmly maintained that exceptional cases1 like those described in the ,pelew ,islands1 were not sufficient to invalidate the theory of subsidence as applied to the widely spread atolls1 encircling reefs1 and barrier-reefs of the ,pacific and ,indian ,oceans4 ,it is worthy of note that to the end of his life ,darwin maintained a friendly correspondence with ,semper concerning the points on which they were at issue4 ,after the appearance of ,semper's work1 ,dr4 ,j4 ,j4 ,rein published an account of the ,bermudas1 in which he opposed the interpretation of the structure of the islands given by ,nelson and other authors1 and maintained that the facts observed in them are opposed to the views of ,darwin4 ,although1 so far as ,i am aware1 ,darwin had no opportunity of studying and considering these particular objections1 it may be mentioned that two ,american geologists have since carefully re-examined the district--,professor ,w4 ,n4 ,rice in #ahhd and ,professor ,a4 ,heilprin in #ahhi--;and they have independently arrived at the conclusion that ,dr4 ,rein's objections cannot be maintained4 ,the most serious opposition to ,darwin's coral-reef theory1 however1 was that which developed itself after the return of ,h4,m4,s4 ,challenger from her famous voyage4 ,mr4 ,john ,murray1 one of the staff of naturalists on board that vessel1 propounded a new theory of coral-reefs1 and maintained that the view that they were formed by subsidence was one that was no longer tenable2 these objections have been supported by ,professor ,alexander ,agassiz in the ,united ,states1 and by ,dr4 ,a4 ,geikie1 and ,dr4 ,h4 ,b4 ,guppy in this country4 ,although ,mr4 ,darwin did not live to bring out a third edition of his 8,coral-,reefs10 ,i know from several conversations with him that he had given the most patient and thoughtful consideration to ,mr4 ,murray's paper on the subject4 ,he admitted to me that had he known1 when he wrote his work1 of the abundant deposition of the remains of calcareous organisms on the sea floor1 he might have regarded this cause as sufficient in a few cases to raise the summits of submerged volcanoes or other mountains to a level at which reef-forming corals can commence to flourish4 ,but he did not think that the admission that under certain favourable conditions1 atolls might be thus formed without subsidence1 necessitated an abandonment of his theory in the case of the innumerable examples of the kind which stud the ,indian and ,pacific ,oceans4 ,a letter written by ,darwin to ,professor ,alexander ,agassiz in ,may #ahha shows exactly the attitude which careful consideration of the subject led him to maintain towards the theory propounded by ,mr4 ,murray3--8,you will have seen10 he writes1 8,mr4 ,murray's views on the formation of atolls and barrier-reefs4 ,before publishing my book1 ,i thought long over the same view1 but only as far as ordinary marine organisms are concerned1 for at that time little was known of the multitude of minute oceanic organisms4 ,i rejected this view1 as from the few dredgings made in the ,beagle1 in the south temperate regions1 ,i concluded that shells1 the smaller corals1 etc41 decayed and were dissolved when not protected by the deposition of sediment1 and sediment could not accumulate in the open ocean4 ,certainly1 shells1 etc41 were in several cases completely rotten1 and crumbled into mud between my fingers2 but you will know whether this is in any degree common4 ,i have expressly said that a bank at the proper depth would give rise to an atoll1 which could not be distinguished from one formed during subsidence4 ,i can1 however1 hardly believe in the existence of as many banks 7there having been no subsidence7 as there are atolls in the great oceans1 within a reasonable depth1 on which minute oceanic organisms could have accumulated to the depth of many hundred feet40 ,darwin's concluding words in the same letter written within a year of his death1 are a striking proof of the candour and openness of mind which he preserved so well to the end1 in this as in other controversies4 8,if ,i am wrong1 the sooner ,i am knocked on the head and annihilated so much the better4 ,it still seems to me a marvellous thing that there should not have been much1 and long-continued1 subsidence in the beds of the great oceans4 ,i wish some doubly rich millionaire would take it into his head to have borings made in some of the ,pacific and ,indian atolls1 and bring home cores for slicing from a depth of #ejj or #fjj feet40 ,it is noteworthy that the objections to ,darwin's theory have for the most part proceeded from zoologists1 while those who have fully appreciated the geological aspect of the question1 have been the staunchest supporters of the theory of subsidence4 ,the desirability of such boring operations in atolls has been insisted upon by several geologists1 and it may be hoped that before many years have passed away1 ,darwin's hopes may be realised1 either with or without the intervention of the 8doubly rich millionaire40 ,three years after the death of ,darwin1 the veteran ,professor ,dana re-entered the lists and contributed a powerful defence of the theory of subsidence in the form of a reply to an essay written by the ablest exponent of the anti-,darwinian views on this subject1 ,dr4 ,a4 ,geikie4 ,while pointing out that the ,darwinian position had been to a great extent misunderstood by its opponents1 he showed that the rival theory presented even greater difficulties than those which it professed to remove4 ,during the last five years1 the whole question of the origin of coral-reefs and islands has been re-opened1 and a controversy has arisen1 into which1 unfortunately1 acrimonious elements have been very unnecessarily introduced4 ,those who desire it1 will find clear and impartial statements of the varied and often mutually destructive views put forward by different authors1 in three works which have made their appearance within the last year1--8,the ,bermuda ,islands10 by ,professor ,angelo ,heilprin2 8,corals and ,coral-,islands10 new edition by ,professor ,j4 ,d4 ,dana2 and the third edition of ,darwin's 8,coral-,reefs10 with ,notes and ,appendix by ,professor ,t4 ,g4 ,bonney4 ,most readers will1 ,i think1 rise from the perusal of these works with the conviction that1 while on certain points of detail it is clear that1 through the want of knowledge concerning the action of marine organisms in the open ocean1 ,darwin was betrayed into some grave errors1 yet the main foundations of his argument have not been seriously impaired by the new facts observed in the deep-sea researches1 or by the severe criticism to which his theory has been subjected during the last ten years4 ,on the other hand1 ,i think it will appear that much misapprehension has been exhibited by some of ,darwin's critics1 as to what his views and arguments really were2 so that the reprint and wide circulation of the book in its original form is greatly to be desired1 and cannot but be attended with advantage to all those who will have the fairness to acquaint themselves with ,darwin's views at first hand1 before attempting to reply to them4 ,,john ,w4 ,,judd4 ,,coral-,,reefs ,,introduction ,the object of this volume is to describe from my own observation and the works of others1 the principal kinds of coral-reefs1 more especially those occurring in the open ocean1 and to explain the origin of their peculiar forms4 ,i do not here treat of the polypifers1 which construct these vast works1 except so far as relates to their distribution1 and to the conditions favourable to their vigorous growth4 ,without any distinct intention to classify coral-reefs1 most voyagers have spoken of them under the following heads3 8lagoon-islands10 or 8atolls10 8barrier0 or 8encircling reefs10 and 8fringing0 or 8shore-reefs40 ,the lagoon-islands have received much the most attention2 and it is not surprising1 for every one must be struck with astonishment1 when he first beholds one of these vast rings of coral-rock1 often many leagues in diameter1 here and there surmounted by a low verdant island with dazzling white shores1 bathed on the outside by the foaming breakers of the ocean1 and on the inside surrounding a calm expanse of water1 which from reflection1 is of a bright but pale green colour4 ,the naturalist will feel this astonishment more deeply after having examined the soft and almost gelatinous bodies of these apparently insignificant creatures1 and when he knows that the solid reef increases only on the outer edge1 which day and night is lashed by the breakers of an ocean never at rest4 ,well did ,franois ,pyrard de ,laval1 in the year #afje1 exclaim1 8,c'est une mrueille de voir chacun de ces atollons1 enuironn d'un grand banc de pierre tout autour1 n'y ayant point d'artifice humain40 ,the accompanying sketch of ,whitsunday island1 in the ,south ,pacific1 taken from ,captain ,beechey's admirable 8,voyage10 although excellent of its kind1 gives but a faint idea of the singular aspect of one of these lagoon-islands4 ,whitsunday ,island is of small size1 and the whole circle has been converted into land1 which is a comparatively rare circumstance4 ,as the reef of a lagoon-island generally supports many separate small islands1 the word 8island10 applied to the whole1 is often the cause of confusion2 hence ,i have invariably used in this volume the term 8atoll10 which is the name given to these circular groups of coral-islets by their inhabitants in the ,indian ,ocean1 and is synonymous with 8lagoon- island40 ,illustration3 ,whitsunday ,island ,barrier-reefs1 when encircling small islands1 have been comparatively little noticed by voyagers2 but they well deserve attention4 ,in their structure they are little less marvellous than atolls1 and they give a singular and most picturesque character to the scenery of the islands they surround4 ,in the accompanying sketch1 taken from the 8,voyage of the ,coquille10 the reef is seen from within1 from one of the high peaks of the island of ,bolabola4#a ,here1 as in ,whitsunday ,island1 the whole of that part of the reef which is visible is converted into land4 ,this is a circumstance of rare occurrence2 more usually a snow-white line of great breakers1 with here and there an islet crowned by cocoa-nut trees1 separates the smooth waters of the lagoon-like channel from the waves of the open sea4 ,the barrier-reefs of ,australia and of ,new ,caledonia1 owing to their enormous dimensions1 have excited much attention3 in structure and form they resemble those encircling many of the smaller islands in the ,pacific ,ocean4 #a ,i have taken the liberty of simplifying the foreground1 and leaving out a mountainous island in the far distance4 ,illustration3 ,island of ,bolabola ,with respect to fringing1 or shore-reefs1 there is little in their structure which needs explanation2 and their name expresses their comparatively small extension4 ,they differ from barrier-reefs in not lying so far from the shore1 and in not having within a broad channel of deep water4 ,reefs also occur around submerged banks of sediment and of worn-down rock2 and others are scattered quite irregularly where the sea is very shallow2 these in most respects are allied to those of the fringing class1 but they are of comparatively little interest4 ,i have given a separate chapter to each of the above classes1 and have described some one reef or island1 on which ,i possessed most information1 as typical2 and have afterwards compared it with others of a like kind4 ,although this classification is useful from being obvious1 and from including most of the coral-reefs existing in the open sea1 it admits of a more fundamental division into barrier and atoll-formed reefs on the one hand1 where there is a great apparent difficulty with respect to the foundation on which they must first have grown2 and into fringing-reefs on the other1 where1 owing to the nature of the slope of the adjoining land1 there is no such difficulty4 ,the two blue tints and the red colour#b on the map 7,plate ,,iii71 represent this main division1 as explained in the beginning of the last chapter4 ,in the ,appendix1 every existing coral-reef1 except some on the coast of ,brazil not included in the map1 is briefly described in geographical order1 as far as ,i possessed information2 and any particular spot may be found by consulting the ,index4 ,several theories have been advanced to explain the origin of atolls or lagoon-islands1 but scarcely one to account for barrier-reefs4 ,from the limited depths at which reef-building polypifers can flourish1 taken into consideration with certain other circumstances1 we are compelled to conclude1 as it will be seen1 that both in atolls and barrier-reefs1 the foundation on which the coral was primarily attached1 has subsided2 and that during this downward movement1 the reefs have grown upwards4 ,this conclusion1 it will be further seen1 explains most satisfactorily the outline and general form of atolls and barrier-reefs1 and likewise certain peculiarities in their structure4 ,the distribution1 also1 of the different kinds of coral-reefs1 and their position with relation to the areas of recent elevation1 and to the points subject to volcanic eruptions1 fully accord with this theory of their origin4#c #b ,replaced by numbers in this edition4 #c ,a brief account of my views on coral formations1 now published in my ,journal of ,researches1 was read ,may #cast1 #ahcg1 before the ,geological ,society1 and an abstract has appeared in the ,proceedings4 ,in the several original surveys1 from which the small plans on this plate have been reduced1 the coral-reefs are engraved in very different styles4 ,for the sake of uniformity1 ,i have adopted the style used in the charts of the ,chagos ,archipelago1 published by the ,east ,indian ,company1 from the survey by ,captain ,moresby and ,lieutenant ,powell4 ,the surface of the reef1 which dries at low water1 is represented by a surface with small crosses3 the coral-islets on the reef are marked by small linear spaces1 on which a few cocoa-nut trees1 out of all proportion too large1 have been introduced for the sake of clearness4 ,the entire annular reef1 which when surrounding an open expanse of water1 forms an 8atoll10 and when surrounding one or more high islands1 forms an encircling 8barrier-reef10 has a nearly uniform structure4 ,the reefs in some of the original surveys are represented merely by a single line with crosses1 so that their breadth is not given2 ,i have had such reefs engraved of the width usually attained by coral-reefs4 ,i have not thought it worth while to introduce all those small and very numerous reefs1 which occur within the lagoons of most atolls and within the lagoon-channels of most barrier-reefs1 and which stand either isolated1 or are attached to the shores of the reef or land4 ,at ,peros ,banhos none of the lagoon-reefs rise to the surface of the water2 a few of them have been introduced1 and are marked by plain dotted circles4 ,a few of the deepest soundings are laid down within each reef2 they are in fathoms1 of six ,english feet4 ,plate ,i--,map showing the resemblance in form between barrier coral-reefs surrounding mountainous islands1 and atolls or lagoon islands4 ,illustration3 ,map showing the resemblance in form4 ,fig4 #a--,,vanikoro1 situated in the western part of the ,south ,pacific2 taken from the survey by ,captain ,d',urville in the ,astrolabe2 the soundings on the southern side of the island1 namely1 from thirty to forty fathoms1 are given from the voyage of the ,chev4 ,dillon2 the other soundings are laid down from the survey by ,d',urville2 height of the summit of the island is #c1jcb feet4 ,the principal small detached reefs within the lagoon-channel have in this instance been represented4 ,the southern shore of the island is narrowly fringed by a reef3 if the engraver had carried this reef entirely round both islands1 this figure would have served 7by leaving out in imagination the barrier-reef7 as a good specimen of an abruptly-sided island1 surrounded by a reef of the fringing class4 ,fig4 #b--;,,hogoleu1 or ,,roug1 in the ,caroline ,archipelago2 taken from the 8,atlas of the ,voyage of the ,astrolabe10 compiled from the surveys of ,captains ,duperrey and ,d',urville2 the depth of the immense lagoon-like space within the reef is not known4 ,fig4 #c--,,raiatea1 in the ,society ,archipelago2 from the map given in the quarto edition of 8,cook's ,first ,voyage20 it is probably not accurate4 ,fig4 #d--;,,bow1 or ,,heyou ,,atoll 7or lagoon-island71 in the ,low ,archipelago1 from the survey by ,captain ,beechey1 ,r4,n42 the lagoon is choked up with reefs1 but the average greatest depth of about twenty fathoms1 is given from the published account of the voyage4 ,fig4 #e--;,,bolabola1 in the ,society ,archipelago1 from the survey of ,captain ,duperrey in the ,coquille3 the soundings in this and the following figures have been altered from ,french feet to ,english fathoms2 height of highest point of the island #d1jbf feet4 ,illustration3 ,map showing the resemblance in form4 ,fig4 #f4--,,maurua1 in the ,society ,archipelago2 from the survey by ,captain ,duperrey in the ,coquille3 height of land about eight hundred feet4 ,fig4 #g--,,pouynip,,te1 or ,,seniavine1 in the ,caroline ,archipelago2 from the survey by ,admiral ,lutk4 ,fig4 #h--;,,gambier ,,islands1 in the southern part of the ,low ,archipelago2 from the survey by ,captain ,beechey2 height of highest island1 #a1bdf feet2 the islands are surrounded by extensive and irregular reefs2 the reef on the southern side is submerged4 ,fig4 #i--,,peros ,,banhos ,,atoll 7or lagoon-island71 in the ,chagos group in the ,indian ,ocean2 from the survey by ,captain ,moresby and ,lieutenant ,powell2 not nearly all the small submerged reefs in the lagoon are represented2 the annular reef on the southern side is submerged4 ,fig4 #aj--,,keeling1 or ,,cocos ,,atoll 7or lagoon-island71 in the ,indian ,ocean2 from the survey by ,captain ,fitzroy2 the lagoon south of the dotted line is very shallow1 and is left almost bare at low water2 the part north of the line is choked up with irregular reefs4 ,the annular reef on the north-west side is broken1 and blends into a shoal sandbank1 on which the sea breaks4 ,chapter ,i ,,atolls ,,or ,,lagoon-,,islands ,section ,i--,,keeling ,,atoll ,corals on the outer margin4--,zone of ,nullipor4--,exterior reef4--,islets4--,coral-conglomerate4--, lagoon4--,calcareous sediment4--,scari and ,holuthuri subsisting on corals4--,changes in the condition of the reefs and islets4--,probable subsidence of the atoll4--,future state of the lagoon4 ,,keeling or ,,cocos atoll is situated in the ,indian ,ocean1 in #ab #e ,s41 and longitude #ij #ee ,e43 a reduced chart of it was made from the survey of ,captain ,fitzroy and the ,officers of ,h4,m4,s4 ,beagle1 is given in ,plate ,i1 ,fig4 #aj4 ,the greatest width of this atoll is nine miles and a half4 ,its structure is in most respects characteristic of the class to which it belongs1 with the exception of the shallowness of the lagoon4 ,the accompanying woodcut represents a vertical section1 supposed to be drawn at low water from the outer coast across one of the low islets 7one being taken of average dimensions7 to within the lagoon4 ,illustration3 ,vertical section of one of the low islets4 ,a4--,level of the sea at low water3 where the letter ,a is placed1 the depth is twenty-five fathoms1 and the distance rather more than one hundred and fifty yards from the edge of the reef4 ,b4--,outer edge of that flat part of the reef1 which dries at low water3 the edge either consists of a convex mound1 as represented1 or of rugged points1 like those a little farther seaward1 beneath the water4 ,c4--,a flat of coral-rock1 covered at high water4 ,d4--,a low projecting ledge of brecciated coral-rock washed by the waves at high water4 ,e4--,a slope of loose fragments1 reached by the sea only during gales3 the upper part1 which is from six to twelve feet high1 is clothed with vegetation4 ,the surface of the islet gently slopes to the lagoon4 ,f4--,level of the lagoon at low water4 ,the section is true to the scale in a horizontal line1 but it could not be made so in a vertical one1 as the average greatest height of the land is only between six and twelve feet above high-water mark4 ,i will describe the section1 commencing with the outer margin4 ,i must first observe that the reef-building polypifers1 not being tidal animals1 require to be constantly submerged or washed by the breakers4 ,i was assured by ,mr4 ,liesk1 a very intelligent resident on these islands1 as well as by some chiefs at ,tahiti 7,otaheite71 that an exposure to the rays of the sun for a very short time invariably causes their destruction4 ,hence it is possible only under the most favourable circumstances1 afforded by an unusually low tide and smooth water1 to reach the outer margin1 where the coral is alive4 ,i succeeded only twice in gaining this part1 and found it almost entirely composed of a living ,porites1 which forms great irregularly rounded masses 7like those of an ,astra1 but larger7 from four to eight feet broad1 and little less in thickness4 ,these mounds are separated from each other by narrow crooked channels1 about six feet deep1 most of which intersect the line of reef at right angles4 ,on the furthest mound1 which ,i was able to reach by the aid of a leaping-pole1 and over which the sea broke with some violence1 although the day was quite calm and the tide low1 the polypifers in the uppermost cells were all dead1 but between three and four inches lower down on its side they were living1 and formed a projecting border round the upper and dead surface4 ,the coral being thus checked in its upward growth1 extends laterally1 and hence most of the masses1 especially those a little further inwards1 had broad flat dead summits4 ,on the other hand ,i could see1 during the recoil of the breakers1 that a few yards further seaward1 the whole convex surface of the ,porites was alive2 so that the point where we were standing was almost on the exact upward and shoreward limit of existence of those corals which form the outer margin of the reef4 ,we shall presently see that there are other organic productions1 fitted to bear a somewhat longer exposure to the air and sun4 ,next1 but much inferior in importance to the ,porites1 is the ,millepora complanata4#a #a ,this ,millepora 7,palmipora of ,blainville71 as well as the ,m4 alcicornis1 possesses the singular property of stinging the skin where it is delicate1 as on the face and arm4 ,it grows in thick vertical plates1 intersecting each other at various angles1 and forms an exceedingly strong honeycombed mass1 which generally affects a circular form1 the marginal plates alone being alive4 ,between these plates and in the protected crevices on the reef1 a multitude of branching zoophytes and other productions flourish1 but the ,porites and ,millepora alone seem able to resist the fury of the breakers on its upper and outer edge3 at the depth of a few fathoms other kinds of stony corals live4 ,mr4 ,liesk1 who was intimately acquainted with every part of this reef1 and likewise with that of ,north ,keeling atoll1 assured me that these corals invariably compose the outer margin4 ,the lagoon is inhabited by quite a distinct set of corals1 generally brittle and thinly branched2 but a ,porites1 apparently of the same species with that on the outside1 is found there1 although it does not seem to thrive1 and certainly does not attain the thousandth part in bulk of the masses opposed to the breakers4 ,the woodcut shows the form of the bottom off the reef3 the water deepens for a space between one and two hundred yards wide1 very gradually to twenty-five fathoms 7,a in section71 beyond which the sides plunge into the unfathomable ocean at an angle of #de4b ,to the depth of ten or twelve fathoms the bottom is exceedingly rugged1 and seems formed of great masses of living coral1 similar to those on the margin4 ,the arming of the lead here invariably came up quite clean1 but deeply indented1 and chains and anchors which were lowered1 in the hopes of tearing up the coral1 were broken4 ,many small fragments1 however1 of ,millepora alcicornis were brought up2 and on the arming from an eight-fathom cast1 there was a perfect impression of an ,astra1 apparently alive4 ,i examined the rolled fragments cast on the beach during gales1 in order further to ascertain what corals grew outside the reef4 ,the fragments consisted of many kinds1 of which the ,porites already mentioned and a ,madrepora1 apparently the ,m4 corymbosa1 were the most abundant4 ,as ,i searched in vain in the hollows on the reef and in the lagoon1 for a living specimen of this ,madrepore1 ,i conclude that it is confined to a zone outside1 and beneath the surface1 where it must be very abundant4 ,fragments of the ,millepora alcicornis and of an ,astra were also numerous2 the former is found1 but not in proportionate numbers1 in the hollows on the reef2 but the ,astra ,i did not see living4 ,hence we may infer1 that these are the kinds of coral which form the rugged sloping surface 7represented in the woodcut by an uneven line71 round and beneath the external margin4 ,between twelve and twenty fathoms the arming came up an equal number of times smoothed with sand1 and indented with coral3 an anchor and lead were lost at the respective depths of thirteen and sixteen fathoms4 ,out of twenty-five soundings taken at a greater depth than twenty fathoms1 every one showed that the bottom was covered with sand2 whereas1 at a less depth than twelve fathoms1 every sounding showed that it was exceedingly rugged1 and free from all extraneous particles4 ,two soundings were obtained at the depth of #cfj fathoms1 and several between two hundred and three hundred fathoms4 ,the sand brought up from these depths consisted of finely triturated fragments of stony zoophytes1 but not1 as far as ,i could distinguish1 of a particle of any lamelliform genus3 fragments of shells were rare4 #b ,the soundings from which this section is laid down were taken with great care by ,captain ,fitzroy himself4 ,he used a bell-shaped lead1 having a diameter of four inches1 and the armings each time were cut off and brought on board for me to examine4 ,the arming is a preparation of tallow1 placed in the concavity at the bottom of the lead4 ,sand1 and even small fragments of rock1 will adhere to it2 and if the bottom be of rock it brings up an exact impression of its surface4 ,at a distance of #b1bjj yards from the breakers1 ,captain ,fitzroy found no bottom with a line of #g1bjj feet in length2 hence the submarine slope of this coral formation is steeper than that of any volcanic cone4 ,off the mouth of the lagoon1 and likewise off the northern point of the atoll1 where the currents act violently1 the inclination1 owing to the accumulation of sediment1 is less4 ,as the arming of the lead from all the greater depths showed a smooth sandy bottom1 ,i at first concluded that the whole consisted of a vast conical pile of calcareous sand1 but the sudden increase of depth at some points1 and the circumstance of the line having been cut1 as if rubbed1 when between five hundred and six hundred fathoms were out1 indicate the probable existence of submarine cliffs4 ,on the margin of the reef1 close within the line where the upper surface of the ,porites and of the ,millepora is dead1 three species of ,nullipora flourish4 ,one grows in thin sheets1 like a lichen on old trees2 the second in stony knobs1 as thick as a man's finger1 radiating from a common centre2 and the third1 which is less common1 in a moss-like reticulation of thin1 but perfectly rigid branches4#c ,the three species occur either separately or mingled together2 and they form by their successive growth a layer two or three feet in thickness1 which in some cases is hard1 but where formed of the lichen-like kind1 readily yields an impression to the hammer3 the surface is of a reddish colour4 ,these ,nullipor1 although able to exist above the limit of true corals1 seem to require to be bathed during the greater part of each tide by breaking water1 for they are not found in any abundance in the protected hollows on the back part of the reef1 where they might be immersed either during the whole or an equal proportional time of each tide4 ,it is remarkable that organic productions of such extreme simplicity1 for the ,nullipor undoubtedly belong to one of the lowest classes of the vegetable kingdom1 should be limited to a zone so peculiarly circumstanced4 ,hence the layer composed by their growth merely fringes the reef for a space of about twenty yards in width1 either under the form of separate mammillated projections1 where the outer masses of coral are separate1 or1 more commonly1 where the corals are united into a solid margin1 as a continuous smooth convex mound 7,b in woodcut71 like an artificial breakwater4 ,both the mound and mammillated projections stand about three feet higher than any other part of the reef1 by which term ,i do not include the islets1 formed by the accumulation of rolled fragments4 ,we shall hereafter see that other coral reefs are protected by a similar thick growth of ,nullipor on the outer margin1 the part most exposed to the breakers1 and this must effectually aid in preserving it from being worn down4 #c ,this last species is of a beautiful bright peach-blossom colour4 ,its branches are about as thick as crow-quills2 they are slightly flattened and knobbed at the extremities4 ,the extremities only are alive and brightly coloured4 ,the two other species are of a dirty purplish-white4 ,the second species is extremely hard2 its short knob-like branches are cylindrical1 and do not grow thicker at their extremities4 ,the woodcut represents a section across one of the islets on the reef1 but if all that part which is above the level of ,c were removed1 the section would be that of a simple reef1 as it occurs where no islet has been formed4 ,it is this reef which essentially forms the atoll4 ,it is a ring1 enclosing the lagoon on all sides except at the northern end1 where there are two open spaces1 through one of which ships can enter4 ,the reef varies in width from two hundred and fifty to five hundred yards1 its surface is level1 or very slightly inclined towards the lagoon1 and at high tide the sea breaks entirely over it3 the water at low tide thrown by the breakers on the reef1 is carried by the many narrow and shoal gullies or channels on its surface1 into the lagoon3 a return stream sets out of the lagoon through the main entrance4 ,the most frequent coral in the hollows on the reef is ,pocillopora verrucosa1 which grows in short sinuous plates1 or branches1 and when alive is of a beautiful pale lake-red3 a ,madrepora1 closely allied or identical with ,m4 pocillifera1 is also common4 ,as soon as an islet is formed1 and the waves are prevented breaking entirely over the reef1 the channels and hollows in it become filled up with cemented fragments1 and its surface is converted into a hard smooth floor 7,c of woodcut71 like an artificial one of freestone4 ,this flat surface varies in width from one hundred to two hundred1 or even three hundred yards1 and is strewed with a few large fragments of coral torn up during gales3 it is uncovered only at low water4 ,i could with difficulty1 and only by the aid of a chisel1 procure chips of rock from its surface1 and therefore could not ascertain how much of it is formed by the aggregation of detritus1 and how much by the outward growth of mounds of corals1 similar to those now living on the margin4 ,nothing can be more singular than the appearance at low tide of this 8flat0 of naked stone1 especially where it is externally bounded by the smooth convex mound of ,nullipor1 appearing like a breakwater built to resist the waves1 which are constantly throwing over it sheets of foaming water4 ,the characteristic appearance of this 8flat0 is shown in the foregoing woodcut of ,whitsunday atoll4 ,the islets on the reef are first formed between two hundred and three hundred yards from its outer edge1 through the accumulation of a pile of fragments1 thrown together by some unusually strong gale4 ,their ordinary width is under a quarter of a mile1 and their length varies from a few yards to several miles4 ,those on the south-east and windward side of the atoll1 increase solely by the addition of fragments on their outer side2 hence the loose blocks of coral1 of which their surface is composed1 as well as the shells mingled with them1 almost exclusively consist of those kinds which live on the outer coast4 ,the highest part of the islets 7excepting hillocks of blown sand1 some of which are thirty feet high71 is close to the outer beach 7,e of the woodcut71 and averages from six to ten feet above ordinary high-water mark4 ,from the outer beach the surface slopes gently to the shores of the lagoon1 which no doubt has been caused by the breakers the further they have rolled over the reef1 having had less power to throw up fragments4 ,the little waves of the lagoon heap up sand and fragments of thinly-branched corals on the inner side of the islets on the leeward side of the atoll2 and these islets are broader than those to windward1 some being even eight hundred yards in width2 but the land thus added is very low4 ,the fragments beneath the surface are cemented into a solid mass1 which is exposed as a ledge 7,d of the woodcut71 projecting some yards in front of the outer shore and from two to four feet high4 ,this ledge is just reached by the waves at ordinary high-water3 it extends in front of all the islets1 and everywhere has a water-worn and scooped appearance4 ,the fragments of coral which are occasionally cast on the 8flat0 are during gales of unusual violence swept together on the beach1 where the waves each day at high-water tend to remove and gradually wear them down2 but the lower fragments having become firmly cemented together by the percolation of calcareous matter1 resist the daily tides longer1 and hence project as a ledge4 ,the cemented mass is generally of a white colour1 but in some few parts reddish from ferruginous matter2 it is very hard1 and is sonorous under the hammer2 it is obscurely divided by seams1 dipping at a small angle seaward2 it consists of fragments of the corals which grow on the outer margin1 some quite and others partially rounded1 some small and others between two and three feet across2 and of masses of previously formed conglomerate1 torn up1 rounded1 and re-cemented2 or it consists of a calcareous sandstone1 entirely composed of rounded particles1 generally almost blended together1 of shells1 corals1 the spines of echini1 and other such organic bodies2 rocks1 of this latter kind1 occur on many shores1 where there are no coral reefs4 ,the structure of the coral in the conglomerate has generally been much obscured by the infiltration of spathose calcareous matter2 and ,i collected a very interesting series1 beginning with fragments of unaltered coral1 and ending with others1 where it was impossible to discover with the naked eye any trace of organic structure4 ,in some specimens ,i was unable1 even with the aid of a lens1 and by wetting them1 to distinguish the boundaries of the altered coral and spathose limestone4 ,many even of the blocks of coral lying loose on the beach1 had their central parts altered and infiltrated4 ,the lagoon alone remains to be described2 it is much shallower than that of most atolls of considerable size4 ,the southern part is almost filled up with banks of mud and fields of coral1 both dead and alive1 but there are considerable spaces1 between three and four fathoms1 and smaller basins1 from eight to ten fathoms deep4 ,probably about half its area consists of sediment1 and half of coral-reefs4 ,the corals composing these reefs have a very different aspect from those on the outside2 they are very numerous in kind1 and most of them are thinly branched4 ,meandrina1 however1 lives in the lagoon1 and great rounded masses of this coral are numerous1 lying quite or almost loose on the bottom4 ,the other commonest kinds consist of three closely allied species of true ,madrepora in thin branches2 of ,seriatapora subulata2 two species of ,porites#d with cylindrical branches1 one of which forms circular clumps1 with the exterior branches only alive2 and lastly1 a coral something like an ,explanaria1 but with stars on both surfaces1 growing in thin1 brittle1 stony1 foliaceous expansions1 especially in the deeper basins of the lagoon4 ,the reefs on which these corals grow are very irregular in form1 are full of cavities1 and have not a solid flat surface of dead rock1 like that surrounding the lagoon2 nor can they be nearly so hard1 for the inhabitants made with crowbars a channel of considerable length through these reefs1 in which a schooner1 built on the ,s4,e4 islet1 was floated out4 ,it is a very interesting circumstance1 pointed out to us by ,mr4 ,liesk1 that this channel1 although made less than ten years before our visit1 was then1 as we saw1 almost choked up with living coral1 so that fresh excavations would be absolutely necessary to allow another vessel to pass through it4 #d ,this ,porites has somewhat the habit of ,p4 clavaria1 but the branches are not knobbed at their ends4 ,when alive it is of a yellow colour1 but after having been washed in fresh water and placed to dry1 a jet-black slimy substance exuded from the entire surface1 so that the specimen now appears as if it had been dipped in ink4 ,the sediment from the deepest parts in the lagoon1 when wet1 appeared chalky1 but when dry1 like very fine sand4 ,large soft banks of similar1 but even finer grained mud1 occur on the ,s4,e4 shore of the lagoon1 affording a thick growth of a ,fucus1 on which turtle feed3 this mud1 although discoloured by vegetable matter1 appears from its entire solution in acids to be purely calcareous4 ,i have seen in the ,museum of the ,geological ,society1 a similar but more remarkable substance1 brought by ,lieutenant ,nelson from the reefs of ,bermuda1 which1 when shown to several experienced geologists1 was mistaken by them for true chalk4 ,on the outside of the reef much sediment must be formed by the action of the surf on the rolled fragments of coral2 but in the calm waters of the lagoon1 this can take place only in a small degree4 ,there are1 however1 other and unexpected agents at work here3 large shoals of two species of ,scarus1 one inhabiting the surf outside the reef and the other the lagoon1 subsist entirely1 as ,i was assured by ,mr4 ,liesk1 the intelligent resident before referred to1 by browsing on the living polypifers4 ,i opened several of these fish1 which are very numerous and of considerable size1 and ,i found their intestines distended by small pieces of coral1 and finely ground calcareous matter4 ,this must daily pass from them as the finest sediment2 much also must be produced by the infinitely numerous vermiform and molluscous animals1 which make cavities in almost every block of coral4 ,dr4 ,j4 ,allan1 of ,forres1 who has enjoyed the best means of observation1 informs me in a letter that the ,holothuri 7a family of ,radiata7 subsist on living coral2 and the singular structure of bone within the anterior extremity of their bodies1 certainly appears well adapted for this purpose4 ,the number of the species of ,holothuria1 and of the individuals which swarm on every part of these coral-reefs1 is extraordinarily great2 and many shiploads are annually freighted1 as is well-known1 for ,china with the trepang1 which is a species of this genus4 ,the amount of coral yearly consumed1 and ground down into the finest mud1 by these several creatures1 and probably by many other kinds1 must be immense4 ,these facts are1 however1 of more importance in another point of view1 as showing us that there are living checks to the growth of coral-reefs1 and that the almost universal law of 8consumed and be consumed10 holds good even with the polypifers forming those massive bulwarks1 which are able to withstand the force of the open ocean4 ,considering that ,keeling atoll1 like other coral formations1 has been entirely formed by the growth of organic beings1 and the accumulation of their detritus1 one is naturally led to inquire how long it has continued1 and how long it is likely to continue1 in its present state4 ,mr4 ,liesk informed me that he had seen an old chart in which the present long island on the ,s4,e4 side was divided by several channels into as many islets2 and he assures me that the channels can still be distinguished by the smaller size of the trees on them4 ,on several islets1 also1 ,i observed that only young cocoa-nut trees were growing on the extremities2 and that older and taller trees rose in regular succession behind them2 which shows that these islets have very lately increased in length4 ,in the upper and south-eastern part of the lagoon1 ,i was much surprised by finding an irregular field of at least a mile square of branching corals1 still upright1 but entirely dead4 ,they consisted of the species already mentioned2 they were of a brown colour1 and so rotten1 that in trying to stand on them ,i sank halfway up the leg1 as if through decayed brushwood4 ,the tops of the branches were barely covered by water at the time of lowest tide4 ,several facts having led me to disbelieve in any elevation of the whole atoll1 ,i was at first unable to imagine what cause could have killed so large a field of coral4 ,upon reflection1 however1 it appeared to me that the closing up of the above-mentioned channels would be a sufficient cause2 for before this1 a strong breeze by forcing water through them into the head of the lagoon1 would tend to raise its level4 ,but now this cannot happen1 and the inhabitants observe that the tide rises to a less height1 during a high ,s4,e4 wind1 at the head than at the mouth of the lagoon4 ,the corals1 which1 under the former condition of things1 had attained the utmost possible limit of upward growth1 would thus occasionally be exposed for a short time to the sun1 and be killed4 ,besides the increase of dry land1 indicated by the foregoing facts1 the exterior solid reef appears to have grown outwards4 ,on the western side of the atoll1 the 8flat0 lying between the margin of the reef and the beach1 is very wide2 and in front of the regular beach with its conglomerate basis1 there is1 in most parts1 a bed of sand and loose fragments with trees growing out of it1 which apparently is not reached even by the spray at high water4 ,it is evident some change has taken place since the waves formed the inner beach2 that they formerly beat against it with violence was evident1 from a remarkably thick and water-worn point of conglomerate at one spot1 now protected by vegetation and a bank of sand2 that they beat against it in the same peculiar manner in which the swell from windward now obliquely curls round the margin of the reef1 was evident from the conglomerate having been worn into a point projecting from the beach in a similarly oblique manner4 ,this retreat in the line of action of the breakers might result1 either from the surface of the reef in front of the islets having been submerged at one time1 and afterward having grown upwards1 or from the mounds of coral on the margin having continued to grow outwards4 ,that an outward growth of this part is in process1 can hardly be doubted from the fact already mentioned of the mounds of ,porites with their summits apparently lately killed1 and their sides only three or four inches lower down thickened by a fresh layer of living coral4 ,but there is a difficulty on this supposition which ,i must not pass over4 ,if the whole1 or a large part of the 8flat10 had been formed by the outward growth of the margin1 each successive margin would naturally have been coated by the ,nullipor1 and so much of the surface would have been of equal height with the existing zone of living ,nullipor3 this is not the case1 as may be seen in the woodcut4 ,it is1 however1 evident from the abraded state of the 8flat10 with its original inequalities filled up1 that its surface has been much modified2 and it is possible that the hinder portions of the zone of ,nullipor1 perishing as the reef grows outwards1 might be worn down by the surf4 ,if this has not taken place1 the reef can in no part have increased outwards in breadth since its formation1 or at least since the ,nullipor formed the convex mound on its margin2 for the zone thus formed1 and which stands between two and three feet above the other parts of the reef1 is nowhere much above twenty yards in width4 ,thus far we have considered facts1 which indicate1 with more or less probability1 the increase of the atoll in its different parts3 there are others having an opposite tendency4 ,on the south-east side1 ,lieutenant ,sulivan1 to whose kindness ,i am indebted for many interesting observations1 found the conglomerate projecting on the reef nearly fifty yards in front of the beach3 we may infer from what we see in all other parts of the atoll1 that the conglomerate was not originally so much exposed1 but formed the base of an islet1 the front and upper part of which has since been swept away4 ,the degree to which the conglomerate1 round nearly the whole atoll1 has been scooped1 broken up1 and the fragments cast on the beach1 is certainly very surprising1 even on the view that it is the office of occasional gales to pile up fragments1 and of the daily tides to wear them away4 ,on the western side1 also1 of the atoll1 where ,i have described a bed of sand and fragments with trees growing out of it1 in front of an old beach1 it struck both ,lieutenant ,sulivan and myself1 from the manner in which the trees were being washed down1 that the surf had lately recommenced an attack on this line of coast4 ,appearances indicating a slight encroachment of the water on the land1 are plainer within the lagoon3 ,i noticed in several places1 both on its windward and leeward shores1 old cocoa-nut trees falling with their roots undermined1 and the rotten stumps of others on the beach1 where the inhabitants assured us the cocoa-nut could not now grow4 ,captain ,fitzroy pointed out to me1 near the settlement1 the foundation posts of a shed1 now washed by every tide1 but which the inhabitants stated1 had seven years before stood above high watermark4 ,in the calm waters of the lagoon1 directly connected with a great1 and therefore stable ocean1 it seems very improbable that a change in the currents1 sufficiently great to cause the water to eat into the land on all sides1 should have taken place within a limited period4 ,from these considerations ,i inferred1 that probably the atoll had lately subsided to a small amount2 and this inference was strengthened by the circumstance1 that in #ahcd1 two years before our visit1 the island had been shaken by a severe earthquake1 and by two slighter ones during the ten previous years4 ,if1 during these subterranean disturbances1 the atoll did subside1 the downward movement must have been very small1 as we must conclude from the fields of dead coral still lipping the surface of the lagoon1 and from the breakers on the western shore not having yet regained the line of their former action4 ,the subsidence must1 also1 have been preceded by a long period of rest1 during which the islets extended to their present size1 and the living margin of the reef grew either upwards1 or as ,i believe outwards1 to its present distance from the beach4 ,whether this view be correct or not1 the above facts are worthy of attention1 as showing how severe a struggle is in progress on these low coral formations between the two nicely balanced powers of land and water4 ,with respect to the future state of ,keeling atoll1 if left undisturbed1 we can see that the islets may still extend in length2 but as they cannot resist the surf until broken by rolling over a wide space1 their increase in breadth must depend on the increasing breadth of the reef2 and this must be limited by the steepness of the submarine flanks1 which can be added to only by sediment derived from the wear and tear of the coral4 ,from the rapid growth of the coral in the channel cut for the schooner1 and from the several agents at work in producing fine sediment1 it might be thought that the lagoon would necessarily become quickly filled up4 ,some of this sediment1 however1 is transported into the open sea1 as appears from the soundings off the mouth of the lagoon1 instead of being deposited within it4 ,the deposition1 moreover1 of sediment1 checks the growth of coral-reefs1 so that these two agencies cannot act together with full effect in filling it up4 ,we know so little of the habits of the many different species of corals1 which form the lagoon-reefs1 that we have no more reasons for supposing that their whole surface would grow up as quickly as the coral did in the schooner-channel1 than for supposing that the whole surface of a peat-moss would increase as quickly as parts are known to do in holes1 where the peat has been cut away4 ,these agencies1 nevertheless1 tend to fill up the lagoon2 but in proportion as it becomes shallower1 so must the polypifers be subject to many injurious agencies1 such as impure water and loss of food4 ,for instance1 ,mr4 ,liesk informed me1 that some years before our visit unusually heavy rain killed nearly all the fish in the lagoon1 and probably the same cause would likewise injure the corals4 ,the reefs also1 it must be remembered1 cannot possibly rise above the level of the lowest spring-tide1 so that the final conversion of the lagoon into land must be due to the accumulation of sediment2 and in the midst of the clear water of the ocean1 and with no surrounding high land1 this process must be exceedingly slow4 ,section ,,ii--,,general ,,description ,,of ,,atolls4 ,general form and size of atolls1 their reefs and islets4--,external slope4--,zone of ,nullipor4--,conglomerate4--,depth of lagoons4--,sediment4--,reefs submerged wholly or in part4--,breaches in the reef4--,ledge-formed shores round certain lagoons4--,conversion of lagoons into land4 ,i will here give a sketch of the general form and structure of the many atolls and atoll-formed reefs which occur in the ,pacific and ,indian ,oceans1 comparing them with ,keeling atoll4 ,the ,maldiva atolls and the ,great ,chagos ,bank differ in so many respects1 that ,i shall devote to them1 besides occasional references1 a third section of this chapter4 ,keeling atoll may be considered as of moderate dimensions and of regular form4 ,of the thirty-two islands surveyed by ,captain ,beechey in the ,low ,archipelago1 the longest was found to be thirty miles1 and the shortest less than a mile2 but ,vliegen atoll1 situated in another part of the same group1 appears to be sixty miles long and twenty broad4 ,most of the atolls in this group are of an elongated form2 thus ,bow ,island is thirty miles in length1 and on an average only six in width 7,see ,fig4 #d1 ,plate ,i71 and ,clermont ,tonnere has nearly the same proportions4 ,in the ,marshall ,archipelago 7the ,ralick and ,radack group of ,kotzebue7 several of the atolls are more than thirty miles in length1 and ,rimsky ,korsacoff is fifty-four long1 and twenty wide1 at the broadest part of its irregular outline4 ,most of the atolls in the ,maldiva ,archipelago are of great size1 one of them 7which1 however1 bears a double name7 measured in a medial and slightly curved line1 is no less than eighty-eight geographical miles long1 its greatest width being under twenty1 and its least only nine and a half miles4 ,some atolls have spurs projecting from them2 and in the ,marshall group there are atolls united together by linear reefs1 for instance ,menchikoff ,island 7,see ,fig4 #c1 ,plate ,,ii71 which is sixty miles in length1 and consists of three loops tied together4 ,in far the greater number of cases an atoll consists of a simple elongated ring1 with its outline moderately regular4 ,the average width of the annular wreath may be taken as about a quarter of a mile4 ,captain ,beechey#e says that in the atolls of the ,low ,archipelago it exceeded in no instance half a mile4 ,the description given of the structure and proportional dimensions of the reef and islets of ,keeling atoll1 appears to apply perfectly to nearly all the atolls in the ,pacific and ,indian ,oceans4 ,the islets are first formed some way back either on the projecting points of the reef1 especially if its form be angular1 or on the sides of the main entrances into the lagoon--that is in both cases1 on points where the breakers can act during gales of wind in somewhat different directions1 so that the matter thrown up from one side may accumulate against that before thrown up from another4 ,in ,lutk's chart of the ,caroline atolls1 we see many instances of the former case2 and the occurrence of islets1 as if placed for beacons1 on the points where there is a gateway or breach through the reef1 has been noticed by several authors4 ,there are some atoll-formed reefs1 rising to the surface of the sea and partly dry at low water1 on which from some cause islets have never been formed2 and there are others on which they have been formed1 but have subsequently been worn away4 ,in atolls of small dimensions the islets frequently become united into a single horse-shoe or ring-formed strip2 but ,diego ,garcia1 although an atoll of considerable size1 being thirteen miles and a half in length1 has its lagoon entirely surrounded1 except at the northern end1 by a belt of land1 on an average a third of a mile in width4 ,to show how small the total area of the annular reef and the land is in islands of this class1 ,i may quote a remark from the voyage of ,lutk1 namely1 that if the forty-three rings1 or atolls1 in the ,caroline ,archipelago1 were put one within another1 and over a steeple in the centre of ,st4 ,petersburg1 the whole world would not cover that city and its suburbs4 #e ,beechey's 8,voyage to the ,pacific and ,beering's ,straits10 chapter viii4 ,the form of the bottom off ,keeling atoll1 which gradually slopes to about twenty fathoms at the distance of between one and two hundred yards from the edge of the reef1 and then plunges at an angle of #de into unfathomable depths1 is exactly the same#f with that of the sections of the atolls in the ,low ,archipelago given by ,captain ,beechey4 ,the nature1 however1 of the bottom seems to differ1 for this officer#g informs me that all the soundings1 even the deepest1 were on coral1 but he does not know whether dead or alive4 ,the slope round ,christmas atoll 7,lat4 #a #d ,n41 #aeg #de ,w471 described by ,cook1#h is considerably less1 at about half a mile from the edge of the reef1 the average depth was about fourteen fathoms on a fine sandy bottom1 and at a mile1 only between twenty and forty fathoms4 ,it has no doubt been owing to this gentle slope1 that the strip of land surrounding its lagoon1 has increased in one part to the extraordinary width of three miles2 it is formed of successive ridges of broken shells and corals1 like those on the beach4 ,i know of no other instance of such width in the reef of an atoll2 but ,mr4 ,f4 ,d4 ,bennett informs me that the inclination of the bottom round ,caroline atoll in the ,pacific1 is like that off ,christmas ,island1 very gentle4 ,off the ,maldiva and ,chagos atolls1 the inclination is much more abrupt2 thus at ,heawandoo ,pholo1 ,lieutenant ,powell#i found fifty and sixty fathoms close to the edge of the reef1 and at #cjj yards distance there was no bottom with a #cjj-yard line4 ,captain ,moresby informs me1 that at #ajj fathoms from the mouth of the lagoon of ,diego ,garcia1 he found no bottom with #aej fathoms2 this is the more remarkable1 as the slope is generally less abrupt in front of channels through a reef1 owing to the accumulation of sediment4 ,at ,egmont ,island1 also1 at #aej fathoms from the reef1 soundings were struck with #aej fathoms4 ,lastly1 at ,cardoo atoll1 only sixty yards from the reef1 no bottom was obtained1 as ,i am informed by ,captain ,moresby1 with a line of #bjj fathoms6 ,the currents run with great force round these atolls1 and where they are strongest1 the inclination appears to be most abrupt4 ,i am informed by the same authority1 that wherever soundings were obtained off these islands1 the bottom was invariably sandy3 nor was there any reason to suspect the existence of submarine cliffs1 as there was at ,keeling ,island4#aj ,here then occurs a difficulty2 can sand accumulate on a slope1 which1 in some cases1 appears to exceed fifty-five degrees8 ,it must be observed1 that ,i speak of slopes where soundings were obtained1 and not of such cases1 as that of ,cardoo1 where the nature of the bottom is unknown1 and where its inclination must be nearly vertical4 ,m4 ,elie de ,beaumont#aa has argued1 and there is no higher authority on this subject1 from the inclination at which snow slides down in avalanches1 that a bed of sand or mud cannot be formed at a greater angle than thirty degrees4 ,considering the number of soundings on sand1 obtained round the ,maldiva and ,chagos atolls1 which appears to indicate a greater angle1 and the extreme abruptness of the sand-banks in the ,west ,indies1 as will be mentioned in the ,appendix1 ,i must conclude that the adhesive property of wet sand counteracts its gravity1 in a much greater ratio than has been allowed for by ,m4 ,elie de ,beaumont4 ,from the facility with which calcareous sand becomes agglutinated1 it is not necessary to suppose that the bed of loose sand is thick4 #f ,the form of the bottom round the ,marshall atolls in the ,northern ,pacific is probably similar3 ,kotzebue 78,first ,voyage10 vol4 ii1 p4 #af7 says3 8,we had at a small distance from the reef1 forty fathoms depth1 which increased a little further so much that we could find no bottom40 #g ,i must be permitted to express my obligation to ,captain ,beechey1 for the very kind manner in which he has given me information on several points1 and to own the great assistance ,i have derived from his excellent published work4 #h ,cook's 8,third ,voyage10 vol4 ii1 chap4 #aj4 #i ,this fact is taken from a ,,ms4 account of these groups lent me by ,captain ,moresby4 ,see also ,captain ,moresby's paper on the ,maldiva atolls in the ,geographical ,journal1 vol4 v1 p4 #dja4 #aj ,off some of the islands in the ,low ,archipelago the bottom appears to descend by ledges4 ,off ,elizabeth ,island1 which1 however1 consists of raised coral1 ,captain ,beechey 7page #de1 #dto ed47 describes three ledges3 the first had an easy slope from the beach to a distance of about fifty yards3 the second extended two hundred yards with twenty-five fathoms on it1 and then ended abruptly1 like the first2 and immediately beyond this there was no bottom with two hundred fathoms4 #aa 8,memoires pour servir une description ,geolog4 de ,france10 tome iv1 p4 #baf4 ,captain ,beechey has observed1 that the submarine slope is much less at the extremities of the more elongated atolls in the ,low ,archipelago1 than at their sides2 in speaking of ,ducie's ,island he says#ab the buttress1 as it may be called1 which 8has the most powerful enemy 7the ,s4,w4 swell7 to oppose1 is carried out much further1 and with less abruptness than the other40 ,in some cases1 the less inclination of a certain part of the external slope1 for instance of the northern extremities of the two ,keeling atolls1 is caused by a prevailing current which there accumulates a bed of sand4 ,where the water is perfectly tranquil1 as within a lagoon1 the reefs generally grow up perpendicularly1 and sometimes even overhang their bases2 on the other hand1 on the leeward side of ,mauritius1 where the water is generally tranquil1 although not invariably so1 the reef is very gently inclined4 ,hence it appears that the exterior angle varies much2 nevertheless in the close similarity in form between the sections of ,keeling atoll and of the atolls in the ,low ,archipelago1 in the general steepness of the reefs of the ,maldiva and ,chagos atolls1 and in the perpendicularity of those rising out of water always tranquil1 we may discern the effects of uniform laws2 but from the complex action of the surf and currents1 on the growing powers of the coral and on the deposition of sediment1 we can by no means follow out all the results4 #ab ,beechey's 8,voyage10 #dto ed41 p4 #dd4 ,where islets have been formed on the reef1 that part which ,i have sometimes called the 8flat0 and which is partly dry at low water1 appears similar in every atoll4 ,in the ,marshall group in the ,north ,pacific1 it may be inferred from ,chamisso's description1 that the reef1 where islets have not been formed on it1 slopes gently from the external margin to the shores of the lagoon2 ,flinders states that the ,australian barrier has a similar inclination inwards1 and ,i have no doubt it is of general occurrence1 although1 according to ,ehrenberg1 the reefs of the ,red ,sea offer an exception4 ,chamisso observes that 8the red colour of the reef 7at the ,marshall atolls7 under the breakers is caused by a ,nullipora1 which covers the stone wherever the waves beat2 and1 under favourable circumstances1 assumes a stalactical form10--a description perfectly applicable to the margin of ,keeling atoll4#ac ,although ,chamisso does not state that the masses of ,nullipor form points or a mound1 higher than the flat1 yet ,i believe that this is the case2 for ,kotzebue1#ad in another part1 speaks of the rocks on the edge of the reef 8as visible for about two feet at low water10 and these rocks we may feel quite certain are not formed of true coral1#ae ,whether a smooth convex mound of ,nullipor1 like that which appears as if artificially constructed to protect the margin of ,keeling ,island1 is of frequent occurrence round atolls1 ,i know not2 but we shall presently meet with it1 under precisely the same form1 on the outer edge of the 8barrier-reefs0 which encircle the ,society ,islands4 #ac ,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #adb4 ,near ,porto ,praya1 in the ,cape de ,verde ,islands1 some basaltic rocks1 lashed by no inconsiderable surf1 were completely enveloped with a layer of ,nullipor4 ,the entire surface over many square inches1 was coloured of a peach-blossomed red2 the layer1 however1 was of no greater thickness than paper4 ,another kind1 in the form of projecting knobs1 grew in the same situation4 ,these ,nullipor are closely related to those described on the coral-reefs1 but ,i believe are of different species4 #ad ,kotzebue's 8,first ,voyage10 vol4 ii1 p4 #af4 ,lieutenant ,nelson1 in his excellent memoir in the ,geological ,transactions 7vol4 ii1 p4 #aje71 alludes to the rocky points mentioned by ,kotzebue1 and infers that they consist of ,serpul1 which compose incrusting masses on the reefs of ,bermudas1 as they likewise do on a sandstone bar off the coast of ,brazil 7which ,i have described in ,london ,phil4 ,journal1 ,october #ahda74 ,these masses of ,serpul hold the same position1 relatively to the action of the sea1 with the ,nullipor on the coral-reefs in the ,indian and ,pacific ,oceans4 #ae ,captain ,moresby1 in his valuable paper 8on the ,northern atolls of ,maldivas0 7,geographical ,journal1 vol4 v71 says that the edges of the reefs there stand above water at low spring-tides4 ,there appears to be scarcely a feature in the structure of ,keeling reef1 which is not of common1 if not of universal occurrence1 in other atolls4 ,thus ,chamisso describes#af a layer of coarse conglomerate1 outside the islets round the ,marshall atolls which 8appears on its upper surface uneven and eaten away40 ,from drawings1 with appended remarks1 of ,diego ,garcia in the ,chagos group and of several of the ,maldiva atolls1 shown me by ,captain ,moresby1#ag it is evident that their outer coasts are subject to the same round of decay and renovation as those of ,keeling atoll4 ,from the description of the atolls in the ,low ,archipelago1 given in ,captain ,beechey's 8,voyage10 it is not apparent that any conglomerate coral-rock was there observed4 #af ,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #add4 #ag ,see also ,moresby on the ,northern atolls of the ,maldivas1 ,geographical ,journal1 vol v1 p4 #djj4 ,the lagoon in ,keeling atoll is shallow2 in the atolls of the ,low ,archipelago the depth varies from #bj to #ch fathoms1 and in the ,marshall ,group1 according to ,chamisso1 from #cj to #ce2 in the ,caroline atolls it is only a little less4 ,within the ,maldiva atolls there are large spaces with #de fathoms1 and some soundings are laid down of #di fathoms4 ,the greater part of the bottom in most lagoons1 is formed of sediment2 large spaces have exactly the same depth1 or the depth varies so insensibly1 that it is evident that no other means1 excepting aqueous deposition1 could have leveled the surface so equally4 ,in the ,maldiva atolls this is very conspicuous1 and likewise in some of the ,caroline and ,marshall ,islands4 ,in the former large spaces consist of sand and soft clay2 and ,kotzebue speaks of clay having been found within one of the ,marshall atolls4 ,no doubt this clay is calcareous mud1 similar to that at ,keeling ,island1 and to that at ,bermuda already referred to1 as undistinguishable from disintegrated chalk1 and which ,lieutenant ,nelson says is called there pipe-clay4#ah #ah ,i may here observe that on the coast of ,brazil1 where there is much coral1 the soundings near the land are described by ,admiral ,roussin1 in the ,pilote du ,brsil1 as siliceous sand1 mingled with much finely comminuted particles of shells and coral4 ,further in the offing1 for a space of #a1cjj miles along the coast1 from the ,abrolhos ,islands to ,maranham1 the bottom in many places is composed of 8tuf blanc1 ml ou form de madrpores broys40 ,this white substance1 probably1 is analogous to that which occurs within the above-mentioned lagoons2 it is sometimes1 according to ,roussin1 firm1 and he compares it to mortar4 ,where the waves act with unequal force on the two sides of an atoll1 the islets appear to be first formed1 and are generally of greater continuity on the more exposed shore4 ,the islets1 also1 which are placed to leeward1 are in most parts of the ,pacific liable to be occasionally swept entirely away by gales1 equalling hurricanes in violence1 which blow in an opposite direction to the ordinary trade-wind4 ,the absence of the islets on the leeward side of atolls1 or when present their lesser dimensions compared with those to windward1 is a comparatively unimportant fact2 but in several instances the reef itself on the leeward side1 retaining its usual defined outline1 does not rise to the surface by several fathoms4 ,this is the case with the southern side of ,peros ,banhos 7,plate #a1 ,fig4 #i7 in the ,chagos group1 with ,mourileu atoll1#ai in the ,caroline ,archipelago1 and with the barrier-reef 7,plate ,i1 ,fig4 #h7 of the ,gambier ,islands4 ,i allude to the latter reef1 although belonging to another class1 because ,captain ,beechey was first led by it to observe the peculiarity in the question4 ,at ,peros ,banhos the submerged part is nine miles in length1 and lies at an average depth of about five fathoms2 its surface is nearly level1 and consists of hard stone1 with a thin covering of loose sand4 ,there is scarcely any living coral on it1 even on the outer margin1 as ,i have been particularly assured by ,captain ,moresby2 it is1 in fact1 a wall of dead coral-rock1 having the same width and transverse section with the reef in its ordinary state1 of which it is a continuous portion4 ,the living and perfect parts terminate abruptly1 and abut on the submerged portions1 in the same manner as on the sides of an ordinary passage through the reef4 ,the reef to leeward in other cases is nearly or quite obliterated1 and one side of the lagoon is left open2 for instance1 at ,oulleay 7,caroline ,archipelago71 where a crescent-formed reef is fronted by an irregular bank1 on which the other half of the annular reef probably once stood4 ,at ,namonouito1 in the same ,archipelago1 both these modifications of the reef concur2 it consists of a great flat bank1 with from twenty to twenty-five fathoms water on it2 for a length of more than forty miles on its southern side it is open and without any reef1 whilst on the other sides it is bounded by a reef1 in parts rising to the surface and perfectly characterised1 in parts lying some fathoms submerged4 ,in the ,chagos group there are annular reefs1 entirely submerged1 which have the same structure as the submerged and defined portions just described4 ,the ,speaker's ,bank offers an excellent example of this structure2 its central expanse1 which is about twenty-two fathoms deep1 is twenty-four miles across2 the external rim is of the usual width of annular reefs1 and is well-defined2 it lies between six and eight fathoms beneath the surface1 and at the same depth there are scattered knolls in the lagoon4 ,captain ,moresby believes the rim consists of dead rock1 thinly covered with sand1 and he is certain this is the case with the external rim of the ,great ,chagos ,bank1 which is also essentially a submerged atoll4 ,in both these cases1 as in the submerged portion of the reef at ,peros ,banhos1 ,captain ,moresby feels sure that the quantity of living coral1 even on the outer edge overhanging the deep-sea water1 is quite insignificant4 ,lastly1 in several parts of the ,pacific and ,indian ,oceans there are banks1 lying at greater depths than in the cases just mentioned1 of the same form and size with the neighbouring atolls1 but with their atoll-like structure wholly obliterated4 ,it appears from the survey of ,freycinet1 that there are banks of this kind in the ,caroline ,archipelago1 and1 as is reported1 in the ,low ,archipelago4 ,when we discuss the origin of the different classes of coral formations1 we shall see that the submerged state of the whole of some atoll-formed reefs1 and of portions of others1 generally but not invariably on the leeward side1 and the existence of more deeply submerged banks now possessing little or no signs of their original atoll-like structure1 are probably the effects of a uniform cause1--namely1 the death of the coral1 during the subsidence of the area1 in which the atolls or banks are situated4 #ai ,frederick ,lutk's 8,voyage autour du ,monde10 vol4 ii1 p4 #bia4 ,see also his account of ,namonouito1 at pp4 #ig and #aje1 and the chart of ,oulleay in the ,atlas4 ,there is seldom1 with the exception of the ,maldiva atolls1 more than two or three channels1 and generally only one leading into the lagoon1 of sufficient depth for a ship to enter4 in small atolls1 there is usually not even one4 ,where there is deep water1 for instance above twenty fathoms1 in the middle of the lagoon1 the channels through the reef are seldom as deep as the centre1--it may be said that the rim only of the saucer-shaped hollow forming the lagoon is notched4 ,mr4 ,lyell#bj has observed that the growth of the coral would tend to obstruct all the channels through a reef1 except those kept open by discharging the water1 which during high tide and the greater part of each ebb is thrown over its circumference4 ,several facts indicate that a considerable quantity of sediment is likewise discharged through these channels2 and ,captain ,moresby informs me that he has observed1 during the change of the monsoon1 the sea discoloured to a distance off the entrances into the ,maldiva and ,chagos atolls4 ,this1 probably1 would check the growth of the coral in them1 far more effectually than a mere current of water4 ,in the many small atolls without any channel1 these causes have not prevented the entire ring attaining the surface4 ,the channels1 like the submerged and effaced parts of the reef1 very generally though not invariably occur on the leeward side of the atoll1 or on that side1 according to ,beechey1#ba which1 from running in the same direction with the prevalent wind1 is not fully exposed to it4 ,passages between the islets on the reef1 through which boats can pass at high water1 must not be confounded with ship-channels1 by which the annular reef itself is breached4 ,the passages between the islets occur1 of course1 on the windward as well as on the leeward side2 but they are more frequent and broader to leeward1 owing to the lesser dimensions of the islets on that side4 #bj 8,principles of ,geology10 vol4 iii1 p4 #bhi4 #ba ,beechey's 8,voyage10 #dto ed41 vol4 i1 p4 #ahi4 ,at ,keeling atoll the shores of the lagoon shelve gradually1 where the bottom is of sediment1 and irregularly or abruptly where there are coral-reefs2 but this is by no means the universal structure in other atolls4 ,chamisso1#bb speaking in general terms of the lagoons in the ,marshall atolls1 says the lead generally sinks 8from a depth of two or three fathoms to twenty or twenty-four1 and you may pursue a line in which on one side of the boat you may see the bottom1 and on the other the azure-blue deep water40 ,the shores of the lagoon-like channel within the barrier-reef at ,vanikoro have a similar structure4 ,captain ,beechey has described a modification of this structure 7and he believes it is not uncommon7 in two atolls in the ,low ,archipelago1 in which the shores of the lagoon descend by a few1 broad1 slightly inclined ledges or steps3 thus at ,matilda atoll1#bc the great exterior reef1 the surface of which is gently inclined towards and beneath the surface of the lagoon1 ends abruptly in a little cliff three fathoms deep2 at its foot1 a ledge forty yards wide extends1 shelving gently inwards like the surface-reef1 and terminated by a second little cliff five fathoms deep2 beyond this1 the bottom of the lagoon slopes to twenty fathoms1 which is the average depth of its centre4 ,these ledges seem to be formed of coral-rock2 and ,captain ,beechey says that the lead often descended several fathoms through holes in them4 ,in some atolls1 all the coral reefs or knolls in the lagoon come to the surface at low water2 in other cases of rarer occurrence1 all lie at nearly the same depth beneath it1 but most frequently they are quite irregular1--some with perpendicular1 some with sloping sides1--some rising to the surface1 and others lying at all intermediate depths from the bottom upwards4 ,i cannot1 therefore1 suppose that the union of such reefs could produce even one uniformly sloping ledge1 and much less two or three1 one beneath the other1 and each terminated by an abrupt wall4 ,at ,matilda ,island1 which offers the best example of the step-like structure1 ,captain ,beechey observes that the coral-knolls within the lagoon are quite irregular in their height4 ,we shall hereafter see that the theory which accounts for the ordinary form of atolls1 apparently includes this occasional peculiarity in their structure4 #bb ,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #adb4 #bc ,beechey's 8,voyage10 #dto ed41 vol4 i1 p4 #afj4 ,at ,whitsunday ,island the bottom of the lagoon slopes gradually towards the centre1 and then deepens suddenly1 the edge of the bank being nearly perpendicular4 ,this bank is formed of coral and dead shells4 ,in the midst of a group of atolls1 there sometimes occur small1 flat1 very low islands of coral formation1 which probably once included a lagoon1 since filled up with sediment and coral-reefs4 ,captain ,beechey entertains no doubt that this has been the case with the two small islands1 which alone of thirty-one surveyed by him in the ,low ,archipelago1 did not contain lagoons4 ,romanzoff ,island 7in lat4 #ae deg ,s47 is described by ,chamisso#bd as formed by a dam of madreporitic rock inclosing a flat space1 thinly covered with trees1 into which the sea on the leeward side occasionally breaks4 ,north ,keeling atoll appears to be in a rather less forward stage of conversion into land2 it consists of a horse-shoe shaped strip of land surrounding a muddy flat1 one mile in its longest axis1 which is covered by the sea only at high water4 ,when describing ,south ,keeling atoll1 ,i endeavoured to show how slow the final process of filling up a lagoon must be2 nevertheless1 as all causes do tend to produce this effect1 it is very remarkable that not one instance1 as ,i believe1 is known of a moderately sized lagoon being filled up even to the low water-line at spring-tides1 much less of such a one being converted into land4 ,it is1 likewise1 in some degree remarkable1 how few atolls1 except small ones1 are surrounded by a single linear strip of land1 formed by the union of separate islets4 ,we cannot suppose that the many atolls in the ,pacific and ,indian ,oceans all have had a late origin1 and yet should they remain at their present level1 subjected only to the action of the sea and to the growing powers of the coral1 during as many centuries as must have elapsed since any of the earlier tertiary epochs1 it cannot1 ,i think1 be doubted that their lagoons and the islets on their reef1 would present a totally different appearance from what they now do4 ,this consideration leads to the suspicion that some renovating agency 7namely subsidence7 comes into play at intervals1 and perpetuates their original structure4 #bd ,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #bba4 ,section ,,iii--,,atolls ,,of ,,the ,,maldiva ,,archipelago--,,great ,,chagos ,,bank ,maldiva ,archipelago4--,ring-formed reefs1 marginal and central4--,great depths in the lagoons of the southern atolls4--,reefs in the lagoons all rising to the surface4--,position of islets and breaches in the reefs1 with respect to the prevalent winds and action of the waves4--,destruction of islets4--,connection in the position and submarine foundation of distinct atolls4--,the apparent disseverment of large atolls4--,the ,great ,chagos ,bank4--,its submerged condition and extraordinary structure4 ,although occasional references have been made to the ,maldiva atolls1 and to the banks in the ,chagos group1 some points of their structure deserve further consideration4 ,my description is derived from an examination of the admirable charts lately published from the survey of ,captain ,moresby and ,lieutenant ,powell1 and more especially from information which ,captain ,moresby has communicated to me in the kindest manner4 ,the ,maldiva ,archipelago is #dgj miles in length1 with an average breadth of about #ej miles4 ,the form and dimensions of the atolls1 and their singular position in a double line1 may be seen1 but not well1 in the greatly reduced chart 7,fig4 #f7 in ,plate ,,ii4 ,the dimensions of the longest atoll in the group 7called by the double name of ,milla-dou-,madou and ,tilla-dou-,matte7 have already been given2 it is #hh miles in a medial and slightly curved line1 and is less than #bj miles in its broadest part4 ,suadiva1 also1 is a noble atoll1 being #dd miles across in one direction1 and #cd in another1 and the great included expanse of water has a depth of between #bej and #cjj feet4 ,the smaller atolls in this group differ in no respect from ordinary ones2 but the larger ones are remarkable from being breached by numerous deep-water channels leading into the lagoon2 for instance1 there are #db channels1 through which a ship could enter the lagoon of ,suadiva4 ,in the three southern large atolls1 the separate portions of reef between these channels have the ordinary structure1 and are linear2 but in the other atolls1 especially the more northern ones1 these portions are ring-formed1 like miniature atolls4 ,other ring-formed reefs rise out of the lagoons1 in the place of those irregular ones which ordinarily occur there4 ,in the reduction of the chart of ,mahlos ,mahdoo 7,plate ,,ii1 ,fig4 #d71 it was not found easy to define the islets and the little lagoons within each reef1 so that the ring-formed structure is very imperfectly shown2 in the large published charts of ,tilla-dou-,matte1 the appearance of these rings1 from standing further apart from each other1 is very remarkable4 ,the rings on the margin are generally elongated2 many of them are three1 and some even five miles1 in diameter2 those within the lagoon are usually smaller1 few being more than two miles across1 and the greater number rather less than one4 ,the depth of the little lagoon within these small annular reefs is generally from five to seven fathoms1 but occasionally more2 and in ,ari atoll many of the central ones are twelve1 and some even more than twelve fathoms deep4 ,these rings rise abruptly from the platform or bank1 on which they are placed2 their outer margin is invariably bordered by living coral#be within which there is a flat surface of coral rock2 of this flat1 sand and fragments have in many cases accumulated and been converted into islets1 clothed with vegetation4 ,i can1 in fact1 point out no essential difference between these little ring-formed reefs 7which1 however1 are larger1 and contain deeper lagoons than many true atolls that stand in the open sea71 and the most perfectly characterised atolls1 excepting that the ring-formed reefs are based on a shallow foundation1 instead of on the floor of the open sea1 and that instead of being scattered irregularly1 they are grouped closely together on one large platform1 with the marginal rings arranged in a rudely formed circle4 #be ,captain ,moresby informs me that ,millepora complanata is one of the commonest kinds on the outer margin1 as it is at ,keeling atoll4 ,the perfect series which can be traced from portions of simple linear reef1 to others including long linear lagoons1 and from these again to oval or almost circular rings1 renders it probable that the latter are merely modifications of the linear or normal state4 ,it is conformable with this view1 that the ring-formed reefs on the margin1 even where most perfect and standing furthest apart1 generally have their longest axes directed in the line which the reef would have held1 if the atoll had been bounded by an ordinary wall4 ,we may also infer that the central ring-formed reefs are modifications of those irregular ones1 which are found in the lagoons of all common atolls4 ,it appears from the charts on a large scale1 that the ring-like structure is contingent on the marginal channels or breaches being wide2 and1 consequently1 on the whole interior of the atoll being freely exposed to the waters of the open sea4 ,when the channels are narrow or few in number1 although the lagoon be of great size and depth 7as in ,suadiva71 there are no ring-formed reefs2 where the channels are somewhat broader1 the marginal portions of reef1 and especially those close to the larger channels1 are ring-formed1 but the central ones are not so2 where they are broadest1 almost every reef throughout the atoll is more or less perfectly ring-formed4 ,although their presence is thus contingent on the openness of the marginal channels1 the theory of their formation1 as we shall hereafter see1 is included in that of the parent atolls1 of which they form the separate portions4 ,the lagoons of all the atolls in the southern part of the ,archipelago are from ten to twenty fathoms deeper than those in the northern part4 ,this is well exemplified in the case of ,addoo1 the southernmost atoll in the group1 for although only nine miles in its longest diameter1 it has a depth of thirty-nine fathoms1 whereas all the other small atolls have comparatively shallow lagoons2 ,i can assign no adequate cause for this difference in depth4 ,in the central and deepest part of the lagoons1 the bottom consists1 as ,i am informed by ,captain ,moresby1 of stiff clay 7probably a calcareous mud72 nearer the border it consists of sand1 and in the channels through the reef1 of hard sand-banks1 sandstone1 conglomerate rubble1 and a little live coral4 ,close outside the reef and the line joining its detached portions 7where intersected by many channels71 the bottom is sandy1 and it slopes abruptly into unfathomable depths4 ,in most lagoons the depth is considerably greater in the centre than in the channels2 but in ,tilla-dou-,matte1 where the marginal ring-formed reefs stand far apart1 the same depth is carried across the entire atoll1 from the deep-water line on one side to that on the other4 ,i cannot refrain from once again remarking on the singularity of these atolls1--a great sandy and generally concave disc rises abruptly from the unfathomable ocean1 with its central expanse studded and its border symmetrically fringed with oval basins of coral-rock1 just lipping the surface of the sea1 sometimes clothed with vegetation1 and each containing a little lake of clear water6 ,in the southern ,maldiva atolls1 of which there are nine large ones1 all the small reefs within the lagoons come to the surface1 and are dry at low water spring-tides2 hence in navigating them1 there is no danger from submarine banks4 ,this circumstance is very remarkable1 as within some atolls1 for instance those of the neighbouring ,chagos group1 not a single reef comes to the surface1 and in most other cases a few only do1 and the rest lie at all intermediate depths from the bottom upwards4 ,when treating of the growth of coral ,i shall again refer to this subject4 ,although in the neighbourhood of the ,maldiva ,archipelago the winds1 during the monsoons1 blow during nearly an equal time from opposite quarters1 and although1 as ,i am informed by ,captain ,moresby1 the westerly winds are the strongest1 yet the islets are almost all placed on the eastern side of the northern atolls1 and on the south-eastern side of the southern atolls4 ,that the formation of the islets is due to detritus thrown up from the outside1 as in the ordinary manner1 and not from the interior of the lagoons1 may1 ,i think be safely inferred from several considerations1 which it is hardly worth while to detail4 ,as the easterly winds are not the strongest1 their action probably is aided by some prevailing swell or current4 ,in groups of atolls1 exposed to a trade-wind1 the ship-channels into the lagoons are almost invariably situated on the leeward or less exposed side of the reef1 and the reef itself is sometimes either wanting there1 or is submerged4 ,a strictly analogous1 but different fact1 may be observed at the ,maldiva atolls--namely1 that where two atolls stand in front of each other1 the breaches in the reef are the most numerous on their near1 and therefore less exposed1 sides4 ,thus on the near sides of ,ari and the two ,nillandoo atolls1 which face ,s4 ,mle1 ,phaleedoo1 and ,moloque atolls1 there are seventy-three deep-water channels1 and only twenty-five on their outer sides2 on the near side of the three latter named atolls there are fifty-six openings1 and only thirty-seven on their outsides4 ,it is scarcely possible to attribute this difference to any other cause than the somewhat different action of the sea on the two sides1 which would ensue from the protection afforded by the two rows of atolls to each other4 ,i may here remark that in most cases1 the conditions favourable to the greater accumulation of fragments on the reef and to its more perfect continuity on one side of the atoll than on the other1 have concurred1 but this has not been the case with the ,maldivas2 for we have seen that the islets are placed on the eastern or south-eastern sides1 whilst the breaches in the reef occur indifferently on any side1 where protected by an opposite atoll4 ,the reef being more continuous on the outer and more exposed sides of those atolls which stand near each other1 accords with the fact1 that the reef of the southern atolls is more continuous than that of the northern ones2 for the former1 as ,i am informed by ,captain ,moresby1 are more constantly exposed than the northern atolls to a heavy surf4 ,the date of the first formation of some of the islets in this ,archipelago is known to the inhabitants2 on the other hand1 several islets1 and even some of those which are believed to be very old1 are now fast wearing away4 ,the work of destruction has1 in some instances1 been completed in ten years4 ,captain ,moresby found on one water-washed reef the marks of wells and graves1 which were excavated when it supported an islet4 ,in ,south ,nillandoo atoll1 the natives say that three of the islets were formerly larger3 in ,north ,nillandoo there is one now being washed away2 and in this latter atoll ,lieutenant ,prentice found a reef1 about six hundred yards in diameter1 which the natives positively affirmed was lately an island covered with cocoa-nut trees4 ,it is now only partially dry at low water spring-tides1 and is 7in ,lieutenant ,prentice's words7 8entirely covered with live coral and madrepore40 ,in the northern part1 also1 of the ,maldiva ,archipelago and in the ,chagos group1 it is known that some of the islets are disappearing4 ,the natives attribute these effects to variations in the currents of the sea4 ,for my own part ,i cannot avoid suspecting that there must be some further cause1 which gives rise to such a cycle of change in the action of the currents of the great and open ocean4 ,several of the atolls in this ,archipelago are so related to each other in form and position1 that at the first glance one is led to suspect that they have originated in the disseverment of a single one4 ,mle consists of three perfectly characterised atolls1 of which the shape and relative position are such1 that a line drawn closely round all three1 gives a symmetrical figure2 to see this clearly1 a larger chart is required than that of the ,archipelago in ,plate ,,ii2 the channel separating the two northern ,male atolls is only little more than a mile wide1 and no bottom was found in it with #ajj fathoms4 ,powell's ,island is situated at the distance of two miles and a half off the northern end of ,mahlos ,mahdoo 7see ,fig4 #d1 ,plate ,,ii71 at the exact point where the two sides of the latter1 if prolonged1 would meet2 no bottom1 however1 was found in the channel with #bjj fathoms2 in the wider channel between ,horsburgh atoll and the southern end of ,mahlos ,mahdoo1 no bottom was found with #bej fathoms4 ,in these and similar cases1 the relation consists only in the form and position of the atolls4 ,but in the channel between the two ,nillandoo atolls1 although three miles and a quarter wide1 soundings were struck at the depth of #bjj fathoms2 the channel between ,ross and ,ari atolls is four miles wide1 and only #aej fathoms deep4 ,here then we have1 besides the relation of form1 a submarine connection4 ,the fact of soundings having been obtained between two separate and perfectly characterised atolls is in itself interesting1 as it has never1 ,i believe1 been effected in any of the many other groups of atolls in the ,pacific and ,indian seas4 ,in continuing to trace the connection of adjoining atolls1 if a hasty glance be taken at the chart 7,fig4 #d1 ,plate ,,ii7 of ,mahlos ,mahdoo1 and the line of unfathomable water be followed1 no one will hesitate to consider it as one atoll4 ,but a second look will show that it is divided by a bifurcating channel1 of which the northern arm is about one mile and three-quarters in width1 with an average depth of #abe fathoms1 and the southern one three-quarters of a mile wide1 and rather less deep4 ,these channels resemble in the slope of their sides and general form1 those which separate atolls in every respect distinct2 and the northern arm is wider than that dividing two of the ,mle atolls4 ,the ring-formed reefs on the sides of this bifurcating channel are elongated1 so that the northern and southern portions of ,mahlos ,mahdoo may claim1 as far as their external outline is concerned1 to be considered as distinct and perfect atolls4 ,but the intermediate portion1 lying in the fork of the channel1 is bordered by reefs less perfect than those which surround any other atoll in the group of equally small dimensions4 ,mahlos ,mahdoo1 therefore1 is in every respect in so intermediate a condition1 that it may be considered either as a single atoll nearly dissevered into three portions1 or as three atolls almost perfect and intimately connected4 ,this is an instance of a very early stage of the apparent disseverment of an atoll1 but a still earlier one in many respects is exhibited at ,tilla-dou-,matte4 ,in one part of this atoll1 the ring-formed reefs stand so far apart from each other1 that the inhabitants have given different names to the northern and southern halves2 nearly all the rings1 moreover1 are so perfect and stand so separate1 and the space from which they rise is so level and unlike a true lagoon1 that we can easily imagine the conversion of this one great atoll1 not into two or three portions1 but into a whole group of miniature atolls4 ,a perfect series such as we have here traced1 impresses the mind with an idea of actual change2 and it will hereafter be seen1 that the theory of subsidence1 with the upward growth of the coral1 modified by accidents of probable occurrence1 will account for the occasional disseverment of large atolls4 ,plate ,,ii--,great ,chagos ,bank1 ,new ,caledonia1 ,menchikoff ,atoll1 etc4 ,illustration3 ,great ,chagos ,bank ,fig4 #a4--;,,great ,,chagos ,,bank1 in the ,indian ,ocean2 taken from the survey by ,captain ,moresby and ,lieutenant ,powell2 the parts which are shaded1 with the exception of two or three islets on the western and northern sides1 do not rise to the surface1 but are submerged from four to ten fathoms2 the banks bounded by the dotted lines lie from fifteen to twenty fathoms beneath the surface1 and are formed of sand2 the central space is of mud1 and from thirty to fifty fathoms deep4 ,fig4 #b4--;,a vertical section1 on the same scale1 in an eastern and western line across the ,great ,chagos ,bank1 given for the sake of exhibiting more clearly its structure4 ,illustration3 ,new ,caledonia1 ,menchikoff ,atoll1 etc4 ,fig4 #c4--,menchikoff ,atoll 7or lagoon-island71 in the ,marshall ,archipelago1 ,northern ,pacific ,ocean2 from ,krusenstern's 8,atlas of the ,pacific20 originally surveyed by ,captain ,hagemeister2 the depth within the lagoons is unknown4 ,fig4 #d4--,,mahlos ,,mahdoo ,,atoll1 together with ,horsburgh atoll1 in the ,maldiva ,archipelago2 from the survey by ,captain ,moresby and ,lieutenant ,powell2 the white spaces in the middle of the separate small reefs1 both on the margin and in the middle part1 are meant to represent little lagoons2 but it was found not possible to distinguish them clearly from the small islets1 which have been formed on these same small reefs2 many of the smaller reefs could not be introduced2 the nautical mark 7dot over a dash7 over the figures #bej and #bjj1 between ,mahlos ,mahdoo and ,horsburgh atoll and ,powell's island1 signifies that soundings were not obtained at these depths4 ,fig4 #e4--,,new ,,caledonia1 in the western part of the ,pacific2 from ,krusenstern's 8,atlas10 compiled from several surveys2 ,i have slightly altered the northern point of the reef1 in accordance with the 8,atlas of the ,voyage of the ,astrolabe40 ,in ,krusenstern's 8,atlas10 the reef is represented by a single line with crosses2 ,i have for the sake of uniformity added an interior line4 ,fig4 #f4--,,maldiva ,,archipelago1 in the ,indian ,ocean2 from the survey by ,captain ,moresby and ,lieutenant ,powell4 ,the ,great ,chagos bank alone remains to be described4 ,in the ,chagos group there are some ordinary atolls1 some annular reefs rising to the surface but without any islets on them1 and some atoll-formed banks1 either quite submerged1 or nearly so4 ,of the latter1 the ,great ,chagos ,bank is much the largest1 and differs in its structure from the others3 a plan of it is given in ,plate ,,ii1 ,fig4 #a1 in which1 for the sake of clearness1 ,i have had the parts under ten fathoms deep finely shaded3 an east and west vertical section is given in ,fig4 #b1 in which the vertical scale has been necessarily exaggerated4 ,its longest axis is ninety nautical miles1 and another line drawn at right angles to the first1 across the broadest part1 is seventy4 ,the central part consists of a level muddy flat1 between forty and fifty fathoms deep1 which is surrounded on all sides1 with the exception of some breaches1 by the steep edges of a set of banks1 rudely arranged in a circle4 ,these banks consist of sand1 with a very little live coral2 they vary in breadth from five to twelve miles1 and on an average lie about sixteen fathoms beneath the surface2 they are bordered by the steep edges of a third narrow and upper bank1 which forms the rim to the whole4 ,this rim is about a mile in width1 and with the exception of two or three spots where islets have been formed1 is submerged between five and ten fathoms4 ,it consists of smooth hard rock1 covered with a thin layer of sand1 but with scarcely any live coral2 it is steep on both sides1 and outwards slopes abruptly into unfathomable depths4 ,at the distance of less than half a mile from one part1 no bottom was found with #aij fathoms2 and off another point1 at a somewhat greater distance1 there was none with #baj fathoms4 ,small steep-sided banks or knolls1 covered with luxuriantly growing coral1 rise from the interior expanse to the same level with the external rim1 which1 as we have seen1 is formed only of dead rock4 ,it is impossible to look at the plan 7,fig4 #a1 ,plate ,,ii71 although reduced to so small a scale1 without at once perceiving that the ,great ,chagos ,bank is1 in the words of ,captain ,moresby1#bf 8nothing more than a half-drowned atoll40 ,but of what great dimensions1 and of how extraordinary an internal structure8 ,we shall hereafter have to consider both the cause of its submerged condition1 a state common to other banks in the group1 and the origin of the singular submarine terraces1 which bound the central expanse3 these1 ,i think1 it can be shown1 have resulted from a cause analogous to that which has produced the bifurcating channel across ,mahlos ,mahdoo4 #bf ,this officer has had the kindness to lend me an excellent ,,ms4 account of the ,chagos ,islands2 from this paper1 from the published charts1 and from verbal information communicated to me by ,captain ,moresby1 the above account of the ,great ,chagos ,bank is taken4 ,chapter ,,ii ,,barrier ,,reefs ,closely resemble in general form and structure atoll-reefs4--,width and depth of the lagoon-channels4--,breaches through the reef in front of valleys1 and generally on the leeward side4--,checks to the filling up of the lagoon-channels4--,size and constitution of the encircled islands4--,number of islands within the same reef4--,barrier-reefs of ,new ,caledonia and ,australia4--,position of the reef relative to the slope of the adjoining land4--,probable great thickness of barrier-reefs4 ,the term 8barrier0 has been generally applied to that vast reef which fronts the ,n4,e4 shore of ,australia1 and by most voyagers likewise to that on the western coast of ,new ,caledonia4 ,at one time ,i thought it convenient thus to restrict the term1 but as these reefs are similar in structure1 and in position relatively to the land1 to those1 which1 like a wall with a deep moat within1 encircle many smaller islands1 ,i have classed them together4 ,the reef1 also1 on the west coast of ,new ,caledonia1 circling round the extremities of the island1 is an intermediate form between a small encircling reef and the ,australian barrier1 which stretches for a thousand miles in nearly a straight line4 ,the geographer ,balbi has in effect described those barrier-reefs1 which encircle moderately sized islands1 by calling them atolls with high land rising from within their central expanse4 ,the general resemblance between the reefs of the barrier and atoll classes may be seen in the small1 but accurately reduced charts on ,plate ,i1#a and this resemblance can be further shown to extend to every part of the structure4 ,beginning with the outside of the reef2 many scattered soundings off ,gambier1 ,oualan1 and some other encircled islands1 show that close to the breakers there exists a narrow shelving margin1 beyond which the ocean becomes suddenly unfathomable2 but off the west coast of ,new ,caledonia1 ,captain ,kent#b found no bottom with #aej fathoms1 at two ships' length from the reef2 so that the slope here must be nearly as precipitous as off the ,maldiva atolls4 #a ,the authorities from which these charts have been reduced1 together with some remarks on them are given in a separately appended page1 descriptive of the ,plates4 #b ,dalrymple1 8,hydrog4 ,mem40 vol4 iii4 ,i can give little information regarding the kinds of corals which live on the outer margin4 ,when ,i visited the reef at ,tahiti1 although it was low water1 the surf was too violent for me to see the living masses2 but1 according to what ,i heard from some intelligent native chiefs1 they resemble in their rounded and branchless forms1 those on the margin of ,keeling atoll4 ,the extreme verge of the reef1 which was visible between the breaking waves at low water1 consisted of a rounded1 convex1 artificial-like breakwater1 entirely coated with ,nullipor1 and absolutely similar to that which ,i have described at ,keeling atoll4 ,from what ,i heard when at ,tahiti1 and from the writings of the ,revs4 ,w4 ,ellis and ,j4 ,williams1 ,i conclude that this peculiar structure is common to most of the encircled islands of the ,society ,archipelago4 ,the reef within this mound or breakwater1 has an extremely irregular surface1 even more so than between the islets on the reef of ,keeling atoll1 with which alone 7as there are no islets on the reef of ,tahiti7 it can properly be compared4 ,at ,tahiti1 the reef is very irregular in width2 but round many other encircled islands1 for instance1 ,vanikoro or ,gambier ,islands 7,figs #a and #h1 ,plate ,i71 it is quite as regular1 and of the same average width1 as in true atolls4 ,most barrier-reefs on the inner side slope irregularly into the lagoon-channel 7as the space of deep water separating the reef from the included land may be called71 but at ,vanikoro the reef slopes only for a short distance1 and then terminates abruptly in a submarine wall1 forty feet high1--a structure absolutely similar to that described by ,chamisso in the ,marshall atolls4 ,in the ,society ,archipelago1 ,ellis#c states1 that the reefs generally lie at the distance of from one to one and a half miles1 and1 occasionally1 even at more than three miles1 from the shore4 ,the central mountains are generally bordered by a fringe of flat1 and often marshy1 alluvial land1 from one to four miles in width4 ,this fringe consists of coral-sand and detritus thrown up from the lagoon-channel1 and of soil washed down from the hills2 it is an encroachment on the channel1 analogous to that low and inner part of the islets in many atolls which is formed by the accumulation of matter from the lagoon4 ,at ,hogoleu 7,fig4 #b1 ,plate ,i71 in the ,caroline ,archipelago1#d the reef on the south side is no less than twenty miles2 on the east side1 five2 and on the north side1 fourteen miles from the encircled high islands4 #c ,consult1 on this and other points1 the 8,polynesian ,researches10 by the ,rev4 ,w4 ,ellis1 an admirable work1 full of curious information4 #d ,see 8,hydrographical ,mem40 and the 8,atlas of the ,voyage of the ,astrolabe10 by ,captain ,dumont ,d',urville1 p4 #dbh4 ,the lagoon channels may be compared in every respect with true lagoons4 ,in some cases they are open1 with a level bottom of fine sand2 in others they are choked up with reefs of delicately branched corals1 which have the same general character as those within the ,keeling atoll4 ,these internal reefs either stand separately1 or more commonly skirt the shores of the included high islands4 ,the depth of the lagoon-channel round the ,society ,islands varies from two or three to thirty fathoms2 in ,cook's#e chart of ,ulieta1 however1 there is one sounding laid down of forty-eight fathoms2 at ,vanikoro there are several of fifty-four and one of fifty-six and a half fathoms 7,english71 a depth which even exceeds by a little that of the interior of the great ,maldiva atolls4 ,some barrier-reefs have very few islets on them2 whilst others are surmounted by numerous ones2 and those round part of ,bolabola 7,plate ,i1 ,fig4 #e7 form a single linear strip4 ,the islets first appear either on the angles of the reef1 or on the sides of the breaches through it1 and are generally most numerous on the windward side4 ,the reef to leeward retaining its usual width1 sometimes lies submerged several fathoms beneath the surface2 ,i have already mentioned ,gambier ,island as an instance of this structure4 ,submerged reefs1 having a less defined outline1 dead1 and covered with sand1 have been observed 7see ,appendix7 off some parts of ,huaheine and ,tahiti4 ,the reef is more frequently breached to leeward than to windward2 thus ,i find in ,krusenstern's 8,memoir on the ,pacific10 that there are passages through the encircling reef on the leeward side of each of the seven ,society ,islands1 which possess ship-harbours2 but that there are openings to windward through the reef of only three of them4 ,the breaches in the reef are seldom as deep as the interior lagoon-like channel2 they generally occur in front of the main valleys1 a circumstance which can be accounted for1 as will be seen in the fourth chapter1 without much difficulty4 ,the breaches being situated in front of the valleys1 which descend indifferently on all sides1 explains their more frequent occurrence through the windward side of barrier-reefs than through the windward side of atolls1--for in atolls there is no included land to influence the position of the breaches4 #e ,see the chart in vol4 i of ,hawkesworth's #dto ed4 of 8,cook's ,first ,voyage40 ,it is remarkable1 that the lagoon-channels round mountainous islands have not in every instance been long ago filled up with coral and sediment2 but it is more easily accounted for than appears at first sight4 ,in cases like that of ,hogoleu and the ,gambier ,islands1 where a few small peaks rise out of a great lagoon1 the conditions scarcely differ from those of an atoll1 and ,i have already shown1 at some length1 that the filling up of a true lagoon must be an extremely slow process4 ,where the channel is narrow1 the agency1 which on unprotected coasts is most productive of sediment1 namely the force of the breakers1 is here entirely excluded1 and the reef being breached in the front of the main valleys1 much of the finer mud from the rivers must be transported into the open sea4 ,as a current is formed by the water thrown over the edge of atoll-formed reefs1 which carries sediment with it through the deep-water breaches1 the same thing probably takes place in barrier-reefs1 and this would greatly aid in preventing the lagoon-channel from being filled up4 ,the low alluvial border1 however1 at the foot of the encircled mountains1 shows that the work of filling up is in progress2 and at ,maura 7,plate ,i1 ,fig4 #f71 in the ,society group1 it has been almost effected1 so that there remains only one harbour for small craft4 ,if we look at a set of charts of barrier-reefs1 and leave out in imagination the encircled land1 we shall find that1 besides the many points already noticed of resemblance1 or rather of identity in structure with atolls1 there is a close general agreement in form1 average dimensions1 and grouping4 ,encircling barrier-reefs1 like atolls1 are generally elongated1 with an irregularly rounded1 though sometimes angular outline4 ,there are atolls of all sizes1 from less than two miles in diameter to sixty miles 7excluding ,tilla-dou-,matte1 as it consists of a number of almost independent atoll-formed reefs72 and there are encircling barrier-reefs from three miles and a half to forty-six miles in diameter1--,turtle ,island being an instance of the former1 and ,hogoleu of the latter4 ,at ,tahiti the encircled island is thirty-six miles in its longest axis1 whilst at ,maurua it is only a little more than two miles4 ,it will be shown1 in the last chapter in this volume1 that there is the strictest resemblance in the grouping of atolls and of common islands1 and consequently there must be the same resemblance in the grouping of atolls and of encircling barrier-reefs4 ,the islands lying within reefs of this class1 are of very various heights4 ,tahiti#f is #g1jjj feet2 ,maurua about #hjj2 ,aitutaki #cfj1 and ,manouai only #ej4 ,the geological nature of the included land varies3 in most cases it is of ancient volcanic origin1 owing apparently to the fact that islands of this nature are most frequent within all great seas2 some1 however1 are of madreporitic limestone1 and others of primary formation1 of which latter kind ,new ,caledonia offers the best example4 ,the central land consists either of one island1 or of several3 thus1 in the ,society group1 ,eimeo stands by itself2 while ,taha and ,raiatea 7,fig4 #c1 ,plate ,i71 both moderately large islands of nearly equal size1 are included in one reef4 ,within the reef of the ,gambier group there are four large and some smaller islands 7,fig4 #h1 ,plate ,i72 within that of ,hogoleu 7,fig4 #b1 ,plate ,i7 nearly a dozen small islands are scattered over the expanse of one vast lagoon4 #f ,the height of ,tahiti is given from ,captain ,beechey2 ,maurua from ,mr4 ,f4 ,d4 ,bennett 7,geograph4 ,journ4 vol4 viii1 p4 #bbj72 ,aitutaki from measurements made on board the ,beagle2 and ,manouai or ,harvey ,island1 from an estimate by the ,rev4 ,j4 ,williams4 ,the two latter islands1 however1 are not in some respects well characterised examples of the encircled class4 ,after the details now given1 it may be asserted that there is not one point of essential difference between encircling barrier-reefs and atolls3 the latter enclose a simple sheet of water1 the former encircle an expanse with one or more islands rising from it4 ,i was much struck with this fact1 when viewing1 from the heights of ,tahiti1 the distant island of ,eimeo standing within smooth water1 and encircled by a ring of snow-white breakers4 ,remove the central land1 and an annular reef like that of an atoll in an early stage of its formation is left2 remove it from ,bolabola1 and there remains a circle of linear coral-islets1 crowned with tall cocoa-nut trees1 like one of the many atolls scattered over the ,pacific and ,indian ,oceans4 ,the barrier-reefs of ,australia and of ,new ,caledonia deserve a separate notice from their great dimensions4 ,the reef on the west coast of ,new ,caledonia 7,fig4 #e1 ,plate ,,ii7 is #djj miles in length2 and for a length of many leagues it seldom approaches within eight miles of the shore2 and near the southern end of the island1 the space between the reef and the land is sixteen miles in width4 ,the ,australian barrier extends1 with a few interruptions1 for nearly a thousand miles2 its average distance from the land is between twenty and thirty miles2 and in some parts from fifty to seventy4 ,the great arm of the sea thus included1 is from ten to twenty-five fathoms deep1 with a sandy bottom2 but towards the southern end1 where the reef is further from the shore1 the depth gradually increases to forty1 and in some parts to more than sixty fathoms4 ,flinders#g has described the surface of this reef as consisting of a hard white agglomerate of different kinds of coral1 with rough projecting points4 ,the outer edge is the highest part2 it is traversed by narrow gullies1 and at rare intervals is breached by ship-channels4 ,the sea close outside is profoundly deep2 but1 in front of the main breaches1 soundings can sometimes be obtained4 ,some low islets have been formed on the reef4 #g ,flinders' 8,voyage to ,terra ,australis10 vol4 ii1 p4 #hh4 ,there is one important point in the structure of barrier-reefs which must here be considered4 ,the accompanying diagrams represent north and south vertical sections1 taken through the highest points of ,vanikoro1 ,gambier1 and ,maurua ,islands1 and through their encircling reefs4 ,the scale both in the horizontal and vertical direction is the same1 namely1 a quarter of an inch to a nautical mile4 ,the height and width of these islands is known2 and ,i have attempted to represent the form of the land from the shading of the hills in the large published charts4 ,it has long been remarked1 even from the time of ,dampier1 that considerable degree of relation subsists between the inclination of that part of the land which is beneath water and that above it2 hence the dotted line in the three sections1 probably1 does not widely differ in inclination from the actual submarine prolongation of the land4 ,if we now look at the outer edge of the reef 7,,aa71 and bear in mind that the plummet on the right hand represents a depth of #a1bjj feet1 we must conclude that the vertical thickness of these barrier coral-reefs is very great4 ,illustration3 ,vertical thickness of ,vanikoro1 ,gambier and ,maurua4 #a4 ,,vanikoro1 from the 8,atlas of the ,voyage of the ,astrolabe10 by ,d4 ,d',urville4 #b4 ,,gambier ,,island1 from ,beechey4 #c4 ,,maurua1 from the 8,atlas of the ,voyage of the ,coquille10 by ,duperrey4 ,the horizontal line is the level of the sea1 from which on the right hand a plummet descends1 representing a depth of #bjj fathoms1 or #a1bjj feet4 ,the vertical shading shows the section of the land1 and the horizontal shading that of the encircling barrier-reef3 from the smallness of the scale1 the lagoon-channel could not be represented4 ,,aa4--,outer edge of the coral-reefs1 where the sea breaks4 ,,bb4--,the shore of the encircled islands4 ,i must observe that if the sections had been taken in any other direction across these islands1 or across other encircled islands1#h the result would have been the same4 ,in the succeeding chapter it will be shown that reef-building polypifers cannot flourish at great depths1--for instance1 it is highly improbable that they could exist at a quarter of the depth represented by the plummet on the right hand of the woodcut4 ,here there is a great apparent difficulty--how were the basal parts of these barrier-reef formed8 ,it will1 perhaps1 occur to some1 that the actual reefs formed of coral are not of great thickness1 but that before their first growth1 the coasts of these encircled islands were deeply eaten into1 and a broad but shallow submarine ledge thus left1 on the edge of which the coral grew2 but if this had been the case1 the shore would have been invariably bounded by lofty cliffs1 and not have sloped down to the lagoon-channel1 as it does in many instances4 ,on this view1#i moreover1 the cause of the reef springing up at such a great distance from the land1 leaving a deep and broad moat within1 remains altogether unexplained4 ,a supposition of the same nature1 and appearing at first more probable is1 that the reefs sprung up from banks of sediment1 which had accumulated round the shore previously to the growth of the coral2 but the extension of a bank to the same distance round an unbroken coast1 and in front of those deep arms of the sea 7as in ,raiatea1 see ,plate ,,ii1 ,fig4 #c7 which penetrate nearly to the heart of some encircled islands1 is exceedingly improbable4 ,and why1 again1 should the reef spring up1 in some cases steep on both sides like a wall1 at a distance of two1 three or more miles from the shore1 leaving a channel often between two hundred and three hundred feet deep1 and rising from a depth which we have reason to believe is destructive to the growth of coral8 ,an admission of this nature cannot possibly be made4 ,the existence1 also1 of the deep channel1 utterly precludes the idea of the reef having grown outwards1 on a foundation slowly formed on its outside1 by the accumulation of sediment and coral detritus4 ,nor1 again1 can it be asserted1 that the reef-building corals will not grow1 excepting at a great distance from the land2 for1 as we shall soon see1 there is a whole class of reefs1 which take their name from growing closely attached 7especially where the sea is deep7 to the beach4 ,at ,new ,caledonia 7see ,plate ,,ii1 ,fig4 #e7 the reefs which run in front of the west coast are prolonged in the same line #aej miles beyond the northern extremity of the island1 and this shows that some explanation1 quite different from any of those just suggested1 is required4 ,the continuation of the reefs on each side of the submarine prolongation of ,new ,caledonia1 is an exceedingly interesting fact1 if this part formerly existed as the northern extremity of the island1 and before the attachment of the coral had been worn down by the action of the sea1 or if it originally existed at its present height1 with or without beds of sediment on each flank1 how can we possibly account for the reefs1 not growing on the crest of this submarine portion1 but fronting its sides1 in the same line with the reefs which front the shores of the lofty island8 ,we shall hereafter see1 that there is one1 and ,i believe only one1 solution of this difficulty4 #h ,in the fifth chapter an east and west section across the ,island of ,bolabola and its barrier-reefs is given1 for the sake of illustrating another point4 ,the unbroken line in it 7woodcut ,no4 #e7 is the section referred to4 ,the scale is 4#eg of an inch to a mile2 it is taken from the 8,atlas of the ,voyage of the ,coquille10 by ,duperrey4 ,the depth of the lagoon-channel is exaggerated4 #i ,the ,rev4 ,d4 ,tyerman and ,mr4 ,bennett 78,journal of ,voyage and ,travels10 vol4 i1 p4 #bae7 have briefly suggested this explanation of the origin of the encircling reefs of the ,society ,islands4 ,one other supposition to account for the position of encircling barrier-reefs remains1 but it is almost too preposterous to be mentioned2 namely1 that they rest on enormous submarine craters1 surrounding the included islands4 ,when the size1 height1 and form of the islands in the ,society group are considered1 together with the fact that all are thus encircled1 such a notion will be rejected by almost every one4 ,new ,caledonia1 moreover1 besides its size1 is composed of primitive formations1 as are some of the ,comoro ,islands2#aj and ,aitutaki consists of calcareous rock4 ,we must1 therefore1 reject these several explanations1 and conclude that the vertical thickness of barrier-reefs1 from their outer edges to the foundation on which they rest 7from ,,aa in the section to the dotted lines7 is really great2 but in this1 there is no difficulty1 for it is not necessary to suppose that the coral has sprung up from an immense depth1 as will be evident when the theory of the upward growth of coral-reefs1 during the slow subsidence of their foundation1 is discussed4 #aj ,i have been informed that this is the case by ,dr4 ,allan of ,forres1 who has visited this group4 ,chapter ,,iii ,,fringing ,,or ,,shore-,,reefs ,reefs of ,mauritius4--,shallow channel within the reef4--,its slow filling up4--,currents of water formed within it4--,upraised reefs4--,narrow fringing-reefs in deep seas4--,reefs on the coast of ,east ,africa and of ,brazil4--,fringing-reefs in very shallow seas1 round banks of sediment and on worn-down islands4--,fringing-reefs affected by currents of the sea4--,coral coating the bottom of the sea1 but not forming reefs4 ,fringing-reefs1 or1 as they have been called by some voyagers1 shore-reefs1 whether skirting an island or part of a continent1 might at first be thought to differ little1 except in generally being of less breadth1 from barrier-reefs4 ,as far as the superficies of the actual reef is concerned this is the case2 but the absence of an interior deep-water channel1 and the close relation in their horizontal extension with the probable slope beneath the sea of the adjoining land1 present essential points of difference4 ,the reefs which fringe the island of ,mauritius offer a good example of this class4 ,they extend round its whole circumference1 with the exception of two or three parts1#a where the coast is almost precipitous1 and where1 if as is probable the bottom of the sea has a similar inclination1 the coral would have no foundation on which to become attached4 ,a similar fact may sometimes be observed even in reefs of the barrier class1 which follow much less closely the outline of the adjoining land2 as1 for instance1 on the south-east and precipitous side of ,tahiti1 where the encircling reef is interrupted4 ,on the western side of the ,mauritius1 which was the only part ,i visited1 the reef generally lies at the distance of about half a mile from the shore2 but in some parts it is distant from one to two1 and even three miles4 ,but even in this last case1 as the coast-land is gently inclined from the foot of the mountains to the sea-beach1 and as the soundings outside the reef indicate an equally gentle slope beneath the water1 there is no reason for supposing that the basis of the reef1 formed by the prolongation of the strata of the island1 lies at a greater depth than that at which the polypifers could begin constructing the reef4 ,some allowance1 however1 must be made for the outward extension of the corals on a foundation of sand and detritus1 formed from their own wear1 which would give to the reef a somewhat greater vertical thickness1 than would otherwise be possible4 #a ,this fact is stated on the authority of the ,officier du ,roi1 in his extremely interesting 8,voyage l',isle de ,france10 undertaken in #agfh4 ,according to ,captain ,carmichael 7,hooker's ,bot4 ,misc4 vol4 ii1 p4 #caf7 on one part of the coast there is a space for sixteen miles without a reef4 ,the outer edge of the reef on the western or leeward side of the island is tolerably well defined1 and is a little higher than any other part4 ,it chiefly consists of large strongly branched corals1 of the genus ,madrepora1 which also form a sloping bed some way out to sea3 the kinds of coral growing in this part will be described in the ensuing chapter4 ,between the outer margin and the beach1 there is a flat space with a sandy bottom and a few tufts of living coral2 in some parts it is so shallow1 that people1 by avoiding the deeper holes and gullies1 can wade across it at low water2 in other parts it is deeper1 seldom however exceeding ten or twelve feet1 so that it offers a safe coasting channel for boats4 ,on the eastern and windward side of the island1 which is exposed to a heavy surf1 the reef was described to me as having a hard smooth surface1 very slightly inclined inwards1 just covered at low-water1 and traversed by gullies2 it appears to be quite similar in structure to the reefs of the barrier and atoll classes4 ,the reef of ,mauritius1 in front of every river and streamlet1 is breached by a straight passage3 at ,grand ,port1 however1 there is a channel like that within a barrier-reef2 it extends parallel to the shore for four miles1 and has an average depth of ten or twelve fathoms2 its presence may probably be accounted for by two rivers which enter at each end of the channel1 and bend towards each other4 ,the fact of reefs of the fringing class being always breached in front of streams1 even of those which are dry during the greater part of the year1 will be explained1 when the conditions unfavourable to the growth of coral are considered4 ,low coral-islets1 like those on barrier-reefs and atolls1 are seldom formed on reefs of this class1 owing apparently in some cases to their narrowness1 and in others to the gentle slope of the reef outside not yielding many fragments to the breakers4 ,on the windward side1 however1 of the ,mauritius1 two or three small islets have been formed4 ,it appears1 as will be shown in the ensuing chapter1 that the action of the surf is favourable to the vigorous growth of the stronger corals1 and that sand or sediment1 if agitated by the waves1 is injurious to them4 ,hence it is probable that a reef on a shelving shore1 like that of ,mauritius1 would at first grow up1 not attached to the actual beach1 but at some little distance from it2 and the corals on the outer margin would be the most vigorous4 ,a shallow channel would thus be formed within the reef1 and as the breakers are prevented acting on the shores of the island1 and as they do not ordinarily tear up many fragments from the outside1 and as every streamlet has its bed prolonged in a straight line through the reef1 this channel could be filled up only very slowly with sediment4 ,but a beach of sand and of fragments of the smaller kinds of coral seems1 in the case of ,mauritius1 to be slowly encroaching on the shallow channel4 ,on many shelving and sandy coasts1 the breakers tend to form a bar of sand a little way from the beach1 with a slight increase of depth within it2 for instance1 ,captain ,grey#b states that the west coast of ,australia1 in latitude #bd1 is fronted by a sand bar about two hundred yards in width1 on which there is only two feet of water2 but within it the depth increases to two fathoms4 ,similar bars1 more or less perfect1 occur on other coasts4 ,in these cases ,i suspect that the shallow channel 7which no doubt during storms is occasionally obliterated7 is scooped out by the flowing away of the water thrown beyond the line1 on which the waves break with the greatest force4 ,at ,pernambuco a bar of hard sandstone1#c which has the same external form and height as a coral-reef1 extends nearly parallel to the coast2 within this bar currents1 apparently caused by the water thrown over it during the greater part of each tide1 run strongly1 and are wearing away its inner wall4 ,from these facts it can hardly be doubted1 that within most fringing-reefs1 especially within those lying some distance from the land1 a return stream must carry away the water thrown over the outer edge2 and the current thus produced1 would tend to prevent the channel being filled up with sediment1 and might even deepen it under certain circumstances4 ,to this latter belief ,i am led1 by finding that channels are almost universally present within the fringing-reefs of those islands which have undergone recent elevatory movements2 and this could hardly have been the case1 if the conversion of the very shallow channel into land had not been counteracted to a certain extent4 #b ,captain ,grey's 8,journal of ,two ,expeditions10 vol4 i1 p4 #cfi4 #c ,i have described this singular structure in the ,lond4 and ,edin4 ,phil4 ,mag41 ,october #ahda4 ,a fringing-reef1 if elevated in a perfect condition above the level of the sea1 ought to present the singular appearance of a broad dry moat within a low mound4 ,the author#d of an interesting pedestrian tour round the ,mauritius1 seems to have met with a structure of this kind3 he says 8,j'observai que l1 o la mer tale1 indpendamment des rescifs du large1 il y terre une espce d'effoncement ou chemin couvert naturel4 ,on y pourrait mettre du canon10 etc4 ,in another place he adds1 8,avant de passer le ,cap1 on remarque un gros banc de corail lev de plus de quinze pieds3 c'est une espce de rescif1 que la mer abandonn1 il regne au pied une longue flaque d'eau1 dont on pourrait faire un bassin pour de petits vaisseaux40 ,but the margin of the reef1 although the highest and most perfect part1 from being most exposed to the surf1 would generally during a slow rise of the land be either partially or entirely worn down to that level1 at which corals could renew their growth on its upper edge4 ,on some parts of the coast-land of ,mauritius there are little hillocks of coral-rock1 which are either the last remnants of a continuous reef1 or of low islets formed on it4 ,i observed that two such hillocks between ,tamarin ,bay and the ,great ,black ,river2 they were nearly twenty feet high1 about two hundred yards from the present beach1 and about thirty feet above its level4 ,they rose abruptly from a smooth surface1 strewed with worn fragments of coral4 ,they consisted in their lower part of hard calcareous sandstone1 and in their upper of great blocks of several species of ,astra and ,madrepora1 loosely aggregated2 they were divided into irregular beds1 dipping seaward1 in one hillock at an angle of #h1 and in the other at #ah4 ,i suspect that the superficial parts of the reefs1 which have been upraised together with the islands they fringe1 have generally been much more modified by the wearing action of the sea1 than those of ,mauritius4 #d 8,voyage l',isle de ,france1 par un ,officier du ,roi10 part i1 pp4 #aib1 #bjj4 ,many islands#e are fringed by reefs quite similar to those of ,mauritius2 but on coasts where the sea deepens very suddenly the reefs are much narrower1 and their limited extension seems evidently to depend on the high inclination of the submarine slope2 a relation1 which1 as we have seen1 does not exist in reefs of the barrier class4 ,the fringing-reefs on steep coasts are frequently not more than from fifty to one hundred yards in width2 they have a nearly smooth1 hard surface1 scarcely uncovered at low water1 and without any interior shoal channel1 like that within those fringing-reefs1 which lie at a greater distance from the land4 ,the fragments torn up during gales from the outer margin are thrown over the reef on the shores of the island4 ,i may give as instances1 ,wateeo1 where the reef is described by ,cook as being a hundred yards wide2 and ,mauti and ,elizabeth#f ,islands1 where it is only fifty yards in width3 the sea round these islands is very deep4 #e ,i may give ,cuba1 as another instance2 ,mr4 ,taylor 7,loudon's ,mag4 of ,nat4 ,hist41 vol4 ix1 p4 #ddi7 has described a reef several miles in length between ,gibara and ,vjaro1 which extends parallel to the shore at the distance of between half and the third part of a mile1 and encloses a space of shallow water1 with a sandy bottom and tufts of coral4 ,outside the edge of the reef1 which is formed of great branching corals1 the depth is six and seven fathoms4 ,this coast has been upheaved at no very distant geological period4 #f ,mauti is described by ,lord ,byron in the voyage of ,h4,m4,s4 ,blonde1 and ,elizabeth ,island by ,captain ,beechey4 ,fringing-reefs1 like barrier-reefs1 both surround islands1 and front the shores of continents4 ,in the charts of the eastern coast of ,africa1 by ,captain ,owen1 many extensive fringing-reefs are laid down2 thus1 for a space of nearly forty miles1 from latitude #a #ae to #a #de ,s41 a reef fringes the shore at an average distance of rather more than one mile1 and therefore at a greater distance than is usual in reefs of this class2 but as the coast-land is not lofty1 and as the bottom shoals very gradually 7the depth being only from eight to fourteen fathoms at a mile and a half outside the reef71 its extension thus far from the land offers no difficulty4 ,the external margin of this reef is described1 as formed of projecting points1 within which there is a space1 from six to twelve feet deep1 with patches of living coral on it4 ,at ,mukdeesha 7lat4 #b #a ,n47 8the port is formed10 it is said1#g 8by a long reef extending eastward1 four or five miles1 within which there is a narrow channel1 with ten to twelve feet of water at low spring-tides20 it lies at the distance of a quarter of a mile from the shore4 ,again1 in the plan of ,mombas 7lat4 #d ,s471 a reef extends for thirty-six miles1 at the distance of from half a mile to one mile and a quarter from the shore2 within it1 there is a channel navigable 8for canoes and small craft10 between six and fifteen feet deep3 outside the reef the depth is about thirty fathoms at the distance of nearly half a mile4 ,part of this reef is very symmetrical1 and has a uniform breadth of two hundred yards4 #g ,owen's 8,africa10 vol4 i1 p4 #ceg1 from which work the foregoing facts are likewise taken4 ,the coast of ,brazil is in many parts fringed by reefs4 ,of these1 some are not of coral formation2 for instance1 those near ,bahia and in front of ,pernambuco2 but a few miles south of this latter city1 the reef follows#h so closely every turn of the shore1 that ,i can hardly doubt it is of coral2 it runs at the distance of three-quarters of a mile from the land1 and within it the depth is from ten to fifteen feet4 ,i was assured by an intelligent pilot that at ,ports ,frances and ,maceio1 the outer part of the reef consists of living coral1 and the inner of a white stone1 full of large irregular cavities1 communicating with the sea4 ,the bottom of the sea off the coast of ,brazil shoals gradually to between thirty and forty fathoms1 at the distance of between nine and ten leagues from the land4 #h ,see ,baron ,roussin's 8,pilote du ,brsil10 and accompanying hydrographical memoir4 ,from the description now given1 we must conclude that the dimensions and structure of fringing-reefs depend entirely on the greater or less inclination of the submarine slope1 conjoined with the fact that reef-building polypifers can exist only at limited depths4 ,it follows from this1 that where the sea is very shallow1 as in the ,persian ,gulf and in parts of the ,east ,indian ,archipelago1 the reefs lose their fringing character1 and appear as separate and irregularly scattered patches1 often of considerable area4 ,from the more vigorous growth of the coral on the outside1 and from the conditions being less favourable in several respects within1 such reefs are generally higher and more perfect in their marginal than in their central parts2 hence these reefs sometimes assume 7and this circumstance ought not to be overlooked7 the appearance of atolls2 but they differ from atolls in their central expanse being much less deep1 in their form being less defined1 and in being based on a shallow foundation4 ,but when in a deep sea reefs fringe banks of sediment1 which have accumulated beneath the surface1 round either islands or submerged rocks1 they are distinguished with difficulty on the one hand from encircling barrier-reefs1 and on the other from atolls4 ,in the ,west ,indies there are reefs1 which ,i should probably have arranged under both these classes1 had not the existence of large and level banks1 lying a little beneath the surface1 ready to serve as the basis for the attachment of coral1 been occasionally brought into view by the entire or partial absence of reefs on them1 and had not the formation of such banks1 through the accumulation of sediment now in progress1 been sufficiently evident4 ,fringing-reefs sometimes coat1 and thus protect the foundations of islands1 which have been worn down by the surf to the level of the sea4 ,according to ,ehrenberg1 this has been extensively the case with the islands in the ,red ,sea1 which formerly ranged parallel to the shores of the mainland1 with deep water within them3 hence the reefs now coating their bases are situated relatively to the land like barrier-reefs1 although not belonging to that class2 but there are1 as ,i believe1 in the ,red ,sea some true barrier-reefs4 ,the reefs of this sea and of the ,west ,indies will be described in the ,appendix4 ,in some cases1 fringing-reefs appear to be considerably modified in outline by the course of the prevailing currents4 ,dr4 ,j4 ,allan informs me that on the east coast of ,madagascar almost every headland and low point of sand has a coral-reef extending from it in a ,s4,w4 and ,n4,e4 line1 parallel to the currents on that shore4 ,i should think the influence of the currents chiefly consisted in causing an extension1 in a certain direction1 of a proper foundation for the attachment of the coral4 ,round many intertropical islands1 for instance the ,abrolhos on the coast of ,brazil surveyed by ,captain ,fitzroy1 and1 as ,i am informed by ,mr4 ,cuming1 round the ,philippines1 the bottom of the sea is entirely coated by irregular masses of coral1 which although often of large size1 do not reach the surface and form proper reefs4 ,this must be owing1 either to insufficient growth1 or to the absence of those kinds of corals which can withstand the breaking of the waves4 ,the three classes1 atoll-formed1 barrier1 and fringing-reefs1 together with the modifications just described of the latter1 include all the most remarkable coral formations anywhere existing4 ,at the commencement of the last chapter in the volume1 where ,i detail the principles on which the map 7,plate ,,iii7 is coloured1 the exceptional cases will be enumerated4 ,chapter ,,iv ,,on ,,the ,,distribution ,,and ,,growth ,,of ,,coral-,,reefs ,in this chapter ,i will give all the facts which ,i have collected1 relating to the distribution of coral-reefs1--to the conditions favourable to their increase1--to the rate of their growth1--and to the depth at which they are formed4 ,these subjects have an important bearing on the theory of the origin of the different classes of coral-reefs4 ,section ,i--,,on ,,the ,,distribution ,,of ,,coral-,,reefs1 ,,and ,,on ,,the ,,conditions ,,favourable ,,to ,,their ,,increase ,with regard to the limits of latitude1 over which coral-reefs extend1 ,i have nothing new to add4 ,the ,bermuda ,islands1 in #cb #ae ,n41 is the point furthest removed from the equator1 in which they appear to exist2 and it has been suggested that their extension so far northward in this instance is owing to the warmth of the ,gulf ,stream4 ,in the ,pacific1 the ,loo ,choo ,islands1 in latitude #bg ,n41 have reefs on their shores1 and there is an atoll in #bh #cj1 situated ,n4,w4 of the ,sandwich ,archipelago4 ,in the ,red ,sea there are coral-reefs in latitude #cj4 ,in the southern hemisphere coral-reefs do not extend so far from the equatorial sea4 ,in the ,southern ,pacific there are only a few reefs beyond the line of the tropics1 but ,houtmans ,abrolhos1 on the western shores of ,australia in latitude #bi ,s41 are of coral formation4 ,the proximity of volcanic land1 owing to the lime generally evolved from it1 has been thought to be favourable to the increase of coral-reefs4 ,there is1 however1 not much foundation for this view2 for nowhere are coral-reefs more extensive than on the shores of ,new ,caledonia1 and of north-eastern ,australia1 which consist of primary formations2 and in the largest groups of atolls1 namely the ,maldiva1 ,chagos1 ,marshall1 ,gilbert1 and ,low ,archipelagoes1 there is no volcanic or other kind of rock1 excepting that formed of coral4 ,the entire absence of coral-reefs in certain large areas within the tropical seas1 is a remarkable fact4 ,thus no coral-reefs were observed1 during the surveying voyages of the ,beagle and her tender on the west coast of ,south ,america south of the equator1 or round the ,galapagos ,islands4 ,it appears1 also1 that there are none#a north of the equator2 ,mr4 ,lloyd1 who surveyed the ,isthmus of ,panama1 remarked to me1 that although he had seen corals living in the ,bay of ,panama1 yet he had never observed any reefs formed by them4 ,i at first attributed this absence of reefs on the coasts of ,peru and of the ,galapagos ,islands1#b to the coldness of the currents from the south1 but the ,gulf of ,panama is one of the hottest pelagic districts in the world4#c ,in the central parts of the ,pacific there are islands entirely free from reefs2 in some few of these cases ,i have thought that this was owing to recent volcanic action2 but the existence of reefs round the greater part of ,hawaii1 one of the ,sandwich ,islands1 shows that recent volcanic action does not necessarily prevent their growth4 #a ,i have been informed that this is the case1 by ,lieutenant ,ryder1 ,r4,n41 and others who have had ample opportunities for observation4 #b ,the mean temperature of the surface sea from observations made by the direction of ,captain ,fitzroy on the shores of the ,galapagos ,islands1 between the #afth of ,september and the #bjth of ,october1 #ahce1 was #fh ,fahr4 ,the lowest temperature observed was #eh4e at the south-west end of ,albemarle ,island2 and on the west coast of this island1 it was several times #fb and #fc4 ,the mean temperature of the sea in the ,low ,archipelago of atolls1 and near ,tahiti1 from similar observations made on board the ,beagle1 was 7although further from the equator7 #gg4e1 the lowest any day being #gf4e4 ,therefore we have here a difference of #i4e in mean temperature1 and #ah in extremes2 a difference doubtless quite sufficient to affect the distribution of organic beings in the two areas4 #c ,humboldt's 8,personal ,narrative10 vol4 vii1 p4 #dcd4 ,in the last chapter ,i stated that the bottom of the sea round some islands is thickly coated with living corals1 which nevertheless do not form reefs1 either from insufficient growth1 or from the species not being adapted to contend with the breaking waves4 ,i have been assured by several people1 that there are no coral-reefs on the west coast of ,africa1#d or round the islands in the ,gulf of ,guinea4 ,this perhaps may be attributed1 in part1 to the sediment brought down by the many rivers debouching on that coast1 and to the extensive mud-banks1 which line great part of it4 ,but the islands of ,st4 ,helena1 ,ascension1 the ,cape ,verdes1 ,st4 ,paul's1 and ,fernando ,noronha1 are1 also1 entirely without reefs1 although they lie far out at sea1 are composed of the same ancient volcanic rocks1 and have the same general form1 with those islands in the ,pacific1 the shores of which are surrounded by gigantic walls of coral-rock4 ,with the exception of ,bermuda1 there is not a single coral-reef in the central expanse of the ,atlantic ,ocean4 ,it will1 perhaps1 be suggested that the quantity of carbonate of lime in different parts of the sea1 may regulate the presence of reefs4 ,but this cannot be the case1 for at ,ascension1 the waves charged to excess precipitate a thick layer of calcareous matter on the tidal rocks2 and at ,st4 ,jago1 in the ,cape ,verdes1 carbonate of lime not only is abundant on the shores1 but it forms the chief part of some upraised post-tertiary strata4 ,the apparently capricious distribution1 therefore1 of coral-reefs1 cannot be explained by any of these obvious causes2 but as the study of the terrestrial and better known half of the world must convince every one that no station capable of supporting life is lost1--nay more1 that there is a struggle for each station1 between the different orders of nature1--we may conclude that in those parts of the intertropical sea1 in which there are no coral-reefs1 there are other organic bodies supplying the place of the reef-building polypifers4 ,it has been shown in the chapter on ,keeling atoll that there are some species of large fish1 and the whole tribe of ,holothuri which prey on the tenderer parts of the corals4 ,on the other hand1 the polypifers in their turn must prey on some other organic beings2 the decrease of which from any cause would cause a proportionate destruction of the living coral4 ,the relations1 therefore1 which determine the formation of reefs on any shore1 by the vigorous growth of the efficient kinds of coral1 must be very complex1 and with our imperfect knowledge quite inexplicable4 ,from these considerations1 we may infer that changes in the condition of the sea1 not obvious to our senses1 might destroy all the coral-reefs in one area1 and cause them to appear in another3 thus1 the ,pacific or ,indian ,ocean might become as barren of coral-reefs as the ,atlantic now is1 without our being able to assign any adequate cause for such a change4 #d ,it might be concluded1 from a paper by ,captain ,owen 7,geograph4 ,journ41 vol4 ii1 p4 #hi71 that the reefs off ,cape ,st4 ,anne and the ,sherboro' ,islands were of coral1 although the author states that they are not purely coralline4 ,but ,i have been assured by ,lieutenant ,holland1 ,r4,n41 that these reefs are not of coral1 or at least that they do not at all resemble those in the ,west ,indies4 ,it has been a question with some naturalists1 which part of a reef is most favourable to the growth of coral4 ,the great mounds of living ,porites and of ,millepora round ,keeling atoll occur exclusively on the extreme verge of the reef1 which is washed by a constant succession of breakers2 and living coral nowhere else forms solid masses4 ,at the ,marshall islands the larger kinds of coral 7chiefly species of ,astra1 a genus closely allied to ,porites7 8which form rocks measuring several fathoms in thickness10 prefer1 according to ,chamisso1#e the most violent surf4 ,i have stated that the outer margin of the ,maldiva atolls consists of living corals 7some of which1 if not all1 are of the same species with those at ,keeling atoll71 and here the surf is so tremendous1 that even large ships have been thrown1 by a single heave of the sea1 high and dry on the reef1 all on board thus escaping with their lives4 ,ehrenberg#f remarks1 that in the ,red ,sea the strongest corals live on the outer reefs1 and appear to love the surf2 he adds1 that the more branched kinds abound a little way within1 but that even these in still more protected places1 become smaller4 ,many other facts having a similar tendency might be adduced4#g ,it has1 however1 been doubted by ,,mm4 ,quoy and ,gaimard1 whether any kind of coral can even withstand1 much less flourish in1 the breakers of an open sea3#h they affirm that the saxigenous lithophytes flourish only where the water is tranquil1 and the heat intense4 ,this statement has passed from one geological work to another2 nevertheless1 the protection of the whole reef undoubtedly is due to those kinds of coral1 which cannot exist in the situations thought by these naturalists to be most favourable to them4 ,for should the outer and living margin perish1 of any one of the many low coral-islands1 round which a line of great breakers is incessantly foaming1 the whole1 it is scarcely possible to doubt1 would be washed away and destroyed1 in less than half a century4 ,but the vital energies of the corals conquer the mechanical power of the waves2 and the large fragments of reef torn up by every storm1 are replaced by the slow but steady growth of the innumerable polypifers1 which form the living zone on its outer edge4 #e ,kotzebue's 8,first ,voyage0 7,eng4 ,trans471 vol4 iii1 pp4 #adb1 #adc1 #cca4 #f ,ehrenberg1 8ber die ,natur und ,bildung der ,corallen ,bnke im rothen ,meere10 p4 #di4 #g ,in the ,west ,indies1 as ,i am informed by ,captain ,bird ,allen1 ,r4,n41 it is the common belief of those1 who are best acquainted with the reefs1 that the coral flourishes most1 where freely exposed to the swell of the open sea4 #h 8,annales des ,sciences ,naturelles10 tome vi1 pp4 #bgf1 #bgh4--8,l o les ondes sont agites1 les ,lytophyts ne peuvent travailler1 parce qu'elles dtruiraient leurs fragiles difices10 etc4 ,from these facts1 it is certain1 that the strongest and most massive corals flourish1 where most exposed4 ,the less perfect state of the reef of most atolls on the leeward and less exposed side1 compared with its state to windward2 and the analogous case of the greater number of breaches on the near sides of those atolls in the ,maldiva ,archipelago1 which afford some protection to each other1 are obviously explained by this circumstance4 ,if the question had been1 under what conditions the greater number of species of coral1 not regarding their bulk and strength1 were developed1 ,i should answer1--probably in the situations described by ,,mm4 ,quoy and ,gaimard1 where the water is tranquil and the heat intense4 ,the total number of species of coral in the circumtropical seas must be very great3 in the ,red ,sea alone1 #abj kinds1 according to ,ehrenberg1#i have been observed4 #i ,ehrenberg1 8ber die ,natur10 etc41 p4 #df4 ,the same author has observed that the recoil of the sea from a steep shore is injurious to the growth of coral1 although waves breaking over a bank are not so4 ,ehrenberg also states1 that where there is much sediment1 placed so as to be liable to be moved by the waves there is little or no coral2 and a collection of living specimens placed by him on a sandy shore died in the course of a few days4#aj ,an experiment1 however1 will presently be related in which some large masses of living coral increased rapidly in size1 after having been secured by stakes on a sandbank4 ,that loose sediment should be injurious to the living polypifers1 appears1 at first sight1 probable2 and accordingly1 in sounding off ,keeling atoll1 and 7as will hereafter be shown7 off ,mauritius1 the arming of the lead invariably came up clean1 where the coral was growing vigorously4 ,this same circumstance has probably given rise to a strange belief1 which1 according to ,captain ,owen1#aa is general amongst the inhabitants of the ,maldiva atolls1 namely that corals have roots1 and therefore that if merely broken down to the surface1 they grow up again2 but1 if rooted out1 they are permanently destroyed4 ,by this means the inhabitants keep their harbours clear2 and thus the ,french ,governor of ,st4 ,mary's in ,madagascar1 8cleared out and made a beautiful little port at that place40 ,for it is probable that sand would accumulate in the hollows formed by tearing out the corals1 but not on the broken and projecting stumps1 and therefore1 in the former case1 the fresh growth of the coral might be thus prevented4 #aj ,ibid41 p4 #di4 #aa ,captain ,owen on the ,geography of the ,maldiva ,islands1 ,geograph4 ,journal1 vol4 ii1 p4 #hh4 ,in the last chapter ,i remarked that fringing-reefs are almost universally breached1 where streams enter the sea4#ab ,most authors have attributed this fact to the injurious effects of the fresh water1 even where it enters the sea only in small quantity1 and during a part of the year4 ,no doubt brackish water would prevent or retard the growth of coral2 but ,i believe that the mud and sand which is deposited1 even by rivulets when flooded1 is a much more efficient check4 ,the reef on each side of the channel leading into ,port ,louis at ,mauritius1 ends abruptly in a wall1 at the foot of which ,i sounded and found a bed of thick mud4 ,this steepness of the sides appears to be a general character in such breaches4 ,cook1#ac speaking of one at ,raiatea1 says1 8like all the rest1 it is very steep on both sides40 ,now1 if it were the fresh water mingling with the salt which prevented the growth of coral1 the reef certainly would not terminate abruptly1 but as the polypifers nearest the impure stream would grow less vigorously than those farther off1 so would the reef gradually thin away4 ,on the other hand1 the sediment brought down from the land would only prevent the growth of the coral in the line of its deposition1 but would not check it on the side1 so that the reefs might increase till they overhung the bed of the channel4 ,the breaches are much fewer in number1 and front only the larger valleys in reefs of the encircling barrier class4 ,they probably are kept open in the same manner as those into the lagoon of an atoll1 namely1 by the force of the currents and the drifting outwards of fine sediment4 ,their position in front of valleys1 although often separated from the land by deep water lagoon-channels1 which it might be thought would entirely remove the injurious effects both of the fresh water and the sediment1 will receive a simple explanation when we discuss the origin of barrier-reefs4 #ab ,lieutenant ,wellstead and others have remarked that this is the case in the ,red ,sea2 ,dr4 ,rppell 78,reise in ,abyss40 ,band4 i1 p4 #adb7 says that there are pear-shaped harbours in the upraised coral-coast1 into which periodical streams enter4 ,from this circumstance1 ,i presume1 we must infer that before the upheaval of the strata now forming the coast-land1 fresh water and sediment entered the sea at these points2 and the coral being thus prevented growing1 the pear-shaped harbours were produced4 #ac ,cook's 8,first ,voyage10 vol4 ii1 p4 #bga 7,hawkesworth's edit47 ,in the vegetable kingdom every different station has its peculiar group of plants1 and similar relations appear to prevail with corals4 ,we have already described the great difference between the corals within the lagoon of an atoll and those on its outer margin4 ,the corals1 also1 on the margin of ,keeling ,island occurred in zones2 thus the ,porites and ,millepora complanata grow to a large size only where they are washed by a heavy sea1 and are killed by a short exposure to the air2 whereas1 three species of ,nullipora also live amidst the breakers1 but are able to survive uncovered for a part of each tide2 at greater depths1 a strong ,madrepora and ,millepora alcicornis are the commonest kinds1 the former appearing to be confined to this part1 beneath the zone of massive corals1 minute encrusting corallines and other organic bodies live4 ,if we compare the external margin of the reef at ,keeling atoll with that on the leeward side of ,mauritius1 which are very differently circumstanced1 we shall find a corresponding difference in the appearance of the corals4 ,at the latter place1 the genus ,madrepora is preponderant over every other kind1 and beneath the zone of massive corals there are large beds of ,seriatopora4 ,there is also a marked difference1 according to ,captain ,moresby1#ad;between the great branching corals of the ,red ,sea1 and those on the reefs of the ,maldiva atolls4 #ad ,captain ,moresby on the ,northern ,maldiva atolls1 ,geograph4 ,journal1 vol4 v1 p4 #dja4 ,these facts1 which in themselves are deserving of notice1 bear1 perhaps1 not very remotely1 on a remarkable circumstance which has been pointed out to me by ,captain ,moresby1 namely1 that with very few exceptions1 none of the coral-knolls within the lagoons of ,peros ,banhos1 ,diego ,garcia1 and the ,great ,chagos ,bank 7all situated in the ,chagos group71 rise to the surface of the water2 whereas all those1 with equally few exceptions1 within ,solomon and ,egmont atolls in the same group1 and likewise within the large southern ,maldiva atolls1 reach the surface4 ,i make these statements1 after having examined the charts of each atoll4 ,in the lagoon of ,peros ,banhos1 which is nearly twenty miles across1 there is only one single reef which rises to the surface2 in ,diego ,garcia there are seven1 but several of these lie close to the margin of the lagoon1 and need scarcely have been reckoned2 in the ,great ,chagos ,bank there is not one4 ,on the other hand1 in the lagoons of some of the great southern ,maldiva atolls1 although thickly studded with reefs1 every one without exception rises to the surface2 and on an average there are less than two submerged reefs in each atoll2 in the northern atolls1 however1 the submerged lagoon-reefs are not quite so rare4 ,the submerged reefs in the ,chagos atolls generally have from one to seven fathoms water on them1 but some have from seven to ten4 ,most of them are small with very steep sides2#ae at ,peros ,banhos they rise from a depth of about thirty fathoms1 and some of them in the ,great ,chagos ,bank from above forty fathoms2 they are covered1 ,captain ,moresby informs me1 with living and healthy coral1 two and three feet high1 consisting of several species4 ,why then have not these lagoon-reefs reached the surface1 like the innumerable ones in the atolls above named8 ,if we attempt to assign any difference in their external conditions1 as the cause of this diversity1 we are at once baffled4 ,the lagoon of ,diego ,garcia is not deep1 and is almost wholly surrounded by its reef2 ,peros ,banhos is very deep1 much larger1 with many wide passages communicating with the open sea4 ,on the other hand1 of those atolls1 in which all or nearly all the lagoon-reefs have reached the surface1 some are small1 others large1 some shallow1 others deep1 some well-enclosed1 and others open4 #ae ,some of these statements were not communicated to me verbally by ,captain ,moresby1 but are taken from the ,,ms4 account before alluded to1 of the ,chagos ,group4 ,captain ,moresby informs me that he has seen a ,french chart of ,diego ,garcia made eighty years before his survey1 and apparently very accurate2 and from it he infers1 that during this interval there has not been the smallest change in the depth on any of the knolls within the lagoon4 ,it is also known that during the last fifty-one years1 the eastern channel into the lagoon has neither become narrower1 nor decreased in depth2 and as there are numerous small knolls of living coral within it1 some change might have been anticipated4 ,moreover1 as the whole reef round the lagoon of this atoll has been converted into land--an unparalleled case1 ,i believe1 in an atoll of such large size1--and as the strip of land is for considerable spaces more than half a mile wide--also a very unusual circumstance1--we have the best possible evidence1 that ,diego ,garcia has remained at its present level for a very long period4 ,with this fact1 and with the knowledge that no sensible change has taken place during eighty years in the coral-knolls1 and considering that every single reef has reached the surface in other atolls1 which do not present the smallest appearance of being older than ,diego ,garcia and ,peros ,banhos1 and which are placed under the same external conditions with them1 one is led to conclude that these submerged reefs1 although covered with luxuriant coral1 have no tendency to grow upwards1 and that they would remain at their present levels for an almost indefinite period4 ,from the number of these knolls1 from their position1 size1 and form1 many of them being only one or two hundred yards across1 with a rounded outline1 and precipitous sides1--it is indisputable that they have been formed by the growth of coral2 and this makes the case much more remarkable4 ,in ,peros ,banhos and in the ,great ,chagos ,bank1 some of these almost columnar masses are #bjj feet high1 and their summits lie only from two to eight fathoms beneath the surface2 therefore1 a small proportional amount more of growth would cause them to attain the surface1 like those numerous knolls1 which rise from an equally great depth within the ,maldiva atolls4 ,we can hardly suppose that time has been wanting for the upward growth of the coral1 whilst in ,diego ,garcia1 the broad annular strip of land1 formed by the continued accumulation of detritus1 shows how long this atoll has remained at its present level4 ,we must look to some other cause than the rate of growth2 and ,i suspect it will be found in the reefs being formed of different species of corals1 adapted to live at different depths4 ,the ,great ,chagos ,bank is situated in the centre of the ,chagos ,group1 and the ,pitt and ,speaker ,banks at its two extreme points4 ,these banks resemble atolls1 except in their external rim being about eight fathoms submerged1 and in being formed of dead rock1 with very little living coral on it3 a portion nine miles long of the annular reef of ,peros ,banhos atoll is in the same condition4 ,these facts1 as will hereafter be shown1 render it very probable that the whole group at some former period subsided seven or eight fathoms2 and that the coral perished on the outer margin of those atolls which are now submerged1 but that it continued alive1 and grew up to the surface on those which are now perfect4 ,if these atolls did subside1 and if from the suddenness of the movement or from any other cause1 those corals which are better adapted to live at a certain depth than at the surface1 once got possession of the knolls1 supplanting the former occupants1 they would exert little or no tendency to grow upwards4 ,to illustrate this1 ,i may observe1 that if the corals of the upper zone on the outer edge of ,keeling atoll were to perish1 it is improbable that those of the lower zone would grow to the surface1 and thus become exposed to conditions for which they do not appear to be adapted4 ,the conjecture1 that the corals on the submerged knolls within the ,chagos atolls have analogous habits with those of the lower zone outside ,keeling atoll1 receives some support from a remark by ,captain ,moresby1 namely1 that they have a different appearance from those on the reefs in the ,maldiva atolls1 which1 as we have seen1 all rise to the surface3 he compares the kind of difference to that of the vegetation under different climates4 ,i have entered at considerable length into this case1 although unable to throw much light on it1 in order to show that an equal tendency to upward growth ought not to be attributed to all coral-reefs1--to those situated at different depths1--to those forming the ring of an atoll or those on the knolls within a lagoon1--to those in one area and those in another4 ,the inference1 therefore1 that one reef could not grow up to the surface within a given time1 because another1 not known to be covered with the same species of corals1 and not known to be placed under conditions exactly the same1 has not within the same time reached the surface1 is unsound4 ,section ,,ii--,,on ,,the ,,rate ,,of ,,growth ,,of ,,coral-,,reefs ,the remark made at the close of the last section1 naturally leads to this division of our subject1 which has not1 ,i think1 hitherto been considered under a right point of view4 ,ehrenberg#af has stated1 that in the ,red ,sea1 the corals only coat other rocks in a layer from one to two feet in thickness1 or at most to a fathom and a half2 and he disbelieves that1 in any case1 they form1 by their own proper growth1 great masses1 stratum over stratum4 ,a nearly similar observation has been made by ,,mm4 ,quoy and ,gaimard1#ag with respect to the thickness of some upraised beds of coral1 which they examined at ,timor and some other places4 ,ehrenberg#ah saw certain large massive corals in the ,red ,sea1 which he imagines to be of such vast antiquity1 that they might have been beheld by ,pharaoh2 and according to ,mr4 ,lyell#ai there are certain corals at ,bermuda1 which are known by tradition1 to have been living for centuries4 ,to show how slowly coral-reefs grow upwards1 ,captain ,beechey#bj has adduced the case of the ,dolphin ,reef off ,tahiti1 which has remained at the same depth beneath the surface1 namely about two fathoms and a half1 for a period of sixty-seven years4 ,there are reefs in the ,red ,sea1 which certainly do not appear#ba to have increased in dimensions during the last half-century1 and from the comparison of old charts with recent surveys1 probably not during the last two hundred years4 ,these1 and other similar facts1 have so strongly impressed many with the belief of the extreme slowness of the growth of corals1 that they have even doubted the possibility of islands in the great oceans having been formed by their agency4 ,others1 again1 who have not been overwhelmed by this difficulty1 have admitted that it would require thousands1 and tens of thousands of years1 to form a mass1 even of inconsiderable thickness2 but the subject has not1 ,i believe1 been viewed in the proper light4 #af ,ehrenberg1 as before cited1 pp4 #ci1 #df1 and #ej4 #ag 8,annales des ,sciences ,nat40 tom4 vi1 p4 #bh4 #ah ,ehrenberg1 ut sup41 p4 #db4 #ai ,lyell's 8,principles of ,geology10 book iii1 ch4 xviii4 #bj ,beechey's 8,voyage to the ,pacific10 ch4 viii4 #ba ,ehrenberg1 ut sup41 p4 #dc4 ,that masses of considerable thickness have been formed by the growth of coral1 may be inferred with certainty from the following facts4 ,in the deep lagoons of ,peros ,banhos and of the ,great ,chagos ,bank1 there are1 as already described1 small steep-sided knolls covered with living coral4 ,there are similar knolls in the southern ,maldiva atolls1 some of which1 as ,captain ,moresby assures me1 are less than a hundred yards in diameter1 and rise to the surface from a depth of between two hundred and fifty and three hundred feet4 ,considering their number1 form1 and position1 it would be preposterous to suppose that they are based on pinnacles of any rock1 not of coral formation2 or that sediment could have been heaped up into such small and steep isolated cones4 ,as no kind of living coral grows above the height of a few feet1 we are compelled to suppose that these knolls have been formed by the successive growth and death of many individuals1--first one being broken off or killed by some accident1 and then another1 and one set of species being replaced by another set with different habits1 as the reef rose nearer the surface1 or as other changes supervened4 ,the spaces between the corals would become filled up with fragments and sand1 and such matter would probably soon be consolidated1 for we learn from ,lieutenant ,nelson1#bb that at ,bermuda a process of this kind takes place beneath water1 without the aid of evaporation4 ,in reefs1 also1 of the barrier class1 we may feel sure1 as ,i have shown1 that masses of great thickness have been formed by the growth of the coral2 in the case of ,vanikoro1 judging only from the depth of the moat between the land and the reef1 the wall of coral-rock must be at least three hundred feet in vertical thickness4 #bb 8,geological ,transactions10 vol4 v1 p4 #aac4 ,it is unfortunate that the upraised coral-islands in the ,pacific have not been examined by a geologist4 ,the cliffs of ,elizabeth ,island1 in the ,low ,archipelago1 are eighty feet high1 and appear1 from ,captain ,beechey's description1 to consist of a homogeneous coral-rock4 ,from the isolated position of this island1 we may safely infer that it is an upraised atoll1 and therefore that it has been formed by masses of coral1 grown together4 ,savage ,island seems1 from the description of the younger ,forster1#bc to have a similar structure1 and its shores are about forty feet high3 some of the ,cook ,islands also appear#bd to be similarly composed4 ,captain ,belcher1 ,r4,n41 in a letter which ,captain ,beaufort showed me at the admiralty1 speaking of ,bow atoll1 says1 8,i have succeeded in boring forty-five feet through coral-sand1 when the auger became jammed by the falling in of the surrounding creamy matter40 ,on one of the ,maldiva atolls1 ,captain ,moresby bored to a depth of twenty-six feet1 when his auger also broke3 he has had the kindness to give me the matter brought up2 it is perfectly white1 and like finely triturated coral-rock4 #bc ,forster's 8,voyage round the ,world with ,cook10 vol4 ii1 pp4 #afc1 #afg4 #bd ,williams's 8,narrative of ,missionary ,enterprise10 p4 #cj4 ,in my description of ,keeling atoll1 ,i have given some facts1 which show that the reef probably has grown outwards2 and ,i have found1 just within the outer margin1 the great mounds of ,porites and of ,millepora1 with their summits lately killed1 and their sides subsequently thickened by the growth of the coral3 a layer1 also1 of ,nullipora had already coated the dead surface4 ,as the external slope of the reef is the same round the whole of this atoll1 and round many other atolls1 the angle of inclination must result from an adaption between the growing powers of the coral1 and the force of the breakers1 and their action on the loose sediment4 ,the reef1 therefore1 could not increase outwards1 without a nearly equal addition to every part of the slope1 so that the original inclination might be preserved1 and this would require a large amount of sediment1 all derived from the wear of corals and shells1 to be added to the lower part4 ,moreover1 at ,keeling atoll1 and probably in many other cases1 the different kinds of corals would have to encroach on each other2 thus the ,nullipor cannot increase outwards without encroaching on the ,porites and ,millepora complanata1 as is now taking place2 nor these latter without encroaching on the strongly branched ,madreporet1 the ,millepora alcicornis1 and some ,astras2 nor these again without a foundation being formed for them within the requisite depth1 by the accumulation of sediment4 ,how slow1 then1 must be the ordinary lateral or outward growth of such reefs4 ,but off ,christmas atoll1 where the sea is much more shallow than is usual1 we have good reason to believe that1 within a period not very remote1 the reef has increased considerably in width4 ,the land has the extraordinary breadth of three miles2 it consists of parallel ridges of shells and broken corals1 which furnish 8an incontestable proof10 as observed by ,cook1#be 8that the island has been produced by accessions from the sea1 and is in a state of increase40 ,the land is fronted by a coral-reef1 and from the manner in which islets are known to be formed1 we may feel confident that the reef was not three miles wide1 when the first1 or most backward ridge1 was thrown up2 and1 therefore1 we must conclude that the reef has grown outwards during the accumulation of the successive ridges4 ,here then1 a wall of coral-rock of very considerable breadth has been formed by the outward growth of the living margin1 within a period during which ridges of shells and corals1 lying on the bare surface1 have not decayed4 ,there can be little doubt1 from the account given by ,captain ,beechey1 that ,matilda atoll1 in the ,low ,archipelago1 has been converted in the space of thirty-four years1 from being1 as described by the crew of a wrecked whaling vessel1 a 8reef of rocks0 into a lagoon-island1 fourteen miles in length1 with 8one of its sides covered nearly the whole way with high trees40#bf ,the islets1 also1 on ,keeling atoll1 it has been shown1 have increased in length1 and since the construction of an old chart1 several of them have become united into one long islet2 but in this case1 and in that of ,matilda atoll1 we have no proof1 and can only infer as probable1 that the reef1 that is the foundation of the islets1 has increased as well as the islets themselves4 #be ,cook's 8,third ,voyage10 book ,,iii1 ch4 x4 #bf ,beechey's 8,voyage to the ,pacific10 ch4 vii and viii4 ,after these considerations1 ,i attach little importance1 as indicating the ordinary and still less the possible rate of outward growth of coral-reefs1 to the fact that certain reefs in the ,red ,sea have not increased during a long interval of time2 or to other such cases1 as that of ,ouluthy atoll in the ,caroline group1 where every islet1 described a thousand years before by ,cantova was found in the same state by ,lutk1#bg--without it could be shown that1 in these cases1 the conditions were favourable to the vigorous and unopposed growth of the corals living in the different zones of depth1 and that a proper basis for the extent of the reef was present4 ,the former conditions must depend on many contingencies1 and in the deep oceans where coral formations most abound1 a basis within the requisite depth can rarely be present4 #bg ,f4 ,lutk's 8,voyage autour du ,monde40 ,in the group ,elato1 however1 it appears that what is now the islet ,falipi1 is called in ,cantova's ,chart1 the ,banc de ,falipi4 ,it is not stated whether this has been caused by the growth of coral1 or by the accumulation of sand4 ,nor do ,i attach any importance to the fact of certain submerged reefs1 as those off ,tahiti1 or those within ,diego ,garcia not now being nearer the surface than they were many years ago1 as an indication of the rate under favourable circumstances of the upward growth of reefs2 after it has been shown1 that all the reefs have grown to the surface in some of the ,chagos atolls1 but that in neighbouring atolls which appear to be of equal antiquity and to be exposed to the same external conditions1 every reef remains submerged2 for we are almost driven to attribute this to a difference1 not in the rate of growth1 but in the habits of the corals in the two cases4 ,in an old-standing reef1 the corals1 which are so different in kind on different parts of it1 are probably all adapted to the stations they occupy1 and hold their places1 like other organic beings1 by a struggle one with another1 and with external nature2 hence we may infer that their growth would generally be slow1 except under peculiarly favourable circumstances4 ,almost the only natural condition1 allowing a quick upward growth of the whole surface of a reef1 would be a slow subsidence of the area in which it stood2 if1 for instance1 ,keeling atoll were to subside two or three feet1 can we doubt that the projecting margin of live coral1 about half an inch in thickness1 which surrounds the dead upper surfaces of the mounds of ,porites1 would in this case form a concentric layer over them1 and the reef thus increase upwards1 instead of1 as at present1 outwards8 ,the ,nullipor are now encroaching on the ,porites and ,millepora1 but in this case might we not confidently expect that the latter would1 in their turn1 encroach on the ,nullipor8 ,after a subsidence of this kind1 the sea would gain on the islets1 and the great fields of dead but upright corals in the lagoon1 would be covered by a sheet of clear water2 and might we not then expect that these reefs would rise to the surface1 as they anciently did when the lagoon was less confined by islets1 and as they did within a period of ten years in the schooner-channel1 cut by the inhabitants8 ,in one of the ,maldiva atolls1 a reef1 which within a very few years existed as an islet bearing cocoa-nut trees1 was found by ,lieutenant ,prentice 8entirely covered with live coral and ,madrepore40 ,the natives believe that the islet was washed away by a change in the currents1 but if1 instead of this1 it had quietly subsided1 surely every part of the island which offered a solid foundation1 would in a like manner have become coated with living coral4 ,through steps such as these1 any thickness of rock1 composed of a singular intermixture of various kinds of corals1 shells1 and calcareous sediment1 might be formed2 but without subsidence1 the thickness would necessarily be determined by the depth at which the reef-building polypifers can exist4 ,if it be asked1 at what rate in years ,i suppose a reef of coral favourably circumstanced could grow up from a given depth2 ,i should answer1 that we have no precise evidence on this point1 and comparatively little concern with it4 ,we see1 in innumerable points over wide areas1 that the rate has been sufficient1 either to bring up the reefs from various depths to the surface1 or1 as is more probable1 to keep them at the surface1 during progressive subsidences2 and this is a much more important standard of comparison than any cycle of years4 ,it may1 however1 be inferred from the following facts1 that the rate in years under favourable circumstances would be very far from slow4 ,dr4 ,allan1 of ,forres1 has1 in his ,,ms4 ,thesis deposited in the library of the ,edinburgh ,university 7extracts from which ,i owe to the kindness of ,dr4 ,malcolmson71 the following account of some experiments1 which he tried during his travels in the years #ahcj to #ahcb on the east coast of ,madagascar4 8,to ascertain the rise and progress of the coral-family1 and fix the number of species met with at ,foul ,point 7latitude #ag #dj7 twenty species of coral were taken off the reef and planted apart on a sand-bank three feet deep at low water4 ,each portion weighed ten pounds1 and was kept in its place by stakes4 ,similar quantities were placed in a clump and secured as the rest4 ,this was done in ,december #ahcj4 ,in ,july following1 each detached mass was nearly level with the sea at low water1 quite immovable1 and several feet long1 stretching as the parent reef1 with the coast current from north to south4 ,the masses accumulated in a clump were found equally increased1 but some of the species in such unequal ratios1 as to be growing over each other40 ,the loss of ,dr4 ,allan's magnificent collection by shipwreck1 unfortunately prevents its being known to what genera these corals belonged2 but from the numbers experimented on1 it is certain that all the more conspicuous kinds must have been included4 ,dr4 ,allan informs me1 in a letter1 that he believes it was a ,madrepora1 which grew most vigorously4 ,one may be permitted to suspect that the level of the sea might possibly have been somewhat different at the two stated periods2 nevertheless1 it is quite evident that the growth of the ten-pound masses1 during the six or seven months1 at the end of which they were found immovably fixed#bh and several feet in length1 must have been very great4 ,the fact of the different kinds of coral1 when placed in one clump1 having increased in extremely unequal ratios1 is very interesting1 as it shows the manner in which a reef1 supporting many species of coral1 would probably be affected by a change in the external conditions favouring one kind more than another4 ,the growth of the masses of coral in ,n4 and ,s4 lines parallel to the prevailing currents1 whether due to the drifting of sediment or to the simple movement of the water1 is1 also1 a very interesting circumstance4 #bh ,it is stated by ,de la ,beche 78,geological ,manual10 p4 #adc71 on the authority of ,mr4 ,lloyd1 who surveyed the ,isthmus of ,panama1 that some specimens of ,polypifers1 placed by him in a sheltered pool of water1 were found in the course of a few days firmly fixed by the secretion of a stony matter1 to the bottom4 ,a fact1 communicated to me by ,lieutenant ,wellstead1 ,i4,n41 in some degree corroborates the result of ,dr4 ,allan's experiments3 it is1 that in the ,persian ,gulf a ship had her copper bottom encrusted in the course of twenty months with a layer of coral1 two feet in thickness1 which it required great force to remove1 when the vessel was docked3 it was not ascertained to what order this coral belonged4 ,the case of the schooner-channel choked up with coral in an interval of less than ten years1 in the lagoon of ,keeling atoll1 should be here borne in mind4 ,we may also infer1 from the trouble which the inhabitants of the ,maldiva atolls take to root out1 as they express it1 the coral-knolls from their harbours1 that their growth can hardly be very slow4#bi #bi ,mr4 ,stutchbury 7,west of ,england ,journal1 ,no4 ,i1 p4 #ej47 has described a specimen of ,agaricia1 8weighing #b lbs4 #i oz41 which surrounds a species of oyster1 whose age could not be more than two years1 and yet is completely enveloped by this dense coral40 ,i presume that the oyster was living when the specimen was procured2 otherwise the fact tells nothing4 ,mr4 ,stutchbury also mentions an anchor1 which had become entirely encrusted with coral in fifty years2 other cases1 however1 are recorded of anchors which have long remained amidst coral-reefs without having become coated4 ,the anchor of the ,beagle1 in #ahcb1 after having been down exactly one month at ,rio de ,janeiro1 was so thickly coated by two species of ,tubularia1 that large spaces of the iron were entirely concealed2 the tufts of this horny zoophyte were between two and three inches in length4 ,it has been attempted to compute1 but ,i believe erroneously1 the rate of growth of a reef1 from the fact mentioned by ,captain ,beechey1 of the ,chama gigas being embedded in coral-rock4 ,but it should be remembered1 that some species of this genus invariably live1 both whilst young and old1 in cavities1 which the animal has the power of enlarging with its growth4 ,i saw many of these shells thus embedded in the outer 8flat0 of ,keeling atoll1 which is composed of dead rock2 and therefore the cavities in this case had no relation whatever with the growth of coral4 ,m4 ,lesson1 also1 speaking of this shell 7,partie ,zoolog4 8,voyage de la ,coquille071 has remarked1 8que constamment ses valves taient engags compltement dans la masse des ,madrepores40 ,from the facts given in this section1 it may be concluded1 first1 that considerable thicknesses of rock have certainly been formed within the present geological area by the growth of coral and the accumulation of its detritus2 and1 secondly1 that the increase of individual corals and of reefs1 both outwards or horizontally and upwards or vertically1 under the peculiar conditions favourable to such increase1 is not slow1 when referred either to the standard of the average oscillations of level in the earth's crust1 or to the more precise but less important one of a cycle of years4 ,section ,,iii--,,on ,,the ,,depths ,,at ,,which ,,reef-,,building ,,polypifers ,,can ,,live ,i have already described in detail1 which might have appeared trivial1 the nature of the bottom of the sea immediately surrounding ,keeling atoll2 and ,i will now describe with almost equal care the soundings off the fringing-reefs of ,mauritius4 ,i have preferred this arrangement1 for the sake of grouping together facts of a similar nature4 ,i sounded with the wide bell-shaped lead which ,captain ,fitzroy used at ,keeling ,island1 but my examination of the bottom was confined to a few miles of coast 7between ,port ,louis and ,tomb ,bay7 on the leeward side of the island4 ,the edge of the reef is formed of great shapeless masses of branching ,madrepores1 which chiefly consist of two species1--apparently ,m4 corymbosa and pocillifera1--mingled with a few other kinds of coral4 ,these masses are separated from each other by the most irregular gullies and cavities1 into which the lead sinks many feet4 ,outside this irregular border of ,madrepores1 the water deepens gradually to twenty fathoms1 which depth generally is found at the distance of from half to three-quarters of a mile from the reef4 ,a little further out the depth is thirty fathoms1 and thence the bank slopes rapidly into the depths of the ocean4 ,this inclination is very gentle compared with that outside ,keeling and other atolls1 but compared with most coasts it is steep4 ,the water was so clear outside the reef1 that ,i could distinguish every object forming the rugged bottom4 ,in this part1 and to a depth of eight fathoms1 ,i sounded repeatedly1 and at each cast pounded the bottom with the broad lead1 nevertheless the arming invariably came up perfectly clean1 but deeply indented4 ,from eight to fifteen fathoms a little calcareous sand was occasionally brought up1 but more frequently the arming was simply indented4 ,in all this space the two ,madrepores above mentioned1 and two species of ,astra1 with rather large#cj stars1 seemed the commonest kinds2 and it must be noticed that twice at the depth of fifteen fathoms1 the arming was marked with a clean impression of an ,astra4 ,besides these lithophytes1 some fragments of the ,millepora alcicornis1 which occurs in the same relative position at ,keeling ,island1 were brought up2 and in the deeper parts there were large beds of a ,seriatopora1 different from ,s4 subulata1 but closely allied to it4 ,on the beach within the reef1 the rolled fragments consisted chiefly of the corals just mentioned1 and of a massive ,porites1 like that at ,keeling atoll1 of a ,meandrina1 ,pocillopora verrucosa1 and of numerous fragments of ,nullipora4 ,from fifteen to twenty fathoms the bottom was1 with few exceptions1 either formed of sand1 or thickly covered with ,seriatopora3 this delicate coral seems to form at these depths extensive beds unmingled with any other kind4 ,at twenty fathoms1 one sounding brought up a fragment of ,madrepora apparently ,m4 pocillifera1 and ,i believe it is the same species 7for ,i neglected to bring specimens from both stations7 which mainly forms the upper margin of the reef2 if so1 it grows in depths varying from #j to #bj fathoms4 ,between #bj and #bc fathoms ,i obtained several soundings1 and they all showed a sandy bottom1 with one exception at #cj fathoms1 when the arming came up scooped out1 as if by the margin of a large ,caryophyllia4 ,beyond #cc fathoms ,i sounded only once2 and from #hf fathoms1 at the distance of one mile and a third from the edge of the reef1 the arming brought up calcareous sand with a pebble of volcanic rock4 ,the circumstance of the arming having invariably come up quite clean1 when sounding within a certain number of fathoms off the reefs of ,mauritius and ,keeling atoll 7eight fathoms in the former case1 and twelve in the latter7 and of its having always come up 7with one exception7 smoothed and covered with sand1 when the depth exceeded twenty fathoms1 probably indicates a criterion1 by which the limits of the vigorous growth of coral might in all cases be readily ascertained4 ,i do not1 however1 suppose that if a vast number of soundings were obtained round these islands1 the limit above assigned would be found never to vary1 but ,i conceive the facts are sufficient to show1 that the exceptions would be few4 ,the circumstance of a gradual change1 in the two cases1 from a field of clean coral to a smooth sandy bottom1 is far more important in indicating the depth at which the larger kinds of coral flourish than almost any number of separate observations on the depth1 at which certain species have been dredged up4 ,for we can understand the gradation1 only as a prolonged struggle against unfavourable conditions4 ,if a person were to find the soil clothed with turf on the banks of a stream of water1 but on going to some distance on one side of it1 he observed the blades of grass growing thinner and thinner1 with intervening patches of sand1 until he entered a desert of sand1 he would safely conclude1 especially if changes of the same kind were noticed in other places1 that the presence of the water was absolutely necessary to the formation of a thick bed of turf3 so may we conclude1 with the same feeling of certainty1 that thick beds of coral are formed only at small depths beneath the surface of the sea4 #cj ,since the preceding pages were printed off1 ,i have received from ,mr4 ,lyell a very interesting pamphlet1 entitled 8,remarks upon ,coral ,formations10 etc41 by ,j4 ,couthouy1 ,boston1 ,united ,states1 #ahdb4 ,there is a statement 7p4 #f71 on the authority of the ,rev4 ,j4 ,williams1 corroborating the remarks made by ,ehrenberg and ,lyell 7p4 #ga of this volume71 on the antiquity of certain individual corals in the ,red ,sea and at ,bermuda2 namely1 that at ,upolu1 one of the ,navigator ,islands1 8particular clumps of coral are known to the fishermen by name1 derived from either some particular configuration or tradition attached to them1 and handed down from time immemorial40 ,with respect to the thickness of masses of coral-rock1 it clearly appears1 from the descriptions given by ,mr4 ,couthouy 7pp4 #cd1 #eh7 that ,mangaia and ,aurora ,islands are upraised atolls1 composed of coral rock3 the level summit of the former is about three hundred feet1 and that of ,aurora ,island is two hundred feet above the sea-level4 ,i have endeavoured to collect every fact1 which might either invalidate or corroborate this conclusion4 ,captain ,moresby1 whose opportunities for observation during his survey of the ,maldiva and ,chagos ,archipelagoes have been unrivalled1 informs me1 that the upper part or zone of the steep-sided reefs1 on the inner and outer coasts of the atolls in both groups1 invariably consists of coral1 and the lower parts of sand4 ,at seven or eight fathoms depth1 the bottom is formed1 as could be seen through the clear water1 of great living masses of coral1 which at about ten fathoms generally stand some way apart from each other1 with patches of white sand between them1 and at a little greater depth these patches become united into a smooth steep slope1 without any coral4 ,captain ,moresby1 also1 informs me in support of his statement1 that he found only decayed coral on the ,padua ,bank 7northern part of the ,laccadive group7 which has an average depth between twenty-five and thirty-five fathoms1 but that on some other banks in the same group with only ten or twelve fathoms water on them 7for instance1 the ,tillacapeni bank71 the coral was living4 ,with regard to the coral-reefs in the ,red ,sea1 ,ehrenberg has the following passage3--8,the living corals do not descend there into great depths4 ,on the edges of islets and near reefs1 where the depth was small1 very many lived2 but we found no more even at six fathoms4 ,the pearl-fishers at ,yemen and ,massaua asserted that there was no coral near the pearl-banks at nine fathoms depth1 but only sand4 ,we were not able to institute any more special researches40#ca ,i am1 however1 assured both by ,captain ,moresby and ,lieutenant ,wellstead1 that in the more northern parts of the ,red ,sea1 there are extensive beds of living coral at a depth of twenty-five fathoms1 in which the anchors of their vessels were frequently entangled4 ,captain ,moresby attributes the less depth1 at which the corals are able to live in the places mentioned by ,ehrenberg1 to the greater quantity of sediment there2 and the situations1 where they were flourishing at the depth of twenty-five fathoms1 were protected1 and the water was extraordinarily limpid4 ,on the leeward side of ,mauritius where ,i found the coral growing at a somewhat greater depth than at ,keeling atoll1 the sea1 owing apparently to its tranquil state1 was likewise very clear4 ,within the lagoons of some of the ,marshall atolls1 where the water can be but little agitated1 there are1 according to ,kotzebue1 living beds of coral in twenty-five fathoms4 ,from these facts1 and considering the manner in which the beds of clean coral off ,mauritius1 ,keeling ,island1 the ,maldiva and ,chagos atolls1 graduated into a sandy slope1 it appears very probable that the depth1 at which reef-building polypifers can exist1 is partly determined by the extent of inclined surface1 which the currents of the sea and the recoiling waves have the power to keep free from sediment4 #ca ,ehrenberg1 8ber die ,natur10 etc41 p4 #ej4 ,,mm4 ,quoy and ,gaimard#cb believe that the growth of coral is confined within very limited depths2 and they state that they never found any fragment of an ,astra 7the genus they consider most efficient in forming reefs7 at a depth above twenty-five or thirty feet4 ,but we have seen that in several places the bottom of the sea is paved with massive corals at more than twice this depth2 and at fifteen fathoms 7or twice this depth7 off the reefs of ,mauritius1 the arming was marked with the distinct impression of a living ,astra4 ,millepora alcicornis lives in from #j to #ab fathoms1 and the genera ,madrepora and ,seriatopora from #j to #bj fathoms4 ,captain ,moresby has given me a specimen of ,sideropora scabra 7,porites of ,lamarck7 brought up alive from #ag fathoms4 ,mr4 ,couthouy#cc states that he has dredged up on the ,bahama banks considerable masses of ,meandrina from #af fathoms1 and he has seen this coral growing in #bj fathoms4 ,a ,caryophyllia1 half an inch in diameter1 was dredged up alive from #hj fathoms off ,juan ,fernandez 7latitude #cc ,s47 by ,captain ,p4 ,p4 ,king3#cd this is the most remarkable fact with which ,i am acquainted1 showing the depth at which a genus of corals often found on reefs1 can exist4#ce ,we ought1 however1 to feel less surprise at this fact1 as ,caryophyllia alone of the lamelliform genera1 ranges far beyond the tropics2 it is found in ,zetland#cf in ,lat4 #fj ,n4 in deep water1 and ,i procured a small species from ,tierra del ,fuego in ,lat4 #ec ,s4 ,captain ,beechey informs me1 that branches of pink and yellow coral were frequently brought up from between twenty and twenty-five fathoms off the ,low atolls2 and ,lieutenant ,stokes1 writing to me from the ,n4,w4 coast of ,australia1 says that a strongly branched coral was procured there from thirty fathoms2 unfortunately it is not known to what genera these corals belong4 #cb 8,annales des ,sci4 ,nat40 tom4 vi4 #cc 8,remarks on ,coral ,formations10 p4 #ab4 #cd ,i am indebted to ,mr4 ,stokes for having kindly communicated this fact to me1 together with much other valuable information4 #ce ,i will record in the form of a note all the facts that ,i have been able to collect on the depths1 both within and without the tropics1 at which those corals and corallines can live1 which there is no reason to suppose ever materially aid in the construction of a reef4 ,ellis 78,nat4 ,hist4 of ,coralline10 p4 #if7 states that ,ombellularia was procured in latitude #gi ,n4 sticking to a line from the depth of #bcf fathoms2 hence this coral either must have been floating loose1 or was entangled in stray line at the bottom4 ,off ,keeling atoll a compound ,ascidia 7,sigillina7 was brought up from #ci fathoms1 and a piece of sponge1 apparently living1 from #gj1 and a fragment of ,nullipora also apparently living from #ib fathoms4 ,at a greater depth than #ij fathoms off this coral island1 the bottom was thickly strewed with joints of ,halimeda and small fragments of other ,nullipor1 but all dead4 ,captain ,b4 ,allen1 ,r4,n41 informs me that in the survey of the ,west ,indies it was noticed that between the depth of #aj and #bjj fathoms1 the sounding lead very generally came up coated with the dead joints of a ,halimeda1 of which he showed me specimens4 ,off ,pernambuco1 in ,brazil1 in about twelve fathoms1 the bottom was covered with fragments dead and alive of a dull red ,nullipora1 and ,i infer from ,roussin's chart1 that a bottom of this kind extends over a wide area4 ,on the beach1 within the coral-reefs of ,mauritius1 vast quantities of fragments of ,nullipor were piled up4 ,from these facts it appears1 that these simply organized bodies are amongst the most abundant productions of the sea4 #cf ,fleming's 8,british ,animals10 genus ,caryophyllia4 ,name of ,zoophyte ,depth in ,fathoms ,country and ,s4 ,latitude ,authority ,sertularia #dj ,cape ,horn #ff 7,where none is given1 the observation is my own47 ,cellaria ,ditto ,ditto ,cellaria4 ,a minute scarlet encrusted species1 found living #aij ,keeling ,atoll #ab ,cellaria4 ,an allied1 small stony sub-generic form #dh ,s4 ,cruz ,river #ej ,a coral allied to ,vincularia1 with eight rows of cells #dj ,cape ,horn ,tubulipora1 near to ,t4 patima ,ditto ,ditto ,ditto #id ,east ,chiloe #dc ,cellepora1 several species1 and allied sub-generic forms #dj ,cape ,horn ,ditto #dj and #eg ,chonos ,arch4 #de ,ditto #dh ,s4 ,cruz #ej ,eschara #cj ,tierra del ,fuego #ec ,ditto #dh ,s4 ,cruz ,r4 #ej ,retepora #dj ,cape ,horn ,ditto #ajj ,c4 ,good ,hope #cd ,quoy and ,gaimard1 ,ann4 ,scien4 ,nat41 t4 vi1 p4 #bhd4 ,millepora1 a strong coral with cylindrical branches1 of a pink colour1 abut two inches high1 resembling in the form of its orifices ,m4 aspera of ,lamarck #id and #cj ,e4 ,chiloe #dc ,tierra del ,fuego #ec ,coralium #abj ,barbary #cc ,n4 ,peyssonel in paper read to ,royal society ,may #ageb4 ,antipathes #af ,chonos #de ,gorgonia 7or an allied form7 #afj ,abrolhos on the coast of ,brazil #ah ,capt4 ,beechey informed me of this fact in a letter4 ,although the limit of depth1 at which each particular kind of coral ceases to exist1 is far from being accurately known2 yet when we bear in mind the manner in which the clumps of coral gradually became infrequent at about the same depth1 and wholly disappeared at a greater depth than twenty fathoms1 on the slope round ,keeling atoll1 on the leeward side of the ,mauritius1 and at rather less depth1 both without and within the atolls of the ,maldiva and ,chagos ,archipelagoes2 and when we know that the reefs round these islands do not differ from other coral formations in their form and structure1 we may1 ,i think1 conclude that in ordinary cases1 reef- building polypifers do not flourish at greater depths than between twenty and thirty fathoms4 ,it has been argued#cg that reefs may possibly rise from very great depths through the means of small corals1 first making a platform for the growth of the stronger kinds4 ,this1 however1 is an arbitrary supposition3 it is not always remembered1 that in such cases there is an antagonist power in action1 namely1 the decay of organic bodies1 when not protected by a covering of sediment1 or by their own rapid growth4 ,we have1 moreover1 no right to calculate on unlimited time for the accumulation of small organic bodies into great masses4 ,every fact in geology proclaims that neither the land1 nor the bed of the sea retain for indefinite periods the same level4 ,as well might it be imagined that the ,british ,seas would in time become choked up with beds of oysters1 or that the numerous small corallines off the inhospitable shores of ,tierra del ,fuego would in time form a solid and extensive coral-reef4 #cg ,journal of the ,royal ,geographical ,society1 #ahca1 p4 #bah4 ,chapter ,v ,,theory ,,of ,,the ,,formation ,,of ,,the ,,different ,,classes ,,of ,,coral-,,reefs ,the atolls of the larger archipelagoes are not formed on submerged craters1 or on banks of sediment4--,immense areas interspersed with atolls4--,their subsidence4--,the effects of storms and earthquakes on atolls4--,recent changes in their state4--,the origin of barrier-reefs and of atolls4--,their relative forms4--,the step-formed ledges and walls round the shores of some lagoons4--,the ring-formed reefs of the ,maldiva atolls4--,the submerged condition of parts or of the whole of some annular reefs4--,the disseverment of large atolls4--,the union of atolls by linear reefs4--,the ,great ,chagos ,bank4--,objections from the area and amount of subsidence required by the theory1 considered4--,the probable composition of the lower parts of atolls4 ,the naturalists who have visited the ,pacific1 seem to have had their attention riveted by the lagoon-islands1 or atolls1--those singular rings of coral-land which rise abruptly out of the unfathomable ocean--and have passed over1 almost unnoticed1 the scarcely less wonderful encircling barrier-reefs4 ,the theory most generally received on the formation of atolls1 is that they are based on submarine craters2 but where can we find a crater of the shape of ,bow atoll1 which is five times as long as it is broad 7,plate ,i1 ,fig4 #d72 or like that of ,menchikoff ,island 7,plate ,,ii1 ,fig4 #c71 with its three loops1 together sixty miles in length2 or like ,rimsky ,korsacoff1 narrow1 crooked1 and fifty-four miles long2 or like the northern ,maldiva atolls1 made up of numerous ring-formed reefs1 placed on the margin of a disc1--one of which discs is eighty-eight miles in length1 and only from ten to twenty in breadth8 ,it is1 also1 not a little improbable1 that there should have existed as many craters of immense size crowded together beneath the sea1 as there are now in some parts atolls4 ,but this theory lies under a greater difficulty1 as will be evident1 when we consider on what foundations the atolls of the larger archipelagoes rest3 nevertheless1 if the rim of a crater afforded a basis at the proper depth1 ,i am far from denying that a reef like a perfectly characterised atoll might not be formed2 some such1 perhaps1 now exist2 but ,i cannot believe in the possibility of the greater number having thus originated4 ,an earlier and better theory was proposed by ,chamisso2#a he supposes that as the more massive kinds of corals prefer the surf1 the outer portions1 in a reef rising from a submarine basis1 would first reach the surface and consequently form a ring4 ,but on this view it must be assumed1 that in every case the basis consists of a flat bank2 for if it were conically formed1 like a mountainous mass1 we can see no reason why the coral should spring up from the flanks1 instead of from the central and highest parts3 considering the number of the atolls in the ,pacific and ,indian ,oceans1 this assumption is very improbable4 ,as the lagoons of atolls are sometimes even more than forty fathoms deep1 it must1 also1 be assumed on this view1 that at a depth at which the waves do not break1 the coral grows more vigorously on the edges of a bank than on its central part2 and this is an assumption without any evidence in support of it4 ,i remarked1 in the third chapter1 that a reef1 growing on a detached bank1 would tend to assume an atoll-like structure2 if1 therefore1 corals were to grow up from a bank1 with a level surface some fathoms submerged1 having steep sides and being situated in a deep sea1 a reef not to be distinguished from an atoll1 might be formed3 ,i believe some such exist in the ,west ,indies4 ,but a difficulty of the same kind with that affecting the crater theory1 runners1 as we shall presently see1 this view inapplicable to the greater number of atolls4 #a ,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #cca4 ,no theory worthy of notice has been advanced to account for those barrier-reefs1 which encircle islands of moderate dimensions4 ,the great reef which fronts the coast of ,australia has been supposed1 but without any special facts1 to rest on the edge of a submarine precipice1 extending parallel to the shore4 ,the origin of the third class or of fringing-reefs presents1 ,i believe1 scarcely any difficulty1 and is simply consequent on the polypifers not growing up from great depths1 and their not flourishing close to gently shelving beaches where the water is often turbid4 ,what cause1 then1 has given to atolls and barrier-reefs their characteristic forms8 ,let us see whether an important deduction will not follow from the consideration of these two circumstances1 first1 the reef-building corals flourishing only at limited depths2 and secondly1 the vastness of the areas interspersed with coral-reefs and coral-islets1 none of which rise to a greater height above the level of the sea1 than that attained by matter thrown up by the waves and winds4 ,i do not make this latter statement vaguely2 ,i have carefully sought for descriptions of every island in the intertropical seas2 and my task has been in some degree abridged by a map of the ,pacific1 corrected in #ahcd by ,,mm4 ,d',urville and ,lottin1 in which the low islands are distinguished from the high ones 7even from those much less than a hundred feet in height7 by being written without a capital letter2 ,i have detected a few errors in this map1 respecting the height of some of the islands1 which will be noticed in the ,appendix1 where ,i treat of coral formations in geographical order4 ,to the ,appendix1 also1 ,i must refer for a more particular account of the data on which the statements on the next page are grounded4 ,i have ascertained1 and chiefly from the writings of ,cook1 ,kotzebue1 ,bellinghausen1 ,duperrey1 ,beechey1 and ,lutk1 regarding the ,pacific2 and from ,moresby#b with respect to the ,indian ,ocean1 that in the following cases the term 8low island0 strictly means land of the height commonly attained by matter thrown up by the winds and the waves of an open sea4 ,if we draw a line 7the plan ,i have always adopted7 joining the external atolls of that part of the ,low ,archipelago in which the islands are numerous1 the figure will be a pointed ellipse 7reaching from ,hood to ,lazaref ,island71 of which the longer axis is #hdj geographical miles1 and the shorter #dbj miles2 in this space#c none of the innumerable islets united into great rings rise above the stated level4 ,the ,gilbert group is very narrow1 and #cjj miles in length4 ,in a prolonged line from this group1 at the distance of #bdj miles1 is the ,marshall ,archipelago1 the figure of which is an irregular square1 one end being broader than the other2 its length is #ebj miles1 with an average width of #bdj2 these two groups together are #a1jdj miles in length1 and all their islets are low4 ,between the southern end of the ,gilbert and the northern end of ,low ,archipelago1 the ocean is thinly strewed with islands1 all of which1 as far as ,i have been able to ascertain1 are low2 so that from nearly the southern end of the ,low ,archipelago1 to the northern end of the ,marshall ,archipelago1 there is a narrow band of ocean1 more than #d1jjj miles in length1 containing a great number of islands1 all of which are low4 ,in the western part of the ,caroline ,archipelago1 there is a space of #dhj miles in length1 and about #ajj broad1 thinly interspersed with low islands4 ,lastly1 in the ,indian ,ocean1 the archipelago of the ,maldivas is #dgj miles in length1 and #fj in breadth2 that of the ,laccadives is #aej by #ajj miles2 as there is a low island between these two groups1 they may be considered as one group of #a1jjj miles in length4 ,to this may be added the ,chagos group of low islands1 situated #bhj miles distant1 in a line prolonged from the southern extremity of the ,maldivas4 ,this group1 including the submerged banks1 is #agj miles in length and #hj in breadth4 ,so striking is the uniformity in direction of these three archipelagoes1 all the islands of which are low1 that ,captain ,moresby1 in one of his papers1 speaks of them as parts of one great chain1 nearly #a1ejj miles long4 ,i am1 then1 fully justified in repeating1 that enormous spaces1 both in the ,pacific and ,indian ,oceans1 are interspersed with islands1 of which not one rises above that height1 to which the waves and winds in an open sea can heap up matter4 ,on what foundations1 then1 have these reefs and islets of coral been constructed8 ,a foundation must originally have been present beneath each atoll at that limited depth1 which is indispensable for the first growth of the reef-building polypifers4 ,a conjecture will perhaps be hazarded1 that the requisite bases might have been afforded by the accumulation of great banks of sediment1 which owing to the action of superficial currents 7aided possibly by the undulatory movement of the sea7 did not quite reach the surface1--as actually appears to have been the case in some parts of the ,west ,indian ,sea4 ,but in the form and disposition of the groups of atolls1 there is nothing to countenance this notion2 and the assumption without any proof1 that a number of immense piles of sediment have been heaped on the floor of the great ,pacific and ,indian ,oceans1 in their central parts far remote from land1 and where the dark blue colour of the limpid water bespeaks its purity1 cannot for one moment be admitted4 #b ,see also ,captain ,owen's and ,lieutenant ,wood's papers in the ,geographical ,journal1 on the ,maldiva and ,laccadive ,archipelagoes4 ,these officers particularly refer to the lowness of the islets2 but ,i chiefly ground my assertion respecting these two groups1 and the ,chagos group1 from information communicated to me by ,captain ,moresby4 #c ,i find from ,mr4 ,couthouy's pamphlet 7p4 #eh7 that ,aurora ,island is about two hundred feet in height2 it consists of coral-rock1 and seems to have been formed by the elevation of an atoll4 ,it lies north-east of ,tahiti1 close without the line bounding the space coloured dark blue in the map appended to this volume4 ,honden ,island1 which is situated in the extreme north-west part of the ,low ,archipelago1 according to measurements made on board the ,beagle1 whilst sailing by1 is #aad feet from the summit of the trees to the water's edge4 ,this island appeared to resemble the other atolls of the group4 ,the many widely-scattered atolls must1 therefore1 rest on rocky bases4 ,but we cannot believe that the broad summit of a mountain lies buried at the depth of a few fathoms beneath every atoll1 and nevertheless throughout the immense areas above-named1 with not one point of rock projecting above the level of the sea2 for we may judge with some accuracy of mountains beneath the sea1 by those on the land2 and where can we find a single chain several hundred miles in length and of considerable breadth1 much less several such chains1 with their many broad summits attaining the same height1 within from #abj to #ahj feet8 ,if the data be thought insufficient1 on which ,i have grounded my belief1 respecting the depth at which the reef-building polypifers can exist1 and it be assumed that they can flourish at a depth of even one hundred fathoms1 yet the weight of the above argument is but little diminished1 for it is almost equally improbable1 that as many submarine mountains1 as there are low islands in the several great and widely separated areas above specified1 should all rise within six hundred feet of the surface of the sea and not one above it1 as that they should be of the same height within the smaller limit of one or two hundred feet4 ,so highly improbable is this supposition1 that we are compelled to believe1 that the bases of the many atolls did never at any one period all lie submerged within the depth of a few fathoms beneath the surface1 but that they were brought into the requisite position or level1 some at one period and some at another1 through movements in the earth's crust4 ,but this could not have been effected by elevation1 for the belief that points so numerous and so widely separated were successively uplifted to a certain level1 but that not one point was raised above that level1 is quite as improbable as the former supposition1 and indeed differs little from it4 ,it will probably occur to those who have read ,ehrenberg's account of the ,reefs of the ,red ,sea1 that many points in these great areas may have been elevated1 but that as soon as raised1 the protuberant parts were cut off by the destroying action of the waves3 a moment's reflection1 however1 on the basin-like form of the atolls1 will show that this is impossible2 for the upheaval and subsequent abrasion of an island would leave a flat disc1 which might become coated with coral1 but not a deeply concave surface2 moreover1 we should expect to see1 in some parts at least1 the rock of the foundation brought to the surface4 ,if1 then1 the foundations of the many atolls were not uplifted into the requisite position1 they must of necessity have subsided into it2 and this at once solves every difficulty1#d for we may safely infer1 from the facts given in the last chapter1 that during a gradual subsidence the corals would be favourably circumstanced for building up their solid frame works and reaching the surface1 as island after island slowly disappeared4 ,thus areas of immense extent in the central and most profound parts of the great oceans1 might become interspersed with coral-islets1 none of which would rise to a greater height than that attained by detritus heaped up by the sea1 and nevertheless they might all have been formed by corals1 which absolutely required for their growth a solid foundation within a few fathoms of the surface4 #d ,the additional difficulty on the crater hypothesis before alluded to1 will now be evident2 for on this view the volcanic action must be supposed to have formed within the areas specified a vast number of craters1 all rising within a few fathoms of the surface1 and not one above it4 ,the supposition that the craters were at different times upraised above the surface1 and were there abraded by the surf and subsequently coated by corals1 is subject to nearly the same objections with those given above in this paragraph2 but ,i consider it superfluous to detail all the arguments opposed to such a notion4 ,chamisso's theory1 from assuming the existence of so many banks1 all lying at the proper depth beneath the water1 is also vitally defective4 ,the same observation applies to an hypothesis of ,lieutenant ,nelson's 78,geolog4 ,trans40 vol4 v1 p4 #abb71 who supposes that the ring-formed structure is caused by a greater number of germs of corals becoming attached to the declivity1 than to the central plateau of a submarine bank3 it likewise applies to the notion formerly entertained 7,forster's 8,observ410 p4 #aea71 that lagoon-islands owe their peculiar form to the instinctive tendencies of the polypifers4 ,according to this latter view1 the corals on the outer margin of the reef instinctively expose themselves to the surf in order to afford protection to corals living in the lagoon1 which belong to other genera1 and to other families6 ,it would be out of place here to do more than allude to the many facts1 showing that the supposition of a gradual subsidence over large areas is by no means improbable4 ,we have the clearest proof that a movement of this kind is possible1 in the upright trees buried under the strata many thousand feet in thickness2 we have also every reason for believing that there are now large areas gradually sinking1 in the same manner as others are rising4 ,and when we consider how many parts of the surface of the globe have been elevated within recent geological periods1 we must admit that there have been subsidences on a corresponding scale1 for otherwise the whole globe would have swollen4 ,it is very remarkable that ,mr4 ,lyell1#e even in the first edition of his 8,principles of ,geology10 inferred that the amount of subsidence in the ,pacific must have exceeded that of elevation1 from the area of land being very small relatively to the agents there tending to form it1 namely1 the growth of coral and volcanic action4 ,but it will be asked1 are there any direct proofs of a subsiding movement in those areas1 in which subsidence will explain a phenomenon otherwise inexplicable8 ,this1 however1 can hardly be expected1 for it must ever be most difficult1 excepting in countries long civilised1 to detect a movement1 the tendency of which is to conceal the part affected4 ,in barbarous and semi-civilised nations how long might not a slow movement1 even of elevation such as that now affecting ,scandinavia1 have escaped attention6 #e 8,principles of ,geology10 sixth edition1 vol4 iii1 p4 #chf4 ,mr4 ,williams#f insists strongly that the traditions of the natives1 which he has taken much pains in collecting1 do not indicate the appearance of any new islands3 but on the theory of a gradual subsidence1 all that would be apparent would be1 the water sometimes encroaching slowly on the land1 and the land again recovering by the accumulation of detritus its former extent1 and perhaps sometimes the conversion of an atoll with coral islets on it1 into a bare or into a sunken annular reef4 ,such changes would naturally take place at the periods when the sea rose above its usual limits1 during a gale of more than ordinary strength2 and the effects of the two causes would be hardly distinguishable4 ,in ,kotzebue's 8,voyage0 there are accounts of islands1 both in the ,caroline and ,marshall ,archipelagoes1 which have been partly washed away during hurricanes2 and ,kadu1 the native who was on board one of the ,russian vessels1 said 8he saw the sea at ,radack rise to the feet of the cocoa-nut trees2 but it was conjured in time40#g ,a storm lately entirely swept away two of the ,caroline islands1 and converted them into shoals2 it partly1 also1 destroyed two other islands4#h ,according to a tradition which was communicated to ,captain ,fitzroy1 it is believed in the ,low ,archipelago1 that the arrival of the first ship caused a great inundation1 which destroyed many lives4 ,mr4 ,stutchbury relates1 that in #ahbe1 the western side of ,chain ,atoll1 in the same group1 was completely devastated by a hurricane1 and not less than #cjj lives lost3 8in this instance it was evident1 even to the natives1 that the hurricane alone was not sufficient to account for the violent agitation of the ocean40#i ,that considerable changes have taken place recently in some of the atolls in the ,low ,archipelago1 appears certain from the case already given of ,matilda ,island3 with respect to ,whitsunday and ,gloucester ,islands in this same group1 we must either attribute great inaccuracy to their discoverer1 the famous circumnavigator ,wallis1 or believe that they have undergone a considerable change in the period of fifty-nine years1 between his voyage and that of ,captain ,beechey's4 ,whitsunday ,island is described by ,wallis as 8about four miles long1 and three wide10 now it is only one mile and a half long4 ,the appearance of ,gloucester ,island1 in ,captain ,beechey's words1#aj 8has been accurately described by its discoverer1 but its present form and extent differ materially40 ,blenheim reef1 in the ,chagos group1 consists of a water-washed annular reef1 thirteen miles in circumference1 surrounding a lagoon ten fathoms deep3 on its surface there were a few worn patches of conglomerate coral-rock1 of about the size of hovels2 and these ,captain ,moresby considered as being1 without doubt1 the last remnants of islets2 so that here an atoll has been converted into an atoll-formed reef4 ,the inhabitants of the ,maldiva ,archipelago1 as long ago as #afje1 declared1 8that the high tides and violent currents were diminishing the number of the islands30#aa and ,i have already shown1 on the authority of ,captain ,moresby1 that the work of destruction is still in progress2 but that on the other hand the first formation of some islets is known to the present inhabitants4 ,in such cases1 it would be exceedingly difficult to detect a gradual subsidence of the foundation1 on which these mutable structures rest4 #f ,williams's 8,narrative of ,missionary ,enterprise10 p4 #ca4 #g ,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #afh4 #h ,m4 ,desmoulins in 8,comptes ,rendus10 #ahdj1 p4 #hcg4 #i ,west of ,england ,journal1 ,no4 ,i1 p4 #ce4 #aj ,beechey's 8,voyage to the ,pacific10 chap4 vii1 and ,wallis's 8,voyage in the ,dolphin10 chap4 iv4 #aa ,see an extract from ,pyrard's ,voyage in ,captain ,owen's paper on the ,maldiva ,archipelago1 in the ,geographical ,journal1 vol4 ii1 p4 #hd4 ,some of the archipelagoes of low coral-islands are subject to earthquakes3 ,captain ,moresby informs me that they are frequent1 though not very strong1 in the ,chagos group1 which occupies a very central position in the ,indian ,ocean1 and is far from any land not of coral formation4 ,one of the islands in this group was formerly covered by a bed of mould1 which1 after an earthquake1 disappeared1 and was believed by the residents to have been washed by the rain through the broken masses of underlying rock2 the island was thus rendered unproductive4 ,chamisso#ab states1 that earthquakes are felt in the ,marshall atolls1 which are far from any high land1 and likewise in the islands of the ,caroline ,archipelago4 ,on one of the latter1 namely ,oulleay atoll1 ,admiral ,lutk1 as he had the kindness to inform me1 observed several straight fissures about a foot in width1 running for some hundred yards obliquely across the whole width of the reef4 ,fissures indicate a stretching of the earth's crust1 and1 therefore1 probably changes in its level2 but these coral-islands1 which have been shaken and fissured1 certainly have not been elevated1 and1 therefore1 probably they have subsided4 ,in the chapter on ,keeling atoll1 ,i attempted to show by direct evidence1 that the island underwent a movement of subsidence1 during the earthquakes lately felt there4 #ab ,see ,chamisso1 in ,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #ahb and #acf4 ,the facts stand thus2--there are many large tracts of ocean1 without any high land1 interspersed with reefs and islets1 formed by the growth of those kinds of corals1 which cannot live at great depths2 and the existence of these reefs and low islets1 in such numbers and at such distant points1 is quite inexplicable1 excepting on the theory1 that the bases on which the reefs first became attached1 slowly and successively sank beneath the level of the sea1 whilst the corals continued to grow upwards4 ,no positive facts are opposed to this view1 and some general considerations render it probable4 ,there is evidence of change in form1 whether or not from subsidence1 on some of these coral-islands2 and there is evidence of subterranean disturbances beneath them4 ,will then the theory1 to which we have thus been led1 solve the curious problem1--what has given to each class of reef its peculiar form8 ,let us in imagination place within one of the subsiding areas1 an island surrounded by a 8fringing-reef10--that kind1 which alone offers no difficulty in the explanation of its origin4 ,let the unbroken lines1 and the oblique shading in the woodcut 7,no4 #d7 represent a vertical section through such an island2 and the horizontal shading will represent the section of the reef4 ,now1 as the island sinks down1 either a few feet at a time or quite insensibly1 we may safely infer from what we know of the conditions favourable to the growth of coral1 that the living masses bathed by the surf on the margin of the reef1 will soon regain the surface4 ,the water1 however1 will encroach1 little by little1 on the shore1 the island becoming lower and smaller1 and the space between the edge of the reef and the beach proportionately broader4 ,a section of the reef and island in this state1 after a subsidence of several hundred feet1 is given by the dotted lines3 coral-islets are supposed to have been formed on the new reef1 and a ship is anchored in the lagoon-channel4 ,this section is in every respect that of an encircling barrier-reef2 it is1 in fact1 a section taken#ac east and west through the highest point of the encircled island of ,bolabola2 of which a plan is given in ,plate ,i1 ,fig4 #e4 ,the same section is more clearly shown in the following woodcut 7,no4 #e7 by the unbroken lines4 ,the width of the reef1 and its slope1 both on the outer and inner side1 will have been determined by the growing powers of the coral1 under the conditions 7for instance the force of the breakers and of the currents7 to which it has been exposed2 and the lagoon-channel will be deeper or shallower1 in proportion to the growth of the delicately branched corals within the reef1 and to the accumulation of sediment1 relatively1 also1 to the rate of subsidence and the length of the intervening stationary periods4 #ac ,the section has been made from the chart given in the 8,atlas of the ,voyage of the ,coquille40 ,the scale is 4#eg of an inch to a mile4 ,the height of the island1 according to ,m4 ,lesson1 is #d1jbf feet4 ,the deepest part of the lagoon-channel is #afb feet2 its depth is exaggerated in the woodcut for the sake of clearness4 ,illustration3 ,vertical section of an island of ,bolabola4 ,,aa--,outer edge of the reef at the level of the sea4 ,,bb--,shores of the island4 ,a,a--,outer edge of the reef1 after its upward growth during a period of subsidence4 ,,cc--,the lagoon-channel between the reef and the shores of the now encircled land4 ,b,b--,the shores of the encircled island4 ,n4,b4--,in this1 and the following woodcut1 the subsidence of the land could only be represented by an apparent rise in the level of the sea4 ,it is evident in this section1 that a line drawn perpendicularly down from the outer edge of the new reef to the foundation of solid rock1 exceeds by as many feet as there have been feet of subsidence1 that small limit of depth at which the effective polypifers can live--the corals having grown up1 as the whole sank down1 from a basis formed of other corals and their consolidated fragments4 ,thus the difficulty on this head1 which before seemed so great1 disappears4 ,as the space between the reef and the subsiding shore continued to increase in breadth and depth1 and as the injurious effects of the sediment and fresh water borne down from the land were consequently lessened1 the greater number of the channels1 with which the reef in its fringing state must have been breached1 especially those which fronted the smaller streams1 will have become choked up with the growth of coral3 on the windward side of the reef1 where the coral grows most vigorously1 the breaches will probably have first been closed4 ,in barrier-reefs1 therefore1 the breaches kept open by draining the tidal waters of the lagoon-channel1 will generally be placed on the leeward side1 and they will still face the mouths of the larger streams1 although removed beyond the influence of their sediment and fresh water2--and this1 it has been shown1 is commonly the case4 ,illustration3 ,vertical section of an island of ,bolabola4 ,a,a--,outer edges of the barrier-reef at the level of the sea4 ,the cocoa-nut trees represent coral-islets formed on the reef4 ,,cc--,the lagoon-channel4 ,b,b--,the shores of the island1 generally formed of low alluvial land and of coral detritus from the lagoon-channel4 ,a,a--,the outer edges of the reef now forming an atoll4 ,c--,the lagoon of the newly formed atoll4 ,according to the scale1 the depth of the lagoon and of the lagoon-channel is exaggerated4 ,referring to the diagram shown above1 in which the newly formed barrier-reef is represented by unbroken lines1 instead of by dots as in the former woodcut1 let the work of subsidence go on1 and the doubly pointed hill will form two small islands 7or more1 according to the number of the hills7 included within one annular reef4 ,let the island continue subsiding1 and the coral-reef will continue growing up on its own foundation1 whilst the water gains inch by inch on the land1 until the last and highest pinnacle is covered1 and there remains a perfect atoll4 ,a vertical section of this atoll is shown in the woodcut by the dotted lines2--a ship is anchored in its lagoon1 but islets are not supposed yet to have been formed on the reef4 ,the depth of the lagoon and the width and slope of the reef1 will depend on the circumstances just referred to under barrier-reefs4 ,any further subsidence will produce no change in the atoll1 except perhaps a diminution in its size1 from the reef not growing vertically upwards2 but should the currents of the sea act violently upon it1 and should the corals perish on part or on the whole of its margin1 changes would result during subsidence which will be presently noticed4 ,i may here observe1 that a bank either of rock or of hardened sediment1 level with the surface of the sea1 and fringed with living coral1 would 7if not so small as to allow the central space to be quickly filled up with detritus7 by subsidence be converted immediately into an atoll1 without passing1 as in the case of a reef fringing the shore of an island1 through the intermediate form of a barrier-reef4 ,if such a bank lay a few fathoms submerged1 the simple growth of the coral 7as remarked in the third chapter7 without the aid of subsidence1 would produce a structure scarcely to be distinguished from a true atoll2 for in all cases the corals on the outer margin of a reef1 from having space and being freely exposed to the open sea1 will grow vigorously and tend to form a continuous ring whilst the growth of the less massive kinds on the central expanse1 will be checked by the sediment formed there1 and by that washed inwards by the breakers2 and as the space becomes shallower1 their growth will1 also1 be checked by the impurities of the water1 and probably by the small amount of food brought by the enfeebled currents1 in proportion to the surface of living reefs studded with innumerable craving mouths3 the subsidence of a reef based on a bank of this kind1 would give depth to its central expanse or lagoon1 steepness to its flanks1 and through the free growth of the coral1 symmetry to its outline3--,i may here repeat that the larger groups of atolls in the ,pacific and ,indian ,oceans cannot be supposed to be founded on banks of this nature4 ,if1 instead of the island in the diagram1 the shore of a continent fringed by a reef had subsided1 a great barrier-reef1 like that on the north-east coast of ,australia1 would have necessarily resulted2 and it would have been separated from the main land by a deep-water channel1 broad in proportion to the amount of subsidence1 and to the less or greater inclination of the neighbouring coast-line4 ,the effect of the continued subsidence of a great barrier-reef of this kind1 and its probable conversion into a chain of separate atolls1 will be noticed1 when we discuss the apparent progressive disseverment of the larger ,maldiva atolls4 ,we now are able to perceive that the close similarity in form1 dimensions1 structure1 and relative position 7which latter point will hereafter be more fully noticed7 between fringing and encircling barrier-reefs1 and between these latter and atolls1 is the necessary result of the transformation1 during subsidence of the one class into the other4 ,on this view1 the three classes of reefs ought to graduate into each other4 ,reefs having intermediate character between those of the fringing and barrier classes do exist2 for instance1 on the south-west coast of ,madagascar1 a reef extends for several miles1 within which there is a broad channel from seven to eight fathoms deep1 but the sea does not deepen abruptly outside the reef4 ,such cases1 however1 are open to some doubts1 for an old fringing-reef1 which had extended itself a little on a basis of its own formation1 would hardly be distinguishable from a barrier-reef1 produced by a small amount of subsidence1 and with its lagoon-channel nearly filled up with sediment during a long stationary period4 ,between barrier-reefs1 encircling either one lofty island or several small low ones1 and atolls including a mere expanse of water1 a striking series can be shown3 in proof of this1 ,i need only refer to the first plate in this volume1 which speaks more plainly to the eye1 than any description could to the ear4 ,the authorities from which the charts have been engraved1 together with some remarks on them and descriptive of the plates1 are given above4 ,at ,new ,caledonia 7,plate ,,ii1 ,fig4 #e47 the barrier-reefs extend for #aej miles on each side of the submarine prolongation of the island2 and at their northern extremity they appear broken up and converted into a vast atoll-formed reef1 supporting a few low coral-islets3 we may imagine that we here see the effects of subsidence actually in progress1 the water always encroaching on the northern end of the island1 towards which the mountains slope down1 and the reefs steadily building up their massive fabrics in the lines of their ancient growth4 ,we have as yet only considered the origin of barrier-reefs and atolls in their simplest form2 but there remain some peculiarities in structure and some special cases1 described in the two first chapters1 to be accounted for by our theory4 ,these consist--in the inclined ledge terminated by a wall1 and sometimes succeeded by a second ledge with a wall1 round the shores of certain lagoons and lagoon-channels2 a structure which cannot1 as ,i endeavoured to show1 be explained by the simple growing powers of the corals1--in the ring or basin-like forms of the central reefs1 as well as of the separate marginal portions of the northern ,maldiva atolls1--in the submerged condition of the whole1 or of parts of certain barrier and atoll-formed reefs2 where only a part is submerged1 this being generally to leeward1--in the apparent progressive disseverment of some of the ,maldiva atolls1--in the existence of irregularly formed atolls1 some being tied together by linear reefs1 and others with spurs projecting from them1--and1 lastly1 in the structure and origin of the ,great ,chagos ,bank4 ,step-formed ledges round certain lagoons4--,if we suppose an atoll to subside at an extremely slow rate1 it is difficult to follow out the complex results4 ,the living corals would grow up on the outer margin2 and likewise probably in the gullies and deeper parts of the bare surface of the annular reef2 the water would encroach on the islets1 but the accumulation of fresh detritus might possibly prevent their entire submergence4 ,after a subsidence of this very slow nature1 the surface of the annular reef sloping gently into the lagoon1 would probably become united with the irregular reefs and banks of sand1 which line the shores of most lagoons4 ,should1 however1 the atoll be carried down by a more rapid movement1 the whole surface of the annular reef1 where there was a foundation of solid matter1 would be favourably circumstanced for the fresh growth of coral2 but as the corals grew upwards on its exterior margin1 and the waves broke heavily on this part1 the increase of the massive polypifers on the inner side would be checked from the want of water4 ,consequently1 the exterior parts would first reach the surface1 and the new annular reef thus formed on the old one1 would have its summit inclined inwards1 and be terminated by a subaqueous wall1 formed by the upward growth of the coral 7before being much checked71 from the inner edge of the solid parts of the old reef4 ,the inner portion of the new reef1 from not having grown to the surface1 would be covered by the waters of the lagoon4 ,should a subsidence of the same kind be repeated1 the corals would again grow up in a wall1 from all the solid parts of the resunken reef1 and1 therefore1 not from within the sandy shores of the lagoon2 and the inner part of the new annular reef would1 from being as before checked in its upward growth1 be of less height than the exterior parts1 and therefore would not reach the surface of the lagoon4 ,in this case the shores of the lagoon would be surrounded by two inclined ledges1 one beneath the other1 and both abruptly terminated by subaqueous cliffs4#ad #ad ,according to ,mr4 ,couthouy 7p4 #bf7 the external reef round many atolls descends by a succession of ledges or terraces4 ,he attempts1 ,i doubt whether successfully1 to explain this structure somewhat in the same manner as ,i have attempted1 with respect to the internal ledges round the lagoons of some atolls4 ,more facts are wanted regarding the nature both of the interior and exterior step-like ledges3 are all the ledges1 or only the upper ones1 covered with living coral8 ,if they are all covered1 are the kinds different on the ledges according to the depth8 ,do the interior and exterior ledges occur together in the same atolls2 if so1 what is their total width1 and is the intervening surface-reef narrow1 etc48 ,the ring or basin-formed reefs of the northern ,maldiva atolls4--,i may first observe1 that the reefs within the lagoons of atolls and within lagoon-channels1 would1 if favourably circumstanced1 grow upwards during subsidence in the same manner as the annular rim2 and1 therefore1 we might expect that such lagoon- reefs1 when not surrounded and buried by an accumulation of sediment more rapid than the rate of subsidence1 would rise abruptly from a greater depth than that at which the efficient polypifers can flourish3 we see this well exemplified in the small abruptly-sided reefs1 with which the deep lagoons of the ,chagos and ,southern ,maldiva atolls are studded4 ,with respect to the ring or basin-formed reefs of the ,northern ,maldiva atolls1 it is evident1 from the perfectly continuous series which exists that the marginal rings1 although wider than the exterior or bounding reef of ordinary atolls1 are only modified portions of such a reef2 it is also evident that the central rings1 although wider than the knolls or reefs which commonly occur in lagoons1 occupy their place4 ,the ring-like structure has been shown to be contingent on the breaches into the lagoon being broad and numerous1 so that all the reefs which are bathed by the waters of the lagoon are placed under nearly the same conditions with the outer coast of an atoll standing in the open sea4 ,hence the exterior and living margins of these reefs must have been favourably circumstanced for growing outwards1 and increasing beyond the usual breadth2 and they must likewise have been favourably circumstanced for growing vigorously upwards1 during the subsiding movements1 to which by our theory the whole archipelago has been subjected2 and subsidence with this upward growth of the margins would convert the central space of each little reef into a small lagoon4 ,this1 however1 could only take place with those reefs1 which had increased to a breadth sufficient to prevent their central spaces from being almost immediately filled up with the sand and detritus driven inwards from all sides3 hence it is that few reefs1 which are less than half a mile in diameter1 even in the atolls where the basin-like structure is most strikingly exhibited1 include lagoons4 ,this remark1 ,i may add1 applies to all coral-reefs wherever found4 ,the basin-formed reefs of the ,maldiva ,archipelago may1 in fact1 be briefly described1 as small atolls formed during subsidence over the separate portions of large and broken atolls1 in the same manner as these latter were formed over the barrier-reefs1 which encircled the islands of a large archipelago now wholly submerged4 ,submerged and dead reefs4--,in the second section of the first chapter1 ,i have shown that there are in the neighbourhood of atolls1 some deeply submerged banks1 with level surfaces2 that there are others1 less deeply but yet wholly submerged1 having all the characters of perfect atolls1 but consisting merely of dead coral-rock2 that there are barrier-reefs and atolls with merely a portion of their reef1 generally on the leeward side1 submerged2 and that such portions either retain their perfect outline1 or they appear to be quite effaced1 their former place being marked only by a bank1 conforming in outline with that part of the reef which remains perfect4 ,these several cases are1 ,i believe1 intimately related together1 and can be explained by the same means4 ,there1 perhaps1 exist some submerged reefs1 covered with living coral and growing upwards1 but to these ,i do not here refer4 ,as we see that in those parts of the ocean1 where coral-reefs are most abundant1 one island is fringed and another neighbouring one is not fringed2 as we see in the same archipelago1 that all the reefs are more perfect in one part of it than in another1 for instance1 in the southern half compared with the northern half of the ,maldiva ,archipelago1 and likewise on the outer coasts compared with the inner coasts of the atolls in this same group1 which are placed in a double row2 as we know that the existence of the innumerable polypifers forming a reef1 depends on their sustenance1 and that they are preyed on by other organic beings2 and1 lastly1 as we know that some inorganic causes are highly injurious to the growth of coral1 it cannot be expected that during the round of change to which earth1 air1 and water are exposed1 the reef-building polypifers should keep alive for perpetuity in any one place2 and still less can this be expected1 during the progressive subsidences1 perhaps at some periods more rapid than at others1 to which by our theory these reefs and islands have been subjected and are liable4 ,it is1 then1 not improbable that the corals should sometimes perish either on the whole or on part of a reef2 if on part1 the dead portion1 after a small amount of subsidence1 would still retain its proper outline and position beneath the water4 ,after a more prolonged subsidence1 it would probably form1 owing to the accumulation of sediment1 only the margin of a flat bank1 marking the limits of the former lagoon4 ,such dead portions of reef would generally lie on the leeward side1#ae for the impure water and fine sediment would more easily flow out from the lagoon over this side of the reef1 where the force of the breakers is less than to windward2 and therefore the corals would be less vigorous on this side1 and be less able to resist any destroying agent4 ,it is likewise owing to this same cause1 that reefs are more frequently breached to leeward by narrow channels1 serving as by ship-channels1 than to windward4 ,if the corals perished entirely1 or on the greater part of the circumference of an atoll1 an atoll-shaped bank of dead rock1 more or less entirely submerged1 would be produced2 and further subsidence1 together with the accumulation of sediment1 would often obliterate its atoll-like structure1 and leave only a bank with a level surface4 #ae ,mr4 ,lyell1 in the first edition of his 8,principles of ,geology10 offered a somewhat different explanation of this structure4 ,he supposes that there has been subsidence2 but he was not aware that the submerged portions of reef were in most cases1 if not in all1 dead2 and he attributes the difference in height in the two sides of most atolls1 chiefly to the greater accumulation of detritus to windward than to leeward4 ,but as matter is accumulated only on the backward part of the reef1 the front part would remain of the same height on both sides4 ,i may here observe that in most cases 7for instance1 at ,peros ,banhos1 the ,gambier group and the ,great ,chagos ,bank71 and ,i suspect in all cases1 the dead and submerged portions do not blend or slope into the living and perfect parts1 but are separated from them by an abrupt line4 ,in some instances small patches of living reef rise to the surface from the middle of the submerged and dead parts4 ,in the ,chagos group of atolls1 within an area of #afj miles by #fj1 there are two atoll-formed banks of dead rock 7besides another very imperfect one71 entirely submerged2 a third1 with merely two or three very small pieces of living reef rising to the surface2 and a fourth1 namely1 ,peros ,banhos 7,plate ,i1 ,fig4 #i71 with a portion nine miles in length dead and submerged4 ,as by our theory this area has subsided1 and as there is nothing improbable in the death1 either from changes in the state of the surrounding sea or from the subsidence being great or sudden1 of the corals on the whole1 or on portions of some of the atolls1 the case of the ,chagos group presents no difficulty4 ,so far indeed are any of the above-mentioned cases of submerged reefs from being inexplicable1 that their occurrence might have been anticipated on our theory1 and as fresh atolls are supposed to be in progressive formation by the subsidence of encircling barrier-reefs1 a weighty objection1 namely that the number of atolls must be increasing infinitely1 might even have been raised1 if proofs of the occasional destruction and loss of atolls could not have been adduced4 ,the disseverment of the larger ,maldiva atolls4--,the apparent progressive disseverment in the ,maldiva ,archipelago of large atolls into smaller ones1 is1 in many respects1 an important consideration1 and requires an explanation4 ,the graduated series which marks1 as ,i believe1 this process1 can be observed only in the northern half of the group1 where the atolls have exceedingly imperfect margins1 consisting of detached basin-formed reefs4 ,the currents of the sea flow across these atolls1 as ,i am informed by ,captain ,moresby1 with considerable force1 and drift the sediment from side to side during the monsoons1 transporting much of it seaward2 yet the currents sweep with greater force round their flanks4 ,it is historically known that these atolls have long existed in their present state2 and we can believe1 that even during a very slow subsidence they might thus remain1 the central expanse being kept at nearly its original depth by the accumulation of sediment4 ,but in the action of such nicely balanced forces during a progressive subsidence 7like that1 to which by our theory this archipelago has been subjected71 it would be strange if the currents of the sea should never make a direct passage across some one of the atolls1 through the many wide breaches in their margins4 ,if this were once effected1 a deep-water channel would soon be formed by the removal of the finer sediment1 and the check to its further accumulation2 and the sides of the channel would be worn into a slope like that on the outer coasts1 which are exposed to the same force of the currents4 ,in fact1 a channel precisely like that bifurcating one which divides ,mahlos ,mahdoo 7,plate ,,ii1 ,fig4 #d71 would almost necessarily be formed4 ,the scattered reefs situated near the borders of the new ocean-channel1 from being favourably placed for the growth of coral1 would1 by their extension1 tend to produce fresh margins to the dissevered portions2 such a tendency is very evident 7as may be seen in the large published chart7 in the elongated reefs on the borders of the two channels intersecting ,mahlos ,mahdoo4 ,such channels would become deeper with continued subsidence1 and probably from the reefs not growing up perpendicularly1 somewhat broader4 ,in this case1 and more especially if the channels had been formed originally of considerable breadth1 the dissevered portions would become perfect and distinct atolls1 like ,ari and ,ross atolls 7,plate ,,ii1 ,fig4 #f71 or like the two ,nillandoo atolls1 which must be considered as distinct1 although related in form and position1 and separated from each other by channels1 which though deep have been sounded4 ,further subsidence would render such channels unfathomable1 and the dissevered portions would then resemble ,phaleedoo and ,moluque atolls1 or ,mahlos ,mahdoo and ,horsburgh atolls 7,plate ,,ii1 ,fig4 #d71 which are related to each other in no respect except in proximity and position4 ,hence1 on the theory of subsidence1 the disseverment of large atolls1 which have imperfect margins 7for otherwise their disseverment would be scarcely possible71 and which are exposed to strong currents1 is far from being an improbable event2 and the several stages1 from close relation to entire isolation in the atolls of the ,maldiva ,archipelago1 are readily explicable4 ,we might go even further1 and assert as not improbable1 that the first formation of the ,maldiva ,archipelago was due to a barrier-reef1 of nearly the same dimensions with that of ,new ,caledonia 7,plate ,,ii1 ,fig4 #e71 for if1 in imagination1 we complete the subsidence of that great island1 we might anticipate from the present broken condition of the northern portion of the reef1 and from the almost entire absence of reefs on the eastern coast1 that the barrier-reef after repeated subsidences1 would become during its upward growth separated into distinct portions2 and these portions would tend to assume an atoll-like structure1 from the coral growing with vigour round their entire circumferences1 when freely exposed to an open sea4 ,as we have some large islands partly submerged with barrier-reefs marking their former limits1 such as ,new ,caledonia1 so our theory makes it probable that there should be other large islands wholly submerged2 and these1 we may now infer1 would be surmounted1 not by one enormous atoll1 but by several large elongated ones1 like the atolls in the ,maldiva group2 and these again1 during long periods of subsidence1 would sometimes become dissevered into smaller atolls4 ,i may add1 that both in the ,marshall and ,caroline ,archipelagoes1 there are atolls standing close together1 which have an evident relationship in form3 we may suppose1 in such cases1 either that two or more encircled islands originally stood close together1 and afforded bases for two or more atolls1 or that one atoll has been dissevered4 ,from the position1 as well as form1 of three atolls in the ,caroline ,archipelago 7the ,namourrek and ,elato group71 which are placed in an irregular circle1 ,i am strongly tempted to believe that they have originated by the process of disseverment4#af #af ,the same remark is1 perhaps1 applicable to the islands of ,ollap1 ,fanadik1 and ,tamatam in the ,caroline ,archipelago1 of which charts are given in the atlas of ,duperrey's voyage3 a line drawn through the linear reefs and lagoons of these three islands forms a semicircle4 ,consult also1 the atlas of ,lutk's voyage2 and for the ,marshall group that of ,kotzebue2 for the ,gilbert group consult the atlas of ,duperrey's voyage4 ,most of the points here referred to may1 however1 be seen in ,krusenstern's general ,atlas of the ,pacific4 ,irregularly formed atolls4--,in the ,marshall group1 ,musquillo atoll consists of two loops united in one point2 and ,menchikoff atoll is formed of three loops1 two of which 7as may be seen in ,fig4 #c1 ,plate ,,ii7 are connected by a mere ribbon-shaped reef1 and the three together are sixty miles in length4 ,in the ,gilbert group some of the atolls have narrow strips of reef1 like spurs1 projecting from them4 ,there occur also in parts of the open sea1 a few linear and straight reefs1 standing by themselves2 and likewise some few reefs in the form of crescents1 with their extremities more or less curled inwards4 ,now1 the upward growth of a barrier-reef which fronted only one side of an island1 or one side of an elongated island with its extremities 7of which cases exist71 would produce after the complete subsidence of the land1 mere strips or crescent or hook-formed reefs3 if the island thus partially fronted became divided during subsidence into two or more islands1 these islands would be united together by linear reefs2 and from the further growth of the coral along their shores together with subsidence1 reefs of various forms might ultimately be produced1 either atolls united together by linear reefs1 or atolls with spurs projecting from them4 ,some1 however1 of the more simple forms above specified1 might1 as we have seen1 be equally well produced by the coral perishing during subsidence on part of the circumference of an atoll1 whilst on the other parts it continued to grow up till it reached the surface4 ,the ,great ,chagos ,bank4--,i have already shown that the submerged condition of the ,great ,chagos ,bank 7,plate ,,ii1 ,fig4 #a1 with its section ,fig4 #b71 and of some other banks in the ,chagos group1 may in all probability be attributed to the coral having perished before or during the movements of subsidence1 to which this whole area by our theory has been subjected4 ,the external rim or upper ledge 7shaded in the chart71 consists of dead coral-rock thinly covered with sand2 it lies at an average depth of between five and eight fathoms1 and perfectly resembles in form the annular reef of an atoll4 ,the banks of the second level1 the boundaries of which are marked by dotted lines in the chart1 lie from about fifteen to twenty fathoms beneath the surface2 they are several miles broad1 and terminate in a very steep slope round the central expanse4 ,this central expanse ,i have already described1 as consisting of a level muddy flat between thirty and forty fathoms deep4 ,the banks of the second level1 might at first sight be thought analogous to the internal step-like ledge of coral-rock which borders the lagoons of some atolls1 but their much greater width1 and their being formed of sand1 are points of essential difference4 ,on the eastern side of the atoll some of the banks are linear and parallel1 resembling islets in a great river1 and pointed directly towards a great breach on the opposite side of the atoll2 these are best seen in the large published chart4 ,i inferred from this circumstance1 that strong currents sometimes set directly across this vast bank2 and ,i have since heard from ,captain ,moresby that this is the case4 ,i observed1 also1 that the channels or breaches through the rim1 were all of the same depth as the central lagoon-like space into which they lead2 whereas the channels into the other atolls of the ,chagos group1 and as ,i believe into most other large atolls1 are not nearly as deep as their lagoons3 for instance at ,peros ,banhos1 the channels are only of the same depth1 namely between ten and twenty fathoms1 as the bottom of the lagoon for a space about a mile and a half in width round its shores1 whilst the central expanse of the lagoon is from thirty-five to forty fathoms deep4 ,now1 if an atoll during a gradual subsidence once became entirely submerged1 like the ,great ,chagos ,bank1 and therefore no longer exposed to the surf1 very little sediment could be formed from it2 and consequently the channels leading into the lagoon from not being filled up with drifted sand and coral detritus1 would continue increasing in depth1 as the whole sank down4 ,in this case1 we might expect that the currents of the open sea1 instead of any longer sweeping round the submarine flanks1 would flow directly through the breaches across the lagoon1 removing in their course the finer sediment1 and preventing its further accumulation4 ,we should then have the submerged reef forming an external and upper rim of rock1 and beneath this portion of the sandy bottom of the old lagoon1 intersected by deep-water channels or breaches1 and thus formed into separate marginal banks2 and these would be cut off by steep slopes1 overhanging the central space1 worn down by the passage of the oceanic currents4 ,by these means1 ,i have scarcely any doubt that the ,great ,chagos ,bank has originated1--a structure which at first appeared to me far more anomalous than any ,i had met with4 ,the process of formation is nearly the same with that1 by which ,mahlos ,mahdoo had been trisected2 but in the ,chagos ,bank the channels of the oceanic currents entering at several different quarters1 have united in a central space4 ,this great atoll-formed bank appears to be in an early stage of disseverment2 should the work of subsidence go on1 from the submerged and dead condition of the whole reef1 and the imperfection of the south-east quarter a mere wreck would probably be left4 ,the ,pitt's ,bank1 situated not far southward1 appears to be precisely in this state2 it consists of a moderately level1 oblong bank of sand1 lying from #aj to #bj fathoms beneath the surface1 with two sides protected by a narrow ledge of rock which is submerged between #e and #h fathoms4 ,a little further south1 at about the same distance as the southern rim of the ,great ,chagos ,bank is from the northern rim1 there are two other small banks with from #aj to #bj fathoms on them2 and not far eastward soundings were struck on a sandy bottom1 with between #aaj and #ade fathoms4 ,the northern portion with its ledge-like margin1 closely resembles any one segment of the ,great ,chagos ,bank1 between two of the deep-water channels1 and the scattered banks1 southward appear to be the last wrecks of less perfect portions4 ,i have examined with care the charts of the ,indian and ,pacific ,oceans1 and have now brought before the reader all the examples1 which ,i have met with1 of reefs differing from the type of the class to which they belong2 and ,i think it has been satisfactorily shown1 that they are all included in our theory1 modified by occasional accidents which might have been anticipated as probable4 ,in this course we have seen1 that in the lapse of ages encircling barrier-reefs are occasionally converted into atolls1 the name of atoll being properly applicable1 at the moment when the last pinnacle of encircled land sinks beneath the surface of the sea4 ,we have1 also1 seen that large atolls during the progressive subsidence of the areas in which they stand1 sometimes become dissevered into smaller ones2 at other times1 the reef-building polypifers having entirely perished1 atolls are converted into atoll-formed banks of dead rock2 and these again through further subsidence and the accumulation of sediment modified by the force of the oceanic currents1 pass into level banks with scarcely any distinguishing character4 ,thus may the history of an atoll be followed from its first origin1 through the occasional accidents of its existence1 to its destruction and final obliteration4 ,objections to the theory of the formation of atolls and barrier-reefs4--,the vast amount of subsidence1 both horizontally or in area1 and vertically or in depth1 necessary to have submerged every mountain1 even the highest1 throughout the immense spaces of ocean interspersed with atolls1 will probably strike most people as a formidable objection to my theory4 ,but as continents1 as large as the spaces supposed to have subsided1 have been raised above the level of the sea1--as whole regions are now rising1 for instance1 in ,scandinavia and ,south ,america1--and as no reason can be assigned1 why subsidences should not have occurred in some parts of the earth's crust on as great a scale both in extent and amount as those of elevation1 objections of this nature strike me as of little force4 ,the remarkable point is that movements to such an extent should have taken place within a period1 during which the polypifers have continued adding matter on and above the same reefs4 ,another and less obvious objection to the theory will perhaps be advanced from the circumstance1 of the lagoons within atolls and within barrier-reefs never having become in any one instance during prolonged subsidences of a greater depth than sixty fathoms1 and seldom more than forty fathoms2 but we already admit1 if the theory be worth considering1 that the rate of subsidence has not exceeded that of the upward growth of the coral on the exterior margin2 we are1 therefore1 only further required to admit1 that the subsidence has not exceeded in rate the filling up of the interior spaces by the growth of the corals living there1 and by the accumulation of sediment4 ,as this filling up must take place very slowly within barrier-reefs lying far from the land1 and within atolls which are of large dimensions and which have open lagoons with very few reefs1 we are led to conclude that the subsidence thus counter-balanced1 must have been slow in an extraordinary degree2 a conclusion which accords with our only means1 namely1 with what is known of the rate and manner of recent elevatory movements1 of judging by analogy what is the probable rate of subsidence4 ,in this chapter it has1 ,i think1 been shown1 that the theory of subsidence1 which we were compelled to receive from the necessity of giving to the corals1 in certain large areas1 foundations at the requisite depth1 explains both the normal structure and the less regular forms of those two great classes of reefs1 which have justly excited the astonishment of all persons who have sailed through the ,pacific and ,indian ,oceans4 ,but further to test the truth of the theory1 a crowd of questions will occur to the reader3 ,do the different kinds of reefs1 which have been produced by the same kind of movement1 generally lie within the same areas8 ,what is their relation of form and position1--for instance1 do adjoining groups of atolls1 and the separate atolls in these groups1 bear the same relation to each other which islands do in common archipelagoes8 ,have we reason to believe1 that where there are fringing-reefs1 there has not lately been subsidence2 or1 for it is almost our only way of ascertaining this point1 are there frequently proofs of recent elevation8 ,can we by this means account for the presence of certain classes of reefs in some large areas1 and their entire absence in others8 ,do the areas which have subsided1 as indicated by the presence of atolls and barrier-reefs1 and the areas which have remained stationary or have been upraised1 as shown by fringing-reefs1 bear any determinate relation to each other2 and are the dimensions of these areas such as harmonise with the greatness of the subterranean changes1 which1 it must be supposed1 have lately taken place beneath them8 ,is there any connection between the movements thus indicated1 and recent volcanic action8 ,all these questions ought to receive answers in accordance with the theory2 and if this can be satisfactorily shown1 not only is the theory confirmed1 but as deductions1 the answers are in themselves important4 ,under this latter point of view1 these questions will be chiefly considered in the following chapter4#ag #ag ,i may take this opportunity of briefly considering the appearances1 which would probably be presented by a vertical and deep section across a coral formation 7referring chiefly to an atoll71 formed by the upward growth of coral during successive subsidences4 ,this is a subject worthy of attention1 as a means of comparison with ancient coral-strata4 ,the circumferential parts would consist of massive species1 in a vertical position1 with their interstices filled up with detritus2 but this would be the part most subject to subsequent denudation and removal4 ,it is useless to speculate how large a portion of the exterior annular reef would consist of upright coral1 and how much of fragmentary rock1 for this would depend on many contingencies1--such as on the rate of subsidence1 occasionally allowing a fresh growth of coral to cover the whole surface1 and on the breakers having force sufficient to throw fragments over this same space4 ,the conglomerate which composes the base of the islets1 would 7if not removed by denudation together with the exterior reef on which it rests7 be conspicuous from the size of the fragments1--the different degrees in which they have been rounded1--the presence of fragments of conglomerate torn up1 rounded1 and recemented1--and from the oblique stratification4 ,the corals which lived in the lagoon-reefs at each successive level1 would be preserved upright1 and they would consist of many kinds1 generally much branched4 ,in this part1 however1 a very large proportion of the rock 7and in some cases nearly all of it7 would be formed of sedimentary matter1 either in an excessively fine1 or in a moderately coarse state1 and with the particles almost blended together4 ,the conglomerate which was formed of rounded pieces of the branched corals1 on the shores of the lagoon1 would differ from that formed on the islets and derived from the outer coast2 yet both might have accumulated very near each other4 ,i have seen a conglomerate limestone from ,devonshire like a conglomerate now forming on the shores of the ,maldiva atolls4 ,the stratification taken as a whole1 would be horizontal2 but the conglomerate beds resting on the exterior reef1 and the beds of sandstone on the shores of the lagoon 7and no doubt on the external flanks7 would probably be divided 7as at ,keeling atoll and at ,mauritius7 by numerous layers dipping at considerable angles in different directions4 ,the calcareous sandstone and coral-rock would almost necessarily contain innumerable shells1 echini1 and the bones of fish1 turtle1 and perhaps of birds2 possibly1 also1 the bones of small saurians1 as these animals find their way to the islands far remote from any continent4 ,the large shells of some species of ,tridacna would be found vertically imbedded in the solid rock1 in the position in which they lived4 ,we might expect also to find a mixture of the remains of pelagic and littoral animals in the strata formed in the lagoon1 for pumice and the seeds of plants are floated from distant countries into the lagoons of many atolls3 on the outer coast of ,keeling atoll1 near the mouth of the lagoon1 the case of a pelagic ,pteropodous animal was brought up on the arming of the sounding lead4 ,all the loose blocks of coral on ,keeling atoll were burrowed by vermiform animals2 and as every cavity1 no doubt1 ultimately becomes filled with spathose limestone1 slabs of the rock taken from a considerable depth1 would1 if polished1 probably exhibit the excavations of such burrowing animals4 ,the conglomerate and fine-grained beds of coral-rock would be hard1 sonorous1 white and composed of nearly pure calcareous matter2 in some few parts1 judging from the specimens at ,keeling atoll1 they would probably contain a small quantity of iron4 ,floating pumice and scori1 and occasionally stones transported in the root of trees 7see my 8,journal of ,researches10 page #edi7 appear the only sources1 through which foreign matter is brought to coral-formations standing in the open ocean4 ,the area over which sediment is transported from coral-reefs must be considerable3 ,captain ,moresby informs me that during the change of monsoons the sea is discoloured to a considerable distance off the ,maldiva and ,chagos atolls4 ,the sediment of fringing and barrier coral-reefs must be mingled with the mud1 which is brought down from the land1 and is transported seaward through the breaches1 which occur in front of almost every valley4 ,if the atolls of the larger archipelagoes were upraised1 the bed of the ocean being converted into land1 they would form flat-topped mountains1 varying in diameter from a few miles 7the smallest atolls being worn away7 to sixty miles2 and from being horizontally stratified and of similar composition1 they would1 as ,mr4 ,lyell has remarked1 falsely appear as if they had originally been united into one vast continuous mass4 ,such great strata of coral-rock would rarely be associated with erupted volcanic matter1 for this could only take place1 as may be inferred from what follows in the next chapter1 when the area1 in which they were situated1 commenced to rise1 or at least ceased to subside4 ,during the enormous period necessary to effect an elevation of the kind just alluded to1 the surface would necessarily be denuded to a great thickness2 hence it is highly improbable that any fringing-reef1 or even any barrier-reef1 at least of those encircling small islands1 would be preserved4 ,from this same cause1 the strata which were formed within the lagoons of atolls and lagoon-channels of barrier-reefs1 and which must consist in a large part of sedimentary matter1 would more often be preserved to future ages1 than the exterior solid reef1 composed of massive corals in an upright position2 although it is on this exterior part that the present existence and further growth of atolls and barrier-reefs entirely depend4 ,plate ,,iii--,map showing the distribution of coral-reefs and active volcanoes4 ,illustration3 ,map showing distribution of coral-reefs and active volcanoes4 ,illustration3 ,map showing distribution of coral-reefs and active volcanoes4 ,illustration3 ,map showing distribution of coral-reefs and active volcanoes4 ,the principles1 on which this map was coloured1 are explained in the beginning of ,chapter ,,vi2 and the authorities for each particular spot are detailed in the ,appendix to ,coral ,reefs4 ,the names not printed in upper case in the ,index refer to the ,appendix4 ,chapter ,,vi ,,on ,,the ,,distribution ,,of ,,coral-,,reefs ,,with ,,reference ,,to ,,the ,,theory ,,of ,,their ,,formation ,description of the coloured map4--,proximity of atolls and barrier-reefs4--,relation in form and position of atolls with ordinary islands4--,direct evidence of subsidence difficult to be detected4--,proofs of recent elevation where fringing-reefs occur4--,oscillations of level4--,absence of active volcanoes in the areas of subsidence4--,immensity of the areas which have been elevated and have subsided4--,their relation to the present distribution of the land4--,areas of subsidence elongated1 their intersection and alternation with those of elevation4--,amount and slow rate of the subsidence4--,recapitulation4 ,it will be convenient to give here a short account of the appended map 7,plate ,,iii73#a a fuller one1 with the data for colouring each spot1 is reserved for the ,appendix2 and every place there referred to may be found in the ,index4 ,a larger chart would have been desirable2 but1 small as the adjoined one is1 it is the result of many months' labour4 ,i have consulted1 as far as ,i was able1 every original voyage and map2 and the colours were first laid down on charts on a larger scale4 ,the same blue colour1 with merely a difference in the depth of tint1 is used for atolls or lagoon-islands1 and barrier-reefs1 for we have seen1 that as far as the actual coral-formation is concerned1 they have no distinguishing character4 ,fringing-reefs have been coloured red1 for between them on the one hand1 and barrier-reefs and atolls on the other1 there is an important distinction with respect to the depth beneath the surface1 at which we are compelled to believe their foundations lie4 ,the two distinct colours1 therefore1 mark two great types of structure4 #a ,inasmuch as the coloured map would have proved too costly to be given in this series1 the indications of colour have been replaced by numbers referring to the dotted groups of reefs1 etc4 ,the author's original wording1 however1 is retained in full1 as it will be easy to refer to the map by the numbers1 and thus the flow of the narrative is undisturbed4 ,the dark blue colour represented by 7#c7 in our plate represents atolls and submerged annular reefs1 with deep water in their centres4 ,i have coloured as atolls1 a few low and small coral-islands1 without lagoons2 but this has been done only when it clearly appeared that they originally contained lagoons1 since filled up with sediment3 when there were not good grounds for this belief1 they have been left uncoloured4 ,the pale blue colour represented by 7#b7 represents barrier-reefs4 ,the most obvious character of reefs of this class is the broad and deep-water moat within the reef3 but this1 like the lagoons of small atolls1 is liable to become filled up with detritus and with reefs of delicately branched corals3 when1 therefore1 a reef round the entire circumference of an island extends very far into a profoundly deep sea1 so that it can hardly be confounded with a fringing-reef which must rest on a foundation of rock within a small depth1 it has been coloured pale blue1 although it does not include a deep-water moat3 but this has only been done rarely1 and each case is distinctly mentioned in the ,appendix4 ,the red colour 7#d7 represents reefs fringing the land quite closely where the sea is deep1 and where the bottom is gently inclined extending to a moderate distance from it1 but not having a deep-water moat or lagoon-like space parallel to the shore4 ,it must be remembered that fringing-reefs are frequently breached in front of rivers and valleys by deepish channels1 where mud has been deposited4 ,a space of thirty miles in width has been coloured round or in front of the reefs of each class1 in order that the colours might be conspicuous on the appended map1 which is reduced to so small a scale4 ,the vermillion spots1 and streaks 7#a7 represent volcanoes now in action1 or historically known to have been so4 ,they are chiefly laid down from ,von ,buch's work on the ,canary ,islands2 and my reasons for making a few alterations are given in the note below4#b #b ,i have also made considerable use of the geological part of ,berghaus' 8,physical ,atlas40 ,beginning at the eastern side of the ,pacific1 ,i have added to the number of the volcanoes in the southern part of the ,cordillera1 and have coloured ,juan ,fernandez according to observations collected during the voyage of the ,beagle 78,geol4 ,trans410 vol4 v1 p4 #fja74 ,i have added a volcano to ,albemarle ,island1 one of the ,galapagos ,archipelago 7the author's 8,journal of ,researches10 p4 #deg74 ,in the ,sandwich group there are no active volcanoes1 except at ,hawaii2 but the ,rev4 ,w4 ,ellis informs me1 there are streams of lava apparently modern on ,maui1 having a very recent appearance1 which can be traced to the craters whence they flowed4 ,the same gentleman informs me1 that there is no reason to believe that any active volcano exists in the ,society ,archipelago2 nor are there any known in the ,samoa or ,navigator group1 although some of the streams of lava and craters there appear recent4 ,in the ,friendly group1 the ,rev4 ,j4 ,williams says 78,narrative of ,missionary ,enterprise10 p4 #bi7 that ,toofoa and ,proby ,islands are active volcanoes4 ,i infer from ,hamilton's 8,voyage in the ,pandora0 7p4 #ie71 that ,proby ,island is synonymous with ,onouafou1 but ,i have not ventured to colour it4 ,there can be no doubt respecting ,toofoa1 and ,captain ,edwards 7,von ,buch1 p4 #chf7 found the lava of recent eruption at ,amargura still smoking4 ,berghaus marks four active volcanoes actually within the ,friendly group2 but ,i do not know on what authority3 ,i may mention that ,maurelle describes ,latte as having a burnt-up appearance3 ,i have marked only ,toofoa and ,amargura4 ,south of the ,new ,hebrides lies ,matthews ,rock1 which is drawn and described as an active crater in the 8,voyage of the ,astrolabe40 ,between it and the volcano on the eastern side of ,new ,zealand1 lies ,brimstone ,island1 which from the high temperature of the water in the crater1 may be ranked as active 7,berghaus 8,vorbemerk10 ,,ii ,lief4 ,s4 #ef74 ,malte ,brun1 vol4 xii1 p4 #bca1 says that there is a volcano near port ,st4 ,vincent in ,new ,caledonia4 ,i believe this to be an error1 arising from a smoke seen on the opposite coast by ,cook 78,second ,voyage10 vol4 ii1 p4 #bc7 which smoke went out at night4 ,the ,mariana ,islands1 especially the northern ones1 contain many craters 7see ,freycinet's 8,hydrog4 ,descript407 which are not active4 ,von ,buch1 however1 states 7p4 #dfb7 on the authority of ,la ,peyrouse1 that there are no less than seven volcanoes between these islands and ,japan4 ,gemelli ,creri 7,churchill's 8,collect40 vol4 iv1 p4 #deh71 says there are two active volcanoes in latitude #bc #cj1 and in latitude #bd3 but ,i have not coloured them4 ,from the statements in ,beechey's 8,voyage0 7p4 #eah1 #dto edit47 ,i have coloured one in the northern part of the ,bonin group4 ,m4 ,s4 ,julien has clearly made out from ,chinese manuscripts not very ancient 78,comptes ,rendus10 #ahdj1 p4 #hcb71 that there are two active volcanoes on the eastern side of ,formosa4 ,in ,torres ,straits1 on ,cap ,island 7#i #dh ,s41 #adb #ci ,e47 a volcano was seen burning with great violence in #agic by ,captain ,bampton 7see ,introduction to ,flinders' 8,voyage10 p4 #da74 ,mr4 ,m',clelland 7,report of ,committee for investigating ,coal in ,india1 p4 #ci7 has shown that the volcanic band passing through ,barren ,island must be extended northwards4 ,it appears by an old chart1 that ,cheduba was once an active volcano 7see also ,silliman's ,north ,american ,journal1 vol4 xxxviii1 p4 #che74 ,in ,berghaus' 8,phys4 ,atlas10 #ahdj1 ,no4 #g of ,geological ,part1 a volcano on the coast of ,pondicherry is said to have burst forth in #ageg4 ,ordinaire 78,hist4 ,nat4 des ,volcans10 p4 #bah7 says that there is one at the mouth of the ,persian ,gulf1 but ,i have not coloured it1 as he gives no particulars4 ,a volcano in ,amsterdam1 or ,st4 ,paul's1 in the southern part of the ,indian ,ocean1 has been seen 7,naut4 ,mag4 #ahch1 p4 #hdb7 in action4 ,dr4 ,j4 ,allan1 of ,forres1 informs me in a letter1 that when he was at ,joanna1 he saw at night flames apparently volcanic1 issuing from the chief ,comoro ,island1 and that the ,arabs assured him that they were volcanic1 adding that the volcano burned more during the wet season4 ,i have marked this as a volcano1 though with some hesitation1 on account of the possibility of the flame arising from gaseous sources4 ,the uncoloured coasts consist1 first and chiefly1 of those1 where there are no coral-reefs1 or such small portions as to be quite insignificant4 ,secondly1 of those coasts where there are reefs1 but where the sea is very shallow1 for in this case the reefs generally lie far from the land1 and become very irregular1 in their forms3 where they have not become irregular1 they have been coloured4 thirdly1 if ,i had the means of ascertaining the fact1 ,i should not colour a reef merely coating the edges of a submarine crater1 or of a level submerged bank2 for such superficial formations differ essentially1 even when not in external appearance1 from reefs whose foundations as well as superficies have been wholly formed by the growth of coral4 ,fourthly1 in the ,red ,sea1 and within some parts of the ,east ,indian ,archipelago 7if the imperfect charts of the latter can be trusted71 there are many scattered reefs1 of small size1 represented in the chart by mere dots1 which rise out of deep water3 these cannot be arranged under either of the three classes3 in the ,red ,sea1 however1 some of these little reefs1 from their position1 seem once to have formed parts of a continuous barrier4 ,there exist1 also1 scattered in the open ocean1 some linear and irregularly formed strips of coral-reef1 which1 as shown in the last chapter1 are probably allied in their origin to atolls2 but as they do not belong to that class1 they have not been coloured2 they are very few in number and of insignificant dimensions4 ,lastly1 some reefs are left uncoloured from the want of information respecting them1 and some because they are of an intermediate structure between the barrier and fringing classes4 ,the value of the map is lessened1 in proportion to the number of reefs which ,i have been obliged to leave uncoloured1 although1 in a theoretical point of view1 few of them present any great difficulty3 but their number is not very great1 as will be found by comparing the map with the statements in the ,appendix4 ,i have experienced more difficulty in colouring fringing-reefs than in colouring barrier-reefs1 as the former1 from their much less dimensions1 have less attracted the attention of navigators4 ,as ,i have had to seek my information from all kinds of sources1 and often from indirect ones1 ,i do not venture to hope that the map is free from many errors4 ,nevertheless1 ,i trust it will give an approximately correct view of the general distribution of the coral-reefs over the whole world 7with the exception of some fringing-reefs on the coast of ,brazil1 not included within the limits of the map71 and of their arrangement into the three great classes1 which1 though necessarily very imperfect from the nature of the objects classified1 have been adopted by most voyagers4 ,i may further remark1 that the dark blue colour represents land entirely composed of coral-rock2 the pale blue1 land with a wide and thick border of coral-rock2 and the red1 a mere narrow fringe of coral-rock4 ,looking now at the map under the theoretical point of view indicated in the last chapter1 the two blue tints signify that the foundations of the reefs thus coloured have subsided to a considerable amount1 at a slower rate than that of the upward growth of the corals1 and that probably in many cases they are still subsiding4 ,the red signifies that the shores which support fringing-reefs have not subsided 7at least to any considerable amount1 for the effects of a subsidence on a small scale would in no case be distinguishable72 but that they have remained nearly stationary since the period when they first became fringed by reefs2 or that they are now rising or have been upraised1 with new lines of reefs successively formed on them3 these latter alternatives are obviously implied1 as newly formed lines of shore1 after elevations of the land1 would be in the same state with respect to the growth of fringing-reefs1 as stationary coasts4 ,if during the prolonged subsidence of a shore1 coral-reefs grew for the first time on it1 or if an old barrier-reef were destroyed and submerged1 and new reefs became attached to the land1 these would necessarily at first belong to the fringing class1 and1 therefore1 be coloured red1 although the coast was sinking3 but ,i have no reason to believe1 that from this source of error1 any coast has been coloured wrongly with respect to movement indicated4 ,well characterised atolls and encircling barrier-reefs1 where several occur in a group1 or a single barrier-reef if of large dimensions1 leave scarcely any doubt on the mind respecting the movement by which they have been produced2 and even a small amount of subsequent elevation is soon betrayed4 ,the evidence from a single atoll or a single encircling barrier-reef1 must be received with some caution1 for the former may possibly be based upon a submerged crater or bank1 and the latter on a submerged margin of sediment1 or of worn-down rock4 ,from these remarks we may with greater certainty infer that the spaces1 especially the larger ones1 tinted blue in the map1 have subsided1 than that the red spaces have remained stationary1 or have been upraised4 ,on the grouping of the different classes of reefs4--,having made these preliminary remarks1 ,i will consider first how far the grouping of the different kinds of coral-islands and reefs is corroborative of the truth of the theory4 ,a glance at the map shows that the reefs1 coloured blue and red1 produced under widely different conditions1 are not indiscriminately mixed together4 ,atolls and barrier-reefs1 on the other hand1 as may be seen by the two blue tints1 generally lie near each other2 and this would be the natural result of both having been produced during the subsidence of the areas in which they stand4 ,thus1 the largest group of encircled islands is that of the ,society ,archipelago2 and these islands are surrounded by atolls1 and only separated by a narrow space from the large group of ,low atolls4 ,in the midst of the ,caroline atolls1 there are three fine encircled islands4 ,the northern point of the barrier-reef of ,new ,caledonia seems itself1 as before remarked1 to form a complete large atoll4 ,the great ,australian barrier is described as including both atolls and small encircled islands4 ,captain ,king#c mentions many atoll-formed and encircling coral-reefs1 some of which lie within the barrier1 and others may be said 7for instance between lat4 #af and #ac7 to form part of it4 ,flinders#d has described an atoll-formed reef in lat4 #aj1 seven miles long and from one to three broad1 resembling a boot in shape1 with apparently very deep water within4 ,eight miles westward of this1 and forming part of the barrier1 lie the ,murray ,islands1 which are high and are encircled4 ,in the ,corallian ,sea1 between the two great barriers of ,australia and ,new ,caledonia1 there are many low islets and coral-reefs1 some of which are annular1 or horse-shoe shaped4 ,observing the smallness of the scale of the map1 the parallels of latitude being nine hundred miles apart1 we see that none of the large groups of reefs and islands supposed to have been produced by long-continued subsidence1 lie near extensive lines of coast coloured red1 which are supposed to have remained stationary since the growth of their reefs1 or to have been upraised and new lines of reefs formed on them4 ,where the red and blue circles do occur near each other1 ,i am able1 in several instances1 to show that there have been oscillations of level1 subsidence having preceded the elevation of the red spots2 and elevation having preceded the subsidence of the blue spots3 and in this case the juxtaposition of reefs belonging to the two great types of structure is little surprising4 ,we may1 therefore1 conclude that the proximity in the same areas of the two classes of reefs1 which owe their origin to the subsidence of the earth's crust1 and their separation from those formed during its stationary or uprising condition1 holds good to the full extent1 which might have been anticipated by our theory4 #c ,sailing directions1 appended to vol4 ii of his 8,surveying ,voyage to ,australia40 #d 8,voyage to ,terra ,australis10 vol4 ii1 p4 #ccf4 ,as groups of atolls have originated in the upward growth1 at each fresh sinking of the land1 of those reefs which primarily fringed the shores of one great island1 or of several smaller ones2 so we might expect that these rings of coral-rock1 like so many rude outline charts1 will still retain some traces of the general form1 or at least general range1 of the land1 round which they were first modelled4 ,that this is the case with the atolls in the ,southern ,pacific as far as their range is concerned1 seems highly probable1 when we observe that the three principal groups are directed in north-west and south-east lines1 and that nearly all the land in the ,s4 ,pacific ranges in this same direction2 namely1 ,n4 ,western ,australia1 ,new ,caledonia1 the northern half of ,new ,zealand1 the ,new ,hebrides1 ,saloman1 ,navigator1 ,society1 ,marquesas1 and ,austral archipelagoes3 in the ,northern ,pacific1 the ,caroline atolls abut against the north-west line of the ,marshall atolls1 much in the same manner as the east and west line of islands from ,ceram to ,new ,britain do on ,new ,ireland3 in the ,indian ,ocean the ,laccadive and ,maldiva atolls extend nearly parallel to the western and mountainous coast of ,india4 ,in most respects1 there is a perfect resemblance with ordinary islands in the grouping of atolls and in their form3 thus the outline of all the larger groups is elongated2 and the greater number of the individual atolls are elongated in the same direction with the group1 in which they stand4 ,the ,chagos group is less elongated than is usual with other groups1 and the individual atolls in it are likewise but little elongated2 this is strikingly seen by comparing them with the neighbouring ,maldiva atolls4 ,in the ,marshall and ,maldiva archipelagoes1 the atolls are ranged in two parallel lines1 like the mountains in a great double mountain-chain4 ,some of the atolls1 in the larger archipelagoes1 stand so near to each other1 and have such an evident relationship in form1 that they compose little sub-groups3 in the ,caroline ,archipelago1 one such sub-group consists of ,pouynipete1 a lofty island encircled by a barrier-reef1 and separated by a channel only four miles and a half wide from ,andeema atoll1 with a second atoll a little further off4 ,in all these respects an examination of a series of charts will show how perfectly groups of atolls resemble groups of common islands4 ,on the direct evidence of the blue spaces in the map having subsided during the upward growth of the reefs so coloured1 and of the red spaces having remained stationary1 or having been upraised4--,with respect to subsidence1 ,i have shown in the last chapter1 that we cannot expect to obtain in countries inhabited only by semi-civilised races1 demonstrative proofs of a movement1 which invariably tends to conceal its own evidence4 ,but on the coral-islands supposed to have been produced by subsidence1 we have proofs of changes in their external appearance--of a round of decay and renovation--of the last vestiges of land on some--of its first commencement on others3 we hear of storms desolating them to the astonishment of their inhabitants3 we know by the great fissures with which some of them are traversed1 and by the earthquakes felt under others1 that subterranean disturbances of some kind are in progress4 ,these facts1 if not directly connected with subsidence1 as ,i believe they are1 at least show how difficult it would be to discover proofs of such movement by ordinary means4 ,at ,keeling atoll1 however1 ,i have described some appearances1 which seem directly to show that subsidence did take place there during the late earthquakes4 ,vanikoro1 according to ,chevalier ,dillon1#e is often violently shaken by earthquakes1 and there1 the unusual depth of the channel between the shore and the reef1--the almost entire absence of islets on the reef1--its wall-like structure on the inner side1 and the small quantity of low alluvial land at the foot of the mountains1 all seem to show that this island has not remained long at its present level1 with the lagoon-channel subjected to the accumulation of sediment1 and the reef to the wear and tear of the breakers4 ,at the ,society ,archipelago1 on the other hand1 where a slight tremor is only rarely felt1 the shoaliness of the lagoon-channels round some of the islands1 the number of islets formed on the reefs of others1 and the broad belt of low land at the foot of the mountains1 indicate that1 although there must have been great subsidence to have produced the barrier-reefs1 there has since elapsed a long stationary period4#f #e ,see ,captain ,dillon's 8,voyage in search of ,la ,peyrouse40 ,m4 ,cordier in his 8,report on the ,voyage of the #8,astrolabe'0 7p4 cxi1 vol4 i71 speaking of ,vanikoro1 says the shores are surrounded by reefs of madrepore1 8qu'on assure etre de formation tout-a-fait moderne40 ,i have in vain endeavoured to learn some further particulars about this remarkable passage4 ,i may here add1 that according to our theory1 the island of ,pouynipete 7,plate ,i1 ,fig4 #g71 in the ,caroline ,archipelago1 being encircled by a barrier-reef1 must have subsided4 ,in the ,new ,s4 ,wales ,lit4 ,advert4 ,february #ahce 7which ,i have seen through the favour of ,dr4 ,lloghtsky71 there is an account of this island 7subsequently confirmed by ,mr4 ,campbell71 in which it is said1 8,at the ,n4,e4 end1 at a place called ,tamen1 there are ruins of a town1 now only accessible by boats1 the waves reaching to the steps of the houses40 ,judging from this passage1 one would be tempted to conclude that the island must have subsided1 since these houses were built4 ,i may1 also1 here append a statement in ,malte ,brun 7vol4 ix1 p4 #gge1 given without any authority71 that the sea gains in an extraordinary manner on the coast of ,cochin ,china1 which lies in front and near the subsiding coral-reefs in the ,china ,sea3 as the coast is granitic1 and not alluvial1 it is scarcely possible that the encroachment of the sea can be owing to the washing away of the land2 and if so1 it must be due to subsidence4 #f ,mr4 ,couthouy states 78,remarks10 p4 #dd7 that at ,tahiti and ,eimeo the space between the reef and the shore has been nearly filled up by the extension of those coral-reefs1 which within most barrier-reefs merely fringe the land4 ,from this circumstance1 he arrives at the same conclusion as ,i have done1 that the ,society ,islands since their subsidence1 have remained stationary during a long period2 but he further believes that they have recently commenced rising1 as well as the whole area of the ,low ,archipelago4 ,he does not give any detailed proofs regarding the elevation of the ,society ,islands1 but ,i shall refer to this subject in another part of this chapter4 ,before making some further comments1 ,i may observe how satisfactory it is to me1 to find ,mr4 ,couthouy affirming1 that 8having personally examined a large number of coral-islands1 and also residing eight months among the volcanic class1 having shore and partially encircling reefs1 ,i may be permitted to state that my own observations have impressed a conviction of the correctness of the theory of ,mr4 ,darwin40 ,this gentleman believes1 that subsequently to the subsidence by which the atolls in the ,low ,archipelago were produced1 the whole area has been elevated to the amount of a few feet2 this would indeed be a remarkable fact2 but as far as ,i am able to judge1 the grounds of his conclusion are not sufficiently strong4 ,he states that he found in almost every atoll which he visited1 the shores of the lagoon raised from eighteen to thirty inches above the sea-level1 and containing imbedded ,tridacn and corals standing as they grew2 some of the corals were dead in their upper parts1 but below a certain line they continued to flourish4 ,in the lagoons1 also1 he frequently met with clusters of ,madrepore1 with their extremities standing from one inch to a foot above the surface of the water4 ,now1 these appearances are exactly what ,i should have expected1 without any subsequent elevation having taken place2 and ,i think ,mr4 ,couthouy has not borne in mind the indisputable fact1 that corals1 when constantly bathed by the surf1 can exist at a higher level than in quite tranquil water1 as in a lagoon4 ,as long1 therefore1 as the waves continued at low water to break entirely over parts of the annular reef of an atoll1 submerged to a small depth1 the corals and shells attached on these parts might continue living at a level above the smooth surface of the lagoon1 into which the waves rolled2 but as soon as the outer edge of the reef grew up to its utmost possible height1 or if the reef were very broad nearly to that height1 the force of the breakers would be checked1 and the corals and shells on the inner parts near the lagoon would occasionally be left dry1 and thus be partially or wholly destroyed4 ,even in atolls1 which have not lately subsided1 if the outer margin of the reef continued to increase in breadth seaward 7each fresh zone of corals rising to the same vertical height as at ,keeling atoll71 the line where the waves broke most heavily would advance outwards1 and therefore the corals1 which when living near the margin1 were washed by the breaking waves during the whole of each tide1 would cease being so1 and would therefore be left on the backward part of the reef standing exposed and dead4 ,the case of the madrepores in the lagoons with the tops of their branches exposed1 seems to be an analogous fact1 to the great fields of dead but upright corals in the lagoon of ,keeling atoll2 a condition of things which ,i have endeavoured to show1 has resulted from the lagoon having become more and more enclosed and choked up with reefs1 so that during high winds1 the rising of the tide 7as observed by the inhabitants7 is checked1 and the corals1 which had formerly grown to the greatest possible height1 are occasionally exposed1 and thus are killed3 and this is a condition of things1 towards which almost every atoll in the intervals of its subsidence must be tending4 ,or if we look to the state of an atoll directly after a subsidence of some fathoms1 the waves would roll heavily over the entire circumference of the reef1 and the surface of the lagoon would1 like the ocean1 never be quite at rest1 and therefore the corals in the lagoon1 from being constantly laved by the rippling water1 might extend their branches to a little greater height than they could1 when the lagoon became enclosed and protected4 ,christmas atoll 7#b ,n4 lat47 which has a very shallow lagoon1 and differs in several respects from most atolls1 possibly may have been elevated recently2 but its highest part appears 7,couthouy1 p4 #df7 to be only ten feet above the sea-level4 ,the facts of a second class1 adduced by ,mr4 ,couthouy1 in support of the alleged recent elevation of the ,low ,archipelago1 are not all 7especially those referring to a shelf of rock7 quite intelligible to me2 he believes that certain enormous fragments of rock on the reef1 must have been moved into their present position1 when the reef was at a lower level2 but here again the force of the breakers on any inner point of the reef being diminished by its outward growth without any change in its level1 has not1 ,i think1 been borne in mind4 ,we should1 also1 not overlook the occasional agency of waves caused by earthquakes and hurricanes4 ,mr4 ,couthouy further argues1 that since these great fragments were deposited and fixed on the reef1 they have been elevated2 he infers this from the greatest amount of erosion not being near their bases1 where they are unceasingly washed by the reflux of the tides1 but at some height on their sides1 near the line of high-water mark1 as shown in an accompanying diagram4 ,my former remark again applies here1 with this further observation1 that as the waves have to roll over a wide space of reef before they reach the fragments1 their force must be greatly increased with the increasing depth of water as the tide rises1 and therefore ,i should have expected that the chief line of present erosion would have coincided with the line of high-water mark2 and if the reef had grown outwards1 that there would have been lines of erosion at greater heights4 ,the conclusion1 to which ,i am finally led by the interesting observations of ,mr4 ,couthouy is1 that the atolls in the ,low ,archipelago have1 like the ,society ,islands1 remained at a stationary level for a long period3 and this probably is the ordinary course of events1 subsidence supervening after long intervals of rest4 ,turning now to the red colour2 as on our map1 the areas which have sunk slowly downwards to great depths are many and large1 we might naturally have been led to conjecture1 that with such great changes of level in progress1 the coasts which have been fringed probably for ages 7for we have no reason to believe that coral-reefs are of short duration71 would not have remained all this time stationary1 but would frequently have undergone movements of elevation4 ,this supposition1 we shall immediately see1 holds good to a remarkable extent2 and although a stationary condition of the land can hardly ever be open to proof1 from the evidence being only negative1 we are1 in some degree1 enabled to ascertain the correctness of the parts coloured red on the map1 by the direct testimony of upraised organic remains of a modern date4 ,before going into the details on this head 7printed in small type71 ,i may mention1 that when reading a memoir on coral formations by ,,mm4 ,quoy and ,gaimard#g ,i was astonished to find1 for ,i knew that they had crossed both the ,pacific and ,indian ,oceans1 that their descriptions were applicable only to reefs of the fringing class2 but my astonishment ended satisfactorily1 when ,i discovered that1 by a strange chance1 all the islands which these eminent naturalists had visited1 though several in number1 namely1 the ,mauritius1 ,timor1 ,new ,guinea1 the ,mariana1 and ,sandwich ,archipelagoes1 could be shown by their own statements to have been elevated within a recent geological era4 #g 8,annales des ,sciences ,nat40 tom4 vi1 p4 #bgi1 etc4 ,in the eastern half of the ,pacific1 the ,sandwich ,islands are all fringed1 and almost every naturalist who has visited them1 has remarked on the abundance of elevated corals and shells1 apparently identical with living species4 ,the ,rev4 ,w4 ,ellis informs me1 that he has noticed round several parts of ,hawaii1 beds of coral-detritus1 about twenty feet above the level of the sea1 and where the coast is low they extend far inland4 ,upraised coral-rock forms a considerable part of the borders of ,oahu2 and at ,elizabeth ,island#h it composes three strata1 each about ten feet thick4 ,nihau1 which forms the northern1 as ,hawaii does the southern end of the group 7#cej miles in length71 likewise seems to consist of coral and volcanic rocks4 ,mr4 ,couthouy#i has lately described with interesting details1 several upraised beaches1 ancient reefs with their surfaces perfectly preserved1 and beds of recent shells and corals1 at the islands of ,maui1 ,morokai1 ,oahu1 and ,tauai 7or ,kauai7 in this group4 ,mr4 ,pierce1 an intelligent resident at ,oahu1 is convinced1 from changes which have taken place within his memory1 during the last sixteen years1 8that the elevation is at present going forward at a very perceptible rate40 ,the natives at ,kauai state that the land is there gaining rapidly on the sea1 and ,mr4 ,couthouy has no doubt1 from the nature of the strata1 that this has been effected by an elevation of the land4 #h 8,zoology of ,captain ,beechey's ,voyage10 p4 #agf4 ,see also ,,mm4 ,quoy and ,gaimard in 8,annales de ,scien4 ,nat40 tom4 vi4 #i 8,remarks on ,coral ,formations10 p4 #ea4 ,in the southern part of the ,low ,archipelago1 ,elizabeth ,island is described by ,captain ,beechey1#aj as being quite flat1 and about eighty feet in height2 it is entirely composed of dead corals1 forming a honeycombed1 but compact rock4 ,in cases like this1 of an island having exactly the appearance1 which the elevation of any one of the smaller surrounding atolls with a shallow lagoon would present1 one is led to conclude 7with little better reason1 however1 than the improbability of such small and low fabrics lasting1 for an immense period1 exposed to the many destroying agents of nature71 that the elevation has taken place at an epoch not geologically remote4 ,when merely the surface of an island of ordinary formation is strewed with marine bodies1 and that continuously1 or nearly so1 from the beach to a certain height1 and not above that height1 it is exceedingly improbable that such organic remains1 although they may not have been specially examined1 should belong to any ancient period4 ,it is necessary to bear these remarks in mind1 in considering the evidence of the elevatory movements in the ,pacific and ,indian ,oceans1 as it does not often rest on specific determinations1 and therefore should be received with caution4 ,six of the ,cook and ,austral ,islands 7,s4,w4 of the ,society group71 are fringed2 of these1 five were described to me by the ,rev4 ,j4 ,williams1 as formed of coral-rock1 associated with some basalt in ,mangaia71 and the sixth as lofty and basaltic4 ,mangaia is nearly three hundred feet high1 with a level summit2 and according to ,mr4 ,s4 ,wilson#aa it is an upraised reef2 8and there are in the central hollow1 formerly the bed of the lagoon1 many scattered patches of coral-rock1 some of them raised to a height of forty feet40 ,these knolls of coral-rock were evidently once separate reefs in the lagoon of an atoll4 ,mr4 ,martens1 at ,sydney1 informed me that this island is surrounded by a terrace-like plain at about the height of a hundred feet1 which probably marks a pause in its elevation4 ,from these facts we may infer1 perhaps1 that the ,cook and ,austral ,islands have been upheaved at a period probably not very remote4 #aj ,beechey's 8,voyage in the ,pacific10 p4 #df1 #dto ed4 #aa ,couthouy's 8,remarks10 p4 #cd4 ,savage ,island 7,s4,e4 of the ,friendly group71 is about forty feet in height4 ,forster#ab describes the plants as already growing out of the dead1 but still upright and spreading trees of coral2 and the younger ,forster#ac believes that an ancient lagoon is now represented by a central plain2 here we cannot doubt that the elevatory forces have recently acted4 ,the same conclusion may be extended1 though with somewhat less certainty1 to the islands of the friendly group1 which have been well described in the second and third voyages of ,cook4 ,the surface of ,tongatabou is low and level1 but with some parts a hundred feet high2 the whole consists of coral-rock1 8which yet shows the cavities and irregularities worn into it by the action of the tides40#ad ,on ,eoua the same appearances were noticed at an elevation of between two hundred and three hundred feet4 ,vavao1 also1 at the opposite or northern end of the group1 consists1 according to the ,rev4 ,j4 ,williams1 of coral-rock4 ,tongatabou1 with its northern extensive reefs1 resembles either an upraised atoll with one half originally imperfect1 or one unequally elevated2 and ,anamouka1 an atoll equally elevated4 ,this latter island contains#ae in its centre a salt-water lake1 about a mile-and-a-half in diameter1 without any communication with the sea1 and around it the land rises gradually like a bank2 the highest part is only between twenty and thirty feet2 but on this part1 as well as on the rest of the land 7which1 as ,cook observes1 rises above the height of true lagoon-islands71 coral-rock1 like that on the beach1 was found4 ,in the ,navigator ,archipelago1 ,mr4 ,couthouy#af found on ,manua many and very large fragments of coral at the height of eighty feet1 8on a steep hill-side1 rising half a mile inland from a low sandy plain abounding in marine remains40 ,the fragments were embedded in a mixture of decomposed lava and sand4 ,it is not stated whether they were accompanied by shells1 or whether the corals resembled recent species2 as these remains were embedded they possibly may belong to a remote epoch2 but ,i presume this was not the opinion of ,mr4 ,couthouy4 ,earthquakes are very frequent in this archipelago4 #ab 8,observations made during ,voyage round the ,world10 p4 #adg4 #ac 8,voyage10 vol4 ii1 p4 #afc4 #ad ,cook's 8,third ,voyage0 7#dto ed471 vol4 i1 p4 #cad4 #ae ,ibid41 vol4 i1 p4 #bce4 #af 8,remarks on ,coral-,formations10 p4 #ej4 ,still proceeding westward we come to the ,new ,hebrides2 on these islands1 ,mr4 ,g4 ,bennett 7author of 8,wanderings in ,new ,south ,wales071 informs me he found much coral at a great altitude1 which he considered of recent origin4 ,respecting ,santa ,cruz1 and the ,solomon ,archipelago1 ,i have no information2 but at ,new ,ireland1 which forms the northern point of the latter chain1 both ,labillardiere and ,lesson have described large beds of an apparently very modern madreporitic rock1 with the form of the corals little altered4 ,the latter author#ag states that this formation composes a newer line of coast1 modelled round an ancient one4 ,there only remains to be described in the ,pacific1 that curved line of fringed islands1 of which the ,marianas form the main part4 ,of these ,guam1 ,rota1 ,tiniam1 ,saypan1 and some islets farther north1 are described by ,quoy and ,gaimard1#ah and ,chamisso1#ai as chiefly composed of madreporitic limestone1 which attains a considerable elevation1 and is in several cases worn into successively rising cliffs3 the two former naturalists seem to have compared the corals and shells with the existing ones1 and state that they are of recent species4 ,fais1 which lies in the prolonged line of the ,marianas1 is the only island in this part of the sea which is fringed2 it is ninety feet high1 and consists entirely of madreporitic rock4#bj #ag 8,voyage de la ,coquille10 ,part4 ,zoolog4 #ah ,freycinet's 8,voyage autour du ,monde40 ,see also the 8,hydrographical ,memoir10 p4 #bae4 #ai ,kotzebue's 8,first ,voyage40 #bj ,lutk's 8,voyage10 vol4 ii1 p4 #cjd4 ,in the ,east ,indian ,archipelago1 many authors have recorded proofs of recent elevation4 ,m4 ,lesson#ba states1 that near ,port ,dory1 on the north coast of ,new ,guinea1 the shores are flanked1 to the height of #aej feet1 by madreporitic strata of modern date4 ,he mentions similar formations at ,waigiou1 ,amboina1 ,bourou1 ,ceram1 ,sonda1 and ,timor3 at this latter place1 ,,mm4 ,quoy and ,gaimard#bb have likewise described the primitive rocks1 as coated to a considerable height with coral4 ,some small islets eastward of ,timor are said in ,kolff's 8,voyage10#bc to resemble small coral islets upraised some feet above the sea4 ,dr4 ,malcolmson informs me that ,dr4 ,hardie found in ,,java an extensive formation1 containing an abundance of shells1 of which the greater part appear to be of existing species4 ,dr4 ,jack#bd has described some upraised shells and corals1 apparently recent1 on ,pulo ,nias off ,sumatra2 and ,marsden relates in his history of this great island1 that the names of many promontories1 show that they were originally islands4 ,on part of the west coast of ,borneo and at the ,sooloo ,islands1 the form of the land1 the nature of the soil1 and the water-washed rocks1 present appearances#be 7although it is doubtful whether such vague evidence is worthy of mention71 of having recently been covered by the sea2 and the inhabitants of the ,sooloo ,islands believe that this has been the case4 ,mr4 ,cuming1 who has lately investigated1 with so much success1 the natural history of the ,philippines1 found near ,cabagan1 in ,luzon1 about fifty feet above the level of the ,r4 ,cagayan1 and seventy miles from its mouth1 a large bed of fossil shells3 these1 he informs me1 are of the same species with those now existing on the shores of the neighbouring islands4 ,from the accounts given us by ,captain ,basil ,hall and ,captain ,beechey#bf of the lines of inland reefs1 and walls of coral-rock worn into caves1 above the present reach of the waves1 at the ,loo ,choo ,islands1 there can be little doubt that they have been upraised at no very remote period4 #ba ,partie ,zoolog41 8,voyage de la ,coquille40 #bb 8,ann4 des ,scien4 ,nat40 tom4 vi1 p4 #bha4 #bc ,translated by ,windsor ,earl1 chapters vi1 vii4 #bd 8,geolog4 ,transact40 #bnd series1 vol4 i1 p4 #djc4 ,on the ,peninsula of ,malacca1 in front of ,pinang1 #e #cj ,n41 ,dr4 ,ward collected some shells1 which ,dr4 ,malcolmson informs me1 although not compared with existing species1 had a recent appearance4 ,dr4 ,ward describes in this neighbourhood 78,trans4 ,asiat4 ,soc40 vol4 xviii1 part ii1 p4 #aff7 a single water-worn rock1 with a conglomerate of sea-shells at its base1 situated six miles inland1 which1 according to the traditions of the natives1 was once surrounded by the sea4 ,captain ,low has also described 7,ibid41 part i1 p4 #aca7 mounds of shells lying two miles inland on this line of coast4 #be 8,notices of the ,east ,indian ,arch40 ,singapore1 #ahbh1 p4 #f1 and ,append41 p4 #dc4 #bf ,captain ,b4 ,hall1 8,voyage to ,loo ,choo10 ,append41 pp4 xxi and xxv4 ,captain ,beechey's 8,voyage10 p4 #dif4 ,dr4 ,davy#bg describes the northern province of ,ceylon as being very low1 and consisting of a limestone with shells and corals of very recent origin2 he adds1 that it does not admit of a doubt that the sea has retired from this district even within the memory of man4 ,there is also some reason for believing that the western shores of ,india1 north of ,ceylon1 have been upraised within the recent period4#bh ,mauritius has certainly been upraised within the recent period1 as ,i have stated in the chapter on fringing-reefs4 ,the northern extremity of ,madagascar is described by ,captain ,owen#bi as formed of madreporitic rock1 as likewise are the shores and outlying islands along an immense space of ,eastern ,africa1 from a little north of the equator for nine hundred miles southward4 ,nothing can be more vague than the expression 8madreporitic rock20 but at the same time it is1 ,i think1 scarcely possible to look at the chart of the linear islets1 which rise to a greater height than can be accounted for by the growth of coral1 in front of the coast1 from the equator to #b ,s41 without feeling convinced that a line of fringing-reefs has been elevated at a period so recent1 that no great changes have since taken place on the surface of this part of the globe4 ,some1 also1 of the higher islands of madreporitic rock on this coast1 for instance ,pemba1 have very singular forms1 which seem to show the combined effect of the growth of coral round submerged banks1 and their subsequent upheaval4 ,dr4 ,allan informs me that he never observed any elevated organic remains on the ,seychelles1 which come under our fringed class4 #bg 8,travels in ,ceylon10 p4 #ac4 ,this madreporitic formation is mentioned by ,m4 ,cordier in his report to the ,institute 7,may #dth1 #ahci71 on the voyage of the ,chevrette1 as one of immense extent1 and belonging to the latest tertiary period4 #bh ,dr4 ,benza1 in his 8,journey through the ,n4 ,circars0 7the ,madras ,lit4 and ,scient4 ,journ4 vol4 v47 has described a formation with recent fresh-water and marine shells1 occurring at the distance of three or four miles from the present shore4 ,dr4 ,benza1 in conversation with me1 attributed their position to a rise of the land4 ,dr4 ,malcolmson1 however 7and there cannot be a higher authority on the geology of ,india7 informs me that he suspects that these beds may have been formed by the mere action of the waves and currents accumulating sediment4 ,from analogy ,i should much incline to ,dr4 ,benza's opinion4 #bi ,owen's 8,africa10 vol4 ii1 p4 #cg1 for ,madagascar2 and for ,s4 ,africa1 vol4 i1 pp4 #dab and #dbf4 ,lieutenant ,boteler's narrative contains fuller particulars regarding the coral-rock1 vol4 i1 p4 #agd1 and vol4 ii1 pp4 #da and #ed4 ,see also ,ruschenberger's 8,voyage round the ,world10 vol4 i1 p4 #fj4 ,the nature of the formations round the shores of the ,red ,sea1 as described by several authors1 shows that the whole of this large area has been elevated within a very recent tertiary epoch4 ,a part of this space in the appended map1 is coloured blue1 indicating the presence of barrier-reefs3 on which circumstance ,i shall presently make some remarks4 ,rppell#cj states that the tertiary formation1 of which he has examined the organic remains1 forms a fringe along the shores with a uniform height of from thirty and forty feet from the mouth of the ,gulf of ,suez to about latitude #bf2 but that south of #bf1 the beds attain only the height of from twelve to fifteen feet4 ,this1 however1 can hardly be quite accurate2 although possibly there may be a decrease in the elevation of the shores in the middle parts of the ,red ,sea1 for ,dr4 ,malcolmson 7as he informs me7 collected from the cliffs of ,camaran ,island 7lat4 #ae #cj ,s47 shells and corals1 apparently recent1 at a height between thirty and forty feet2 and ,mr4 ,salt 78,travels in ,abyssinia07 describes a similar formation a little southward on the opposite shore at ,amphila4 ,moreover1 near the mouth of the ,gulf of ,suez1 although on the coast opposite to that on which ,dr4 ,rppell says that the modern beds attain a height of only thirty to forty feet1 ,mr4 ,burton#ca found a deposit replete with existing species of shells1 at the height of #bjj feet4 ,in an admirable series of drawings by ,captain ,moresby1 ,i could see how continuously the cliff-bounded low plains of this formation extended with a nearly equable height1 both on the eastern and western shores4 ,the southern coast of ,arabia seems to have been subjected to the same elevatory movement1 for ,dr4 ,malcolmson found at ,sahar low cliffs containing shells and corals1 apparently of recent species4 #cj ,rppell1 8,reise in ,abyssinien10 ,band i41 s4 #ada4 #ca ,lyell's 8,principles of ,geology10 #eth ed41 vol4 iv1 p4 #be4 ,the ,persian ,gulf abounds with coral-reefs2 but as it is difficult to distinguish them from sand-banks in this shallow sea1 ,i have coloured only some near the mouth2 towards the head of the gulf ,mr4 ,ainsworth#cb says that the land is worn into terraces1 and that the beds contain organic remains of existing forms4 ,the ,west ,indian ,archipelago of 8fringed0 islands1 alone remains to be mentioned2 evidence of an elevation within a late tertiary epoch of nearly the whole of this great area1 may be found in the works of almost all the naturalists who have visited it4 ,i will give some of the principal references in a note4#cc #cb ,ainsworth's 8,assyria and ,babylon10 p4 #bag4 #cc ,on ,florida and the north shores of the ,gulf of ,mexico1 ,rogers' 8,report to ,brit4 ,assoc40 vol4 iii1 p4 #ad4--,on the shores of ,mexico1 ,humboldt1 8,polit4 ,essay on ,new ,spain10 vol4 i1 p4 #fb4 7,i have also some corroborative facts with respect to the shores of ,mexico47--,honduras and the ,antilles1 ,lyell's 8,principles10 #eth ed41 vol4 iv1 p4 #bb4--,santa ,cruz and ,barbadoes1 ,prof4 ,hovey1 8,silliman's ,journal01 vol4 xxxv1 p4 #gd4--,st4 ,domingo1 ,courrojolles1 8,journ de ,phys40 tom4 liv41 p4 #ajf4--;,bahamas1 8,united ,service ,journal01 ,no4 lxxi1 pp4 #bah and #bbd4 ,jamaica1 ,de la ,beche1 8,geol4 ,man40 p4 #adb4--;,cuba1 ,taylor in 8,lond4 and ,edin4 ,mag40 vol4 xi1 p4 #ag4 ,dr4 ,daubeny also1 at a meeting of the ,geolog4 ,soc41 orally described some very modern beds lying on the ,n4,w4 parts of ,cuba4 ,i might have added many other less important references4 ,it is very remarkable on reviewing these details1 to observe in how many instances fringing-reefs round the shores1 have coincided with the existence on the land of upraised organic remains1 which seem1 from evidence more or less satisfactory1 to belong to a late tertiary period4 ,it may1 however1 be objected1 that similar proofs of elevation1 perhaps1 occur on the coasts coloured blue in our map3 but this certainly is not the case with the few following and doubtful exceptions4 ,the entire area of the ,red ,sea appears to have been upraised within a modern period2 nevertheless ,i have been compelled 7though on unsatisfactory evidence1 as given in the ,appendix7 to class the reefs in the middle part1 as barrier-reefs2 should1 however1 the statements prove accurate to the less height of the tertiary bed in this middle part1 compared with the northern and southern districts1 we might well suspect that it had subsided subsequently to the general elevation by which the whole area has been upraised4 ,several authors#cd have stated that they have observed shells and corals high up on the mountains of the ,society ,islands1--a group encircled by barrier-reefs1 and1 therefore1 supposed to have subsided3 at ,tahiti ,mr4 ,stutchbury found on the apex of one of the highest mountains1 between #e1jjj and #g1jjj feet above the level of the sea1 8a distinct and regular stratum of semi-fossil coral40 ,at ,tahiti1 however1 other naturalists1 as well as myself1 have searched in vain at a low level near the coast1 for upraised shells or masses of coral-reef1 where if present they could hardly have been overlooked4 ,from this fact1 ,i concluded that probably the organic remains strewed high up on the surface of the land1 had originally been embedded in the volcanic strata1 and had subsequently been washed out by the rain4 ,i have since heard from the ,rev4 ,w4 ,ellis1 that the remains which he met with1 were 7as he believes7 interstratified with an argillaceous tuff2 this likewise was the case with the shells observed by the ,rev4 ,d4 ,tyerman at ,huaheine4 ,these remains have not been specifically examined2 they may1 therefore1 and especially the stratum observed by ,mr4 ,stutchbury at an immense height1 be contemporaneous with the first formation of the ,society ,islands1 and be of any degree of antiquity2 or they may have been deposited at some subsequent1 but probably not very recent1 period of elevation2 for if the period had been recent1 the entire surface of the coast land of these islands1 where the reefs are so extensive1 would have been coated with upraised coral1 which certainly is not the case4 ,two of the ,harvey1 or ,cook ,islands1 namely1 ,aitutaki and ,manouai1 are encircled by reefs1 which extend so far from the land1 that ,i have coloured them blue1 although with much hesitation1 as the space within the reef is shallow1 and the outline of the land is not abrupt4 ,these two islands consist of coral-rock2 but ,i have no evidence of their recent elevation1 besides1 the improbability of ,mangaia1 a fringed island in the same group 7but distant #agj miles71 having retained its nearly perfect atoll-like structure1 during any immense lapse of time after its upheaval4 ,the ,red ,sea1 therefore1 is the only area in which we have clear proofs of the recent elevation of a district1 which1 by our theory 7although the barrier-reefs are there not well characterised71 has lately subsided4 ,but we have no reason to be surprised at oscillation1 of level of this kind having occasionally taken place4 ,there can be scarcely any doubt that ,savage1 ,aurora1#ce and ,mangaia ,islands1 and several of the islands in the ,friendly group1 existed originally as atolls1 and these have undoubtedly since been upraised to some height above the level of the sea2 so that by our theory1 there has here1 also1 been an oscillation of level1--elevation having succeeded subsidence1 instead of1 as in the middle part of the ,red ,sea and at the ,harvey ,islands1 subsidence having probably succeeded recent elevation4 #cd ,ellis1 in his 8,polynesian ,researches10 was the first to call attention to these remains 7vol4 i1 p4 #ch71 and the tradition of the natives concerning them4 ,see also ,williams1 8,nar4 of ,missionary ,enterprise10 p4 #ba2 also ,tyerman and ,g4 ,bennett1 8,journal of ,voyage10 vol4 i1 p4 #bac2 also ,mr4 ,couthouy's 8,remarks10 p4 #ea2 but this principal fact1 namely1 that there is a mass of upraised coral on the narrow peninsula of ,tiarubu1 is from hearsay evidence2 also ,mr4 ,stutchbury1 ,west of ,england ,journal1 ,no4 i1 p4 #ed4 ,there is a passage in ,von ,zach1 8,corres4 ,astronom40 vol4 x1 p4 #bff1 inferring an uprising at ,tahiti1 from a footpath now used1 which was formerly impassable2 but ,i particularly inquired from several native chiefs1 whether they knew of any change of this kind1 and they were unanimous in giving me an answer in the negative4 #ce ,aurora ,island is described by ,mr4 ,couthouy 78,remarks10 p4 #eh72 it lies #abj miles north-east of ,tahiti2 it is not coloured in the appended map1 because it does not appear to be fringed by living reefs4 ,mr4 ,couthouy describes its summit as 8presenting a broad table-land which declines a few feet towards the centre1 where we may suppose the lagoon to have been placed40 ,it is about two hundred feet in height1 and consists of reef-rock and conglomerate1 with existing species of coral embedded in it4 ,the island has been elevated at two successive periods2 the cliffs being marked halfway up with a horizontal water-worn line of deep excavations4 ,aurora ,island seems closely to resemble in structure ,elizabeth ,island1 at the southern end of the ,low ,archipelago4 ,it is an interesting fact1 that ,fais1 which1 from its composition1 form1 height1 and situation at the western end of the ,caroline ,archipelago1 one is strongly induced to believe existed before its upheaval as an atoll1 lies exactly in the prolongation of the curved line of the ,mariana group1 which we know to be a line of recent elevation4 ,i may add1 that ,elizabeth ,island1 in the southern part of the ,low ,archipelago1 which seems to have had the same kind of origin as ,fais1 lies near ,pitcairn ,island1 the only one in this part of the ocean which is high1 and at the same time not surrounded by an encircling barrier-reef4 ,on the absence of active volcanoes in the areas of subsidence1 and on their frequent presence in the areas of elevation4--,before making some concluding remarks on the relations of the spaces coloured blue and red1 it will be convenient to consider the position on our map of the volcanoes historically known to have been in action4 ,it is impossible not to be struck1 first with the absence of volcanoes in the great areas of subsidence tinted pale and dark blue1--namely1 in the central parts of the ,indian ,ocean1 in the ,china ,sea1 in the sea between the barriers of ,australia and ,new ,caledonia1 in the ,caroline1 ,marshall1 ,gilbert1 and ,low ,archipelagoes2 and1 secondly1 with the coincidence of the principal volcanic chains with the parts coloured red1 which indicates the presence of fringing-reefs2 and1 as we have just seen1 the presence in most cases of upraised organic remains of a modern date4 ,i may here remark that the reefs were all coloured before the volcanoes were added to the map1 or indeed before ,i knew of the existence of several of them4 ,the volcano in ,torres ,strait1 at the northern point of ,australia1 is that which lies nearest to a large subsiding area1 although situated #abe miles within the outer margin of the actual barrier-reef4 ,the ,great ,comoro ,island1 which probably contains a volcano1 is only twenty miles distant from the barrier-reef of ,mohila2 ,ambil volcano1 in the ,philippines1 is distant only a little more than sixty miles from the atoll-formed ,appoo reef3 and there are two other volcanoes in the map within ninety miles of circles coloured blue4 ,these few cases1 which thus offer partial exceptions to the rule1 of volcanoes being placed remote from the areas of subsidence1 lie either near single and isolated atolls1 or near small groups of encircled islands2 and these by our theory can have1 in few instances1 subsided to the same amount in depth or area1 as groups of atolls4 ,there is not one active volcano within several hundred miles of an archipelago1 or even a small group of atolls4 ,it is1 therefore1 a striking fact that in the ,friendly ,archipelago1 which owes its origin to the elevation of a group of atolls1 two volcanoes1 and1 perhaps1 others are known to be in action3 on the other hand1 on several of the encircled islands in the ,pacific1 supposed by our theory to have subsided1 there are old craters and streams of lava1 which show the effects of past and ancient eruptions4 ,in these cases1 it would appear as if the volcanoes had come into action1 and had become extinguished on the same spots1 according as the elevating or subsiding movements prevailed4 ,there are some other coasts on the map1 where volcanoes in a state of action concur with proofs of recent elevation1 besides those coloured red from being fringed by coral-reefs4 ,thus ,i hope to show in a future volume1 that nearly the whole line of the west coast of ,south ,america1 which forms the greatest volcanic chain in the world1 from near the equator for a space of between #b1jjj and #c1jjj miles southward1 has undergone an upward movement during a late geological period4 ,the islands on the north-western shores of the ,pacific1 which form the second greatest volcanic chain1 are very imperfectly known2 but ,luzon1 in the ,philippines1 and the ,loo ,choo ,islands1 have been recently elevated2 and at ,kamtschatka#cf there are extensive tertiary beds of modern date4 ,evidence of the same nature1 but not very satisfactory1 may be detected in ,northern ,new ,zealand where there are two volcanoes4 ,the co-existence in other parts of the world of active volcanoes1 with upraised beds of a modern tertiary origin1 will occur to every geologist4#cg ,nevertheless1 until it could be shown that volcanoes were inactive1 or did not exist in subsiding areas1 the conclusion that their distribution depended on the nature of the subterranean movements in progress1 would have been hazardous4 ,but now1 viewing the appended map1 it may1 ,i think1 be considered as almost established1 that volcanoes are often 7not necessarily always7 present in those areas where the subterranean motive power has lately forced1 or is now forcing outwards1 the crust of the earth1 but that they are invariably absent in those1 where the surface has lately subsided or is still subsiding4#ch #cf ,at ,sedanka1 in latitude #eh ,n4 7,von ,buch's 8,descrip4 des ,isles ,canaries10 p4 #dee74 ,in a forthcoming part1 ,i shall give the evidence referred to with respect to the elevation of ,new ,zealand4 #cg ,during the subterranean disturbances which took place in ,chile1 in #ahce1 ,i have shown 78,geolog4 ,trans40 #bnd ,ser41 vol4 v1 p4 #fjf7 that at the same moment that a large district was upraised1 volcanic matter burst forth at widely separated points1 through both new and old vents4 #ch ,we may infer from this rule1 that in any old deposit1 which contains interstratified beds of erupted matter1 there was at the period1 and in the area of its formation1 a tendency to an upward movement in the earth's surface1 and certainly no movement of subsidence4 ,on the relations of the areas of subsidence and elevation4--,the immense surfaces on the map1 which1 both by our theory and by the plain evidence of upraised marine remains1 have undergone a change of level either downwards or upwards during a late period1 is a most remarkable fact4 ,the existence of continents shows that the areas have been immense which at some period have been upraised2 in ,south ,america we may feel sure1 and on the north-western shores of the ,indian ,ocean we may suspect1 that this rising is either now actually in progress1 or has taken place quite recently4 ,by our theory1 we may conclude that the areas are likewise immense which have lately subsided1 or1 judging from the earthquakes occasionally felt and from other appearances1 are now subsiding4 ,the smallness of the scale of our map should not be overlooked3 each of the squares on it contains 7not allowing for the curvature of the earth7 #haj1jjj square miles4 ,look at the space of ocean from near the southern end of the ,low ,archipelago to the northern end of the ,marshall ,archipelago1 a length of #d1ejj miles1 in which1 as far as is known1 every island1 except ,aurora which lies just without the ,low ,archipelago1 is atoll-formed4 ,the eastern and western boundaries of our map are continents1 and they are rising areas3 the central spaces of the great ,indian and ,pacific ,oceans1 are mostly subsiding2 between them1 north of ,australia1 lies the most broken land on the globe1 and there the rising parts are surrounded and penetrated by areas of subsidence1#ci so that the prevailing movements now in progress1 seem to accord with the actual states of surface of the great divisions of the world4 #ci ,i suspect that the ,arru and ,timor-laut ,islands present an included small area of subsidence1 like that of the ,china ,sea1 but ,i have not ventured to colour them from my imperfect information1 as given in the ,appendix4 ,the blue spaces on the map are nearly all elongated2 but it does not necessarily follow from this 7a caution1 for which ,i am indebted to ,mr4 ,lyell71 that the areas of subsidence were likewise elongated2 for the subsidence of a long1 narrow space of the bed of the ocean1 including in it a transverse chain of mountains1 surmounted by atolls1 would only be marked on the map by a transverse blue band4 ,but where a chain of atolls and barrier-reefs lies in an elongated area1 between spaces coloured red1 which therefore have remained stationary or have been upraised1 this must have resulted either from the area of subsidence having originally been elongated 7owing to some tendency in the earth's crust thus to subside71 or from the subsiding area having originally been of an irregular figure1 or as broad as long1 and having since been narrowed by the elevation of neighbouring districts4 ,thus the areas1 which subsided during the formation of the great north and south lines of atolls in the ,indian ,ocean1--of the east and west line of the ,caroline atolls1--and of the north-west and south-east line of the barrier-reefs of ,new ,caledonia and ,louisiade1 must have originally been elongated1 or if not so1 they must have since been made elongated by elevations1 which we know to belong to a recent period4 ,i infer from ,mr4 ,hopkins' researches1#dj that for the formation of a long chain of mountains1 with few lateral spurs1 an area elongated in the same direction with the chain1 must have been subjected to an elevatory movement4 ,mountain-chains1 however1 when already formed1 although running in very different directions1 it seems#da may be raised together by a widely-acting force3 so1 perhaps1 mountain-chains may subside together4 ,hence1 we cannot tell1 whether the ,caroline and ,marshall ,archipelagoes1 two groups of atolls running in different directions and meeting each other1 have been formed by the subsidence of two areas1 or of one large area1 including two distinct lines of mountains4 ,we have1 however1 in the southern prolongation of the ,mariana ,islands1 probable evidence of a line of recent elevation having intersected one of recent subsidence4 ,a view of the map will show that1 generally1 there is a tendency to alternation in the parallel areas undergoing opposite kinds of movement2 as if the sinking of one area balanced the rising of another4 #dj 8,researches in ,physical ,geology10 ,transact4 ,cambridge ,phil4 ,soc41 vol4 vi1 part i4 #da ,for instance in ,s4 ,america from lat4 #cd1 for very many degrees southward there are upraised beds containing recent species of shells1 on both the ,atlantic and ,pacific side of the continent1 and from the gradual ascent of the land1 although with very unequal slopes1 on both sides towards the ,cordillera1 ,i think it can hardly be doubted that the entire width has been upraised in mass within the recent period4 ,in this case the two ,w4,n4,w4 and ,e4,s4,e4 mountain-lines1 namely the ,sierra ,ventana and the ,s4 ,tapalguen1 and the great north and south line of the ,cordillera have been together raised4 ,in the ,west ,indies the ,n4 and ,s4 line of the ,eastern ,antilles1 and the ,e4 and ,w4 line of ,jamaica1 appear both to have been upraised within the latest geological period4 ,the existence in many parts of the world of high table-land1 proves that large surfaces have been upraised in mass to considerable heights above the level of the ocean2 although the highest points in almost every country consist of upturned strata1 or erupted matter3 and from the immense spaces scattered with atolls1 which indicate that land originally existed there1 although not one pinnacle now remains above the level of the sea1 we may conclude that wide areas have subsided to an amount1 sufficient to bury not only any formerly existing table-land1 but even the heights formed by fractured strata1 and erupted matter4 ,the effects produced on the land by the later elevatory movements1 namely1 successively rising cliffs1 lines of erosion1 and beds of literal shells and pebbles1 all requiring time for their production1 prove that these movements have been very slow2 we can1 however1 infer this with safety1 only with respect to the few last hundred feet of rise4 ,but with reference to the whole vast amount of subsidence1 necessary to have produced the many atolls widely scattered over immense spaces1 it has already been shown 7and it is1 perhaps1 the most interesting conclusion in this volume71 that the movements must either have been uniform and exceedingly slow1 or have been effected by small steps1 separated from each other by long intervals of time1 during which the reef-constructing polypifers were able to bring up their solid frameworks to the surface4 ,we have little means of judging whether many considerable oscillations of level have generally occurred during the elevation of large tracts2 but we know1 from clear geological evidence1 that this has frequently taken place2 and we have seen on our map1 that some of the same islands have both subsided and been upraised4 ,i conclude1 however1 that most of the large blue spaces1 have subsided without many and great elevatory oscillations1 because only a few upraised atolls have been observed3 the supposition that such elevations have taken place1 but that the upraised parts have been worn down by the surf1 and thus have escaped observation1 is overruled by the very considerable depth of the lagoons of all the larger atolls2 for this could not have been the case1 if they had suffered repeated elevations and abrasion4 ,from the comparative observations made in these latter pages1 we may finally conclude1 that the subterranean changes which have caused some large areas to rise1 and others to subside1 have acted in a very similar manner4 ,recapitulation4--,in the three first chapters1 the principal kinds of coral-reefs were described in detail1 and they were found to differ little1 as far as relates to the actual surface of the reef4 ,an atoll differs from an encircling barrier-reef only in the absence of land within its central expanse2 and a barrier-reef differs from a fringing-reef1 in being placed at a much greater distance from the land with reference to the probable inclination of its submarine foundation1 and in the presence of a deep-water lagoon-like space or moat within the reef4 ,in the fourth chapter the growing powers of the reef-constructing polypifers were discussed2 and it was shown1 that they cannot flourish beneath a very limited depth4 ,in accordance with this limit1 there is no difficulty respecting the foundations on which fringing-reefs are based2 whereas1 with barrier-reefs and atolls1 there is a great apparent difficulty on this head2 in barrier-reefs from the improbability of the rock of the coast or of banks of sediment extending1 in every instance1 so far seaward within the required depth2--and in atolls1 from the immensity of the spaces over which they are interspersed1 and the apparent necessity for believing that they are all supported on mountain-summits1 which although rising very near to the surface-level of the sea1 in no one instance emerge above it4 ,to escape this latter most improbable admission1 which implies the existence of submarine chains of mountains of almost the same height1 extending over areas of many thousand square miles1 there is but one alternative2 namely1 the prolonged subsidence of the foundations1 on which the atolls were primarily based1 together with the upward growth of the reef-constructing corals4 ,on this view every difficulty vanishes2 fringing reefs are thus converted into barrier-reefs2 and barrier-reefs1 when encircling islands1 are thus converted into atolls1 the instant the last pinnacle of land sinks beneath the surface of the ocean4 ,thus the ordinary forms and certain peculiarities in the structure of atolls and barrier-reefs can be explained2--namely1 the wall-like structure on their inner sides1 the basin or ring-like shape both of the marginal and central reefs in the ,maldiva atolls--the union of some atolls as if by a ribbon--the apparent disseverment of others--and the occurrence1 in atolls as well as in barrier-reefs1 of portions of reef1 and of the whole of some reefs1 in a dead and submerged state1 but retaining the outline of living reefs4 ,thus can be explained the existence of breaches through barrier-reefs in front of valleys1 though separated from them by a wide space of deep water2 thus1 also1 the ordinary outline of groups of atolls and the relative forms of the separate atolls one to another2 thus can be explained the proximity of the two kinds of reefs formed during subsidence1 and their separation from the spaces where fringing-reefs abound4 ,on searching for other evidence of the movements supposed by our theory1 we find marks of change in atolls and in barrier-reefs1 and of subterranean disturbances under them2 but from the nature of things1 it is scarcely possible to detect any direct proofs of subsidence1 although some appearances are strongly in favour of it4 ,on the fringed coasts1 however1 the presence of upraised marine bodies of a recent epoch1 plainly show1 that these coasts1 instead of having remained stationary1 which is all that can be directly inferred from our theory1 have generally been elevated4 ,finally1 when the two great types of structure1 namely barrier-reefs and atolls on the one hand1 and fringing-reefs on the other1 were laid down in colours on our map1 a magnificent and harmonious picture of the movements1 which the crust of the earth has within a late period undergone1 is presented to us4 ,we there see vast areas rising1 with volcanic matter every now and then bursting forth through the vents or fissures with which they are traversed4 ,we see other wide spaces slowly sinking without any volcanic outburst1 and we may feel sure1 that this sinking must have been immense in amount as well as in area1 thus to have buried over the broad face of the ocean every one of those mountains1 above which atolls now stand like monuments1 marking the place of their former existence4 ,reflecting how powerful an agent with respect to denudation1 and consequently to the nature and thickness of the deposits in accumulation1 the sea must ever be1 when acting for prolonged periods on the land1 during either its slow emergence or subsidence2 reflecting1 also1 on the final effects of these movements in the interchange of land and ocean-water on the climate of the earth1 and on the distribution of organic beings1 ,i may be permitted to hope1 that the conclusions derived from the study of coral-formations1 originally attempted merely to explain their peculiar forms1 may be thought worthy of the attention of geologists4 ,,appendix4 ,,containing ,a ,,detailed ,,description ,,of ,,the ,,reefs ,,and ,,islands ,,in ,,plate ,,iii4 ,in the beginning of the last chapter ,i stated the principles on which the map is coloured4 ,there only remains to be said1 that it is an exact copy of one by ,m4 ,c4 ,gressier1 published by the ,dpt ,gnral de la ,marine1 in #ahce4 ,the names have been altered into ,english1 and the longitude has been reduced to that of ,greenwich4 ,the colours were first laid down on accurate charts1 on a large scale4 ,the data1 on which the volcanoes historically known to have been in action1 have been marked with vermillion1 were given in a note to the last chapter4 ,i will commence my description on the eastern side of the map1 and will describe each group of islands consecutively1 proceeding westward across the ,pacific and ,indian ,oceans1 but ending with the ,west ,indies4 ,the ,,western ,,shores ,,of ,,america appear to be entirely without coral-reefs2 south of the equator the survey of the ,beagle1 and north of it1 the published charts show that this is the case4 ,even in the ,bay of ,panama1 where corals flourish1 there are no true coral-reefs1 as ,i have been informed by ,mr4 ,lloyd4 ,there are no coral-reefs in the ,galapagos ,archipelago1 as ,i know from personal inspection2 and ,i believe there are none on the ,cocos1 ,revilla-gigedo1 and other neighbouring islands4 ,clipperton rock1 #aj ,n41 #aji ,w41 has lately been surveyed by ,captain ,belcher2 in form it is like the crater of a volcano4 ,from a drawing appended to the ,,ms4 plan in the ,admiralty1 it evidently is not an atoll4 ,the eastern parts of the ,pacific present an enormous area1 without any islands1 except ,e1 and ,sala1 and ,gomez ,islands1 which do not appear to be surrounded by reefs4 ,the ,,low ,,archipelago4--,this group consists of about eighty atolls3 it will be quite superfluous to refer to descriptions of each4 ,in ,d',urville and ,lottin's chart1 one island 7,wolchonsky7 is written with a capital letter1 signifying1 as explained in a former chapter1 that it is a high island2 but this must be a mistake1 as the original chart by ,bellinghausen shows that it is a true atoll4 ,captain ,beechey says of the thirty-two groups which he examined 7of the greater number of which ,i have seen beautiful ,,ms4 charts in the ,admiralty71 that twenty-nine now contain lagoons1 and he believes the other three originally did4 ,bellinghausen 7see an account of his ,russian voyage1 in the 8,biblioth4 des ,voyages10 #ahcd1 p4 #ddc7 says1 that the seventeen islands which he discovered resembled each other in structure1 and he has given charts on a large scale of all of them4 ,kotzebue has given plans of several2 ,cook and ,bligh mention others2 a few were seen during the voyage of the ,beagle2 and notices of other atolls are scattered through several publications4 ,the ,acton group in this archipelago has lately been discovered 7,geograph4 ,journ41 vol4 vii1 p4 #ded72 it consists of three small and low islets1 one of which has a lagoon4 ,another lagoon-island has been discovered 7,naut4 ,mag41 #ahci1 p4 #ggj71 in #bb #d ,s41 and #acf #bj ,w4 ,towards the ,s4,e4 part of the group1 there are some islands of different formation3 ,elizabeth ,island is described by ,beechey 7p4 #df1 #dto ed47 as fringed by reefs1 at the distance of between two and three hundred yards2 coloured red4 ,pitcairn ,island1 in the immediate neighbourhood1 according to the same authority1 has no reefs of any kind1 although numerous pieces of coral are thrown up on the beach2 the sea close to its shore is very deep 7see 8,zool4 of ,beechey's ,voyage10 p4 #afd72 it is left uncoloured4 ,gambier ,islands 7see ,plate ,i ,fig4 #h71 are encircled by a barrier-reef2 the greatest depth within is thirty-eight fathoms2 coloured pale blue4 ,aurora ,island1 which lies ,n4,e4 of ,tahiti close to the large space coloured dark blue in the map1 has been already described in a note 771 on the authority of ,mr4 ,couthouy2 it is an upraised atoll1 but as it does not appear to be fringed by living reefs1 it is left uncoloured4 ,the ,,society ,arch4 is separated by a narrow space from the ,low ,archipelago2 and in their parallel direction they manifest some relation to each other4 ,i have already described the general character of the reefs of these fine encircled islands4 ,in the 8,atlas of the ,coquille's ,voyage0 there is a good general chart of the group1 and separate plans of some of the islands4 ,tahiti1 the largest island in the group1 is almost surrounded1 as seen in ,cook's chart1 by a reef from half a mile to a mile and a half from the shore1 with from ten to thirty fathoms within it4 ,some considerable submerged reefs lying parallel to the shore1 with a broad and deep space within1 have lately been discovered 7,naut4 ,mag41 #ahcf1 p4 #bfd7 on the ,n4,e4 coast of the island1 where none are laid down by ,cook4 ,at ,eimeo the reef 8which like a ring surrounds it1 is in some places one or two miles distant from the shore1 in others united to the beach0 7,ellis1 8,polynesian ,researches10 vol4 i1 p4 #ah1 #abmo edition74 ,cook found deep water 7twenty fathoms7 in some of the harbours within the reef4 ,mr4 ,couthouy1 however1 states 78,remarks10 p4 #de7 that both at ,tahiti and ,eimeo1 the space between the barrier-reef and the shore1 has been almost filled up1--8a nearly continuous fringing-reef surrounding the island1 and varying from a few yards to rather more than a mile in width1 the lagoons merely forming canals between this and the sea-reef10 that is the barrier-reef4 ,tapamanoa is surrounded by a reef at a considerable distance from the shore2 from the island being small it is breached1 as ,i am informed by the ,rev4 ,w4 ,ellis1 only by a narrow and crooked boat channel4 ,this is the lowest island in the group1 its height probably not exceeding #ejj feet4 ,a little way north of ,tahiti1 the low coral-islets of ,teturoa are situated2 from the description of them given me by the ,rev4 ,j4 ,williams 7the author of the 8,narrative of ,missionary ,enterprise071 ,i should have thought they had formed a small atoll1 and likewise from the description given by the ,rev4 ,d4 ,tyerman and ,g4 ,bennett 78,journal of ,voyage and ,travels10 vol4 i1 p4 #ahc71 who say that ten low coral-islets 8are comprehended within one general reef1 and separated from each other by interjacent lagoons20 but as ,mr4 ,stutchbury 7,west of ,england ,journal1 vol4 i1 p4 #ed7 describes it as consisting of a mere narrow ridge1 ,i have left it uncoloured4 ,maitea1 eastward of the group1 is classed by ,forster as a high encircled island2 but from the account given by the ,rev4 ,d4 ,tyerman and ,g4 ,bennett 7vol4 i1 p4 #eg7 it appears to be an exceedingly abrupt cone1 rising from the sea without any reef2 ,i have left it uncoloured4 ,it would be superfluous to describe the northern islands in this group1 as they may be well seen in the chart accompanying the #dto edition of ,cook's 8,voyages10 and in the 8,atlas of the ,coquille's ,voyage40 ,maurua is the only one of the northern islands1 in which the water within the reef is not deep1 being only four and a half fathoms2 but the great width of the reef1 stretching three miles and a half southward of the land 7which is represented in the drawing in the 8,atlas of the ,coquille's ,voyage0 as descending abruptly to the water7 shows1 on the principle explained in the beginning of the last chapter1 that it belongs to the barrier class4 ,i may here mention1 from information communicated to me by the ,rev4 ,w4 ,ellis1 that on the ,n4,e4 side of ,huaheine there is a bank of sand1 about a quarter of a mile wide1 extending parallel to the shore1 and separated from it by an extensive and deep lagoon2 this bank of sand rests on coral-rock1 and undoubtedly was originally a living reef4 ,north of ,bolabola lies the atoll of ,toubai 7,motou-iti of the 8,coquille's ,atlas07 which is coloured dark blue2 the other islands1 surrounded by barrier-reefs1 are pale blue2 three of them are represented in ,figs #c1 #d1 and #e1 in ,plate ,i4 ,there are three low coral-groups lying a little ,e4 of the ,society ,archipelago1 and almost forming part of it1 namely ,bellinghausen1 which is said by ,kotzebue 78,second ,voyage10 vol4 ii1 p4 #bee71 to be a lagoon-island2 ,mopeha1 which1 from ,cook's description 78,second ,voyage10 book iii1 chap4 i71 no doubt is an atoll2 and the ,scilly ,islands1 which are said by ,wallis 78,voyage10 chap4 ix7 to form a group of low islets and shoals1 and1 therefore1 probably1 they compose an atoll3 the two former have been coloured blue1 but not the latter4 ,,mendana ,,or ,,marquesas ,group4--,these islands are entirely without reefs1 as may be seen in ,krusenstern's ,atlas1 making a remarkable contrast with the adjacent group of the ,society ,islands4 ,mr4 ,f4 ,d4 ,bennett has given some account of this group1 in the seventh volume of the ,geograph4 ,journ4 ,he informs me that all the islands have the same general character1 and that the water is very deep close to their shores4 ,he visited three of them1 namely1 ,dominicana1 ,christiana1 and ,roapoa2 their beaches are strewed with rounded masses of coral1 and although no regular reefs exist1 yet the shore is in many places lined by coral-rock1 so that a boat grounds on this formation4 ,hence these islands ought probably to come within the class of fringed islands and be coloured red2 but as ,i am determined to err on the cautious side1 ,i have left them uncoloured4 ,,cook or ,,harvey and ,,austral ,,island4--,palmerston ,island is minutely described as an atoll by ,captain ,cook during his voyage in #aggd2 coloured blue4 ,aitutaki was partially surveyed by the ,beagle 7see map accompanying 8,voyages of ,adventure and ,beagle072 the land is hilly1 sloping gently to the beach2 the highest point is #cfj feet2 on the southern side the reef projects five miles from the land3 off this point the ,beagle found no bottom with #bgj fathoms3 the reef is surmounted by many low coral-islets4 ,although within the reef the water is exceedingly shallow1 not being more than a few feet deep1 as ,i am informed by the ,rev4 ,j4 ,williams1 nevertheless1 from the great extension of this reef into a profoundly deep ocean1 this island probably belongs1 on the principle lately adverted to1 to the barrier class1 and ,i have coloured it pale blue2 although with much hesitation4--,manouai or ,harvey ,island4 ,the highest point is about fifty feet3 the ,rev4 ,j4 ,williams informs me that the reef here1 although it lies far from the shore1 is less distant than at ,aitutaki1 but the water within the reef is rather deeper3 ,i have also coloured this pale blue with many doubts4--,round ,mitiaro ,island1 as ,i am informed by ,mr4 ,williams1 the reef is attached to the shore2 coloured red4--,mauki or ,maouti2 the reef round this island 7under the name of ,parry ,island1 in the 8,voyage of ,h4,m4,s4 ,blonde10 p4 #bji7 is described as a coral-flat1 only fifty yards wide1 and two feet under water4 ,this statement has been corroborated by ,mr4 ,williams1 who calls the reef attached2 coloured red4--,aitu1 or ,wateeo2 a moderately elevated hilly island1 like the others of this group4 ,the reef is described in ,cook's 8,voyage10 as attached to the shore1 and about one hundred yards wide2 coloured red4--,fenoua-iti2 ,cook describes this island as very low1 not more than six or seven feet high 7vol4 i1 book ii1 chap4 iii1 #aggg72 in the chart published in the 8,coquille's ,atlas10 a reef is engraved close to the shore3 this island is not mentioned in the list given by ,mr4 ,williams 7page #af7 in the 8,narrative of ,missionary ,enterprise20 nature doubtful4 ,as it is so near ,atiu1 it has been unavoidably coloured red4--,rarotonga2 ,mr4 ,williams informs me that it is a lofty basaltic island with an attached reef2 coloured red4--,there are three islands1 ,rourouti1 ,roxburgh1 and ,hull1 of which ,i have not been able to obtain any account1 and have left them uncoloured4 ,hull ,island1 in the ,french chart1 is written with small letters as being low4--,mangaia2 height about three hundred feet2 8the surrounding reef joins the shore0 7,williams1 8,narrative10 p4 #ah72 coloured red4--,rimetara2 ,mr4 ,williams informs me that the reef is rather close to the shore2 but1 from information given me by ,mr4 ,ellis1 the reef does not appear to be quite so closely attached to it as in the foregoing cases3 the island is about three hundred feet high 7,naut4 ,mag41 #ahci1 p4 #gch72 coloured red4--,rurutu2 ,mr4 ,williams and ,mr4 ,ellis inform me that this island has an attached reef2 coloured red4 ,it is described by ,cook under the name of ,oheteroa3 he says it is not surrounded1 like the neighbouring islands by a reef2 he must have meant a distant reef4--,toubouai2 in ,cook's chart 78,second ,voyage10 vol4 ii1 p4 #b7 the reef is laid down in part one mile1 and in part two miles from the shore4 ,mr4 ,ellis 78,polynes4 ,res40 vol4 iii1 p4 #cha7 says the low land round the base of the island is very extensive2 and this gentleman informs me that the water within the reef appears deep2 coloured blue4--,raivaivai1 or ,vivitao2 ,mr4 ,williams informs me that the reef is here distant3 ,mr4 ,ellis1 however1 says that this is certainly not the case on one side of the island2 and he believes that the water within the reef is not deep2 hence ,i have left it uncoloured4--,lancaster ,reef1 described in ,naut4 ,mag41 #ahcc 7p4 #fic71 as an extensive crescent-formed coral-reef4 ,i have not coloured it4--,rapa1 or ,oparree2 from the accounts given of it by ,ellis and ,vancouver1 there does not appear to be any reef4--,i4 de ,bass is an adjoining island1 of which ,i cannot find any account4--,kemin ,island2 ,krusenstern seems hardly to know its position1 and gives no further particulars4 ,,islands ,,between the ,low and ,gilbert ,archipelagoes4 ,caroline ,island 7#aj ,s41 #aej deg ,w47 is described by ,mr4 ,f4 ,d4 ,bennett 7,geograph4 ,journ41 vol4 vii1 p4 #bbe7 as containing a fine lagoon2 coloured blue4--,flint ,island 7#aa ,s41 #aea ,w472 ,krusenstern believes that it is the same with ,peregrino1 which is described by ,quiros 7,burney's 8,chron4 ,hist410 vol4 ii1 p4 #bhc7 as 8a cluster of small islands connected by a reef1 and forming a lagoon in the middle20 coloured blue4--,wostock is an island a little more than half a mile in diameter1 and apparently quite flat and low1 and was discovered by ,bellinghausen2 it is situated a little west of ,caroline ,island1 but it is not placed on the ,french charts2 ,i have not coloured it1 although ,i entertain little doubt from the chart of ,bellinghausen1 that it originally contained a small lagoon4--,penrhyn ,island 7#i ,s41 #aeh ,w472 a plan of it in the 8,atlas of the ,first ,voyage0 of ,kotzebue1 shows that it is an atoll2 blue4--,slarbuck ,island 7#e ,s41 #aef ,w47 is described in ,byron's 8,voyage in the ,blonde0 7p4 #bjf7 as formed of a flat coral-rock1 with no trees2 the height not given2 not coloured4--,malden ,island 7#d ,s41 #aed ,w472 in the same voyage 7p4 #bje7 this island is said to be of coral formation1 and no part above forty feet high2 ,i have not ventured to colour it1 although1 from being of coral-formation1 it is probably fringed2 in which case it should be red4--,jarvis1 or ,bunker ,island 7#j #bj ,s41 #afj ,w47 is described by ,mr4 ,f4 ,d4 ,bennett 7,geograph4 ,journ41 vol4 vii1 p4 #bbg7 as a narrow1 low strip of coral-formation2 not coloured4--,brook1 is a small low island between the two latter2 the position1 and perhaps even the existence of it is doubtful2 not coloured4--,pescado and ,humphrey ,islands2 ,i can find out nothing about these islands1 except that the latter appears to be small and low2 not coloured4--,rearson1 or ,grand ,duke ,alexander's 7#aj ,s41 #afa ,w472 an atoll1 of which a plan is given by ,bellinghausen2 blue4--,souvoroff ,islands 7#ac ,s41 #afc ,w472 ,admiral ,krusenstern1 in the most obliging manner1 obtained for me an account of these islands from ,admiral ,lazareff1 who discovered them4 ,they consist of five very low islands of coral-formation1 two of which are connected by a reef1 with deep water close to it4 ,they do not surround a lagoon1 but are so placed that a line drawn through them includes an oval space1 part of which is shallow2 these islets1 therefore1 probably once 7as is the case with some of the islands in the ,caroline ,archipelago7 formed a single atoll2 but ,i have not coloured them4--,danger ,island 7#aj ,s41 #aff ,w472 described as low by ,commodore ,byron1 and more lately surveyed by ,bellinghausen2 it is a small atoll with three islets on it2 blue4--,clarence ,island 7#i ,s41 #agb ,w472 discovered in the ,pandora 7,g4 ,hamilton's 8,voyage10 p4 #ge73 it is said1 8in running along the land1 we saw several canoes crossing the lagoons20 as this island is in the close vicinity of other low islands1 and as it is said1 that the natives make reservoirs of water in old cocoa-nut trees 7which shows the nature of the land71 ,i have no doubt it is an atoll1 and have coloured it blue4 ,york ,island 7#h ,s41 #agb ,w47 is described by ,commodore ,byron 7chap4 x of his 8,voyage07 as an atoll2 blue4--,sydney ,island 7#d ,s41 #agb ,w47 is about three miles in diameter1 with its interior occupied by a lagoon 7,captain ,tromelin1 8,annal4 ,marit40 #ahbi1 p4 #big72 blue4--,phoenix ,island 7#d ,s41 #aga ,w47 is nearly circular1 low1 sandy1 not more than two miles in diameter1 and very steep outside 7,tromelin1 8,annal4 ,marit40 #ahbi1 p4 #big72 it may be inferred that this island originally contained a lagoon1 but ,i have not coloured it4--,new ,nantucket 7#j #ae ,n41 #agd ,w474 ,from the ,french chart it must be a low island2 ,i can find nothing more about it or about ,mary ,island2 both uncoloured4--,gardner ,island 7#e ,s41 #agd ,w47 from its position is certainly the same as ,kemin ,island described 7,krusenstern1 p4 #dce1 ,appen4 to ,mem41 published #ahbg7 as having a lagoon in its centre2 blue4 ,,islands ,,south of the ,sandwich ,archipelago4 ,christmas ,island 7#b ,n41 #aeg ,w474 ,captain ,cook1 in his 8,third ,voyage0 7vol4 ii1 chap4 x71 has given a detailed account of this atoll4 ,the breadth of the islets on the reef is unusually great1 and the sea near it does not deepen so suddenly as is generally the case4 ,it has more lately been visited by ,mr4 ,f4 ,d4 ,bennett 7,geograph4 ,journ41 vol4 vii1 p4 #bbf72 and he assures me that it is low and of coral-formation3 ,i particularly mention this1 because it is engraved with a capital letter1 signifying a high island1 in ,d',urville and ,lottin's chart4 ,mr4 ,couthouy1 also1 has given some account of it 78,remarks10 p4 #df7 from the ,hawaiian 8,spectator02 he believes it has lately undergone a small elevation1 but his evidence does not appear to me satisfactory2 the deepest part of the lagoon is said to be only ten feet2 nevertheless1 ,i have coloured it blue4--,fanning ,island 7#d ,n41 #aeh ,w47 according to ,captain ,tromelin 78,ann4 ,maritim410 #ahbi1 p4 #bhc71 is an atoll3 his account as observed by ,krusenstern1 differs from that given in ,fanning's 8,voyage0 7p4 #bbd71 which1 however1 is far from clear2 coloured blue4--,washington ,island 7#d ,n41 #aei ,w47 is engraved as a low island in ,d',urville's chart1 but is described by ,fanning 7p4 #bbf7 as having a much greater elevation than ,fanning ,island1 and hence ,i presume it is not an atoll2 not coloured4--,palmyra ,island 7#f ,n41 #afb ,w47 is an atoll divided into two parts 7,krusenstern's 8,mem4 ,suppl410 p4 #ej1 also ,fanning's 8,voyage10 p4 #bcc72 blue4--,smyth's or ,johnston's ,islands 7#ag ,n41 #agj ,w474 ,captain ,smyth1 ,r4,n41 has had the kindness to inform me that they consist of two very low1 small islands1 with a dangerous reef off the east end of them4 ,captain ,smyth does not recollect whether these islets1 together with the reef1 surrounded a lagoon2 uncoloured4 ,,sandwich ,,archipelago4--,hawaii2 in the chart in ,freycinet's 8,atlas10 small portions of the coast are fringed by reefs2 and in the accompanying 8,hydrog4 ,memoir10 reefs are mentioned in several places1 and the coral is said to injure the cables4 ,on one side of the islet of ,kohaihai there is a bank of sand and coral with five feet water on it1 running parallel to the shore1 and leaving a channel of about fifteen feet deep within4 ,i have coloured this island red1 but it is very much less perfectly fringed than others of the group4--,maui2 in ,freycinet's chart of the anchorage of ,raheina1 two or three miles of coast are seen to be fringed2 and in the 8,hydrog4 ,memoir10 8banks of coral along shore0 are spoken of4 ,mr4 ,f4 ,d4 ,bennett informs me that the reefs1 on an average1 extend about a quarter of a mile from the beach2 the land is not very steep1 and outside the reefs the sea does not become deep very suddenly2 coloured red4--,morotoi1 ,i presume1 is fringed3 ,freycinet speaks of the breakers extending along the shore at a little distance from it4 ,from the chart1 ,i believe it is fringed2 coloured red4--,oahu2 ,freycinet1 in his 8,hydrog4 ,memoir10 mentions some of the reefs4 ,mr4 ,f4 ,d4 ,bennett informs me that the shore is skirted for forty or fifty miles in length4 ,there is even a harbour for ships formed by the reefs1 but it is at the mouth of a valley2 red4--,atooi1 in ,la ,peyrouse's charts1 is represented as fringed by a reef1 in the same manner as ,oahu and ,morotoi2 and this1 as ,i have been informed by ,mr4 ,ellis1 on part at least of the shore1 is of coral-formation3 the reef does not leave a deep channel within2 red4--,oneehow2 ,mr4 ,ellis believes that this island is also fringed by a coral-reef3 considering its close proximity to the other islands1 ,i have ventured to colour it red4 ,i have in vain consulted the works of ,cook1 ,vancouver1 ,la ,peyrouse1 and ,lisiansky1 for any satisfactory account of the small islands and reefs1 which lie scattered in a ,n4,w4 line prolonged from the ,sandwich group1 and hence have left them uncoloured1 with one exception2 for ,i am indebted to ,mr4 ,f4 ,d4 ,bennett for informing me of an atoll-formed reef1 in latitude #bh #bb1 longitude #agh #cj ,w41 on which the ,gledstanes was wrecked in #ahcg4 ,it is apparently of large size1 and extends in a ,n4,w4 and ,s4,e4 line3 very few islets have been formed on it4 ,the lagoon seems to be shallow2 at least1 the deepest part which was surveyed was only three fathoms4 ,mr4 ,couthouy 78,remarks10 p4 #ch7 describes this island under the name of ,ocean island4 ,considerable doubts should be entertained regarding the nature of a reef of this kind1 with a very shallow lagoon1 and standing far from any other atoll1 on account of the possibility of a crater or flat bank of rock lying at the proper depth beneath the surface of the water1 thus affording a foundation for a ring-formed coral-reef4 ,i have1 however1 thought myself compelled1 from its large size and symmetrical outline1 to colour it blue4 ,,samoa or ,,navigator ,,group4--,kotzebue1 in his 8,second ,voyage10 contrasts the structure of these islands with many others in the ,pacific1 in not being furnished with harbours for ships1 formed by distant coral-reefs4 ,the ,rev4 ,j4 ,williams1 however1 informs me1 that coral-reefs do occur in irregular patches on the shores of these islands2 but that they do not form a continuous band1 as round ,mangaia1 and other such perfect cases of fringed islands4 ,from the charts accompanying ,la ,peyrouse's 8,voyage10 it appears that the north shore of ,savaii1 ,maouna1 ,orosenga1 and ,manua1 are fringed by reefs4 ,la ,peyrouse1 speaking of ,maouna 7p4 #abf71 says that the coral-reef surrounding its shores1 almost touches the beach2 and is breached in front of the little coves and streams1 forming passages for canoes1 and probably even for boats4 ,further on 7p4 #aei71 he extends the same observation to all the islands which he visited4 ,mr4 ,williams in his 8,narrative10 speaks of a reef going round a small island attached to ,oyolava1 and returning again to it3 all these islands have been coloured red4--,a chart of ,rose ,island1 at the extreme west end of the group1 is given by ,freycinet1 from which ,i should have thought that it had been an atoll2 but according to ,mr4 ,couthouy 78,remarks10 p4 #dc71 it consists of a reef1 only a league in circuit1 surmounted by a very few low islets2 the lagoon is very shallow1 and is strewed with numerous large boulders of volcanic rock4 ,this island1 therefore1 probably consists of a bank of rock1 a few feet submerged1 with the outer margin of its upper surface fringed with reefs2 hence it cannot be properly classed with atolls1 in which the foundations are always supposed to lie at a depth1 greater than that at which the reef-constructing polypifers can live2 not coloured4 ,beveridge ,reef1 #bj ,s41 #afg ,w41 is described in the ,naut4 ,mag4 7,may #ahcc1 p4 #ddb7 as ten miles long in a ,n4 and ,s4 line1 and eight wide2 8in the inside of the reef there appears deep water20 there is a passage near the ,s4,w4 corner3 this therefore seems to be a submerged atoll1 and is coloured blue4 ,savage ,island1 #ai ,s41 #agj ,w41 has been described by ,cook and ,forster4 ,the younger ,forster 7vol4 ii1 p4 #afc7 says it is about forty feet high3 he suspects that it contains a low plain1 which formerly was the lagoon4 ,the ,rev4 ,j4 ,williams informs me that the reef fringing its shores1 resembles that round ,mangaia2 coloured red4 ,,friendly ,,archipelago4--,pylstaart ,island4 ,judging from the chart in ,freycinet's 8,atlas10 ,i should have supposed that it had been regularly fringed2 but as nothing is said in the 8,hydrog4 ,memoir0 7or in the 8,voyage0 of ,tasman1 the discoverer7 about coral-reefs1 ,i have left it uncoloured4--,tongatabou3 ,in the 8,atlas of the ,voyage of the ,astrolabe10 the whole south side of the island is represented as narrowly fringed by the same reef which forms an extensive platform on the northern side4 ,the origin of this latter reef1 which might have been mistaken for a barrier-reef1 has already been attempted to be explained1 when giving the proofs of the recent elevation of this island4--,in ,cook's charts the little outlying island also of ,eoaigee1 is represented as fringed2 coloured red4--,eoua4 ,i cannot make out from ,captain ,cook's charts and descriptions1 that this island has any reef1 although the bottom of the neighbouring sea seems to be corally1 and the island itself is formed of coral-rock4 ,forster1 however1 distinctly 78,observations10 p4 #ad7 classes it with high islands having reefs1 but it certainly is not encircled by a barrier-reef and the younger ,forster 78,voyage10 vol4 i1 p4 #dbf7 says1 that 8a bed of coral-rocks surrounded the coast towards the landing-place40 ,i have therefore classed it with the fringed islands and coloured it red4 ,the several islands lying ,n4,w4 of ,tongatabou1 namely ,anamouka1 ,komango1 ,kotou1 ,lefouga1 ,foa1 etc41 are seen in ,captain ,cook's chart to be fringed by reefs1 in several of them are connected together4 ,from the various statements in the first volume of ,cook's 8,third ,voyage10 and especially in the fourth and sixth chapters1 it appears that these reefs are of coral-formation1 and certainly do not belong to the barrier class2 coloured red4--,toufoa and ,kao1 forming the western part of the group1 according to ,forster have no reefs2 the former is an active volcano4--,vavao4 ,there is a chart of this singularly formed island1 by ,espinoza3 according to ,mr4 ,williams it consists of coral-rock3 the ,chevalier ,dillon informs me that it is not fringed2 not coloured4 ,nor are the islands of ,latte and ,amargura1 for ,i have not seen plans on a large scale of them1 and do not know whether they are fringed4 ,niouha1 #af ,s41 #agd ,w41 or ,keppel ,island of ,wallis1 or ,cocos ,island4 ,from a view and chart of this island given in ,wallis's 8,voyage0 7#dto ed47 it is evidently encircled by a reef2 coloured blue3 it is however remarkable that ,boscawen ,island1 immediately adjoining1 has no reef of any kind2 uncoloured4 ,wallis ,island1 #ac ,s41 #agf ,w41 a chart and view of this island in ,wallis's 8,voyage0 7#dto ed47 shows that it is encircled4 ,a view of it in the ,naut4 ,mag41 ,july #ahcc1 p4 #cgf1 shows the same fact2 blue4 ,alloufatou1 or ,horn ,island1 ,onouafu1 or ,proby ,island1 and ,hunter ,islands1 lie between the ,navigator and ,fidji groups4 ,i can find no distinct accounts of them4 ,,fidji or ,,viti ,,group4--,the best chart of the numerous islands of this group1 will be found in the 8,atlas of the ,astrolabe's ,voyage40 ,from this1 and from the description given in the 8,hydrog4 ,memoir10 accompanying it1 it appears that many of these islands are bold and mountainous1 rising to the height of between #c1jjj and #d1jjj feet4 ,most of the islands are surrounded by reefs1 lying far from the land1 and outside of which the ocean appears very deep4 ,the ,astrolabe sounded with ninety fathoms in several places about a mile from the reefs1 and found no bottom4 ,although the depth within the reef is not laid down1 it is evident from several expressions1 that ,captain ,d',urville believes that ships could anchor within1 if passages existed through the outer barriers4 ,the ,chevallier ,dillon informs me that this is the case3 hence ,i have coloured this group blue4 ,in the ,s4,e4 part lies ,batoa1 or ,turtle ,island of ,cook 78,second ,voyage10 vol4 ii1 p4 #bc1 and chart1 #dto ed47 surrounded by a coral-reef1 8which in some places extends two miles from the shore20 within the reef the water appears to be deep1 and outside it is unfathomable2 coloured pale blue4 ,at the distance of a few miles1 ,captain ,cook 7,ibid41 p4 #bd7 found a circular coral-reef1 four or five leagues in circuit1 with deep water within2 8in short1 the bank wants only a few little islets to make it exactly like one of the half-drowned isles so often mentioned10--namely1 atolls4 ,south of ,batoa1 lies the high island of ,ono1 which appears in ,bellinghausen's 8,atlas0 to be encircled2 as do some other small islands to the south2 coloured pale blue2 near ,ono1 there is an annular reef1 quite similar to the one just described in the words of ,captain ,cook2 coloured dark blue4 ,rotoumah1 #ac ,s41 #agi ,e4--,from the chart in ,duperrey's 8,atlas10 ,i thought this island was encircled1 and had coloured it blue1 but the ,chevallier ,dillon assures me that the reef is only a shore or fringing one2 red4 ,independence ,island1 #aj ,s41 #agi ,e41 is described by ,mr4 ,g4 ,bennett1 7,united ,service ,journ41 #ahca1 part ii1 p4 #aig7 as a low island of coral-formation1 it is small1 and does not appear to contain a lagoon1 although an opening through the reef is referred to4 ,a lagoon probably once existed1 and has since been filled up2 left uncoloured4 ,,ellice ,,group4--,oscar1 ,peyster1 and ,ellice ,islands are figured in ,arrowsmith's 8,chart of the ,pacific0 7corrected to #ahcb7 as atolls1 and are said to be very low2 blue4--,nederlandisch ,island4 ,i am greatly indebted to the kindness of ,admiral ,krusenstern1 for sending me the original documents concerning this island4 ,from the plans given by ,captains ,eeg and ,khremtshenko1 and from the detailed account given by the former1 it appears that it is a narrow coral-island1 about two miles long1 containing a small lagoon4 ,the sea is very deep close to the shore1 which is fronted by sharp coral-rocks4 ,captain ,eeg compares the lagoon with that of other coral-islands2 and he distinctly says1 the land is 8very low40 ,i have therefore coloured it blue4 ,admiral ,krusenstern 78,memoir on the ,pacific10 ,append41 #ahce7 states that its shores are eighty feet high2 this probably arose from the height of the cocoa-nut trees1 with which it is covered1 being mistaken for land4--,gran ,cocal is said in ,krusenstern's 8,memoir10 to be low1 and to be surrounded by a reef2 it is small1 and therefore probably once contained a lagoon2 uncoloured4--,st4 ,augustin4 ,from a chart and view of it1 given in the 8,atlas of the ,coquille's ,voyage10 it appears to be a small atoll1 with its lagoon partly filled up2 coloured blue4 ,,gilbert ,,group4--,the chart of this group1 given in the 8,atlas of the ,coquille's ,voyage10 at once shows that it is composed of ten well characterised atolls4 ,in ,d',urville and ,lottin's chart1 ,sydenham is written with a capital letter1 signifying that it is high2 but this certainly is not the case1 for it is a perfectly characterised atoll1 and a sketch1 showing how low it is1 is given in the 8,coquille's ,atlas40 ,some narrow strip-like reefs project from the southern side of ,drummond atoll1 and render it irregular4 ,the southern island of the group is called ,chase 7in some charts1 ,rotches72 of this ,i can find no account1 but ,mr4 ,f4 ,d4 ,bennett discovered 7,geograph4 ,journ41 vol4 vii1 p4 #bbi71 a low extensive island in nearly the same latitude1 about three degrees westward of the longitude assigned to ,rotches1 but very probably it is the same island4 ,mr4 ,bennett informs me that the man at the masthead reported an appearance of lagoon-water in the centre2 and1 therefore1 considering its position1 ,i have coloured it blue4--,pitt ,island1 at the extreme northern point of the group1 is left uncoloured1 as its exact position and nature is not known4--,byron ,island1 which lies a little to the eastward1 does not appear to have been visited since ,commodore ,byron's voyage1 and it was then seen only from a distance of eighteen miles2 it is said to be low2 uncoloured4 ,ocean1 ,pleasant1 and ,atlantic ,islands all lie considerably to the west of the ,gilbert group3 ,i have been unable to find any distinct account of them4 ,ocean ,island is written with small letters in the ,french chart1 but in ,krusenstern's 8,memoir0 it is said to be high4 ,,marshall ,,group4--,we are well acquainted with this group from the excellent charts of the separate islands1 made during the two voyages of ,kotzebue3 a reduced one of the whole group may be easily seen in ,krusenstern's 8,atlas10 and in ,kotzebue's 8,second ,voyage40 ,the group consists 7with the exception of two little islands which probably have had their lagoon filled up7 of a double row of twenty-three large and well-characterised atolls1 from the examination of which ,chamisso has given us his well-known account of coral-formations4 ,i include ,gaspar ,rico1 or ,cornwallis ,island in this group1 which is described by ,chamisso 7,kotzebue's 8,first ,voyage10 vol4 iii1 p4 #agi7 8as a low sickle-formed group1 with mould only on the windward side40 ,gaspard ,island is considered by some geographers as a distinct island lying ,n4,e4 of the group1 but it is not entered in the chart by ,krusenstern2 left uncoloured4 ,in the ,s4,w4 part of this group lies ,baring ,island1 of which little is known 7see ,krusenstern's 8,appendix10 #ahce1 p4 #adi74 ,i have left it uncoloured2 but ,boston ,island ,i have coloured blue1 as it is described 7,ibid47 as consisting of fourteen small islands1 which1 no doubt1 enclose a lagoon1 as represented in a chart in the 8#8,coquille's' ,atlas40--,two islands1 ,aur ,kawen and ,gaspar ,rico1 are written in the ,french chart with capital letters2 but this is an error1 for from the account given by ,chamisso in ,kotzebue's 8,first ,voyage10 they are certainly low4 ,the nature1 position1 and even existence1 of the shoals and small islands north of the ,marshall group1 are doubtful4 ,,new ,,hebrides4--,any chart1 on even a small scale1 of these islands1 will show that their shores are almost without reefs1 presenting a remarkable contrast with those of ,new ,caledonia on the one hand1 and the ,fidji group on the other4 ,nevertheless1 ,i have been assured by ,mr4 ,g4 ,bennett1 that coral grows vigorously on their shores2 as indeed1 will be further shown in some of the following notices4 ,as1 therefore1 these islands are not encircled1 and as coral grows vigorously on their shores1 we might almost conclude1 without further evidence1 that they were fringed1 and hence ,i have applied the red colour with rather greater freedom than in other instances4--,matthew's ,rock1 an active volcano1 some way south of the group 7of which a plan is given in the 8,atlas of the ,astrolabe's ,voyage07 does not appear to have reefs of any kind about it4--,annatom1 the southernmost of the ,hebrides2 from a rough woodcut given in the ,united ,service ,journal 7#ahca1 part iii1 p4 #aij71 accompanying a paper by ,mr4 ,bennett1 it appears that the shore is fringed2 coloured red4--,tanna4 ,forster1 in his 8,observations0 7p4 #bb71 says ,tanna has on its shores coral-rock and madrepores2 and the younger ,forster1 in his account 7vol4 ii1 p4 #bfi7 speaking of the harbour says1 the whole ,s4,e4 side consists of coral-reefs1 which are overflowed at high-water2 part of the southern shore in ,cook's chart is represented as fringed2 coloured red4--,immer is described 7,united ,service ,journal1 #ahca1 part iii1 p4 #aib7 by ,mr4 ,bennett as being of moderate elevation1 with cliffs appearing like sandstone3 coral grows in patches on its shore1 but ,i have not coloured it2 and ,i mention these facts1 because ,immer might have been thought from ,forster's classification 78,observations10 p4 #ad71 to have been a low island or even an atoll4--,erromango ,island2 ,cook 78,second ,voyage10 vol4 ii1 p4 #de1 #dto ed47 speaks of rocks everywhere lining the coast1 and the natives offered to haul his boat over the breakers to the sandy beach3 ,mr4 ,bennett1 in a letter to the ,editor of the ,singapore ,chron41 alludes to the reefs on its shores4 ,it may1 ,i think1 be safely inferred from these passages that the shore is fringed in parts by coral-reefs2 coloured red4--,sandwich ,island4 ,the east coast is said 7,cook's 8,second ,voyage10 vol4 ii1 p4 #da7 to be low1 and to be guarded by a chain of breakers4 ,in the accompanying chart it is seen to be fringed by a reef2 coloured red4--,mallicollo4 ,forster speaks of the reef-bounded shore3 the reef is about thirty yards wide1 and so shallow that a boat cannot pass over it4 ,forster also 78,observations10 p4 #bc7 says1 that the rocks of the sea-shore consist of madrepore4 ,in the plan of ,sandwich harbour1 the headlands are represented as fringed2 coloured red4--,aurora and ,pentecost ,islands1 according to ,bougainville1 apparently have no reefs2 nor has the large island of ,s4 ,espiritu1 nor ,bligh ,island or ,banks' ,islands1 which latter lie to the ,n4,e4 of the ,hebrides4 ,but in none of these cases1 have ,i met with any detailed account of their shores1 or seen plans on a large scale2 and it will be evident1 that a fringing-reef of only thirty or even a few hundred yards in width1 is of so little importance to navigation1 that it will seldom be noticed1 excepting by chance2 and hence ,i do not doubt that several of these islands1 now left uncoloured1 ought to be red4 ,,santa ,,cruz ,,group4--,vanikoro 7,fig4 #a1 ,plate ,i7 offers a striking example of a barrier- reef3 it was first described by the ,chevalier ,dillon1 in his voyage1 and was surveyed in the ,astrolabe2 coloured pale blue4--,tikopia and ,fataka ,islands appear1 from the descriptions of ,dillon and ,d',urville1 to have no reefs2 ,anouda is a low1 flat island1 surrounded by cliffs 78,astrolabe ,hydrog40 and ,krusenstern1 8,mem40 vol4 ii1 p4 #dcb72 these are uncoloured4 ,toupoua 7,otooboa of ,dillon7 is stated by ,captain ,tromelin 78,annales ,marit40 #ahbi1 p4 #bhi7 to be almost entirely included in a reef1 lying at the distance of two miles from the shore4 ,there is a space of three miles without any reef1 which1 although indented with bays1 offers no anchorage from the extreme depth of the water close to the shore3 ,captain ,dillon also speaks of the reefs fronting this island2 coloured blue4--,santa-,cruz4 ,i have carefully examined the works of ,carteret1 ,d',entrecasteaux1 ,wilson1 and ,tromelin1 and ,i cannot discover any mention of reefs on its shores2 left uncoloured4--,tinakoro is a constantly active volcano without reefs4--,mendana ,isles 7mentioned by ,dillon under the name of ,mammee1 etc472 said by ,krusenstern to be low1 and intertwined with reefs4 ,i do not believe they include a lagoon2 ,i have left them uncoloured4--,duff's ,islands compose a small group directed in a ,n4,w4 and ,s4,e4 band2 they are described by ,wilson 7p4 #bif1 8,miss4 ,voy40 #dto ed471 as formed by bold-peaked land1 with the islands surrounded by coral-reefs1 extending about half a mile from the shore2 at a distance of a mile from the reefs he found only seven fathoms4 ,as ,i have no reason for supposing there is deep water within these reefs1 ,i have coloured them red4 ,kennedy ,island1 ,n4,e4 of ,duff's4 ,i have been unable to find any account of it4 ,,new ,,caledonia4--,the great barrier-reefs on the shores of this island have already been described 7,fig4 #e1 ,plate ,,ii74 ,they have been visited by ,labillardiere1 ,cook1 and the northern point by ,d',urville2 this latter part so closely resembles an atoll that ,i have coloured it dark blue4 ,the ,loyalty group is situated eastward of this island2 from the chart and description given in the 8,voyage of the ,astrolabe10 they do not appear to have any reefs2 north of this group1 there are some extensive low reefs 7called ,astrolabe and ,beaupr17 which do not seem to be atoll-formed2 these are left uncoloured4 ,,australian ,,barrier-,,reef4--,the limits of this great reef1 which has already been described1 have been coloured from the charts of ,flinders and ,king4 ,in the northern parts1 an atoll-formed reef1 lying outside the barrier1 has been described by ,bligh1 and is coloured dark blue4 ,in the space between ,australia and ,new ,caledonia1 called by ,flinders the ,corallian ,sea1 there are numerous reefs4 ,of these1 some are represented in ,krusenstern's 8,atlas0 as having an atoll-like structure2 namely1 ,bampton shoal1 ,frederic1 ,vine or ,horse-shoe1 and ,alert reefs2 these have been coloured dark blue4 ,,louisiade4--,the dangerous reefs which front and surround the western1 southern1 and northern coasts of this so-called peninsula and archipelago1 seem evidently to belong to the barrier class4 ,the land is lofty1 with a low fringe on the coast2 the reefs are distant1 and the sea outside them profoundly deep4 ,nearly all that is known of this group is derived from the labours of ,d',entrecasteaux and ,bougainville3 the latter has represented one continuous reef ninety miles long1 parallel to the shore1 and in places as much as ten miles from it2 coloured pale blue4 ,a little distance northward we have the ,laughlan ,islands1 the reefs round which are engraved in the 8,atlas of the ,voyage of the ,astrolabe10 in the same manner as in the encircled islands of the ,caroline ,archipelago1 the reef is1 in parts1 a mile and a half from the shore1 to which it does not appear to be attached2 coloured blue4 ,at some little distance from the extremity of the ,louisiade lies the ,wells reef1 described in ,g4 ,hamilton's 8,voyage in ,h4,m4,s4 ,pandora0 7p4 #ajj73 it is said1 8,we found we had got embayed in a double reef1 which will soon be an island40 ,as this statement is only intelligible on the supposition of the reef being crescent or horse-shoe formed1 like so many other submerged annular reefs1 ,i have ventured to colour it blue4 ,,solomon ,,archipelago4--,the chart in ,krusenstern's 8,atlas0 shows that these islands are not encircled1 and as coral appears from the works of ,surville1 ,bougainville1 and ,labillardiere1 to grow on their shores1 this circumstance1 as in the case of the ,new ,hebrides1 is a presumption that they are fringed4 ,i cannot find out anything from ,d',entrecasteaux's 8,voyage10 regarding the southern islands of the group1 so have left them uncoloured4--,malayta ,island in a rough ,,ms4 chart in the ,admiralty has its northern shore fringed4--,ysabel ,island1 the ,n4,e4 part of this island1 in the same chart1 is also fringed3 ,mendana1 speaking 7,burney1 vol4 i1 p4 #bhj7 of an islet adjoining the northern coast1 says it is surrounded by reefs2 the shores1 also of ,port ,praslin appear regularly fringed4--,choiseul ,island4 ,in ,bougainville's 8,chart of ,choiseul ,bay10 parts of the shores are fringed by coral-reefs4--,bougainville ,island4 ,according to ,d',entrecasteaux the western shore abounds with coral-reefs1 and the smaller islands are said to be attached to the larger ones by reefs2 all the before-mentioned islands have been coloured red4--,bouka ,islands4 ,captain ,duperrey has kindly informed me in a letter that he passed close round the northern side of this island 7of which a plan is given in his 8,atlas of the ,coquille's ,voyage071 and that it was 8garnie d'une bande de rcifs fleur d'eau adherentes au rivage20 and he infers1 from the abundance of coral on the islands north and south of ,bouka1 that the reef probably is of coral2 coloured red4 ,off the north coast of the ,solomon ,archipelago there are several small groups which are little known2 they appear to be low1 and of coral-formation2 and some of them probably have an atoll-like structure2 the ,chevallier ,dillon1 however1 informs me that this is not the case with the ,b4 de ,candelaria4--,outong ,java1 according to the ,spanish navigator1 ,maurelle1 is thus characterised2 but this is the only one which ,i have ventured to colour blue4 ,,new ,,ireland4--,the shores of the ,s4,w4 point of this island and some adjoining islets1 are fringed by reefs1 as may be seen in the 8,atlases of the ,voyages of the ,coquille and ,astrolabe40 ,m4 ,lesson observes that the reefs are open in front of each streamlet4 ,the ,duke of ,york's ,island is also fringed2 but with regard to the other parts of ,new ,ireland1 ,new ,hanover1 and the small islands lying northward1 ,i have been unable to obtain any information4 ,i will only add that no part of ,new ,ireland appears to be fronted by distant reefs4 ,i have coloured red only the above specified portions4 ,,new ,,britain ,,and ,,the ,,northern ,,shore ,,of ,,new ,,guinea4--,from the charts in the 8,voyage of the ,astrolabe10 and from the 8,hydrog4 ,memoir10 it appears that these coasts are entirely without reefs1 as are the ,schouten ,islands1 lying close to the northern shore of ,new ,guinea4 ,the western and south-western parts of ,new ,guinea1 will be treated of when we come to the islands of the ,east ,indian ,archipelago4 ,,admiralty ,,group4--,from the accounts by ,bougainville1 ,maurelle1 ,d',entrecasteaux1 and the scattered notices collected by ,horsburgh1 it appears1 that some of the many islands composing it1 are high1 with a bold outline2 and others are very low1 small and interlaced with reefs4 ,all the high islands appear to be fronted by distant reefs rising abruptly from the sea1 and within some of which there is reason to believe that the water is deep4 ,i have therefore little doubt they are of the barrier class4--,in the southern part of the group we have ,elizabeth ,island1 which is surrounded by a reef at the distance of a mile2 and two miles eastward of it 7,krusenstern1 8,append40 #ahce1 p4 #db7 there is a little island containing a lagoon4--,near here1 also lies ,circular-reef 7,horsburgh1 8,direct410 vol4 i1 p4 #fia1 #dth ed471 8three or four miles in diameter having deep water inside with an opening at the ,n4,n4,w4 part1 and on the outside steep to40 ,i have from these data1 coloured the group pale blue1 and ,circular-reef dark blue4--,the ,anachorites1 ,echequier1 and ,hermites1 consist of innumerable low islands of coral-formation1 which probably have atoll-like forms2 but not being able to ascertain this1 ,i have not coloured them1 nor ,durour ,island1 which is described by ,carteret as low4 ,the ,,caroline ,,archipelago is now well-known1 chiefly from the hydrographical labours of ,lutk2 it contains about forty groups of atolls1 and three encircled islands1 two of which are engraved in ,figs #b and #g1 ,plate ,i4 ,commencing with the eastern part2 the encircling reef round ,ualen appears to be only about half a mile from the shore2 but as the land is low and covered with mangroves 78,voyage autour du ,monde10 par ,f4 ,lutk1 vol4 i1 p4 #cci71 the real margin has not probably been ascertained4 ,the extreme depth in one of the harbours within the reef is thirty-three fathoms 7see charts in 8,atlas of ,coquille's ,voyage071 and outside at half a mile distant from the reef1 no bottom was obtained with two hundred and fifty fathoms4 ,the reef is surmounted by many islets1 and the lagoon-like channel within is mostly shallow1 and appears to have been much encroached on by the low land surrounding the central mountains2 these facts show that time has allowed much detritus to accumulate2 coloured pale blue4--,pouynipte1 or ,seniavine4 ,in the greater part of the circumference of this island1 the reef is about one mile and three quarters distant2 on the north side it is five miles off the included high islets4 ,the reef is broken in several places2 and just within it1 the depth in one place is thirty fathoms1 and in another1 twenty-eight1 beyond which1 to all appearance1 there was 8un porte vaste et sur0 7,lutk1 vol4 ii1 p4 #d72 coloured pale blue4--,hogoleu or ,roug4 ,this wonderful group contains at least sixty-two islands1 and its reef is one hundred and thirty-five miles in circuit4 ,of the islands1 only a few1 about six or eight 7see 8,hydrog4 ,descrip40 p4 #dbh1 of the 8,voyage of the ,astrolabe10 and the large accompanying chart taken chiefly from that given by ,duperrey7 are high1 and the rest are all small1 low1 and formed on the reef4 ,the depth of the great interior lake has not been ascertained2 but ,captain ,d',urville appears to have entertained no doubt about the possibility of taking in a frigate4 ,the reef lies no less than fourteen miles distant from the northern coasts of the interior high islands1 seven from their western sides1 and twenty from the southern2 the sea is deep outside4 ,this island is a likeness on a grand scale to the ,gambier group in the ,low ,archipelago4 ,of the groups of low#a islands forming the chief part of the ,caroline ,archipelago1 all those of larger size1 have the true atoll-structure 7as may be seen in the 8,atlas0 by ,captain ,lutk71 and some even of the very small ones1 as ,macaskill and ,duperrey1 of which plans are given in the 8,atlas of the ,coquille's ,voyage40 ,there are1 however1 some low small islands of coral-formation1 namely ,ollap1 ,tamatam1 ,bigali1 ,satahoual1 which do not contain lagoons2 but it is probable that lagoons originally existed1 but have since filled up3 ,lutk 7vol4 ii1 p4 #cjd7 seems to have thought that all the low islands1 with only one exception1 contained lagoons4 ,from the sketches1 and from the manner in which the margins of these islands are engraved in the 8,atlas of the ,voyage of the ,coquille10 it might have been thought that they were not low2 but by a comparison with the remarks of ,lutk 7vol4 ii1 p4 #ajg1 regarding ,bigali7 and of ,freycinet 78,hydrog4 ,memoir ,l',uranie ,voyage10 p4 #ahh1 regarding ,tamatam1 ,ollap1 etc471 it will be seen that the artist must have represented the land incorrectly4 ,the most southern island in the group1 namely ,piguiram1 is not coloured1 because ,i have found no account of it4 ,nougouor1 or ,monte ,verdison1 which was not visited by ,lutk1 is described and figured by ,mr4 ,bennett 7,united ,service ,journal1 ,january #ahcb7 as an atoll4 ,all the above-mentioned islands have been coloured blue4 #a ,in ,d',urville and ,lottin's chart1 ,peserare is written with capital letters2 but this evidently is an error1 for it is one of the low islets on the reef of ,namonouyto 7see ,lutk's charts7--a regular atoll4 ,,western ,,part ,,of ,,the ,,caroline ,,archipelago4--,fais ,island is ninety feet high1 and is surrounded1 as ,i have been informed by ,admiral ,lutk1 by a narrow reef of living coral1 of which the broadest part1 as represented in the charts1 is only #aej yards2 coloured red4--,philip ,island41 ,i believe1 is low2 but ,hunter1 in his 8,historical ,journal10 gives no clear account of it2 uncoloured4--,elivi2 from the manner in which the islets on the reefs are engraved1 in the 8,atlas of the ,astrolabe's ,voyage10 ,i should have thought they were above the ordinary height1 but ,admiral ,lutk assures me this is not the case3 they form a regular atoll2 coloured blue4--,gouap 7,eap of ,chamisso71 is a high island with a reef 7see chart in 8,voyage of the ,astrolabe071 more than a mile distant in most parts from the shore1 and two miles in one part4 ,captain ,d',urville thinks that there would be anchorage 78,hydrog4 ,descript4 ,astrolabe ,voyage10 p4 #dcf7 for ships within the reef1 if a passage could be found2 coloured pale blue4--,goulou1 from the chart in the 8,astrolabe's ,atlas10 appears to be an atoll4 ,d',urville 78,hydrog4 ,descript40 p4 #dcg7 speaks of the low islets on the reef2 coloured dark blue4 ,,pelew ,,islands4--,krusenstern speaks of some of the islands being mountainous2 the reefs are distant from the shore1 and there are spaces within them1 and not opposite valleys1 with from ten to fifteen fathoms4 ,according to a ,,ms4 chart of the group by ,lieutenant ,elmer in the ,admiralty1 there is a large space within the reef with deepish water2 although the high land does not hold a central position with respect to the reefs1 as is generally the case1 ,i have little doubt that the reefs of the ,pelew ,islands ought to be ranked with the barrier class1 and ,i have coloured them pale blue4 ,in ,lieutenant ,elmer's chart there is a horseshoe-formed shoal1 laid down thirteen miles ,n4,w4 of ,pelew1 with fifteen fathoms within the reef1 and some dry banks on it2 coloured dark blue4--,spanish1 ,martires1 ,sanserot1 ,pulo ,anna and ,mariere ,islands are not coloured1 because ,i know nothing about them1 excepting that according to ,krusenstern1 the second1 third1 and fourth mentioned1 are low1 placed on coral-reefs1 and therefore1 perhaps1 contain lagoons2 but ,pulo ,mariere is a little higher4 ,,mariana ,,archipelago1 or ,,ladrones4--,guahan4 ,almost the whole of this island is fringed by reefs1 which extend in most parts about a third of a mile from the land4 ,even where the reefs are most extensive1 the water within them is shallow4 ,in several parts there is a navigable channel for boats and canoes within the reefs4 ,in ,freycinet's 8,hydrog4 ,mem40 there is an account of these reefs1 and in the 8,atlas10 a map on a large scale2 coloured red4--,rota4 8,l'ile est presque entirement entoure des rcifs0 7p4 #bab1 ,freycinet's 8,hydrog4 ,mem4074 ,these reefs project about a quarter of a mile from the shore2 coloured red4--,tinian4 ,the eastern coast is precipitous1 and is without reefs2 but the western side is fringed like the last island2 coloured red4--,saypan4 ,the ,n4,e4 coast1 and likewise the western shores appear to be fringed2 but there is a great1 irregular1 horn-like reef projecting far from this side2 coloured red4--,farallon de ,medinilla1 appears so regularly and closely fringed in ,freycinet's charts1 that ,i have ventured to colour it red1 although nothing is said about reefs in the 8,hydrographical ,memoir40 ,the several islands which form the northern part of the group are volcanic 7with the exception perhaps of ,torres1 which resembles in form the madreporitic island of ,medinilla71 and appear to be without reefs4--,mangs1 however1 is described 7by ,freycinet1 p4 #bai1 8,hydrog407 from some ,spanish charts1 as formed of small islands placed 8au milieu des nombreux rcifs20 and as these reefs in the general chart of the group do not project so much as a mile2 and as there is no appearance from a double line1 of the existence of deep water within1 ,i have ventured1 although with much hesitation1 to colour them red4 ,respecting ,folger and ,marshall ,islands which lie some way east of the ,marianas1 ,i can find out nothing1 excepting that they are probably low4 ,krusenstern says this of ,marshall ,island2 and ,folger ,island is written with small letters in ,d',urville's chart2 uncoloured4 ,,bonin ,,or ,,arzobispo ,,group4--,peel ,island has been examined by ,captain ,beechey1 to whose kindness ,i am much indebted for giving me information regarding it3 8,at ,port ,lloyd there is a great deal of coral2 and the inner harbour is entirely formed by coral-reefs1 which extend outside the port along the coast40 ,captain ,beechey1 in another part of his letter to me1 alludes to the reefs fringing the island in all directions2 but at the same time it must be observed that the surf washes the volcanic rocks of the coast in the greater part of its circumference4 ,i do not know whether the other islands of the ,archipelago are fringed2 ,i have coloured ,peel ,island red4--,grampus ,island to the eastward1 does not appear 7,meare's 8,voyage10 p4 #ie7 to have any reefs1 nor does ,rosario ,island 7from ,lutk's chart71 which lies to the westward4 ,respecting the few other islands in this part of the sea1 namely the ,sulphur ,islands1 with an active volcano1 and those lying between ,bonin and ,japan 7which are situated near the extreme limit in latitude1 at which reefs are formed71 ,i have not been able to find any clear account4 ,,west ,,end ,,of ,,new ,,guinea4--,port ,dory4 ,from the charts in the 8,voyage of the ,coquille10 it would appear that the coast in this part is fringed by coral-reefs2 ,m4 ,lesson1 however1 remarks that the coral is sickly2 coloured red4--,waigiou4 ,a considerable portion of the northern shores of these islands is seen in the charts 7on a large scale7 in ,freycinet's 8,atlas0 to be fringed by coral-reefs4 ,forrest 7p4 #ba1 8,voyage to ,new ,guinea07 alludes to the coral-reefs lining the heads of ,piapis ,bay2 and ,horsburgh 7vol4 ii1 p4 #eii1 #dth edit471 speaking of the islands in ,dampier ,strait1 says 8sharp coral-rocks line their shores20 coloured red4--,in the sea north of these islands1 we have ,guedes 7or ,freewill1 or ,st4 ,david's71 which from the chart given in the #dto edit4 of ,carteret's 8,voyage10 must be an atoll4 ,krusenstern says the islets are very low2 coloured blue4--,carteret's ,shoals1 in #b #ec ,n41 are described as circular1 with stony points showing all round1 with deeper water in the middle2 coloured blue4--,aiou2 the plan of this group1 given in the 8,atlas of the ,voyage of the ,astrolabe10 shows that it is an atoll2 and1 from a chart in ,forrest's 8,voyage10 it appears that there is twelve fathoms within the circular reef2 coloured blue4--,the ,s4,w4 coast of ,new ,guinea appears to be low1 muddy1 and devoid of reefs4 ,the ,arru1 ,timor-laut1 and ,tenimber groups have lately been examined by ,captain ,kolff1 the ,,ms4 translation of which1 by ,mr4 ,w4 ,earl1 ,i have been permitted to read1 through the kindness of ,captain ,washington1 ,r4,n4 ,these islands are mostly rather low1 and are surrounded by distant reefs 7the ,ki ,islands1 however1 are lofty1 and1 from ,mr4 ,stanley's survey1 appear without reefs72 the sea in some parts is shallow1 in others profoundly deep 7as near ,larrat74 ,from the imperfection of the published charts1 ,i have been unable to decide to which class these reefs belong4 ,from the distance to which they extend from the land1 where the sea is very deep1 ,i am strongly inclined to believe they ought to come within the barrier class1 and be coloured blue2 but ,i have been forced to leave them uncoloured4--,the last-mentioned groups are connected with the east end of ,ceram by a chain of small islands1 of which the small groups of ,ceram-laut1 ,goram and ,keffing are surrounded by very extensive reefs1 projecting into deep water1 which1 as in the last case1 ,i strongly suspect belong to the barrier class2 but ,i have not coloured them4 ,from the south side of ,keffing1 the reefs project five miles 7,windsor ,earl's 8,sailing ,direct4 for the ,arafura ,sea10 p4 #i74 ,,ceram4--,in various charts which ,i have examined1 several parts of the coast are represented as fringed by reefs4--,manipa ,island1 between ,ceram and ,bourou1 in an old ,,ms4 chart in the ,admiralty1 is fringed by a very irregular reef1 partly dry at low water1 which ,i do not doubt is of coral-formation2 both islands coloured red4--,bourou2 parts of this island appear fringed by coral-reefs1 namely1 the eastern coast1 as seen in ,freycinet's chart2 and ,cajeli ,bay1 which is said by ,horsburgh 7vol4 ii1 p4 #fcj7 to be lined by coral-reefs1 that stretch out a little way1 and have only a few feet water on them4 ,in several charts1 portions of the islands forming the ,,amboina ,,group are fringed by reefs2 for instance1 ,noessa1 ,harenca1 and ,ucaster1 in ,freycinet's charts4 ,the above-mentioned islands have been coloured red1 although the evidence is not very satisfactory4--,north of ,bourou the parallel line of the ,xulla ,isles extends3 ,i have not been able to find out anything about them1 excepting that ,horsburgh 7vol4 ii1 p4 #edc7 says that the northern shore is surrounded by a reef at the distance of two or three miles2 uncoloured4--,mysol ,group2 the ,kanary ,islands are said by ,forrest 78,voyage10 p4 #acj7 to be divided from each other by deep straits1 and are lined with coral-rocks2 coloured red4--,guebe1 lying between ,waigiou and ,gilolo1 is engraved as if fringed2 and it is said by ,freycinet1 that all the soundings under five fathoms were on coral2 coloured red4--,gilolo4 ,in a chart published by ,dalrymple1 the numerous islands on the western1 southern 7,batchian and the ,strait of ,patientia71 and eastern sides appear fringed by narrow reefs2 these reefs1 ,i suppose1 are of coral1 for it is said in 8,malte ,brun0 7vol4 xii1 p4 #aef71 8,sur les ctes 7of ,batchian7 comme dans les plupart des iles de cet archipel1 il y a de rocs de mdrepores d'une beaut et d'une varit infimies40 ,forrest1 also 7p4 #ej71 says ,seland1 near ,batchian1 is a little island with reefs of coral2 coloured red4--,morty ,island 7north of ,gilolo74 ,horsburgh 7vol4 ii1 p4 #ejf7 says the northern coast is lined by reefs1 projecting one or two miles1 and having no soundings close to them2 ,i have left it uncoloured1 although1 as in some former cases1 it ought probably to be pale blue4--,celebes4 ,the western and northern coasts appear in the charts to be bold and without reefs4 ,near the extreme northern point1 however1 an islet in the ,straits of ,limbe1 and parts of the adjoining shore1 appear to be fringed3 the east side of the bay of ,manado1 has deep water1 and is fringed by sand and coral 78,astrol4 ,voyage10 ,hydrog4 ,part1 pp4 #dec-d72 this extreme point1 therefore1 ,i have coloured red4--,of the islands leading from this point to ,magindanao1 ,i have not been able to find any account1 except of ,serangani1 which appears surrounded by narrow reefs2 and ,forrest 78,voyage10 p4 #afd7 speaks of coral on its shores2 ,i have1 therefore1 coloured this island red4 ,to the eastward of this chain lie several islands2 of which ,i cannot find any account1 except of ,karkalang1 which is said by ,horsburgh 7vol4 ii1 p4 #ejd7 to be lined by a dangerous reef1 projecting several miles from the northern shore2 not coloured4 ,,islands ,,near ,,timor4--,the account of the following islands is taken from ,captain ,d4 ,kolff's 8,voyage10 in #ahbe1 translated by ,mr4 ,w4 ,earl1 from the ,dutch4--,lette has 8reefs extending along shore at the distance of half a mile from the land40--,moa has reefs on the ,s4,w4 part4--,lakor has a reef lining its shore2 these islands are coloured red4--,still more eastward1 ,luan has1 differently from the last-mentioned islands1 an extensive reef2 it is steep outside1 and within there is a depth of twelve feet2 from these facts1 it is impossible to decide to which class this island belongs4--,kissa1 off the point of ,timor1 has its 8shore fronted by a reef1 steep too on the outer side1 over which small proahs can go at the time of high water20 coloured red4--,timor2 most of the points1 and some considerable spaces of the northern shore1 are seen in ,freycinet's chart to be fringed by coral-reefs2 and mention is made of them in the accompanying 8,hydrog4 ,memoir20 coloured red4--,savu1 ,s4,e4 of ,timor1 appears in ,flinders' chart to be fringed2 but ,i have not coloured it1 as ,i do not know that the reefs are of coral4--,sandalwood ,island has1 according to ,horsburgh 7vol4 ii1 p4 #fjg71 a reef on its southern shore1 four miles distant from the land2 as the neighbouring sea is deep1 and generally bold1 this probably is a barrier-reef1 but ,i have not ventured to colour it4 ,n4,w4 ,,coast ,,of ,,australia4--,it appears1 in ,captain ,king's ,sailing ,directions 78,narrative of ,survey10 vol4 ii1 pp4 #cbe-cfi71 that there are many extensive coral-reefs skirting1 often at considerable distances1 the ,n4,w4 shores1 and encompassing the small adjoining islets4 ,deep water1 in no instance1 is represented in the charts between these reefs and the land2 and1 therefore1 they probably belong to the fringing class4 ,but as they extend far into the sea1 which is generally shallow1 even in places where the land seems to be somewhat precipitous2 ,i have not coloured them4 ,houtman's ,abrolhos 7lat4 #bh ,s4 on west coast7 have lately been surveyed by ,captain ,wickham 7as described in ,naut4 ,mag4 #ahda1 p4 #eaa73 they lie on the edge of a steeply shelving bank1 which extends about thirty miles seaward1 along the whole line of coast4 ,the two southern reefs1 or islands1 enclose a lagoon-like space of water1 varying in depth from five to fifteen fathoms1 and in one spot with twenty-three fathoms4 ,the greater part of the island has been formed on their inland sides1 by the accumulation of fragments of coral2 the seaward face consisting of nearly bare ledges of rock4 ,some of the specimens1 brought home by ,captain ,wickham1 contained fragments of marine shells1 but others did not2 and these closely resembled a formation at ,king ,george's ,sound1 principally due to the action of the wind on calcareous dust1 which ,i shall describe in a forthcoming part4 ,from the extreme irregularity of these reefs with their lagoons1 and from their position on a bank1 the usual depth of which is only thirty fathoms1 ,i have not ventured to class them with atolls1 and hence have left them uncoloured4--,rowley ,shoals4 ,these lie some way from the ,n4,w4 coast of ,australia3 according to ,captain ,king 78,narrative of ,survey10 vol4 i1 p4 #fj71 they are of coral-formation4 ,they rise abruptly from the sea1 and ,captain ,king had no bottom with #agj fathoms close to them4 ,three of them are crescent-shaped2 they are mentioned by ,mr4 ,lyell1 on the authority of ,captain ,king1 with reference to the direction of their open sides4 8,a third oval reef of the same group is entirely submerged0 78,principles of ,geology10 book iii1 chap4 xviii72 coloured blue4--,scott's ,reefs1 lying north of ,rowley ,shoals1 are briefly described by ,captain ,wickham 7,naut4 ,mag4 #ahda1 p4 #ddj73 they appear to be of great size1 of a circular form1 and 8with smooth water within1 forming probably a lagoon of great extent40 ,there is a break on the western side1 where there probably is an entrance3 the water is very deep off these reefs2 coloured blue4 ,proceeding westward along the great volcanic chain of the ,east ,indian ,archipelago1 ,solor ,strait is represented in a chart published by ,dalrymple from a ,dutch ,,ms41 as fringed2 as are parts of ,flores1 of ,adenara1 and of ,solor4 ,horsburgh speaks of coral growing on these shores2 and therefore ,i have no doubt that the reefs are of coral1 and accordingly have coloured them red4 ,we hear from ,horsburgh 7vol4 ii1 p4 #fjb7 that a coral-flat bounds the shores of ,sapy ,bay4 ,from the same authority it appears 7p4 #faj7 that reefs fringe the island of ,timor-,young1 on the ,n4 shore of ,sumbawa2 and1 likewise 7p4 #fjj71 that ,bally town in ,lombock1 is fronted by a reef1 stretching along the shore at a distance of a hundred fathoms1 with channels through it for boats2 these places1 therefore1 have been coloured red4--,bally ,island4 ,in a ,dutch ,,ms4 chart on a large scale of ,java1 which was brought from that island by ,dr4 ,horsfield1 who had the kindness to show it me at the ,india ,house1 its western1 northern1 and southern shores appear very regularly fringed by a reef 7see also ,horsburgh1 vol4 ii1 p4 #eic72 and as coral is found abundantly there1 ,i have not the least doubt that the reef is of coral1 and therefore have coloured it red4 ,,java4--,my information regarding the reefs of this great island is derived from the chart just mentioned4 ,the greater part of ,maduara is represented in it as regularly fringed1 and likewise portions of the coast of ,java immediately south of it4 ,dr4 ,horsfield informs me that coral is very abundant near ,sourabaya4 ,the islets and parts of the ,n4 coast of ,java1 west of ,point ,buang1 or ,japara1 are fringed by reefs1 said to be of coral4 ,lubeck1 or ,bavian ,islands1 lying at some distance from the shore of ,java1 are regularly fringed by coral-reefs4 ,carimon ,java appears equally so1 though it is not directly said that the reefs are of coral2 there is a depth between thirty and forty fathoms round these islands4 ,parts of the shores of ,sunda ,strait1 where the water is from forty to eighty fathoms deep1 and the islets near ,batavia appear in several charts to be fringed4 ,in the ,dutch chart the southern shore1 in the narrowest part of the island1 is in two places fringed by reefs of coral4 ,west of ,segorrowodee ,bay1 and the extreme ,s4,e4 and ,e4 portions are likewise fringed by coral-reefs2 all the above-mentioned places coloured red4 ,macassar ,strait2 the east coast of ,borneo appears1 in most parts1 free from reefs1 and where they occur1 as on the east coast of ,pamaroong1 the sea is very shallow2 hence no part is coloured4 ,in ,macassar ,strait itself1 in about lat4 #b ,s41 there are many small islands with coral-shoals projecting far from them4 ,there are also 7old charts by ,dalrymple7 numerous little flats of coral1 not rising to the surface of the water1 and shelving suddenly from five fathoms to no bottom with fifty fathoms2 they do not appear to have a lagoon-like structure4 ,there are similar coral-shoals a little farther south2 and in lat4 #d #ee there are two1 which are engraved from modern surveys1 in a manner which might represent an annular reef with deep water inside3 ,captain ,moresby1 however1 who was formerly in this sea1 doubts this fact1 so that ,i have left them uncoloured3 at the same time ,i may remark1 that these two shoals make a nearer approach to the atoll-like structure than any other within the ,e4 ,indian ,archipelago4 ,southward of these shoals there are other low islands and irregular coral-reefs2 and in the space of sea1 north of the great volcanic chain1 from ,timor to ,java1 we have also other islands1 such as the ,postillions1 ,kalatoa1 ,tokan-,bessees1 etc41 which are chiefly low1 and are surrounded by very irregular and distant reefs4 ,from the imperfect charts ,i have seen1 ,i have not been able to decide whether they belong to the atoll or barrier-classes1 or whether they merely fringe submarine banks1 and gently sloping land4 ,in the ,bay of ,bonin1 between the two southern arms of ,celebes1 there are numerous coral-reefs2 but none of them seem to have an atoll-like structure4 ,i have1 therefore1 not coloured any of the islands in this part of the sea2 ,i think it1 however1 exceedingly probable that some of them ought to be blue4 ,i may add that there is a harbour on the ,s4,e4 coast of ,bouton which1 according to an old chart1 is formed by a reef1 parallel to the shore1 with deep water within2 and in the 8,voyage of the ,coquille10 some neighbouring islands are represented with reefs a good way distant1 but ,i do not know whether with deep water within4 ,i have not thought the evidence sufficient to permit me to colour them4 ,,sumatra4--,commencing with the west coast and outlying islands1 ,engano ,island is represented in the published chart as surrounded by a narrow reef1 and ,napier1 in his 8,sailing ,directions10 speaks of the reef being of coral 7also ,horsburgh1 vol4 ii1 p4 #aae72 coloured red4--,rat ,island 7#c #ea ,s47 is surrounded by reefs of coral1 partly dry at low water1 7,horsburgh1 vol4 ii1 p4 #if74--,trieste ,island 7#d #b ,s474 ,the shore is represented in a chart which ,i saw at the ,india ,house1 as fringed in such a manner1 that ,i feel sure the fringe consists of coral2 but as the island is so low1 that the sea sometimes flows quite over it 7,dampier1 8,voyage10 vol4 i1 p4 #dgd71 ,i have not coloured it4--,pulo ,dooa 7lat4 #c74 ,in an old chart it is said there are chasms in the reefs round the island1 admitting boats to the watering-place1 and that the southern islet consists of a mass of sand and coral4--,pulo ,pisang2 ,horsburgh 7vol4 ii1 p4 #hf7 says that the rocky coral-bank1 which stretches about forty yards from the shore1 is steep to all round3 in a chart1 also1 which ,i have seen1 the island is represented as regularly fringed4--,pulo ,mintao is lined with reefs on its west side 7,horsburgh1 vol4 ii1 p4 #ajg74--,pulo ,baniak2 the same authority 7vol4 ii1 p4 #aje71 speaking of a part1 says it is faced with coral-rocks4--,minguin 7#c #cf ,n474 ,a coral-reef fronts this place1 and projects into the sea nearly a quarter of a mile 78,notices of the ,indian ,arch40 published at ,singapore1 p4 #aje74--,pulo ,brassa 7#e #df ,n474 ,a reef surrounds it at a cable's length 7,horsburgh1 vol4 ii1 p4 #fj74 ,i have coloured all the above-specified points red4 ,i may here add1 that both ,horsburgh and ,mr4 ,moor 7in the 8,notices0 just alluded to7 frequently speak of the numerous reefs and banks of coral on the west coast of ,sumatra2 but these nowhere have the structure of a barrier-reef1 and ,marsden 78,history of ,sumatra07 states1 that where the coast is flat1 the fringing-reefs extend furthest from it4 ,the northern and southern points1 and the greater part of the east coast1 are low1 and faced with mud banks1 and therefore without coral4 ,,nicobar ,,islands4--,the chart represents the islands of this group as fringed by reefs4 ,with regard to ,great ,nicobar1 ,captain ,moresby informs me1 that it is fringed by reefs of coral1 extending between two and three hundred yards from the shore4 ,the ,northern ,nicobars appear so regularly fringed in the published charts1 that ,i have no doubt the reefs are of coral4 ,this group1 therefore1 is coloured red4 ,,andaman ,,islands4--,from an examination of the ,,ms4 chart1 on a large scale1 of this island1 by ,captain ,arch4 ,blair1 in the ,admiralty1 several portions of the coast appear fringed2 and as ,horsburgh speaks of coral-reefs being numerous in the vicinity of these islands1 ,i should have coloured them red1 had not some expressions in a paper in the 8,asiatic ,researches0 7vol4 iv1 p4 #djb7 led me to doubt the existence of reefs2 uncoloured4 ,the coast of ,malacca1 ,tenasserim and the coasts northward1 appear in the greater part to be low and muddy3 where reefs occur1 as in parts of ,malacca ,straits1 and near ,singapore1 they are of the fringing kind2 but the water is so shoal1 that ,i have not coloured them4 ,in the sea1 however1 between ,malacca and the west coast of ,borneo1 where there is a greater depth from forty to fifty fathoms1 ,i have coloured red some of the groups1 which are regularly fringed4 ,the northern ,natunas and the ,anambas ,islands are represented in the charts on a large scale1 published in the 8,atlas of the ,voyage of the ,favourite10 as fringed by reefs of coral1 with very shoal water within them4--,tumbelan and ,bunoa ,islands 7#a ,n47 are represented in the ,english charts as surrounded by a very regular fringe4--,st4 ,barbes 7#j #ae ,n47 is said by ,horsburgh 7vol4 ii1 p4 #bgi7 to be fronted by a reef1 over which boats can land only at high water4--,the shore of ,borneo at ,tunjong ,apee is also fronted by a reef1 extending not far from the land 7,horsburgh1 vol4 ii1 p4 #dfh74 ,these places ,i have coloured red2 although with some hesitation1 as the water is shallow4 ,i might perhaps have added ,pulo ,leat1 in ,gaspar ,strait1 ,lucepara1 and ,carimata2 but as the sea is confined and shallow1 and the reefs not very regular1 ,i have left them uncoloured4 ,the water shoals gradually towards the whole west coast of ,borneo3 ,i cannot make out that it has any reefs of coral4 ,the islands1 however1 off the northern extremity1 and near the ,s4,w4 end of ,palawan1 are fringed by very distant coral-reefs2 thus the reefs in the case of ,balabac are no less than five miles from the land2 but the sea1 in the whole of this district1 is so shallow1 that the reefs might be expected to extend very far from the land4 ,i have not1 therefore1 thought myself authorised to colour them4 ,the ,n4,e4 point of ,borneo1 where the water is very shoal1 is connected with ,magindanao by a chain of islands called the ,sooloo ,archipelago1 about which ,i have been able to obtain very little information2 ,pangootaran1 although ten miles long1 entirely consists of a bed of coral-rock 78,notices of ,e4 ,indian ,arch40 p4 #eh73 ,i believe from ,horsburgh that the island is low2 not coloured4--,tahow ,bank1 in some old charts1 appears like a submerged atoll2 not coloured4 ,forrest 78,voyage10 p4 #ba7 states that one of the islands near ,sooloo is surrounded by coral-rocks2 but there is no distant reef4 ,near the ,s4 end of ,basselan1 some of the islets in the chart accompanying ,forrest's 8,voyage10 appear fringed with reefs2 hence ,i have coloured1 though unwillingly1 parts of the ,sooloo group red4 ,the sea between ,sooloo and ,palawan1 near the shoal coast of ,borneo1 is interspersed with irregular reefs and shoal patches2 not coloured3 but in the northern part of this sea1 there are two low islets1 ,cagayanes and ,cavilli1 surrounded by extensive coral-reefs2 the breakers round the latter 7,horsburgh1 vol4 ii1 p4 #eac7 extend five or six miles from a sandbank1 which forms the only dry part2 these breakers are steep to outside2 there appears to be an opening through them on one side1 with four or five fathoms within3 from this description1 ,i strongly suspect that ,cavilli ought to be considered an atoll2 but1 as ,i have not seen any chart of it1 on even a moderately large scale1 ,i have not coloured it4 ,the islets off the northern end of ,palawan1 are in the same case as those off the southern end1 namely they are fringed by reefs1 some way distant from the shore1 but the water is exceedingly shallow2 uncoloured4 ,the western shore of ,palawan will be treated of under the head of ,china ,sea4 ,,philippine ,,archipelago4--,a chart on a large scale of ,appoo ,shoal1 which lies near the ,s4,e4 coast of ,mindoro1 has been executed by ,captain ,d4 ,ross3 it appears atoll-formed1 but with rather an irregular outline2 its diameter is about ten miles2 there are two well-defined passages leading into the interior lagoon1 which appears open2 close outside the reef all round1 there is no bottom with seventy fathoms2 coloured blue4--,mindoro3 the ,n4,w4 coast is represented in several charts1 as fringed by a reef1 and ,luban ,island is said1 by ,horsburgh 7vol4 ii1 p4 #dcf71 to be 8lined by a reef40--,luzon3 ,mr4 ,cuming1 who has lately investigated with so much success the ,natural ,history of the ,philippines1 informs me1 that about three miles of the shore north of ,point ,st4 ,jago1 is fringed by a reef2 as are 7,horsburgh1 vol4 ii1 p4 #dcg7 the ,three ,friars off ,silanguin ,bay4 ,between ,point ,capones and ,playa ,honda1 the coast is 8lined by a coral-reef1 stretching out nearly a mile in some places10 7,horsburgh72 and ,mr4 ,cuming visited some fringing-reefs on parts of this coast1 namely1 near ,puebla1 ,iba1 and ,mansinglor4 ,in the neighbourhood of ,solon-solon ,bay1 the shore is lined 7,horsburgh1 ii1 p4 #dci7 by coral-reefs1 stretching out a great way3 there are also reefs about the islets off ,solamague2 and as ,i am informed by ,mr4 ,cuming1 near ,st4 ,catalina1 and a little north of it4 ,the same gentleman informs me there are reefs on the ,s4,e4 point of this island in front of ,samar1 extending from ,malalabon to ,bulusan4 ,these appear to be the principal fringing-reefs on the coasts of ,luzon2 and they have all been coloured red4 ,mr4 ,cuming informs me that none of them have deep water within2 although it appears from ,horsburgh that some few extend to a considerable distance from the shore4 ,within the ,philippine ,archipelago1 the shores of the islands do not appear to be commonly fringed1 with the exception of the ,s4 shore of ,masbate1 and nearly the whole of ,bohol2 which are both coloured red4 ,on the ,s4 shore of ,magindanao1 ,bunwoot ,island is surrounded 7according to ,forrest1 8,voyage10 p4 #bec71 by a coral-reef1 which in the chart appears one of the fringing class4 ,with respect to the eastern coasts of the whole ,archipelago1 ,i have not been able to obtain any account4 ,,babuyan ,,islands4--,horsburgh says 7vol4 ii1 p4 #ddb71 coral-reefs line the shores of the harbour in ,fuga2 and the charts show there are other reefs about these islands4 ,camiguin has its shore in parts lined by coral-rock 7,horsburgh1 p4 #ddc72 about a mile off shore there is between thirty and thirty-five fathoms4 ,the plan of ,port ,san ,pio ,quinto shows that its shores are fringed with coral2 coloured red4--,,bashee ,,islands3 ,horsburgh1 speaking of the southern part of the group 7vol4 ii1 p4 #dde7 says the shores of both islands are fortified by a reef1 and through some of the gaps in it1 the natives can pass in their boats in fine weather2 the bottom near the land is coral-rock4 ,from the published charts1 it is evident that several of these islands are most regularly fringed2 coloured red4 ,the northern islands are left uncoloured1 as ,i have been unable to find any account of them4--,,formosa4 ,the shores1 especially the western one1 seem chiefly composed of mud and sand1 and ,i cannot make out that they are anywhere lined by reefs2 except in a harbour 7,horsburgh1 vol4 ii1 p4 #ddi7 at the extreme northern point3 hence1 of course1 the whole of this island is left uncoloured4 ,the small adjoining islands are in the same case4--,,patchow1 ,,or ,,madjiko-,,sima ,,groups4 ,patchuson has been described by ,captain ,broughton 78,voy4 to the ,n4 ,pacific10 p4 #aia72 he says1 the boats1 with some difficulty1 found a passage through the coral-reefs1 which extend along the coast1 nearly half a mile off it4 ,the boats were well sheltered within the reef2 but it does not appear that the water is deep there4 ,outside the reef the depth is very irregular1 varying from five to fifty fathoms2 the form of the land is not very abrupt2 coloured red4--,taypin-san2 from the description given 7p4 #aie7 by the same author1 it appears that a very irregular reef extends1 to the distance of several miles1 from the southern island2 but whether it encircles a space of deep water is not evident2 nor1 indeed1 whether these outlying reefs are connected with those more immediately adjoining the land2 left uncoloured4 ,i may here just add that the shore of ,kumi 7lying west of ,patchow71 has a narrow reef attached to it in the plan of it1 in ,la ,peyrouse's 8,atlas20 but it does not appear in the account of the voyage that it is of coral2 uncoloured4--,,loo ,,choo4 ,the greater part of the coast of this moderately hilly island1 is skirted by reefs1 which do not extend far from the shore1 and which do not leave a channel of deep water within them1 as may be seen in the charts accompanying ,captain ,b4 ,hall's voyage to ,loo ,choo 7see also remarks in ,appendix1 pp4 xxi4 and xxv474 ,there are1 however1 some ports with deep water1 formed by reefs in front of valleys1 in the same manner as happens at ,mauritius4 ,captain ,beechey1 in a letter to me1 compares these reefs with those encircling the ,society ,islands2 but there appears to me a marked difference between them1 in the less distance at which the ,loo ,choo reefs lie from the land with relation to the probable submarine inclination1 and in the absence of an interior deep water-moat or channel1 parallel to the land4 ,hence1 ,i have classed these reefs with fringing-reefs1 and coloured them red4--,,pescadores 7west of ,formosa74 ,dampier 7vol4 i1 p4 #daf71 has compared the appearance of the land to the southern parts of ,england4 ,the islands are interlaced with coral-reefs2 but as the water is very shoal1 and as spits of sand and gravel 7,horsburgh1 vol4 ii1 p4 #dej7 extend far out from them1 it is impossible to draw any inferences regarding the nature of the reefs4 ,,china ,,sea4--,proceeding from north to south1 we first meet the ,pratas ,shoal 7lat4 #bj ,n47 which1 according to ,horsburgh 7vol4 ii1 p4 #cce71 is composed of coral1 is of a circular form1 and has a low islet on it4 ,the reef is on a level with the water's edge1 and when the sea runs high1 there are breakers mostly all round1 8but the water within seems pretty deep in some places2 although steep-to in most parts outside1 there appear to be several parts where a ship might find anchorage outside the breakers20 coloured blue4--,the ,paracells have been accurately surveyed by ,captain ,d4 ,ross1 and charts on a large scale published3 but few low islets have been formed on these shoals1 and this seems to be a general circumstance in the ,china ,sea2 the sea close outside the reefs is very deep2 several of them have a lagoon-like structure2 or separate islets 7,prattle1 ,robert1 ,drummond1 etc47 are so arranged round a moderately shallow space1 as to appear as if they had once formed one large atoll4--,bombay ,shoal 7one of the ,paracells7 has the form of an annular reef1 and is 8apparently deep within20 it seems to have an entrance 7,horsburgh1 vol4 ii1 p4 #ccb7 on its west side2 it is very steep outside4--,discovery ,shoal1 also is of an oval form1 with a lagoon-like space within1 and three openings leading into it1 in which there is a depth from two to twenty fathoms4 ,outside1 at the distance 7,horsburgh1 vol4 ii1 p4 #ccc7 of only twenty yards from the reef1 soundings could not be obtained4 ,the ,paracells are coloured blue4--,macclesfield ,bank3 this is a coral-bank of great size1 lying east of the ,paracells2 some parts of the bank are level1 with a sandy bottom1 but1 generally1 the depth is very irregular4 ,it is intersected by deep cuts or channels4 ,i am not able to perceive in the published charts 7its limits1 however1 are not very accurately known7 whether the central part is deeper1 which ,i suspect is the case1 as in the ,great ,chagos ,bank1 in the ,indian ,ocean2 not coloured4--,scarborough ,shoal3 this coral-shoal is engraved with a double row of crosses1 forming a circle1 as if there was deep water within the reef3 close outside there was no bottom1 with a hundred fathoms2 coloured blue4--,the sea off the west coast of ,palawan and the northern part of ,borneo is strewed with shoals3 ,swallow ,shoal1 according to ,horsburgh 7vol4 ii1 p4 #dca7 8is formed1 like most of the shoals hereabouts1 of a belt of coral-rocks1 8with a basin of deep water within40--,half-,moon ,shoal has a similar structure2 ,captain ,d4 ,ross describes it1 as a narrow belt of coral-rock1 with a basin of deep water in the centre10 and deep sea close outside4--,bombay ,shoal appears 7,horsburgh1 vol4 ii1 p4 #dcb7 8to be a basin of smooth water surrounded by breakers40 ,these three shoals ,i have coloured blue4--,the ,paraquas ,shoals are of a circular form1 with deep gaps running through them2 not coloured4--,a bank gradually shoaling to the depth of thirty fathoms1 extends to a distance of about twenty miles from the northern part of ,borneo1 and to thirty miles from the northern part of ,palawan4 ,near the land this bank appears tolerably free from danger1 but a little further out it is thickly studded with coral-shoals1 which do not generally rise quite to the surface2 some of them are very steep-to1 and others have a fringe of shoal-water round them4 ,i should have thought that these shoals had level surfaces1 had it not been for the statement made by ,horsburgh 8that most of the shoals hereabouts are formed of a belt of coral40 ,but1 perhaps that expression was more particularly applied to the shoals further in the offing4 ,if these reefs of coral have a lagoon-like structure1 they should have been coloured blue1 and they would have formed an imperfect barrier in front of ,palawan and the northern part of ,borneo4 ,but1 as the water is not very deep1 these reefs may have grown up from inequalities on the bank3 ,i have not coloured them4--,the coast of ,china1 ,tonquin1 and ,cochin-,china1 forming the western boundary of the ,china ,sea1 appear to be without reefs3 with regard to the two last-mentioned coasts1 ,i speak after examining the charts on a large scale in the 8,atlas of the ,voyage of the ,favourite40 ,,indian ,,ocean4--,south ,keeling atoll has been specially described4 ,nine miles north of it lies ,north ,keeling1 a very small atoll1 surveyed by the ,beagle1 the lagoon of which is dry at low water4--,christmas ,island1 lying to the east1 is a high island1 without1 as ,i have been informed by a person who passed it1 any reefs at all4--,,ceylon3 a space about eighty miles in length of the south-western and southern shores of these islands has been described by ,mr4 ,twynam 7,naut4 ,mag4 #ahcf1 pp4 #cfe and #eah72 parts of this space appear to be very regularly fringed by coral-reefs1 which extend from a quarter to half a mile from the shore4 ,these reefs are in places breached1 and afford safe anchorage for the small trading craft4 ,outside1 the sea gradually deepens2 there is forty fathoms about six miles off shore3 this part ,i have coloured red4 ,in the published charts of ,ceylon there appear to be fringing-reefs in several parts of the south-eastern shores1 which ,i have also coloured red4--,at ,venloos ,bay the shore is likewise fringed4 ,north of ,trincomalee there are also reefs of the same kind4 ,the sea off the northern part of ,ceylon is exceedingly shallow2 and therefore ,i have not coloured the reefs which fringe portions of its shores1 and the adjoining islets1 as well as the ,indian promontory of ,madura4 ,,chagos1 ,,maldiva1 ,,and ,,laccadive ,,archipelagoes4--,these three great groups which have already been often noticed1 are now well-known from the admirable surveys of ,captain ,moresby and ,lieutenant ,powell4 ,the published charts1 which are worthy of the most attentive examination1 at once show that the ,chagos and ,maldiva groups are entirely formed of great atolls1 or lagoon-formed reefs1 surmounted by islets4 ,in the ,laccadive group1 this structure is less evident2 the islets are low1 not exceeding the usual height of coral-formations 7see ,lieutenant ,wood's account1 ,geograph4 ,journ41 vol4 vi1 p4 #bi71 and most of the reefs are circular1 as may be seen in the published charts2 and within several of them1 as ,i am informed by ,captain ,moresby1 there is deepish water2 these1 therefore1 have been coloured blue4 ,directly north1 and almost forming part of this group1 there is a long1 narrow1 slightly curved bank1 rising out of the depths of the ocean1 composed of sand1 shells1 and decayed coral1 with from twenty-three to thirty fathoms on it4 ,i have no doubt that it has had the same origin with the other ,laccadive banks2 but as it does not deepen towards the centre ,i have not coloured it4 ,i might have referred to other authorities regarding these three archipelagoes2 but after the publication of the charts by ,captain ,moresby1 to whose personal kindness in giving me much information ,i am exceedingly indebted1 it would have been superfluous4 ,sahia de ,malha bank consists of a series of narrow banks1 with from eight to sixteen fathoms on them2 they are arranged in a semicircular manner1 round a space about forty fathoms deep1 which slopes on the ,s4,e4 quarter to unfathomable depths2 they are steep-to on both sides1 but more especially on the ocean-side4 ,hence this bank closely resembles in structure1 and ,i may add from ,captain ,moresby's information in composition1 the ,pitt's ,bank in the ,chagos group2 and the ,pitt's ,bank1 must1 after what has been shown of the ,great ,chagos ,bank1 be considered as a sunken1 half-destroyed atoll2 hence coloured blue4--,cargados ,carajos ,bank4 ,its southern portion consists of a large1 curved1 coral-shoal1 with some low islets on its eastern edge1 and likewise some on the western side1 between which there is a depth of about twelve fathoms4 ,northward1 a great bank extends4 ,i cannot 7probably owing to the want of perfect charts7 refer this reef and bank to any class2--therefore not coloured4--,ile de ,sable is a little island1 lying west of ,c4 ,carajos1 only some toises in height 78,voyage of the ,favourite10 vol4 i1 p4 #acj72 it is surrounded by reefs2 but its structure is unintelligible to me4 ,there are some small banks north of it1 of which ,i can find no clear account4--,mauritius4 ,the reefs round this island have been described in the chapter on fringing-reefs2 coloured red4--,rodriguez4 ,the coral-reefs here are exceedingly extensive2 in one part they project even five miles from the shore4 ,as far as ,i can make out1 there is no deep-water moat within them2 and the sea outside does not deepen very suddenly4 ,the outline1 however1 of the land appears to be 78,life of ,sir ,j4 ,makintosh10 vol4 ii1 p4 #afe7 hilly and rugged4 ,i am unable to decide whether these reefs belong to the barrier class2 as seems probable from their great extension1 or to the fringing class2 uncoloured4--,bourbon4 ,the greater part of the shores of this island are without reefs2 but ,captain ,carmichael 7,hooker's 8,bot4 ,misc407 states that a portion1 fifteen miles in length1 on the ,s4,e4 side1 is imperfectly fringed with coral reefs3 ,i have not thought this sufficient to colour the island4 ,,seychelles4--,the rocky islands of primary formation1 composing this group1 rise from a very extensive and tolerably level bank1 having a depth between twenty and forty fathoms4 ,in ,captain ,owen's chart1 and in that in the 8,atlas of the ,voyage of the ,favourite10 it appears that the east side of ,mahe and the adjoining islands of ,st4 ,anne and ,cerf1 are regularly fringed by coral-reefs4 ,a portion of the ,s4,e4 part of ,curieuse ,island1 the ,n41 and part of the ,s4,w4 shore of ,praslin ,island1 and the whole west side of ,digue ,island1 appear fringed4 ,from a ,,ms4 account of these islands by ,captain ,f4 ,moresby1 in the ,admiralty1 it appears that ,silhouette is also fringed2 he states that all these islands are formed of granite and quartz1 that they rise abruptly from the sea1 and that 8coral-reefs have grown round them1 and project for some distance40 ,dr4 ,allan1 of ,forres1 who visited these islands1 informs me that there is no deep water between the reefs and the shore4 ,the above specified points have been coloured red4 ,amirantes ,islands3 ,the small islands of this neighbouring group1 according to the ,,ms4 account of them by ,captain ,f4 ,moresby1 are situated on an extensive bank2 they consist of the debris of corals and shells2 are only about twenty feet in height1 and are environed by reefs1 some attached to the shore1 and some rather distant from it4--,i have taken great pains to procure plans and information regarding the several islands lying between ,s4,e4 and ,s4,w4 of the ,amirantes1 and the ,seychelles2 relying chiefly on ,captain ,f4 ,moresby and ,dr4 ,allan1 it appears that the greater number1 namely--,platte1 ,alphonse1 ,coetivi1 ,galega1 ,providence1 ,st4 ,pierre1 ,astova1 ,assomption1 and ,glorioso1 are low1 formed of sand or coral-rock1 and irregularly shaped2 they are situated on very extensive banks1 and are connected with great coral-reefs4 ,galega is said by ,dr4 ,allan1 to be rather higher than the other islands2 and ,st4 ,pierre is described by ,captain ,f4 ,moresby1 as being cavernous throughout1 and as not consisting of either limestone or granite4 ,these islands1 as well as the ,amirantes1 certainly are not atoll-formed1 and they differ as a group from every other group with which ,i am acquainted2 ,i have not coloured them2 but probably the reefs belong to the fringing class4 ,their formation is attributed1 both by ,dr4 ,allan and ,captain ,f4 ,moresby1 to the action of the currents1 here exceedingly violent1 on banks1 which no doubt have had an independent geological origin4 ,they resemble in many respects some islands and banks in the ,west ,indies1 which owe their origin to a similar agency1 in conjunction with an elevation of the entire area4 ,in close vicinity to the several islands1 there are three others of an apparently different nature3 first1 ,juan de ,nova1 which appears from some plans and accounts to be an atoll2 but from others does not appear to be so2 not coloured4 ,secondly ,cosmoledo2 8this group consists of a ring of coral1 ten leagues in circumference1 and a quarter of a mile broad in some places1 enclosing a magnificent lagoon1 into which there did not appear a single opening0 7,horsburgh1 vol4 i1 p4 #aea72 coloured blue4 ,thirdly1 ,aldabra2 it consists of three islets1 about twenty-five feet in height1 with red cliffs 7,horsburgh1 vol4 i1 p4 #agf7 surrounding a very shallow basin or lagoon4 ,the sea is profoundly deep close to the shore4 ,viewing this island in a chart1 it would be thought an atoll2 but the foregoing description shows that there is something different in its nature2 ,dr4 ,allan also states that it is cavernous1 and that the coral-rock has a vitrified appearance4 ,is it an upheaved atoll1 or the crater of a volcano8--uncoloured4 ,,comoro ,,group4--,mayotta1 according to ,horsburgh 7vol4 i1 p4 #baf1 #dth ed471 is completely surrounded by a reef1 which runs at the distance of three1 four1 and in some places even five miles from the land2 in an old chart1 published by ,dalrymple1 a depth in many places of thirty-six and thirty-eight fathoms is laid down within the reef4 ,in the same chart1 the space of open water within the reef in some parts is even more than three miles wide3 the land is bold and peaked2 this island1 therefore1 is encircled by a well-characterised barrier-reef1 and is coloured pale blue4--,johanna2 ,horsburgh says 7vol4 i1 p4 #bag7 this island from the ,n4,w4 to the ,s4,w4 point1 is bounded by a reef1 at the distance of two miles from the shore2 in some parts1 however1 the reef must be attached1 since ,lieutenant ,boteler 78,narr40 vol4 i1 p4 #afa7 describes a passage through it1 within which there is room only for a few boats4 ,its height1 as ,i am informed by ,dr4 ,allan1 is about #c1ejj feet2 it is very precipitous1 and is composed of granite1 greenstone1 and quartz2 coloured blue4--,mohilla2 on the ,s4 side of this island there is anchorage1 in from thirty to forty-five fathoms1 between a reef and the shore 7,horsburgh1 vol4 i1 p4 #bad72 in ,captain ,owen's chart of ,madagascar1 this island is represented as encircled2 coloured blue4--,great ,comoro ,island is1 as ,i am informed by ,dr4 ,allan1 about #h1jjj feet high1 and apparently volcanic2 it is not regularly encircled2 but reefs of various shapes and dimensions1 jut out from every headland on the ,w41 ,s41 and ,s4,e4 coasts1 inside of which reefs there are channels1 often parallel with the shore1 with deep water4 ,on the north-western coasts the reefs appear attached to the shores4 ,the land near the coast is in some places bold1 but generally speaking it is flat2 ,horsburgh says 7vol4 i1 p4 #bad7 the water is profoundly deep close to the shore1 from which expression ,i presume some parts are without reefs4 ,from this description ,i apprehend the reef belongs to the barrier class2 but ,i have not coloured it1 as most of the charts which ,i have seen1 represent the reefs round it as very much less extensive than round the other islands in the group4 ,,madagascar4--,my information is chiefly derived from the published charts by ,captain ,owen1 and the accounts given by him and by ,lieutenant ,boteler4 ,commencing at the ,s4,w4 extremity of the island2 towards the northern part of the ,star ,bank 7in lat4 #be ,s47 the coast for ten miles is fringed by a reef2 coloured red4 ,the shore immediately ,s4 of ,st4 ,augustine's ,bay appears fringed2 but ,tullear ,harbour1 directly ,n4 of it1 is formed by a narrow reef ten miles long1 extending parallel to the shore1 with from four to ten fathoms within it4 ,if this reef had been more extensive1 it must have been classed as a barrier-reef2 but as the line of coast falls inwards here1 a submarine bank perhaps extends parallel to the shore1 which has offered a foundation for the growth of the coral2 ,i have left this part uncoloured4 ,from lat4 #bb #af to #ba #cg1 the shore is fringed by coral-reefs 7see ,lieutenant ,boteler's 8,narrative10 vol4 ii1 p4 #ajf71 less than a mile in width1 and with shallow water within4 ,there are outlying coral-shoals in several parts of the offing1 with about ten fathoms between them and the shore1 and the depth of the sea one mile and a half seaward1 is about thirty fathoms4 ,the part above specified is engraved on a large scale2 and as in the charts on rather a smaller scale the same fringe of reef extends as far as lat4 #cc #ae2 ,i have coloured the whole of this part of the coast red4 ,the islands of ,juan de ,nova 7in lat4 #ag ,s47 appear in the charts on a large scale to be fringed1 but ,i have not been able to ascertain whether the reefs are of coral2 uncoloured4 ,the main part of the west coast appears to be low1 with outlying sandbanks1 which1 ,lieutenant ,boteler 7vol4 ii1 p4 #ajf7 says1 8are faced on the edge of deep water by a line of sharp-pointed coral-rocks40 ,nevertheless ,i have not coloured this part1 as ,i cannot make out by the charts that the coast itself is fringed4 ,the headlands of ,narrenda and ,passandava ,bays 7#ad #dj7 and the islands in front of ,radama ,harbour are represented in the plans as regularly fringed1 and have accordingly been coloured red4 ,with respect to the ,east coast of ,madagascar1 ,dr4 ,allan informs me in a letter1 that the whole line of coast1 from ,tamatave1 in #ah #ab1 to ,c4 ,amber1 at the extreme northern point of the island1 is bordered by coral-reefs4 ,the land is low1 uneven1 and gradually rising from the coast4 ,from ,captain ,owen's charts1 also1 the existence of these reefs1 which evidently belong to the fringing class1 on some parts1 namely ,n4 of ,british ,sound1 and near ,ngoncy1 of the above line of coast might have been inferred4 ,lieutenant ,boteler 7vol4 i1 p4 #aee7 speaks of 8the reef surrounding the island of ,st4 ,mary's at a small distance from the shore40 ,in a previous chapter ,i have described1 from the information of ,dr4 ,allan1 the manner in which the reefs extend in ,n4,e4 lines from the headlands on this coast1 thus sometimes forming rather deep channels within them1 this seems caused by the action of the currents1 and the reefs spring up from the submarine prolongations of the sandy headlands4 ,the above specified portion of the coast is coloured red4 ,the remaining ,s4,e4 portions do not appear in any published chart to possess reefs of any kind2 and the ,rev4 ,w4 ,ellis1 whose means of information regarding this side of ,madagascar have been extensive1 informs me he believes there are none4 ,,east ,,coast ,,of ,,africa4--,proceeding from the northern part1 the coast appears1 for a considerable space1 without reefs4 ,my information1 ,i may here observe1 is derived from the survey by ,captain ,owen1 together with his narrative2 and that by ,lieutenant ,boteler4 ,at ,mukdeesha 7#aj #a ,n47 there is a coral-reef extending four or five miles along the shore 7,owen's 8,narr40 vol4 i1 p4 #ceg7 which in the chart lies at the distance of a quarter of a mile from the shore1 and has within it from six to ten feet water3 this then is a fringing-reef1 and is coloured red4 ,from ,juba1 a little ,s4 of the equator1 to ,lamoo 7in #b #bj ,s47 8the coast and islands are formed of madrepore0 7,owen's 8,narrative10 vol4 i1 p4 #cfc74 ,the chart of this part 7entitled ,dundas ,islands71 presents an extraordinary appearance2 the coast of the mainland is quite straight and it is fronted at the average distance of two miles1 by exceedingly narrow1 straight islets1 fringed with reefs4 ,within the chain of islets1 there are extensive tidal flats and muddy bays1 into which many rivers enter2 the depths of these spaces varies from one to four fathoms--the latter depth not being common1 and about twelve feet the average4 ,outside the chain of islets1 the sea1 at the distance of a mile1 varies in depth from eight to fifteen fathoms4 ,lieutenant ,boteler 78,narr410 vol4 i1 p4 #cfi7 describes the muddy bay of ,patta1 which seems to resemble other parts of this coast1 as fronted by small1 narrow1 level islets formed of decomposing coral1 the margin of which is seldom of greater height than twelve feet1 overhanging the rocky surface from which the islets rise4 ,knowing that the islets are formed of coral1 it is1 ,i think1 scarcely possible to view the coast1 and not at once conclude that we here see a fringing-reef1 which has been upraised a few feet3 the unusual depth of from two to four fathoms within some of these islets1 is probably due to muddy rivers having prevented the growth of coral near the shore4 ,there is1 however1 one difficulty on this view1 namely1 that before the elevation took place1 which converted the reef into a chain of islets1 the water must apparently have been still deeper2 on the other hand it may be supposed that the formation of a nearly perfect barrier in front1 of so large an extent of coast1 would cause the currents 7especially in front of the rivers71 to deepen their muddy beds4 ,when describing in the chapter on fringing-reefs1 those of ,mauritius1 ,i have given my reasons for believing that the shoal spaces within reefs of this kind1 must1 in many instances1 have been deepened4 ,however this may be1 as several parts of this line of coast are undoubtedly fringed by living reefs1 ,i have coloured it red4--,maleenda 7#c #bj ,s474 ,in the plan of the harbour1 the south headland appears fringed2 and in ,owen's chart on a larger scale1 the reefs are seen to extend nearly thirty miles southward2 coloured red4--,mombas 7#d #e ,s474 ,the island which forms the harbour1 8is surrounded by cliffs of madrepore1 capable of being rendered almost impregnable0 7,owen's 8,narr410 vol4 i1 p4 #dab74 ,the shore of the mainland ,n4 and ,s4 of the harbour1 is most regularly fringed by a coral-reef at a distance from half a mile to one mile and a quarter from the land2 within the reef the depth is from nine to fifteen feet2 outside the reef the depth at rather less than half a mile is thirty fathoms4 ,from the charts it appears that a space about thirty-six miles in length1 is here fringed2 coloured red4--,pemba 7#e ,s47 is an island of coral-formation1 level1 and about two hundred feet in height 7,owen's 8,narr410 vol4 i1 p4 #dbe72 it is thirty-five miles long1 and is separated from the mainland by a deep sea4 ,the outer coast is represented in the chart as regularly fringed2 coloured red4 ,the mainland in front of ,pemba is likewise fringed2 but there also appear to be some outlying reefs with deep water between them and the shore4 ,i do not understand their structure1 either from the charts or the description1 therefore have not coloured them4--,zanzibar resembles ,pemba in most respects2 its southern half on the western side and the neighbouring islets are fringed2 coloured red4 ,on the mainland1 a little ,s4 of ,zanzibar1 there are some banks parallel to the coast1 which ,i should have thought had been formed of coral1 had it not been said 7,boteler's 8,narr410 vol4 ii1 p4 #ci7 that they were composed of sand2 not coloured4--,latham's ,bank is a small island1 fringed by coral-reefs2 but being only ten feet high1 it has not been coloured4--,monfeea is an island of the same character as ,pemba2 its outer shore is fringed1 and its southern extremity is connected with ,keelwa ,point on the mainland by a chain of islands fringed by reefs2 coloured red4 ,the four last-mentioned islands resemble in many respects some of the islands in the ,red ,sea1 which will presently be described4--,keelwa4 ,in a plan of the shore1 a space of twenty miles ,n4 and ,s4 of this place is fringed by reefs1 apparently of coral3 these reefs are prolonged still further southward in ,owen's general chart4 ,the coast in the plans of the rivers ,lindy and ,monghow 7#i #ei and #aj #g ,s47 has the same structure2 coloured red4--,querimba ,islands 7from #aj #dj to #ac ,s474 ,a chart on a large scale is given of these islands2 they are low1 and of coral-formation 7,boteler's 8,narr410 vol4 ii1 p4 #ed72 and generally have extensive reefs projecting from them which are dry at low water1 and which on the outside rise abruptly from a deep sea3 on their insides they are separated from the continent by a channel1 or rather a succession of bays1 with an average depth of ten fathoms4 ,the small headlands on the continent also have coral-banks attached to them2 and the ,querimba islands and banks are placed on the lines of prolongation of these headlands1 and are separated from them by very shallow channels4 ,it is evident that whatever cause1 whether the drifting of sediment or subterranean movements1 produced the headlands1 likewise produced1 as might have been expected1 submarine prolongations to them2 and these towards their outer extremities1 have since afforded a favourable basis for the growth of coral-reefs1 and subsequently for the formation of islets4 ,as these reefs clearly belong to the fringing class1 the ,querimba islands have been coloured red4--,monabila 7#ac #cb ,s474 ,in the plan of this harbour1 the headlands outside are fringed by reefs apparently of coral2 coloured red4--,mozambique 7#aej ,s47 ,the outer part of the island on which the city is built1 and the neighbouring islands1 are fringed by coral-reefs2 coloured red4 ,from the description given in ,owen's 8,narr40 7vol4 i1 p4 #afb71 the shore from ,mozambique to ,delagoa ,bay appears to be low and sandy2 many of the shoals and islets off this line of coast are of coral-formation2 but from their small size and lowness1 it is not possible1 from the charts1 to know whether they are truly fringed4 ,hence this portion of coast is left uncoloured1 as are likewise those parts more northward1 of which no mention has been made in the foregoing pages from the want of information4 ,,persian ,,gulf4--,from the charts lately published on a large scale by the ,east ,india ,company1 it appears that several parts1 especially the southern shores of this gulf1 are fringed by coral-reefs2 but as the water is very shallow1 and as there are numerous sandbanks1 which are difficult to distinguish on the chart from reefs1 ,i have not coloured the upper part red4 ,towards the mouth1 however1 where the water is rather deeper1 the islands of ,ormuz and ,larrack1 appear so regularly fringed1 that ,i have coloured them red4 ,there are certainly no atolls in the ,persian ,gulf4 ,the shores of ,immaum1 and of the promontory forming the southern headland of the ,persian ,gulf1 seem to be without reefs4 ,the whole ,s4,w4 part 7except one or two small patches7 of ,arabia ,felix1 and the shores of ,socotra appear from the charts and memoir of ,captain ,haines 7,geograph4 ,journ41 #ahci1 p4 #abe7 to be without any reefs4 ,i believe there are no extensive coral-reefs on any part of the coasts of ,india1 except on the low promontory of ,madura 7as already mentioned7 in front of ,ceylon4 ,,red ,,sea4--,my information is chiefly derived from the admirable charts published by the ,east ,india ,company in #ahcf1 from personal communication with ,captain ,moresby1 one of the surveyors1 and from the excellent memoir1 8ber die ,natur der ,corallen-,bnken des ,rothen ,meeres10 by ,ehrenberg4 ,the plains immediately bordering the ,red ,sea seem chiefly to consist of a sedimentary formation of the newer tertiary period4 ,the shore is1 with the exception of a few parts1 fringed by coral-reefs4 ,the water is generally profoundly deep close to the shore2 but this fact1 which has attracted the attention of most voyagers1 seems to have no necessary connection with the presence of reefs2 for ,captain ,moresby particularly observed to me1 that1 in lat4 #bd #aj on the eastern side1 there is a piece of coast1 with very deep water close to it1 without any reefs1 but not differing in other respects from the usual nature of the coast-line4 ,the most remarkable feature in the ,red ,sea is the chain of submerged banks1 reefs1 and islands1 lying some way from the shore1 chiefly on the eastern side2 the space within being deep enough to admit a safe navigation in small vessels4 ,the banks are generally of an oval form1 and some miles in width2 but some of them are very long in proportion to their width4 ,captain ,moresby informs me that any one1 who had not made actual plans of them1 would be apt to think that they were much more elongated than they really are4 ,many of them rise to the surface1 but the greater number lie from five to thirty fathoms beneath it1 with irregular soundings on them4 ,they consist of sand and living coral2 coral on most of them1 according to ,captain ,moresby1 covering the greater part of their surface4 ,they extend parallel to the shore1 and they are not unfrequently connected in their middle parts by short transverse banks with the mainland4 ,the sea is generally profoundly deep quite close to them1 as it is near most parts of the coast of the mainland2 but this is not universally the case1 for between lat4 #ae and #ag the water deepens quite gradually from the banks1 both on the eastern and western shores1 towards the middle of the sea4 ,islands in many parts arise from these banks2 they are low1 flat-topped1 and consist of the same horizontally stratified formation with that forming the plain-like margin of the mainland4 ,some of the smaller and lower islands consist of mere sand4 ,captain ,moresby informs me1 that small masses of rock1 the remnants of islands1 are left on many banks where there is now no dry land4 ,ehrenberg also asserts that most of the islets1 even the lowest1 have a flat abraded basis1 composed of the same tertiary formation3 he believes that as soon as the surf wears down the protuberant parts of a bank1 just beneath the level of the sea1 the surface becomes protected from further abrasion by the growth of coral1 and he thus accounts for the existence of so many banks standing on a level with the surface of this sea4 ,it appears that most of the islands are certainly decreasing in size4 ,the form of the banks and islands is most singular in the part just referred to1 namely1 from lat4 #ae to #ag1 where the sea deepens quite gradually3 the ,dhalac group1 on the western coast1 is surrounded by an intricate archipelago of islets and shoals2 the main island is very irregularly shaped1 and it includes a bay seven miles long1 by four across1 in which no bottom was found with #beb feet3 there is only one entrance into this bay1 half a mile wide1 and with an island in front of it4 ,the submerged banks on the eastern coast1 within the same latitudes1 round ,farsan ,island1 are1 likewise1 penetrated by many narrow creeks of deep water2 one is twelve miles long1 in the form of a hatchet1 in which1 close to its broad upper end1 soundings were not struck with #cfj feet1 and its entrance is only half a mile wide3 in another creek of the same nature1 but even with a more irregular outline1 there was no bottom with #dhj feet4 ,the island of ,farsan1 itself1 has as singular a form as any of its surrounding banks4 ,the bottom of the sea round the ,dhalac and ,farsan ,islands consists chiefly of sand and agglutinated fragments1 but1 in the deep and narrow creeks1 it consists of mud2 the islands themselves consist of thin1 horizontally stratified1 modern tertiary beds1 containing but little broken coral1#b their shores are fringed by living coral-reefs4 #b ,rppell1 8,reise in ,abyssinie10 ,band4 i1 ,s4 #bdg4 ,from the account given by ,rppell#c of the manner in which ,dhalac has been rent by fissures1 the opposite sides of which have been unequally elevated 7in one instance to the amount of fifty feet71 it seems probable that its irregular form1 as well as probably that of ,farsan1 may have been partly caused by unequal elevations2 but1 considering the general form of the banks1 and of the deep-water creeks1 together with the composition of the land1 ,i think their configuration is more probably due in great part to strong currents having drifted sediment over an uneven bottom3 it is almost certain that their form cannot be attributed to the growth of coral4 ,whatever may have been the precise origin of the ,dhalac and ,farsan ,archipelagoes1 the greater number of the banks on the eastern side of the ,red ,sea seem to have originated through nearly similar means4 ,i judge of this from their similarity in configuration 7in proof of which ,i may instance a bank on the east coast in lat4 #bb2 and although it is true that the northern banks generally have a less complicated outline71 and from their similarity in composition1 as may be observed in their upraised portions4 ,the depth within the banks northward of lat4 #ag1 is usually greater1 and their outer sides shelve more abruptly 7circumstances which seem to go together7 than in the ,dhalac and ,farsan ,archipelagoes2 but this might easily have been caused by a difference in the action of the currents during their formation3 moreover1 the greater quantity of living coral1 which1 according to ,captain ,moresby1 exists on the northern banks1 would tend to give them steeper margins4 #c ,ibid41 ,s4 #bde4 ,from this account1 brief and imperfect as it is1 we can see that the great chain of banks on the eastern coast1 and on the western side in the southern portion1 differ greatly from true barrier-reefs wholly formed by the growth of coral4 ,it is indeed the direct conclusion of ,ehrenberg 78ber die10 etc41 pp4 #de and #ea71 that they are connected in their origin quite secondarily with the growth of coral2 and he remarks that the islands off the coast of ,norway1 if worn down level with the sea1 and merely coated with living coral1 would present a nearly similar appearance4 ,i cannot1 however1 avoid suspecting1 from information given me by ,dr4 ,malcolmson and ,captain ,moresby1 that ,ehrenberg has rather under-rated the influence of corals1 in some places at least1 on the formation of the tertiary deposits of the ,red ,sea4 ,the west coast of the ,red ,sea between lat4 #ai and #bb4--,there are1 in this space1 reefs1 which1 if ,i had known nothing of those in other parts of the ,red ,sea1 ,i should unhesitatingly have considered as barrier-reefs2 and1 after deliberation1 ,i have come to the same conclusion4 ,one of these reefs1 in #bj #ae1 is twenty miles long1 less than a mile in width 7but expanding at the northern end into a disc71 slightly sinuous1 and extending parallel to the mainland at the distance of five miles from it1 with very deep water within2 in one spot soundings were not obtained with #bje fathoms4 ,some leagues further south1 there is another linear reef1 very narrow1 ten miles long1 with other small portions of reef1 north and south1 almost connected with it2 and within this line of reefs 7as well as outside7 the water is profoundly deep4 ,there are also some small linear and sickle-formed reefs1 lying a little way out at sea4 ,all these reefs are covered1 as ,i am informed by ,captain ,moresby1 by living corals4 ,here1 then1 we have all the characters of reefs of the barrier class2 and in some outlying reefs we have an approach to the structure of atolls4 ,the source of my doubts about the classification of these reefs1 arises from having observed in the ,dhalac and ,farsan groups the narrowness and straightness of several spits of sand and rock3 one of these spits in the ,dhalac group is nearly fifteen miles long1 only two broad1 and it is bordered on each side with deep water2 so that1 if worn down by the surf1 and coated with living corals1 it would form a reef nearly similar to those within the space under consideration4 ,there is1 also1 in this space 7lat4 #ba7 a peninsula1 bordered by cliffs1 with its extremity worn down to the level of the sea1 and its basis fringed with reefs3 in the line of prolongation of this peninsula1 there lies the island of ,macowa 7formed1 according to ,captain ,moresby1 of the usual tertiary deposit71 and some smaller islands1 large parts of which likewise appear to have been worn down1 and are now coated with living corals4 ,if the removal of the strata in these several cases had been more complete1 the reefs thus formed would have nearly resembled those barrier-like ones now under discussion4 ,notwithstanding these facts1 ,i cannot persuade myself that the many very small1 isolated1 and sickle-formed reefs and others1 long1 nearly straight1 and very narrow1 with the water unfathomably deep close round them1 could possibly have been formed by corals merely coating banks of sediment1 or the abraded surfaces of irregularly shaped islands4 ,i feel compelled to believe that the foundations of these reefs have subsided1 and that the corals1 during their upward growth1 have given to these reefs their present forms3 ,i may remark that the subsidence of narrow and irregularly-shaped peninsulas and islands1 such as those existing on the coasts of the ,red ,sea1 would afford the requisite foundations for the reefs in question4 ,the west coast from lat4 #bb to #bd4--,this part of the coast 7north of the space coloured blue on the map7 is fronted by an irregularly shelving bank1 from about ten to thirty fathoms deep2 numerous little reefs1 some of which have the most singular shapes1 rise from this bank4 ,it may be observed1 respecting one of them1 in lat4 #bc #aj1 that if the promontory in lat4 #bd were worn down to the level of the sea1 and coated with corals1 a very similar and grotesquely formed reef would be produced4 ,many of the reefs on this part of the coast may thus have originated2 but there are some sickle1 and almost atoll-formed reefs lying in deep water off the promontory in lat4 #bd1 which lead me to suppose that all these reefs are more probably allied to the barrier or atoll classes4 ,i have not1 however1 ventured to colour this portion of coast4 ,on the west coast from lat4 #ai to #ag 7south of space coloured blue on the map71 there are many low islets of very small dimensions1 not much elongated1 and rising out of great depths at a distance from the coast2 these cannot be classed either with atolls1 or barrier- or fringing-reefs4 ,i may here remark that the outlying reefs on the west coast1 between lat4 #ai and #bd1 are the only ones in the ,red ,sea1 which approach in structure to the true atolls of the ,indian and ,pacific ,oceans1 but they present only imperfect miniature likenesses of them4 ,eastern coast4--,i have felt the greatest doubt about colouring any portion of this coast1 north of the fringing-reefs round the ,farsan ,islands in #af #aj4 ,there are many small outlying coral-reefs along the whole line of coast2 but as the greater number rise from banks not very deeply submerged 7the formation of which has been shown to be only secondarily connected with the growth of coral71 their origin may be due simply to the growth of knolls of corals1 from an irregular foundation situated within a limited depth4 ,but between lat4 #ah and #bj1 there are so many linear1 elliptic1 and extremely small reefs1 rising abruptly out of profound depths1 that the same reasons1 which led me to colour blue a portion of the west coast1 have induced me to do the same in this part4 ,there exist some small outlying reefs rising from deep water1 north of lat4 #bj 7the northern limit coloured blue71 on the east coast2 but as they are not very numerous and scarcely any of them linear1 ,i have thought it right to leave them uncoloured4 ,in the southern parts of the ,red ,sea1 considerable spaces of the mainland1 and of some of the ,dhalac islands1 are skirted by reefs1 which1 as ,i am informed by ,captain ,moresby1 are of living coral1 and have all the characters of the fringing class4 ,as in these latitudes1 there are no outlying linear or sickle-formed reefs1 rising out of unfathomable depths1 ,i have coloured these parts of the coast red4 ,on similar grounds1 ,i have coloured red the northern parts of the western coast 7north of lat4 #bd #cj71 and likewise the shores of the chief part of the ,gulf of ,suez4 ,in the ,gulf of ,acaba1 as ,i am informed by ,captain ,moresby there are no coral-reefs1 and the water is profoundly deep4 ,,west ,,indies4--,my information regarding the reefs of this area1 is derived from various sources1 and from an examination of numerous charts2 especially of those lately executed during the survey under ,captain ,owen1 ,r4,n4 ,i lay under particular obligation to ,captain ,bird ,allen1 ,r4,n41 one of the members of the late survey1 for many personal communications on this subject4 ,as in the case of the ,red ,sea1 it is necessary to make some preliminary remarks on the submerged banks of the ,west ,indies1 which are in some degree connected with coral-reefs1 and cause considerable doubts in their classification4 ,that large accumulations of sediment are in progress on the ,west ,indian shores1 will be evident to any one who examines the charts of that sea1 especially of the portion north of a line joining ,yucutan and ,florida4 ,the area of deposition seems less intimately connected with the debouchement of the great rivers1 than with the course of the sea-currents2 as is evident from the vast extension of the banks from the promontories of ,yucutan and ,mosquito4 ,besides the coast-banks1 there are many of various dimensions which stand quite isolated2 these closely resemble each other1 they lie from two or three to twenty or thirty fathoms under water1 and are composed of sand1 sometimes firmly agglutinated1 with little or no coral2 their surfaces are smooth and nearly level1 shelving only to the amount of a few fathoms1 very gradually all round towards their edges1 where they plunge abruptly into the unfathomable sea4 ,this steep inclination of their sides1 which is likewise characteristic of the coast-banks1 is very remarkable3 ,i may give as an instance1 the ,misteriosa ,bank1 on the edges of which the soundings change in #bej fathoms horizontal distance1 from #aa to #baj fathoms2 off the northern point of the bank of ,old ,providence1 in #bjj fathoms horizontal distance1 the change is from #ai to #aeb fathoms2 off the ,great ,bahama ,bank1 in #afj fathoms horizontal distance1 the inclination is in many places from #aj fathoms to no bottom with #aij fathoms4 ,on coasts in all parts of the world1 where sediment is accumulating1 something of this kind may be observed2 the banks shelve very gently far out to sea1 and then terminate abruptly4 ,the form and composition of the banks standing in the middle parts of the ,w4 ,indian ,sea1 clearly show that their origin must be chiefly attributed to the accumulation of sediment2 and the only obvious explanation of their isolated position is the presence of a nucleus1 round which the currents have collected fine drift matter4 ,any one who will compare the character of the bank surrounding the hilly island of ,old ,providence1 with those banks in its neighbourhood which stand isolated1 will scarcely doubt that they surround submerged mountains4 ,we are led to the same conclusion by examining the bank called ,thunder ,knoll1 which is separated from the ,great ,mosquito ,bank by a channel only seven miles wide1 and #ade fathoms deep4 ,there cannot be any doubt that the ,mosquito ,bank has been formed by the accumulation of sediment round the promontory of the same name2 and ,thunder ,knoll resembles the ,mosquito ,bank1 in the state of its surface submerged twenty fathoms1 in the inclinations of its sides1 in composition1 and in every other respect4 ,i may observe1 although the remark is here irrelevant1 that geologists should be cautious in concluding that all the outlyers of any formation have once been connected together1 for we here see that deposits1 doubtless of exactly the same nature1 may be deposited with large valley-like spaces between them4 ,linear strips of coral-reefs and small knolls project from many of the isolated1 as well as coast-banks2 sometimes they occur quite irregularly placed1 as on the ,mosquito ,bank1 but more generally they form crescents on the windward side1 situated some little distance within the outer edge of the banks3--thus on the ,serranilla ,bank they form an interrupted chain which ranges between two and three miles within the windward margin3 generally they occur1 as on ,roncador1 ,courtown1 and ,anegada ,banks1 nearer the line of deep water4 ,their occurrence on the windward side is conformable to the general rule1 of the efficient kinds of corals flourishing best where most exposed2 but their position some way within the line of deep water ,i cannot explain1 without it be1 that a depth somewhat less than that close to the outer margin of the banks1 is most favourable to their growth4 ,where the corals have formed a nearly continuous rim1 close to the windward edge of a bank some fathoms submerged1 the reef closely resembles an atoll2 but if the bank surrounds an island 7as in the case of ,old ,providence71 the reef resembles an encircling barrier-reef4 ,i should undoubtedly have classed some of these fringed banks as imperfect atolls1 or barrier-reefs1 if the sedimentary nature of their foundations had not been evident from the presence of other neighbouring banks1 of similar forms and of similar composition1 but without the crescent-like marginal reef3 in the third chapter1 ,i observed that probably some atoll-like reefs did exist1 which had originated in the manner here supposed4 ,proofs of elevation within recent tertiary periods abound1 as referred to in the sixth chapter1 over nearly the whole area of the ,west ,indies4 ,hence it is easy to understand the origin of the low land on the coasts1 where sediment is now accumulating2 for instance on the northern part of ,yucutan1 and on the ,n4,e4 part of ,mosquito1 where the land is low1 and where extensive banks appear to be in progressive formation4 ,hence1 also1 the origin of the ,great ,bahama ,banks1 which are bordered on their western and southern edges by very narrow1 long1 singularly shaped islands1 formed of sand1 shells1 and coral-rock1 and some of them about a hundred feet in height1 is easily explained by the elevation of banks fringed on their windward 7western and southern7 sides by coral-reefs4 ,on this view1 however1 we must suppose either that the chief part of the surfaces of the great ,bahama sandbanks were all originally deeply submerged1 and were brought up to their present level by the same elevatory action1 which formed the linear islands2 or that during the elevation of the banks1 the superficial currents and swell of the waves continued wearing them down and keeping them at a nearly uniform level3 the level is not quite uniform2 for1 in proceeding from the ,n4,w4 end of the ,bahama group towards the ,s4,e4 end1 the depth of the banks increases1 and the area of land decreases1 in a very gradual and remarkable manner4 ,the latter view1 namely1 that these banks have been worn down by the currents and swell during their elevation1 seems to me the most probable one4 ,it is1 also1 ,i believe1 applicable to many banks1 situated in widely distant parts of the ,west ,indian ,sea1 which are wholly submerged2 for1 on any other view1 we must suppose1 that the elevatory forces have acted with astonishing uniformity4 ,the shores of the ,gulf of ,mexico1 for the space of many hundred miles1 is formed by a chain of lagoons1 from one to twenty miles in breadth 78,columbian ,navigator10 p4 #agh1 etc471 containing either fresh or salt water1 and separated from the sea by linear strips of sand4 ,great spaces of the shores of ,southern ,brazil1#d and of the ,united ,states from ,long ,island 7as observed by ,professor ,rogers7 to ,florida have the same character4 ,professor ,rogers1 in his 8,report to the ,british ,association0 7vol4 iii1 p4 #ac71 speculates on the origin of these low1 sandy1 linear islets2 he states that the layers of which they are composed are too homogeneous1 and contain too large a proportion of shells1 to permit the common supposition of their formation being simply due to matter thrown up1 where it now lies1 by the surf3 he considers these islands as upheaved bars or shoals1 which were deposited in lines where opposed currents met4 ,it is evident that these islands and spits of sand parallel to the coast1 and separated from it by shallow lagoons1 have no necessary connection with coral-formations4 ,but in ,southern ,florida1 from the accounts ,i have received from persons who have resided there1 the upraised islands seem to be formed of strata1 containing a good deal of coral1 and they are extensively fringed by living reefs2 the channels within these islands are in some places between two and three miles wide1 and five or six fathoms deep1 though generally#e they are less in depth than width4 ,after having seen how frequently banks of sediment in the ,west ,indian ,sea are fringed by reefs1 we can readily conceive that bars of sediment might be greatly aided in their formation along a line of coast1 by the growth of corals2 and such bars would1 in that case1 have a deceptive resemblance with true barrier-reefs4 #d ,in the ,london and ,edinburgh ,philosophical ,journal1 #ahda1 p4 #beg1 ,i have described a singular bar of sandstone lying parallel to the coast off ,pernambuco in ,brazil1 which probably is an analogous formation4 #e ,in the ordinary sea-charts1 no lagoons appear on the coast of ,florida1 north of #bf2 but ,major ,whiting 7,silliman's ,journal1 vol4 xxxv1 p4 #ed7 says that many are formed by sand thrown up along the whole line of coast from ,st4 ,augustine's to ,jupiter ,inlet4 ,having now endeavoured to remove some sources of doubt in classifying the reefs of the ,west ,indies1 ,i will give my authorities for colouring such portions of the coast as ,i have thought myself warranted in doing4 ,captain ,bird ,allen informs me1 that most of the islands on the ,bahama ,banks are fringed1 especially on their windward sides1 with living reefs2 and hence ,i have coloured those1 which are thus represented in ,captain ,owen's late chart1 red4 ,the same officer informs me1 that the islands along the southern part of ,florida are similarly fringed2 coloured red4 ,,cuba3 ,proceeding along the northern coast1 at the distance of forty miles from the extreme ,s4,e4 point1 the shores are fringed by reefs1 which extend westward for a space of #afj miles1 with only a few breaks4 ,parts of these reefs are represented in the plans of the harbours on this coast by ,captain ,owen2 and an excellent description is given of them by ,mr4 ,taylor 7,loudon's 8,mag4 of ,nat4 ,hist410 vol4 ix1 p4 #ddi72 he states that they enclosed a space called the 8baxo10 from half to three-quarters of a mile in width1 with a sandy bottom1 and a little coral4 ,in most parts people can wade1 at low water1 to the reef2 but in some parts the depth is between two and three fathoms4 ,close outside the reef1 the depth is between six and seven fathoms2 these well-characterised fringing-reefs are coloured red4 ,westward of longitude #gg #cj1 on the northern side of ,cuba1 a great bank commences1 which extends along the coast for nearly four degrees of longitude4 ,in the place of its commencement1 in its structure1 and in the 8cays10 or low islands on its edge1 there is a marked correspondence 7as observed by ,humboldt1 8,pers4 ,narr410 vol4 vii1 p4 #hh7 between it and the ,great ,bahama and ,sal ,banks1 which lie directly in front4 ,hence one is led to attribute the same origin to both these sets of banks2 namely1 the accumulation of sediment1 conjoined with an elevatory movement1 and the growth of coral on their outward edges2 those parts which appear fringed by living reefs are coloured red4 ,westward of these banks1 there is a portion of coast apparently without reefs1 except in the harbours1 the shores of which seem in the published plans to be fringed4 ,the ,colorado ,shoals 7see ,captain ,owen's charts71 and the low land at the western end of ,cuba1 correspond as closely in relative position and structure to the banks at the extreme point of ,florida1 as the banks above described on the north side of ,cuba1 do to the ,bahamas1 the depth within the islets and reefs on the outer edge of the ,colorados1 is generally between two and three fathoms1 increasing to twelve fathoms in the southern part1 where the bank becomes nearly open1 without islets or coral-reefs2 the portions which are fringed are coloured red4 ,the southern shore of ,cuba is deeply concave1 and the included space is filled up with mud and sandbanks1 low islands and coral-reefs4 ,between the mountainous ,isle of ,pines and the southern shore of ,cuba1 the general depth is only between two and three fathoms2 and in this part small islands1 formed of fragmentary rock and broken madrepores 7,humboldt1 8,pers4 ,narr410 vol4 vii1 pp4 #ea1 #hf to #ij1 #bia1 #cji1 #cbj71 rise abruptly1 and just reach the surface of the sea4 ,from some expressions used in the 8,columbian ,navigator0 7vol4 i1 pt ii1 p4 #id71 it appears that considerable spaces along the outer coast of ,southern ,cuba are bounded by cliffs of coral-rock1 formed probably by the upheaval of coral-reefs and sandbanks4 ,the charts represent the southern part of the ,isle of ,pines as fringed by reefs1 which the 8,columb4 ,navig40 says extend some way from the coast1 but have only from nine to twelve feet water on them2 these are coloured red4--,i have not been able to procure any detailed description of the large groups of banks and 8cays0 further eastward on the southern side of ,cuba2 within them there is a large expanse1 with a muddy bottom1 from eight to twelve fathoms deep2 although some parts of this line of coast are represented in the general charts of the ,west ,indies1 as fringed1 ,i have not thought it prudent to colour them4 ,the remaining portion of the south coast of ,cuba appears to be without coral-reefs4 ,,yucutan4--,the ,n4,e4 part of the promontory appears in ,captain ,owen's charts to be fringed2 coloured red4 ,the eastern coast1 from #bj to #ah is fringed4 ,south of lat4 #ah1 there commences the most remarkable reef in the ,west ,indies3 it is about one hundred and thirty miles in length1 ranging in a ,n4 and ,s4 line1 at an average distance of fifteen miles from the coast4 ,the islets on it are all low1 as ,i have been informed by ,captain ,b4 ,allen2 the water deepens suddenly on the outside of the reef1 but not more abruptly than off many of the sedimentary banks3 within its southern extremity 7off ,honduras7 the depth is twenty-five fathoms2 but in the more northern parts1 the depth soon increases to ten fathoms1 and within the northernmost part1 for a space of twenty miles1 the depth is only from one to two fathoms4 ,in most of these respects we have the characteristics of a barrier-reef2 nevertheless1 from observing1 first1 that the channel within the reef is a continuation of a great irregular bay1 which penetrates the mainland to the depth of fifty miles2 and secondly1 that considerable spaces of this barrier-like reef are described in the charts 7for instance1 in lat4 #af #de and #af #ab7 as formed of pure sand2 and thirdly1 from knowing that sediment is accumulating in many parts of the ,west ,indies in banks parallel to the shore2 ,i have not ventured to colour this reef as a barrier1 without further evidence that it has really been formed by the growth of corals1 and that it is not merely in parts a spit of sand1 and in other parts a worn down promontory1 partially coated and fringed by reefs2 ,i lean1 however1 to the probability of its being a barrier-reef1 produced by subsidence4 ,to add to my doubts1 immediately on the outside of this barrier-like reef1 ,turneffe1 ,lighthouse1 and ,glover reefs are situated1 and these reefs have so completely the form of atolls1 that if they had occurred in the ,pacific1 ,i should not have hesitated about colouring them blue4 ,turneffe ,reef seems almost entirely filled up with low mud islets2 and the depth within the other two reefs is only from one to three fathoms4 ,from this circumstance and from their similarity in form1 structure1 and relative position1 both to the bank called ,northern ,triangles1 on which there is an islet between seventy and eighty feet1 and to ,cozumel ,island1 the level surface of which is likewise between seventy and eighty feet in height1 ,i consider it more probable that the three foregoing banks are the worn down bases of upheaved shoals1 fringed with corals1 than that they are true atolls1 wholly produced by the growth of coral during subsidence2 left uncoloured4 ,in front of the eastern ,mosquito coast1 there are between lat4 #ab and #af some extensive banks 7already mentioned1 p4 #adh71 with high islands rising from their centres2 and there are other banks wholly submerged1 both of which kinds of banks are bordered1 near their windward margins1 by crescent-shaped coral-reefs4 ,but it can hardly be doubted1 as was observed in the preliminary remarks1 that these banks owe their origin1 like the great bank extending from the ,mosquito promontory1 almost entirely to the accumulation of sediment1 and not to the growth of corals2 hence ,i have not coloured them4 ,cayman ,island3 this island appears in the charts to be fringed2 and ,captain ,b4 ,allen informs me that the reefs extend about a mile from the shore1 and have only from five to twelve feet water within them2 coloured red4--,jamaica3 judging from the charts1 about fifteen miles of the ,s4,e4 extremity1 and about twice that length on the ,s4,w4 extremity1 and some portions on the ,s4 side near ,kingston and ,port ,royal1 are regularly fringed1 and therefore are coloured red4 ,from the plans of some harbours on the ,n4 side of ,jamaica1 parts of the coast appear to be fringed2 but as these are not represented in the charts of the whole island1 ,i have not coloured them4--,st4 ,domingo3 ,i have not been able to obtain sufficient information1 either from plans of the harbours1 or from general charts1 to enable me to colour any part of the coast1 except sixty miles from ,port de ,plata westward1 which seems very regularly fringed2 many other parts1 however1 of the coast are probably fringed1 especially towards the eastern end of the island4--,puerto ,rico3 considerable portions of the southern1 western1 and eastern coasts1 and some parts of the northern coast1 appear in the charts to be fringed2 coloured red4--,some miles in length of the southern side of the ,island of ,st4 ,thomas is fringed2 most of the ,virgin ,gorda ,islands1 as ,i am informed by ,mr4 ,schomburgk1 are fringed2 the shores of ,anegada1 as well as the bank on which it stands1 are likewise fringed2 these islands have been coloured red4 ,the greater part of the southern side of ,santa ,cruz appears in the ,danish survey to be fringed 7see also ,prof4 ,hovey's account of this island1 in ,silliman's ,journal1 vol4 xxxv1 p4 #gd72 the reefs extend along the shore for a considerable space1 and project rather more than a mile2 the depth within the reef is three fathoms2 coloured red4--,the ,antilles1 as remarked by ,von ,buch 78,descrip4 ,iles ,canaries10 p4 #did71 may be divided into two linear groups1 the western row being volcanic1 and the eastern of modern calcareous origin2 my information is very defective on the whole group4 ,of the eastern islands1 ,barbuda and the western coasts of ,antigua and ,mariagalante appear to be fringed3 this is also the case with ,barbadoes1 as ,i have been informed by a resident2 these islands are coloured red4 ,on the shores of the ,western ,antilles1 of volcanic origin1 very few coral-reefs appear to exist4 ,the island of ,martinique1 of which there are beautifully executed ,french charts1 on a very large scale1 alone presents any appearance worthy of special notice4 ,the south-western1 southern1 and eastern coasts1 together forming about half the circumference of the island1 are skirted by very irregular banks1 projecting generally rather less than a mile from the shore1 and lying from two to five fathoms submerged4 ,in front of almost every valley1 they are breached by narrow1 crooked1 steep-sided passages4 ,the ,french engineers ascertained by boring1 that these submerged banks consisted of madreporitic rocks1 which were covered in many parts by thin layers of mud or sand4 ,from this fact1 and especially from the structure of the narrow breaches1 ,i think there can be little doubt that these banks once formed living reefs1 which fringed the shores of the island1 and like other reefs probably reached the surface4 ,from some of these submerged banks reefs of living coral rise abruptly1 either in small detached patches1 or in lines parallel to1 but some way within the outer edges of the banks on which they are based4 ,besides the above banks which skirt the shores of the island1 there is on the eastern side a range of linear banks1 similarly constituted1 twenty miles in length1 extending parallel to the coast line1 and separated from it by a space between two and four miles in width1 and from five to fifteen fathoms in depth4 ,from this range of detached banks1 some linear reefs of living coral likewise rise abruptly2 and if they had been of greater length 7for they do not front more than a sixth part of the circumference of the island71 they would necessarily from their position have been coloured as barrier-reefs2 as the case stands they are left uncoloured4 ,i suspect that after a small amount of subsidence1 the corals were killed by sand and mud being deposited on them1 and the reefs being thus prevented from growing upwards1 the banks of madreporitic rock were left in their present submerged condition4 ,,the ,,bermuda ,,islands have been carefully described by ,lieutenant ,nelson1 in an excellent ,memoir in the 8,geological ,transactions0 7vol4 v1 part i1 p4 #ajc74 ,in the form of the bank or reef1 on one side of which the islands stand1 there is a close general resemblance to an atoll2 but in the following respects there is a considerable difference1--first1 in the margin of the reef not forming 7as ,i have been informed by ,mr4 ,chaffers1 ,r4,n47 a flat1 solid surface1 laid bare at low water1 and regularly bounding the internal space of shallow water or lagoon2 secondly1 in the border of gradually shoaling water1 nearly a mile and a half in width1 which surrounds the entire outside of the reef 7as is laid down in ,captain ,hurd's chart72 and thirdly1 in the size1 height1 and extraordinary form of the islands1 which present little resemblance to the long1 narrow1 simple islets1 seldom exceeding half a mile in breadth1 which surmount the annular reefs of almost all the atolls in the ,indian and ,pacific ,oceans4 ,moreover1 there are evident proofs 7,nelson1 ,ibid41 p4 #aah71 that islands similar to the existing ones1 formerly extended over other parts of the reef4 ,it would1 ,i believe1 be difficult to find a true atoll with land exceeding thirty feet in height2 whereas1 ,mr4 ,nelson estimates the highest point of the ,bermuda ,islands to be #bfj feet2 if1 however1 ,mr4 ,nelson's view1 that the whole of the land consists of sand drifted by the winds1 and agglutinated together1 were proved correct1 this difference would be immaterial2 but1 from his own account 7p4 #aah71 there occur in one place1 five or six layers of red earth1 interstratified with the ordinary calcareous rock1 and including stones too heavy for the wind to have moved1 without having at the same time utterly dispersed every grain of the accompanying drifted matter4 ,mr4 ,nelson attributes the origin of these several layers1 with their embedded stones1 to as many violent catastrophes2 but further investigation in such cases has generally succeeded in explaining phenomena of this kind by ordinary and simpler means4 ,finally1 ,i may remark1 that these islands have a considerable resemblance in shape to ,barbuda in the ,west ,indies1 and to ,pemba on the eastern coast of ,africa1 which latter island is about two hundred feet in height1 and consists of coral-rock4 ,i believe that the ,bermuda ,islands1 from being fringed by living reefs1 ought to have been coloured red2 but ,i have left them uncoloured1 on account of their general resemblance in external form to a lagoon-island or atoll4 ,,geological ,,observations ,,on ,,volcanic ,,islands4 ,,critical ,,introduction4 ,the preparation of the series of works published under the general title 8,geology of the ,voyage of the ,beagle0 occupied a great part of ,darwin's time during the ten years that followed his return to ,england4 ,the second volume of the series1 entitled 8,geological ,observations on ,volcanic ,islands1 with ,brief ,notices on the ,geology of ,australia and the ,cape of ,good ,hope10 made its appearance in #ahdd4 ,the materials for this volume were collected in part during the outward voyage1 when the ,beagle called at ,st4 ,jago in the ,cape de ,verde ,islands1 and ,st4 ,paul's ,rocks1 and at ,fernando ,noronha1 but mainly during the homeward cruise2 then it was that the ,galapagos ,islands were surveyed1 the ,low ,archipelago passed through1 and ,tahiti visited2 after making calls at the ,bay of ,islands1 in ,new ,zealand1 and also at ,sydney1 ,hobart ,town and ,king ,george's ,sound in ,australia1 the ,beagle sailed across the ,indian ,ocean to the little group of the ,keeling or ,cocos ,islands1 which ,darwin has rendered famous by his observations1 and thence to ,mauritius2 calling at the ,cape of ,good ,hope on her way1 the ship then proceeded successively to ,st4 ,helena and ,ascension1 and revisited the ,cape de ,verde ,islands before finally reaching ,england4 ,although ,darwin was thus able to gratify his curiosity by visits to a great number of very interesting volcanic districts1 the voyage opened for him with a bitter disappointment4 ,he had been reading ,humboldt's 8,personal ,narrative0 during his last year's residence in ,cambridge1 and had copied out from it long passages about ,teneriffe4 ,he was actually making inquiries as to the best means of visiting that island1 when the offer was made to him to accompany ,captain ,fitzroy in the ,beagle4 ,his friend ,henslow too1 on parting with him1 had given him the advice to procure and read the recently published first volume of the 8,principles of ,geology10 though he warned him against accepting the views advocated by its author4 ,during the time the ,beagle was beating backwards and forwards when the voyage commenced1 ,darwin1 although hardly ever able to leave his berth1 was employing all the opportunities which the terrible sea-sickness left him1 in studying ,humboldt and ,lyell4 ,we may therefore form an idea of his feelings when1 on the ship reaching ,santa ,cruz1 and the ,peak of ,teneriffe making its appearance among the clouds1 they were suddenly informed that an outbreak of cholera would prevent any landing6 ,ample compensation for this disappointment was found1 however1 when the ship reached ,porta ,praya in ,st4 ,jago1 the largest of the ,cape de ,verde ,islands4 ,here he spent three most delightful weeks1 and really commenced his work as a geologist and naturalist4 ,writing to his father he says1 8,geologising in a volcanic country is most delightful2 besides the interest attached to itself1 it leads you into most beautiful and retired spots4 ,nobody but a person fond of ,natural ,history can imagine the pleasure of strolling under cocoa-nuts in a thicket of bananas and coffee-plants1 and an endless number of wild flowers4 ,and this island1 that has given me so much instruction and delight1 is reckoned the most uninteresting place that we perhaps shall touch at during our voyage4 ,it certainly is generally very barren1 but the valleys are more exquisitely beautiful1 from the very contrast4 ,it is utterly useless to say anything about the scenery2 it would be as profitable to explain to a blind man colours1 as to a person who has not been out of ,europe1 the total dissimilarity of a tropical view4 ,whenever ,i enjoy anything1 ,i always look forward to writing it down1 either in my log-book 7which increases in bulk71 or in a letter2 so you must excuse raptures1 and those raptures badly expressed4 ,i find my collections are increasing wonderfully1 and from ,rio ,i think ,i shall be obliged to send a cargo home40 ,the indelible impression made on ,darwin's mind by this first visit to a volcanic island1 is borne witness to by a remarkable passage in the 8,autobiography0 written by him in #ahgf4 8,the geology of ,st4 ,jago is very striking1 yet simple2 a stream of lava formerly flowed over the bed of the sea1 formed of triturated recent shells and corals1 which it has baked into a hard white rock4 ,since then the whole island has been upheaved4 ,but the line of white rock revealed to me a new and important fact1 namely that there had been afterwards subsidence round the craters which had since been in action1 and had poured forth lava4 ,it then first dawned on me that ,i might perhaps write a book on the geology of the various countries visited1 and this made me thrill with delight4 ,that was a memorable hour to me1 and how distinctly ,i can call to mind the low cliff of lava beneath which ,i rested1 with the sun glaring hot1 a few strange desert plants growing near and with living corals in the tidal pools at my feet40 ,only five years before1 when listening to poor ,professor ,jameson's lectures on the effete ,wernerianism1 which at that time did duty for geological teaching1 ,darwin had found them 8incredibly dull10 and he declared that 8the sole effect they produced on me was a determination never so long as ,i lived to read a book on ,geology1 or in any way to study the science40 ,what a contrast we find in the expressions which he makes use of in referring to ,geological ,science1 in his letters written home from the ,beagle6 ,after alluding to the delight of collecting and studying marine animals1 he exclaims1 8,but ,geology carries the day60 ,writing to ,henslow he says1 8,i am quite charmed with ,geology1 but1 like the wise animal between two bundles of hay1 ,i do not know which to like best2 the old crystalline group of rocks1 or the softer and more fossiliferous beds40 ,and just as the long voyage is about to come to a close he again writes1 8,i find in ,geology a never-failing interest2 as it has been remarked1 it creates the same grand ideas respecting this world which ,astronomy does for the ,universe40 ,in this passage ,darwin doubtless refers to a remark of ,sir ,john ,herschel's in his admirable 8,preliminary ,discourse on the ,study of ,natural ,philosophy10--a book which exercised a most remarkable and beneficial influence on the mind of the young naturalist4 ,if there cannot be any doubt as to the strong predilection in ,darwin's mind for geological studies1 both during and after the memorable voyage1 there is equally little difficulty in perceiving the school of geological thought which1 in spite of the warnings of ,sedgwick and ,henslow1 had obtained complete ascendancy over his mind4 ,he writes in #ahgf3 8,the very first place which ,i examined1 namely ,st4 ,jago in the ,cape de ,verde ,islands1 showed me clearly the wonderful superiority of ,lyell's manner of treating ,geology1 compared with that of any other author1 whose works ,i had with me1 or ever afterwards read40 ,and again1 8,the science of ,geology is enormously indebted to ,lyell--more so1 as ,i believe1 than to any other man who ever lived 4 4 4 ,i am proud to remember that the first place1 namely1 ,st4 ,jago1 in the ,cape de ,verde ,archipelago1 in which ,i geologised1 convinced me of the infinite superiority of ,lyell's views over those advocated in any other work known to me40 ,the passages ,i have cited will serve to show the spirit in which ,darwin entered upon his geological studies1 and the perusal of the following pages will furnish abundant proofs of the enthusiasm1 acumen1 and caution with which his researches were pursued4 ,large collections of rocks and minerals were made by ,darwin during his researches1 and sent home to ,cambridge1 to be kept under the care of his faithful friend ,henslow4 ,after visiting his relations and friends1 ,darwin's first care on his return to ,england was to unpack and examine these collections4 ,he accordingly1 at the end of #ahcf1 took lodgings for three months in ,fitzwilliam ,street1 ,cambridge1 so as to be near ,henslow2 and in studying and determining his geological specimens received much valuable aid from the eminent crystallographer and mineralogist1 ,professor ,william ,hallows ,miller4 ,the actual writing of the volume upon volcanic islands was not commenced till #ahdc1 when ,darwin had settled in the spot which became his home for the rest of his life--the famous house at ,down1 in ,kent4 ,writing to his friend ,mr4 ,fox1 on ,march #bhth1 #ahdc1 he says1 8,i am very slowly progressing with a volume1 or rather pamphlet1 on the volcanic islands which we visited3 ,i manage only a couple of hours per day1 and that not very regularly4 ,it is uphill work writing books1 which cost money in publishing1 and which are not read even by geologists40 ,the work occupied ,darwin during the whole of the year #ahdc1 and was issued in the spring of the following year1 the actual time engaged in preparing it being recorded in his diary as 8from the summer of #ahdb to ,january #ahdd20 but the author does not appear to have been by any means satisfied with the result when the book was finished4 ,he wrote to ,lyell1 8,you have pleased me much by saying that you intend looking through my #8,volcanic ,islands2' it cost me eighteen months666 and ,i have heard of very few who have read it4 ,now ,i shall feel1 whatever little 7and little it is7 there is confirmatory of old work1 or new1 will work its effect and not be lost40 ,to ,sir ,joseph ,hooker he wrote1 8,i have just finished a little volume on the volcanic islands which we visited4 ,i do not know how far you care for dry simple geology1 but ,i hope you will let me send you a copy40 ,every geologist knows how full of interest and suggestiveness is this book of ,darwin's on volcanic islands4 ,probably the scant satisfaction which its author seemed to find in it may be traced to the effect of a contrast which he felt between the memory of glowing delights he had experienced when1 hammer in hand1 he roamed over new and interesting scenes1 and the slow1 laborious1 and less congenial task of re-writing and arranging his notes in book-form4 ,in #ahgd1 in writing an account of the ancient volcanoes of the ,hebrides1 ,i had frequent occasion to quote ,mr4 ,darwin's observations on the ,atlantic volcanoes1 in illustration of the phenomena exhibited by the relics of still older volcanoes in our own islands4 ,darwin1 in writing to his old friend ,sir ,charles ,lyell upon the subject1 says1 8,i was not a little pleased to see my volcanic book quoted1 for ,i thought it was completely dead and forgotten40 ,two years later the original publishers of this book and of that on ,south ,america proposed to re-issue them4 ,darwin at first hesitated1 for he seemed to think there could be little of abiding interest in them2 he consulted me upon the subject in one of the conversations which ,i used to have with him at that time1 and ,i strongly urged upon him the reprint of the works4 ,i was much gratified when he gave way upon the point1 and consented to their appearing just as originally issued4 ,in his preface he says1 8,owing to the great progress which ,geology has made in recent times1 my views on some few points may be somewhat antiquated1 but ,i have thought it best to leave them as they originally appeared40 ,it may be interesting to indicate1 as briefly as possible1 the chief geological problem upon which the publication of ,darwin's 8,volcanic ,islands0 threw new and important light4 ,the merit of the work consisted in supplying interesting observations1 which in some cases have proved of crucial value in exploding prevalent fallacies2 in calling attention to phenomena and considerations that had been quite overlooked by geologists1 but have since exercised an important influence in moulding geological speculation2 and lastly in showing the importance which attaches to small and seemingly insignificant causes1 some of which afford a key to the explanation of very curious geological problems4 ,visiting as he did the districts in which ,von ,buch and others had found what they thought to be evidence of the truth of 8,elevation-craters10 ,darwin was able to show that the facts were capable of a totally different interpretation4 ,the views originally put forward by the old ,german geologist and traveller1 and almost universally accepted by his countrymen1 had met with much support from ,elie de ,beaumont and ,dufrenoy1 the leaders of geological thought in ,france4 ,they were1 however1 stoutly opposed by ,scrope and ,lyell in this country1 and by ,constant ,prevost and ,virlet on the other side of the channel4 ,darwin1 in the work before us1 shows how little ground there is for the assumption that the great ring-craters of the ,atlantic islands have originated in gigantic blisters of the earth's surface which1 opening at the top1 have given origin to the craters4 ,admitting the influence of the injection of lava into the structure of the volcanic cones1 in increasing their bulk and elevation1 he shows that1 in the main1 the volcanoes are built up by repeated ejections causing an accumulation of materials around the vent4 ,while1 however1 agreeing on the whole with ,scrope and ,lyell1 as to the explosive origin of ordinary volcanic craters1 ,darwin clearly saw that1 in some cases1 great craters might be formed or enlarged1 by the subsidence of the floors after eruptions4 ,the importance of this agency1 to which too little attention has been directed by geologists1 has recently been shown by ,professor ,dana1 in his admirable work on ,kilauea and the other great volcanoes of the ,hawaiian ,archipelago4 ,the effects of subsidence at a volcanic centre in producing a downward dip of the strata around it1 was first pointed out by ,darwin1 as the result of his earliest work in the ,cape de ,verde ,islands4 ,striking illustrations of the same principle have since been pointed out by ,m4 ,robert and others in ,iceland1 by ,mr4 ,heaphy in ,new ,zealand1 and by myself in the ,western ,isles of ,scotland4 ,darwin again and again called attention to the evidence that volcanic vents exhibit relations to one another which can only be explained by assuming the existence of lines of fissure in the earth's crust1 along which the lavas have made their way to the surface4 ,but he1 at the same time1 clearly saw that there was no evidence of the occurrence of great deluges of lava along such fissures2 he showed how the most remarkable plateaux1 composed of successive lava sheets1 might be built up by repeated and moderate ejections from numerous isolated vents2 and he expressly insists upon the rapidity with which the cinder-cones around the orifices of ejection and the evidences of successive outflows of lava would be obliterated by denudation4 ,one of the most striking parts of the book is that in which he deals with the effects of denudation in producing 8basal wrecks0 or worn down stumps of volcanoes4 ,he was enabled to examine a series of cases in which could be traced every gradation1 from perfect volcanic cones down to the solidified plugs which had consolidated in the vents from which ejections had taken place4 ,darwin's observations on these points have been of the greatest value and assistance to all who have essayed to study the effects of volcanic action during earlier periods of the earth's history4 ,like ,lyell1 he was firmly persuaded of the continuity of geological history1 and ever delighted in finding indications1 in the present order of nature1 that the phenomena of the past could be accounted for by means of causes which are still in operation4 ,lyell's last work in the field was carried on about his home in ,forfarshire1 and only a few months before his death he wrote to ,darwin3 8,all the work which ,i have done has confirmed me in the belief that the only difference between ,palozoic and recent volcanic rocks is no more than we must allow for1 by the enormous time to which the products of the oldest volcanoes have been subjected to chemical changes40 ,darwin was greatly impressed1 as the result of his studies of volcanic phenomena1 followed by an examination of the great granite-masses of the ,andes1 with the relations between the so-called ,plutonic rocks and those of undoubtedly volcanic origin4 ,it was indeed a fortunate circumstance1 that after studying some excellent examples of recent volcanic rocks1 he proceeded to examine in ,south ,america many fine illustrations of the older igneous rock-masses1 and especially of the most highly crystalline types of the same1 and then on his way home had opportunities of reviving the impression made upon him by the fresh and unaltered volcanic rocks4 ,some of the general considerations suggested by these observations were discussed in a paper read by him before the ,geological ,society1 on ,march #gth1 #ahch1 under the title 8,on the ,connection of ,certain ,volcanic ,phenomena1 and ,on the ,formation of ,mountain-chains1 and the ,effect of ,continental ,elevations40 ,the exact bearing of these two classes of facts upon one another are more fully discussed in his book on ,south ,american geology4 ,the proofs of recent elevation around many of the volcanic islands led ,darwin to conclude that volcanic areas were1 as a rule1 regions in which upward movements were taking place1 and he was naturally led to contrast them with the areas in which1 as he showed1 the occurrence of atolls1 encircling reefs1 and barrier-reefs afford indication of subsidence4 ,in this way he was able to map out the oceanic areas in different zones1 along which opposite kinds of movement were taking place4 ,his conclusions on this subject were full of novelty and suggestiveness4 ,very clearly did ,darwin recognise the importance of the fact that most of the oceanic islands appear to be of volcanic origin1 though he was careful to point out the remarkable exceptions which somewhat invalidate the generalisation4 ,in his 8,origin of ,species0 he has elaborated the idea and suggested the theory of the permanence of ocean-basins1 a suggestion which has been adopted and pushed farther by subsequent authors1 than we think its originator would have approved4 ,his caution and fairness of mind on this and similar speculative questions was well-known to all who were in the habit of discussing them with him4 ,some years before the voyage of the ,beagle1 ,mr4 ,poulett ,scrope had pointed out the remarkable analogies that exist between certain igneous rocks of banded structure1 as seen in the ,ponza ,islands1 and the foliated crystalline schists4 ,it does not appear that ,darwin was acquainted with this remarkable memoir1 but quite independently he called attention to the same phenomena when he came to study some very similar rocks which occur in the island of ,ascension4 ,coming fresh from the study of the great masses of crystalline schist in the ,south ,american continent1 he was struck by the circumstance that in the undoubtedly igneous rocks of ,ascension we find a similar separation of the constituent minerals along parallel 8folia40 ,these observations led ,darwin to the same conclusion as that arrived at some time before by ,scrope--namely that when crystallisation takes place in rock masses under the influence of great deforming stresses1 a separation and parallel arrangement of the constituent minerals will result4 ,this is a process which is now fully recognised as having been a potent factor in the production of the metamorphic rock1 and has been called by more recent writers 8dynamo-metamorphism40 ,in this1 and in many similar discussions1 in which exact mineralogical knowledge was required1 it is remarkable how successful ,darwin was in making out the true facts with regard to the rocks he studied by the simple aid of a penknife and pocket-lens1 supplemented by a few chemical tests and the constant use of the blowpipe4 ,since his day1 the method of study of rocks by thin sections under the microscope has been devised1 and has become a most efficient aid in all petrographical inquiries4 ,during the voyage of ,h4,m4,s4 ,challenger1 many of the islands studied by ,darwin have been revisited and their rocks collected4 ,the results of their study by one of the greatest masters of the science of micropetrography--,professor ,renard of ,brussels--have been recently published in one of the volumes of 8,reports on the ,challenger ,expedition40 ,while much that is new and valuable has been contributed to geological science by these more recent investigations1 and many changes have been made in nomenclature and other points of detail1 it is interesting to find that all the chief facts described by ,darwin and his friend ,professor ,miller have stood the test of time and further study1 and remain as a monument of the acumen and accuracy in minute observation of these pioneers in geological research4 ,,john ,w4 ,,judd4 ,chapter ,i ,,st4 ,,jago1 ,,in ,,the ,,cape ,,de ,,verde ,,archipelago4 ,rocks of the lowest series4--,a calcareous sedimentary deposit1 with recent shells1 altered by the contact of superincumbent lava1 its horizontality and extent4--,subsequent volcanic eruptions1 associated with calcareous matter in an earthy and fibrous form1 and often enclosed within the separate cells of the scori4--,ancient and obliterated orifices of eruption of small size4 ,difficulty of tracing over a bare plain recent streams of lava4--,inland hills of more ancient volcanic rock4--,decomposed olivine in large masses4 ,feldspathic rocks beneath the upper crystalline basaltic strata4 ,uniform structure and form of the more ancient volcanic hills4--,form of the valleys near the coast4 ,conglomerate now forming on the sea beach4 ,the island of ,st4 ,jago extends in a ,n4,n4,w4 and ,s4,s4,e4 direction1 thirty miles in length by about twelve in breadth4 ,my observations1 made during two visits1 were confined to the southern portion within the distance of a few leagues from ,porto ,praya4 ,the country1 viewed from the sea1 presents a varied outline3 smooth conical hills of a reddish colour 7like ,red ,hill in ,fig4 #aa7 and others less regular1 flat-topped1 and of a blackish colour 7like ,a1 ,b1 ,c17 rise from successive1 step-formed plains of lava4 ,at a distance1 a chain of mountains1 many thousand feet in height1 traverses the interior of the island4 ,there is no active volcano in ,st4 ,jago1 and only one in the group1 namely at ,fogo4 ,the island since being inhabited has not suffered from destructive earthquakes4 #a ,the outline of the coast1 the position of the villages1 streamlets1 and of most of the hills in this woodcut1 are copied from the chart made on board ,h4,m4,s4 ,leven4 ,the square-topped hills 7,a1 ,b1 ,c1 etc47 are put in merely by eye1 to illustrate my description4 ,the lowest rocks exposed on the coast near ,porto ,praya1 are highly crystalline and compact2 they appear to be of ancient1 submarine1 volcanic origin2 they are unconformably covered by a thin1 irregular1 calcareous deposit1 abounding with shells of a late tertiary period2 and this again is capped by a wide sheet of basaltic lava1 which has flowed in successive streams from the interior of the island1 between the square-topped hills marked ,a1 ,b1 ,c1 etc4 ,still more recent streams of lava have been erupted from the scattered cones1 such as ,red and ,signal ,post ,hills4 ,the upper strata of the square-topped hills are intimately related in mineralogical composition1 and in other respects1 with the lowest series of the coast-rocks1 with which they seem to be continuous4 ,illustration3 ,part of ,st4 ,jago1 one of the ,cape de ,verde islands4 ,part of ,st4 ,jago1 one of the ,cape de ,verde islands4 ,mineralogical description of the rocks of the lowest series4--,these rocks possess an extremely varying character2 they consist of black1 brown1 and grey1 compact1 basaltic bases1 with numerous crystals of augite1 hornblende1 olivine1 mica1 and sometimes glassy feldspar4 ,a common variety is almost entirely composed of crystals of augite with olivine4 ,mica1 it is known1 seldom occurs where augite abounds2 nor probably does the present case offer a real exception1 for the mica 7at least in my best characterised specimen1 in which one nodule of this mineral is nearly half an inch in length7 is as perfectly rounded as a pebble in a conglomerate1 and evidently has not been crystallised in the base1 in which it is now enclosed1 but has proceeded from the fusion of some pre-existing rock4 ,these compact lavas alternate with tuffs1 amygdaloids1 and wacke1 and in some places with coarse conglomerate4 ,some of the argillaceous wackes are of a dark green colour1 others1 pale yellowish-green1 and others nearly white2 ,i was surprised to find that some of the latter varieties1 even where whitest1 fused into a jet black enamel1 whilst some of the green varieties afforded only a pale gray bead4 ,numerous dikes1 consisting chiefly of highly compact augitic rocks1 and of gray amygdaloidal varieties1 intersect the strata1 which have in several places been dislocated with considerable violence1 and thrown into highly inclined positions4 ,one line of disturbance crosses the northern end of ,quail ,island 7an islet in the ,bay of ,porto ,praya71 and can be followed to the mainland4 ,these disturbances took place before the deposition of the recent sedimentary bed2 and the surface1 also1 had previously been denuded to a great extent1 as is shown by many truncated dikes4 ,description of the calcareous deposit overlying the foregoing volcanic rocks4--,this stratum is very conspicuous from its white colour1 and from the extreme regularity with which it ranges in a horizontal line for some miles along the coast4 ,its average height above the sea1 measured from the upper line of junction with the superincumbent basaltic lava1 is about sixty feet2 and its thickness1 although varying much from the inequalities of the underlying formation1 may be estimated at about twenty feet4 ,it consists of quite white calcareous matter1 partly composed of organic dbris1 and partly of a substance which may be aptly compared in appearance with mortar4 ,fragments of rock and pebbles are scattered throughout this bed1 often forming1 especially in the lower part1 a conglomerate4 ,many of the fragments of rock are whitewashed with a thin coating of calcareous matter4 ,at ,quail ,island1 the calcareous deposit is replaced in its lowest part by a soft1 brown1 earthy tuff1 full of ,turritell2 this is covered by a bed of pebbles1 passing into sandstone1 and mixed with fragments of echini1 claws of crabs1 and shells2 the oyster-shells still adhering to the rock on which they grew4 ,numerous white balls appearing like pisolitic concretions1 from the size of a walnut to that of an apple1 are embedded in this deposit2 they usually have a small pebble in their centres4 ,although so like concretions1 a close examination convinced me that they were ,nullipor1 retaining their proper forms1 but with their surfaces slightly abraded3 these bodies 7plants as they are now generally considered to be7 exhibit under a microscope of ordinary power1 no traces of organisation in their internal structure4 ,mr4 ,george ,r4 ,sowerby has been so good as to examine the shells which ,i collected3 there are fourteen species in a sufficiently perfect condition for their characters to be made out with some degree of certainty1 and four which can be referred only to their genera4 ,of the fourteen shells1 of which a list is given in the ,appendix1 eleven are recent species2 one1 though undescribed1 is perhaps identical with a species which ,i found living in the harbour of ,porto ,praya2 the two remaining species are unknown1 and have been described by ,mr4 ,sowerby4 ,until the shells of this ,archipelago and of the neighbouring coasts are better known1 it would be rash to assert that even these two latter shells are extinct4 ,the number of species which certainly belong to existing kinds1 although few in number1 are sufficient to show that the deposit belongs to a late tertiary period4 ,from its mineralogical character1 from the number and size of the embedded fragments1 and from the abundance of ,patell1 and other littoral shells1 it is evident that the whole was accumulated in a shallow sea1 near an ancient coast-line4 ,effects produced by the flowing of the superincumbent basaltic lava over the calcareous deposit4--,these effects are very curious4 ,the calcareous matter is altered to the depth of about a foot beneath the line of junction2 and a most perfect gradation can be traced1 from loosely aggregated1 small1 particles of shells1 corallines1 and ,nullipor1 into a rock1 in which not a trace of mechanical origin can be discovered1 even with a microscope4 ,where the metamorphic change has been greatest1 two varieties occur4 ,the first is a hard1 compact1 white1 fine-grained rock1 striped with a few parallel lines of black volcanic particles1 and resembling a sandstone1 but which1 upon close examination1 is seen to be crystallised throughout1 with the cleavages so perfect that they can be readily measured by the reflecting goniometer4 ,in specimens1 where the change has been less complete1 when moistened and examined under a strong lens1 the most interesting gradation can be traced1 some of the rounded particles retaining their proper forms1 and others insensibly melting into the granulo-crystalline paste4 ,the weathered surface of this stone1 as is so frequently the case with ordinary limestones1 assumes a brick-red colour4 ,the second metamorphosed variety is likewise a hard rock1 but without any crystalline structure4 ,it consists of a white1 opaque1 compact1 calcareous stone1 thickly mottled with rounded1 though regular1 spots of a soft1 earthy1 ochraceous substance4 ,this earthy matter is of a pale yellowish-brown colour1 and appears to be a mixture of carbonate of lime with iron2 it effervesces with acids1 is infusible1 but blackens under the blowpipe1 and becomes magnetic4 ,the rounded form of the minute patches of earthy substance1 and the steps in the progress of their perfect formation1 which can be followed in a suit of specimens1 clearly show that they are due either to some power of aggregation in the earthy particles amongst themselves1 or more probably to a strong attraction between the atoms of the carbonate of line1 and consequently to the segregation of the earthy extraneous matter4 ,i was much interested by this fact1 because ,i have often seen quartz rocks 7for instance1 in the ,falkland ,islands1 and in the lower ,silurian strata of the ,stiper-stones in ,shropshire71 mottled in a precisely analogous manner1 with little spots of a white1 earthy substance 7earthy feldspar872 and these rocks1 there was good reason to suppose1 had undergone the action of heat1--a view which thus receives confirmation4 ,this spotted structure may possibly afford some indication in distinguishing those formations of quartz1 which owe their present structure to igneous action1 from those produced by the agency of water alone2 a source of doubt1 which ,i should think from my own experience1 that most geologists1 when examining arenaceo-quartzose districts must have experienced4 ,the lowest and most scoriaceous part of the lava1 in rolling over the sedimentary deposit at the bottom of the sea1 has caught up large quantities of calcareous matter1 which now forms a snow-white1 highly crystalline basis to a breccia1 including small pieces of black1 glossy scori4 ,a little above this1 where the lime is less abundant1 and the lava more compact1 numerous little balls1 composed of spicula of calcareous spar1 radiating from common centres1 occupy the interstices4 ,in one part of ,quail ,island1 the lime has thus been crystallised by the heat of the superincumbent lava1 where it is only thirteen feet in thickness2 nor had the lava been originally thicker1 and since reduced by degradation1 as could be told from the degree of cellularity of its surface4 ,i have already observed that the sea must have been shallow in which the calcareous deposit was accumulated4 ,in this case1 therefore1 the carbonic acid gas has been retained under a pressure1 insignificant compared with that 7a column of water1 #a1gjh feet in height7 originally supposed by ,sir ,james ,hall to be requisite for this end3 but since his experiments1 it has been discovered that pressure has less to do with the retention of carbonic acid gas1 than the nature of the circumjacent atmosphere2 and hence1 as is stated to be the case by ,mr4 ,faraday1#b masses of limestone are sometimes fused and crystallised even in common limekilns4 ,carbonate of lime can be heated to almost any degree1 according to ,faraday1 in an atmosphere of carbonic acid gas1 without being decomposed2 and ,gay-,lussac found that fragments of limestone1 placed in a tube and heated to a degree1 not sufficient by itself to cause their decomposition1 yet immediately evolved their carbonic acid1 when a stream of common air or steam was passed over them3 ,gay-,lussac attributes this to the mechanical displacement of the nascent carbonic acid gas4 ,the calcareous matter beneath the lava1 and especially that forming the crystalline spicula between the interstices of the scori1 although heated in an atmosphere probably composed chiefly of steam1 could not have been subjected to the effects of a passing stream2 and hence it is1 perhaps1 that they have retained their carbonic acid1 under a small amount of pressure4 #b ,i am much indebted to ,mr4 ,e4 ,w4 ,brayley in having given me the following references to papers on this subject3 ,faraday in the ,edinburgh ,new ,philosophical ,journal1 vol4 xv1 p4 #cih2 ,gay-,lussac in ,annales de ,chem4 et ,phys41 tome lxiii1 p4 #bai1 translated in the ,london and ,edinburgh ,philosophical ,magazine1 vol4 x1 p4 #dif4 ,the fragments of scori1 embedded in the crystalline calcareous basis1 are of a jet black colour1 with a glossy fracture like pitchstone4 ,their surfaces1 however1 are coated with a layer of a reddish-orange1 translucent substance1 which can easily be scratched with a knife2 hence they appear as if overlaid by a thin layer of rosin4 ,some of the smaller fragments are partially changed throughout into this substance3 a change which appears quite different from ordinary decomposition4 ,at the ,galapagos ,archipelago 7as will be described in a future chapter71 great beds are formed of volcanic ashes and particles of scori1 which have undergone a closely similar change4 ,the extent and horizontality of the calcareous stratum4--,the upper line of surface of the calcareous stratum1 which is so conspicuous from being quite white and so nearly horizontal1 ranges for miles along the coast1 at the height of about sixty feet above the sea4 ,the sheet of basalt1 by which it is capped1 is on an average eighty feet in thickness4 ,westward of ,porto ,praya beyond ,red ,hill1 the white stratum with the superincumbent basalt is covered up by more recent streams4 ,northward of ,signal ,post ,hill1 ,i could follow it with my eye1 trending away for several miles along the sea cliffs4 ,the distance thus observed is about seven miles2 but ,i cannot doubt from its regularity that it extends much farther4 ,in some ravines at right angles to the coast1 it is seen gently dipping towards the sea1 probably with the same inclination as when deposited round the ancient shores of the island4 ,i found only one inland section1 namely1 at the base of the hill marked ,a1 where1 at the height of some hundred feet1 this bed was exposed2 it here rested on the usual compact augitic rock associated with wacke1 and was covered by the widespread sheet of modern basaltic lava4 ,some exceptions occur to the horizontality of the white stratum3 at ,quail ,island1 its upper surface is only forty feet above the level of the sea2 here also the capping of lava is only between twelve and fifteen feet in thickness2 on the other hand1 at the north-east side of ,porto ,praya harbour1 the calcareous stratum1 as well as the rock on which it rests1 attain a height above the average level3 the inequality of level in these two cases is not1 as ,i believe1 owing to unequal elevation1 but to original irregularities at the bottom of the sea4 ,of this fact1 at ,quail ,island1 there was clear evidence in the calcareous deposit being in one part of much greater than the average thickness1 and in another part being entirely absent2 in this latter case1 the modern basaltic lavas rested directly on those of more ancient origin4 ,fig4 #b ,illustration3 ,signal ,post ,hill ,,signal ,,post ,,hill ,a--,ancient volcanic rocks4 ,b--,calcareous stratum4 ,c--,upper balastic lava4 ,under ,signal ,post ,hill1 the white stratum dips into the sea in a remarkable manner4 ,this hill is conical1 #dej feet in height1 and retains some traces of having had a crateriform structure2 it is composed chiefly of matter erupted posteriorly to the elevation of the great basaltic plain1 but partly of lava of apparently submarine origin and of considerable antiquity4 ,the surrounding plain1 as well as the eastern flank of this hill1 has been worn into steep precipices1 overhanging the sea4 ,in these precipices1 the white calcareous stratum may be seen1 at the height of about seventy feet above the beach1 running for some miles both northward and southward of the hill1 in a line appearing to be perfectly horizontal2 but for a space of a quarter of a mile directly under the hill1 it dips into the sea and disappears4 ,on the south side the dip is gradual1 on the north side it is more abrupt1 as is shown in ,fig4 #b4 ,as neither the calcareous stratum1 nor the superincumbent basaltic lava 7as far as the latter can be distinguished from the more modern ejections71 appears to thicken as it dips1 ,i infer that these strata were not originally accumulated in a trough1 the centre of which afterwards became a point of eruption2 but that they have subsequently been disturbed and bent4 ,we may suppose either that ,signal ,post ,hill subsided after its elevation with the surrounding country1 or that it never was uplifted to the same height with it4 ,this latter seems to me the most probable alternative1 for during the slow and equable elevation of this portion of the island1 the subterranean motive power1 from expending part of its force in repeatedly erupting volcanic matter from beneath this point1 would1 it is likely1 have less force to uplift it4 ,something of the same kind seems to have occurred near ,red ,hill1 for when tracing upwards the naked streams of lava from near ,porto ,praya towards the interior of the island1 ,i was strongly induced to suspect1 that since the lava had flowed1 the slope of the land had been slightly modified1 either by a small subsidence near ,red ,hill1 or by that portion of the plain having been uplifted to a less height during the elevation of the whole area4 ,the basaltic lava1 superincumbent on the calcareous deposit4--,this lava is of a pale grey colour1 fusing into a black enamel2 its fracture is rather earthy and concretionary2 it contains olivine in small grains4 ,the central parts of the mass are compact1 or at most crenulated with a few minute cavities1 and are often columnar4 ,at ,quail ,island this structure was assumed in a striking manner2 the lava in one part being divided into horizontal lamin1 which became in another part split by vertical fissures into five-sided plates2 and these again1 being piled on each other1 insensibly became soldered together1 forming fine symmetrical columns4 ,the lower surface of the lava is vesicular1 but sometimes only to the thickness of a few inches2 the upper surface1 which is likewise vesicular1 is divided into balls1 frequently as much as three feet in diameter1 made up of concentric layers4 ,the mass is composed of more than one stream2 its total thickness being1 on an average1 about eighty feet3 the lower portion has certainly flowed beneath the sea1 and probably likewise the upper portion4 ,the chief part of this lava has flowed from the central districts1 between the hills marked ,a1 ,b1 ,c1 etc41 in the woodcut-map4 ,the surface of the country1 near the coast1 is level and barren2 towards the interior1 the land rises by successive terraces1 of which four1 when viewed from a distance1 could be distinctly counted4 ,volcanic eruptions subsequent to the elevation of the coastland2 the ejected matter associated with earthy lime4--,these recent lavas have proceeded from those scattered1 conical1 reddish-coloured hills1 which rise abruptly from the plain-country near the coast4 ,i ascended some of them1 but will describe only one1 namely1 ,red ,hill1 which may serve as a type of its class1 and is remarkable in some especial respects4 ,its height is about six hundred feet2 it is composed of bright red1 highly scoriaceous rock of a basaltic nature2 on one side of its summit there is a hollow1 probably the last remnant of a crater4 ,several of the other hills of this class1 judging from their external forms1 are surmounted by much more perfect craters4 ,when sailing along the coast1 it was evident that a considerable body of lava had flowed from ,red ,hill1 over a line of cliff about one hundred and twenty feet in height1 into the sea3 this line of cliff is continuous with that forming the coast1 and bounding the plain on both sides of this hill2 these streams1 therefore1 were erupted1 after the formation of the coast-cliffs1 from ,red ,hill1 when it must have stood1 as it now does1 above the level of the sea4 ,this conclusion accords with the highly scoriaceous condition of all the rock on it1 appearing to be of subaerial formation3 and this is important1 as there are some beds of calcareous matter near its summit1 which might1 at a hasty glance1 have been mistaken for a submarine deposit4 ,these beds consist of white1 earthy1 carbonate of lime1 extremely friable so as to be crushed with the least pressure2 the most compact specimens not resisting the strength of the fingers4 ,some of the masses are as white as quicklime1 and appear absolutely pure2 but on examining them with a lens1 minute particles of scori can always be seen1 and ,i could find none which1 when dissolved in acids1 did not leave a residue of this nature4 ,it is1 moreover1 difficult to find a particle of the lime which does not change colour under the blowpipe1 most of them even becoming glazed4 ,the scoriaceous fragments and the calcareous matter are associated in the most irregular manner1 sometimes in obscure beds1 but more generally as a confused breccia1 the lime in some parts and the scori in others being most abundant4 ,sir ,h4 ,de la ,beche has been so kind as to have some of the purest specimens analysed1 with a view to discover1 considering their volcanic origin1 whether they contained much magnesia2 but only a small portion was found1 such as is present in most limestones4 ,fragments of the scori embedded in the calcareous mass1 when broken1 exhibit many of their cells lined and partly filled with a white1 delicate1 excessively fragile1 moss-like1 or rather conferva-like1 reticulation of carbonate of lime4 ,these fibres1 examined under a lens of one-tenth of an inch focal distance1 appear cylindrical2 they are rather above one-thousandth of an inch in diameter2 they are either simply branched1 or more commonly united into an irregular mass of network1 with the meshes of very unequal sizes and of unequal numbers of sides4 ,some of the fibres are thickly covered with extremely minute spicula1 occasionally aggregated into little tuffs2 and hence they have a hairy appearance4 ,these spicula are of the same diameter throughout their length2 they are easily detached1 so that the object-glass of the microscope soon becomes scattered over with them4 ,within the cells of many fragments of the scoria1 the lime exhibits this fibrous structure1 but generally in a less perfect degree4 ,these cells do not appear to be connected with one another4 ,there can be no doubt1 as will presently be shown1 that the lime was erupted1 mingled with the lava in its fluid state1 and therefore ,i have thought it worth while to describe minutely this curious fibrous structure1 of which ,i know nothing analogous4 ,from the earthy condition of the fibres1 this structure does not appear to be related to crystallisation4 ,other fragments of the scoriaceous rock from this hill1 when broken1 are often seen marked with short and irregular white streaks1 which are owing to a row of separate cells being partly1 or quite1 filled with white calcareous powder4 ,this structure immediately reminded me of the appearance in badly kneaded dough1 of balls and drawn-out streaks of flour1 which have remained unmixed with the paste2 and ,i cannot doubt that small masses of the lime1 in the same manner remaining unmixed with the fluid lava1 have been drawn out when the whole was in motion4 ,i carefully examined1 by trituration and solution in acids1 pieces of the scori1 taken from within half-an-inch of those cells which were filled with the calcareous powder1 and they did not contain an atom of free lime4 ,it is obvious that the lava and lime have on a large scale been very imperfectly mingled2 and where small portions of the lime have been entangled within a piece of the viscid lava1 the cause of their now occupying1 in the form of a powder or of a fibrous reticulation1 the vesicular cavities1 is1 ,i think1 evidently due to the confined gases having most readily expanded at the points where the incoherent lime rendered the lava less adhesive4 ,a mile eastward of the town of ,praya1 there is a steep-sided gorge1 about one hundred and fifty yards in width1 cutting through the basaltic plain and underlying beds1 but since filled up by a stream of more modern lava4 ,this lava is dark grey1 and in most parts compact and rudely columnar2 but at a little distance from the coast1 it includes in an irregular manner a brecciated mass of red scori mingled with a considerable quantity of white1 friable1 and in some parts1 nearly pure earthy lime1 like that on the summit of ,red ,hill4 ,this lava1 with its entangled lime1 has certainly flowed in the form of a regular stream2 and1 judging from the shape of the gorge1 towards which the drainage of the country 7feeble though it now be7 still is directed1 and from the appearance of the bed of loose water-worn blocks with their interstices unfilled1 like those in the bed of a torrent1 on which the lava rests1 we may conclude that the stream was of subaerial origin4 ,i was unable to trace it to its source1 but1 from its direction1 it seemed to have come from ,signal ,post ,hill1 distant one mile and a quarter1 which1 like ,red ,hill1 has been a point of eruption subsequent to the elevation of the great basaltic plain4 ,it accords with this view1 that ,i found on ,signal ,post ,hill1 a mass of earthy1 calcareous matter of the same nature1 mingled with scori4 ,i may here observe that part of the calcareous matter forming the horizontal sedimentary bed1 especially the finer matter with which the embedded fragments of rock are whitewashed1 has probably been derived from similar volcanic eruptions1 as well as from triturated organic remains3 the underlying1 ancient1 crystalline rocks1 also1 are associated with much carbonate of lime1 filling amygdaloidal cavities1 and forming irregular masses1 the nature of which latter ,i was unable to understand4 ,considering the abundance of earthy lime near the summit of ,red ,hill1 a volcanic cone six hundred feet in height1 of subaerial growth1--considering the intimate manner in which minute particles and large masses of scori are embedded in the masses of nearly pure lime1 and on the other hand1 the manner in which small kernels and streaks of the calcareous powder are included in solid pieces of the scori1--considering1 also1 the similar occurrence of lime and scori within a stream of lava1 also supposed1 with good reason1 to have been of modern subaerial origin1 and to have flowed from a hill1 where earthy lime also occurs3 ,i think1 considering these facts1 there can be no doubt that the lime has been erupted1 mingled with the molten lava4 ,i am not aware that any similar case has been described3 it appears to me an interesting one1 inasmuch as most geologists must have speculated on the probable effects of a volcanic focus1 bursting through deep-seated beds of different mineralogical composition4 ,the great abundance of free silex in the trachytes of some countries 7as described by ,beudant in ,hungary1 and by ,p4 ,scrope in the ,panza ,islands71 perhaps solves the inquiry with respect to deep-seated beds of quartz2 and we probably here see it answered1 where the volcanic action has invaded subjacent masses of limestone4 ,one is naturally led to conjecture in what state the now earthy carbonate of lime existed1 when ejected with the intensely heated lava3 from the extreme cellularity of the scori on ,red ,hill1 the pressure cannot have been great1 and as most volcanic eruptions are accompanied by the emission of large quantities of steam and other gases1 we here have the most favourable conditions1 according to the views at present entertained by chemists1 for the expulsion of the carbonic acid4#c ,has the slow re-absorption of this gas1 it may be asked1 given to the lime in the cells of the lava1 that peculiar fibrous structure1 like that of an efflorescing salt8 ,finally1 ,i may remark on the great contrast in appearance between this earthy lime1 which must have been heated in a free atmosphere of steam and other gases1 while the white1 crystalline1 calcareous spar1 produced by a single thin sheet of lava 7as at ,quail ,island7 rolling over similar earthy lime and the dbris of organic remains1 at the bottom of a shallow sea4 #c ,whilst deep beneath the surface1 the carbonate of lime was1 ,i presume1 in a fluid state4 ,hutton1 it is known1 thought that all amygdaloids were produced by drops of molten limestone floating in the trap1 like oil in water3 this no doubt is erroneous1 but if the matter forming the summit of ,red ,hill had been cooled under the pressure of a moderately deep sea1 or within the walls of a dike1 we should1 in all probability1 have had a trap rock associated with large masses of compact1 crystalline1 calcareous spar1 which1 according to the views entertained by many geologists1 would have been wrongly attributed to subsequent infiltration4 ,signal ,post ,hill4--,this hill has already been several times mentioned1 especially with reference to the remarkable manner in which the white calcareous stratum1 in other parts so horizontal 7,fig4 #b71 dips under it into the sea4 ,it has a broad summit1 with obscure traces of a crateriform structure1 and is composed of basaltic rocks1#d some compact1 others highly cellular with inclined beds of loose scori1 of which some are associated with earthy lime4 ,like ,red ,hill1 it has been the source of eruptions1 subsequently to the elevation of the surrounding basaltic plain2 but unlike that hill1 it has undergone considerable denudation1 and has been the seat of volcanic action at a remote period1 when beneath the sea4 ,i judge of this latter circumstance from finding on its inland flank the last remnants of three small points of eruption4 ,these points are composed of glossy scori1 cemented by crystalline calcareous spar1 exactly like the great submarine calcareous deposit1 where the heated lava has rolled over it3 their demolished state can1 ,i think1 be explained only by the denuding action of the waves of the sea4 ,i was guided to the first orifice by observing a sheet of lava1 about two hundred yards square1 with steepish sides1 superimposed on the basaltic plain with no adjoining hillock1 whence it could have been erupted2 and the only trace of a crater which ,i was able to discover1 consisted of some inclined beds of scori at one of its corners4 ,at the distance of fifty yards from a second level-topped patch of lava1 but of much smaller size1 ,i found an irregular circular group of masses of cemented1 scoriaceous breccia1 about six feet in height1 which doubtless had once formed the point of eruption4 ,the third orifice is now marked only by an irregular circle of cemented scori1 about four yards in diameter1 and rising in its highest point scarcely three feet above the level of the plain1 the surface of which1 close all round1 exhibits its usual appearance3 here we have a horizontal basal section of a volcanic spiracle1 which1 together with all its ejected matter1 has been almost totally obliterated4 #d ,of these1 one common variety is remarkable for being full of small fragments of a dark jasper-red earthy mineral1 which1 when examined carefully1 shows an indistinct cleavage2 the little fragments are elongated in form1 are soft1 are magnetic before and after being heated1 and fuse with difficulty into a dull enamel4 ,this mineral is evidently closely related to the oxides of iron1 but ,i cannot ascertain what it exactly is4 ,the rock containing this mineral is crenulated with small angular cavities1 which are lined and filled with yellowish crystals of carbonate of lime4 ,the stream of lava1 which fills the narrow gorge#e eastward of the town of ,praya1 judging from its course1 seems1 as before remarked1 to have come from ,signal ,post ,hill1 and to have flowed over the plain1 after its elevation3 the same observation applies to a stream 7possibly part of the same one7 capping the sea cliffs1 a little eastward of the gorge4 ,when ,i endeavoured to follow these streams over the stony level plain1 which is almost destitute of soil and vegetation1 ,i was much surprised to find1 that although composed of hard basaltic matter1 and not having been exposed to marine denudation1 all distant traces of them soon became utterly lost4 ,but ,i have since observed at the ,galapagos ,archipelago1 that it is often impossible to follow even great deluges of quite recent lava across older streams1 except by the size of the bushes growing on them1 or by the comparative states of glossiness of their surfaces1--characters which a short lapse of time would be sufficient quite to obscure4 ,i may remark1 that in a level country1 with a dry climate1 and with the wind blowing always in one direction 7as at the ,cape de ,verde ,archipelago71 the effects of atmospheric degradation are probably much greater than would at first be expected2 for soil in this case accumulates only in a few protected hollows1 and being blown in one direction1 it is always travelling towards the sea in the form of the finest dust1 leaving the surface of the rocks bare1 and exposed to the full effects of renewed meteoric action4 #e ,the sides of this gorge1 where the upper basaltic stratum is intersected1 are almost perpendicular4 ,the lava1 which has since filled it up1 is attached to these sides1 almost as firmly as a dike is to its walls4 ,in most cases1 where a stream of lava has flowed down a valley1 it is bounded on each side by loose scoriaceous masses4 ,inland hills of more ancient volcanic rocks4--,these hills are laid down by eye1 and marked as ,a1 ,b1 ,c1 etc41 in ,map #a4 ,they are related in mineralogical composition1 and are probably directly continuous with the lowest rocks exposed on the coast4 ,these hills1 viewed from a distance1 appear as if they had once formed part of an irregular tableland1 and from their corresponding structure and composition this probably has been the case4 ,they have flat1 slightly inclined summits1 and are1 on an average1 about six hundred feet in height2 they present their steepest slope towards the interior of the island1 from which point they radiate outwards1 and are separated from each other by broad and deep valleys1 through which the great streams of lava1 forming the coast-plains1 have descended4 ,their inner and steeper escarpments are ranged in an irregular curve1 which rudely follows the line of the shore1 two or three miles inland from it4 ,i ascended a few of these hills1 and from others1 which ,i was able to examine with a telescope1 ,i obtained specimens1 through the kindness of ,mr4 ,kent1 the assistant-surgeon of the ,beagle2 although by these means ,i am acquainted with only a part of the range1 five or six miles in length1 yet ,i scarcely hesitate1 from their uniform structure1 to affirm that they are parts of one great formation1 stretching round much of the circumference of the island4 ,the upper and lower strata of these hills differ greatly in composition4 ,the upper are basaltic1 generally compact1 but sometimes scoriaceous and amygdaloidal1 with associated masses of wacke3 where the basalt is compact1 it is either fine-grained or very coarsely crystallised2 in the latter case it passes into an augitic rock1 containing much olivine2 the olivine is either colourless1 or of the usual yellow and dull reddish shades4 ,on some of the hills1 beds of calcareous matter1 both in an earthy and in a crystalline form1 including fragments of glossy scori1 are associated with the basaltic strata4 ,these strata differ from the streams of basaltic lava forming the coast-plains1 only in being more compact1 and in the crystals of augite1 and in the grains of olivine being of much greater size2--characters which1 together with the appearance of the associated calcareous beds1 induce me to believe that they are of submarine formation4 ,some considerable masses of wacke1 which are associated with these basaltic strata1 and which likewise occur in the basal series on the coast1 especially at ,quail ,island1 are curious4 ,they consist of a pale yellowish-green argillaceous substance1 of a crumbling texture when dry1 but unctuous when moist3 in its purest form1 it is of a beautiful green tint1 with translucent edges1 and occasionally with obscure traces of an original cleavage4 ,under the blowpipe it fuses very readily into a dark grey1 and sometimes even black bead1 which is slightly magnetic4 ,from these characters1 ,i naturally thought that it was one of the pale species1 decomposed1 of the genus augite2--a conclusion supported by the unaltered rock being full of large separate crystals of black augite1 and of balls and irregular streaks of dark grey augitic rock4 ,as the basalt ordinarily consists of augite1 and of olivine often tarnished and of a dull red colour1 ,i was led to examine the stages of decomposition of this latter mineral1 and ,i found1 to my surprise1 that ,i could trace a nearly perfect gradation from unaltered olivine to the green wacke4 ,part of the same grain under the blowpipe would in some instances behave like olivine1 its colour being only slightly changed1 and part would give a black magnetic bead4 ,hence ,i can have no doubt that the greenish wacke originally existed as olivine2 but great chemical changes must have been effected during the act of decomposition thus to have altered a very hard1 transparent1 infusible mineral1 into a soft1 unctuous1 easily melted1 argillaceous substance4#f #f ,d',aubuisson 8,trait de ,gognosie0 7tome ii1 p4 #efi7 mentions1 on the authority of ,m4 ,marcel de ,serres1 masses of green earth near ,montpellier1 which are supposed to be due to the decomposition of olivine4 ,i do not1 however1 find1 that the action of this mineral under the blowpipe being entirely altered1 as it becomes decomposed1 has been noticed2 and the knowledge of this fact is important1 as at first it appears highly improbable that a hard1 transparent1 refractory mineral should be changed into a soft1 easily fused clay1 like this of ,st4 ,jago4 ,i shall hereafter describe a green substance1 forming threads within the cells of some vesicular basaltic rocks in ,van ,diemen's ,land1 which behave under the blowpipe like the green wacke of ,st4 ,jago2 but its occurrence in cylindrical threads1 shows it cannot have resulted from the decomposition of olivine1 a mineral always existing in the form of grains or crystals4 ,the basal strata of these hills1 as well as some neighbouring1 separate1 bare1 rounded hillocks1 consist of compact1 fine-grained1 non-crystalline 7or so slightly as scarcely to be perceptible71 ferruginous1 feldspathic rocks1 and generally in a state of semi-decomposition4 ,their fracture is exceedingly irregular1 and splintery2 yet small fragments are often very tough4 ,they contain much ferruginous matter1 either in the form of minute grains with a metallic lustre1 or of brown hair-like threads3 the rock in this latter case assuming a pseudo-brecciated structure4 ,these rocks sometimes contain mica and veins of agate4 ,their rusty brown or yellowish colour is partly due to the oxides of iron1 but chiefly to innumerable1 microscopically minute1 black specks1 which1 when a fragment is heated1 are easily fused1 and evidently are either hornblende or augite4 ,these rocks1 therefore1 although at first appearing like baked clay or some altered sedimentary deposit1 contain all the essential ingredients of trachyte2 from which they differ only in not being harsh1 and in not containing crystals of glassy feldspar4 ,as is so often the case with trachytic formation1 no stratification is here apparent4 ,a person would not readily believe that these rocks could have flowed as lava2 yet at ,st4 ,helena there are well-characterised streams 7as will be described in an ensuing chapter7 of nearly similar composition4 ,amidst the hillocks composed of these rocks1 ,i found in three places1 smooth conical hills of phonolite1 abounding with fine crystals of glassy feldspar1 and with needles of hornblende4 ,these cones of phonolite1 ,i believe1 bear the same relation to the surrounding feldspathic strata which some masses of coarsely crystallised augitic rock1 in another part of the island1 bear to the surrounding basalt1 namely1 that both have been injected4 ,the rocks of a feldspathic nature being anterior in origin to the basaltic strata1 which cap them1 as well as to the basaltic streams of the coast-plains1 accords with the usual order of succession of these two grand divisions of the volcanic series4 ,the strata of most of these hills in the upper part1 where alone the planes of division are distinguishable1 are inclined at a small angle from the interior of the island towards the sea-coast4 ,the inclination is not the same in each hill2 in that marked ,a it is less than in ,b1 ,d1 or ,e2 in ,c the strata are scarcely deflected from a horizontal plane1 and in ,f 7as far as ,i could judge without ascending it7 they are slightly inclined in a reverse direction1 that is1 inwards and towards the centre of the island4 ,notwithstanding these differences of inclination1 their correspondence in external form1 and in the composition both of their upper and lower parts1--their relative position in one curved line1 with their steepest sides turned inwards1--all seem to show that they originally formed parts of one platform2 which platform1 as before remarked1 probably extended round a considerable portion of the circumference of the island4 ,the upper strata certainly flowed as lava1 and probably beneath the sea1 as perhaps did the lower feldspathic masses3 how then come these strata to hold their present position1 and whence were they erupted8 ,in the centre of the island#g there are lofty mountains1 but they are separated from the steep inland flanks of these hills by a wide space of lower country3 the interior mountains1 moreover1 seem to have been the source of those great streams of basaltic lava which1 contracting as they pass between the bases of the hills in question1 expand into the coast-plains4 ,round the shores of ,st4 ,helena there is a rudely formed ring of basaltic rocks1 and at ,mauritius there are remnants of another such a ring round part1 if not round the whole1 of the island2 here again the same question immediately occurs1 how came these masses to hold their present position1 and whence were they erupted8 ,the same answer1 whatever it may be1 probably applies in these three cases2 and in a future chapter we shall recur to this subject4 #g ,i saw very little of the inland parts of the island4 ,near the village of ,st4 ,domingo1 there are magnificent cliffs of rather coarsely crystallised basaltic lava4 ,following the little stream in this valley1 about a mile above the village1 the base of the great cliff was formed of a compact fine-grained basalt1 conformably covered by a bed of pebbles4 ,near ,fuentes1 ,i met with pap-formed hills of the compact feldspathic series of rocks4 ,valleys near the coast4--,these are broad1 very flat1 and generally bounded by low cliff-formed sides4 ,portions of the basaltic plain are sometimes nearly or quite isolated by them2 of which fact1 the space on which the town of ,praya stands offers an instance4 ,the great valley west of the town has its bottom filled up to a depth of more than twenty feet by well-rounded pebbles1 which in some parts are firmly cemented together by white calcareous matter4 ,there can be no doubt1 from the form of these valleys1 that they were scooped out by the waves of the sea1 during that equable elevation of the land1 of which the horizontal calcareous deposit1 with its existing species of marine remains1 gives evidence4 ,considering how well shells have been preserved in this stratum1 it is singular that ,i could not find even a single small fragment of shell in the conglomerate at the bottom of the valleys4 ,the bed of pebbles in the valley west of the town is intersected by a second valley joining it as a tributary1 but even this valley appears much too wide and flat-bottomed to have been formed by the small quantity of water1 which falls only during one short wet season2 for at other times of the year these valleys are absolutely dry4 ,recent conglomerate4--,on the shores of ,quail ,island1 ,i found fragments of brick1 bolts of iron1 pebbles1 and large fragments of basalt1 united by a scanty base of impure calcareous matter into a firm conglomerate4 ,to show how exceedingly firm this recent conglomerate is1 ,i may mention1 that ,i endeavoured with a heavy geological hammer to knock out a thick bolt of iron1 which was embedded a little above low-water mark1 but was quite unable to succeed4 ,chapter ,,ii ,,fernando ,,noronha2 ,,terceira2 ,,tahiti1 ,,etc4 ,,fernando ,,noronha4--,precipitous hill of phonolite4 ,,terceira4--,trachytic rocks3 their singular decomposition by steam of high temperature4 ,,tahiti4--,passage from wacke into trap2 singular volcanic rock with the vesicles half-filled with mesotype4 ,,mauritius4--,proofs of its recent elevation4 ,structure of its more ancient mountains2 similarity with ,st4 ,jago4 ,,st4 ,,paul',s ,,rocks4--,not of volcanic origin4 ,their singular mineralogical composition4 ,fernando ,noronha4--,during our short visit at this and the four following islands1 ,i observed very little worthy of description4 ,fernando ,noronha is situated in the ,atlantic ,ocean1 in lat4 #c #ej ,s41 and #bcj miles distant from the coast of ,south ,america4 ,it consists of several islets1 together nine miles in length by three in breadth4 ,the whole seems to be of volcanic origin2 although there is no appearance of any crater1 or of any one central eminence4 ,the most remarkable feature is a hill #a1jjj feet high1 of which the upper #djj feet consist of a precipitous1 singularly shaped pinnacle1 formed of columnar phonolite1 containing numerous crystals of glassy feldspar1 and a few needles of hornblende4 ,from the highest accessible point of this hill1 ,i could distinguish in different parts of the group several other conical hills1 apparently of the same nature4 ,at ,st4 ,helena there are similar1 great1 conical1 protuberant masses of phonolite1 nearly one thousand feet in height1 which have been formed by the injection of fluid feldspathic lava into yielding strata4 ,if this hill has had1 as is probable1 a similar origin1 denudation has been here effected on an enormous scale4 ,near the base of this hill1 ,i observed beds of white tuff1 intersected by numerous dikes1 some of amygdaloidal basalt and others of trachyte2 and beds of slaty phonolite with the planes of cleavage directed ,n4,w4 and ,s4,e4 ,parts of this rock1 where the crystals were scanty1 closely resembled common clay-slate1 altered by the contact of a trap-dike4 ,the lamination of rocks1 which undoubtedly have once been fluid1 appears to me a subject well deserving attention4 ,on the beach there were numerous fragments of compact basalt1 of which rock a distant faade of columns seemed to be formed4 ,terceira in the ,azores4--,the central parts of this island consist of irregularly rounded mountains of no great elevation1 composed of trachyte1 which closely resembles in general character the trachyte of ,ascension1 presently to be described4 ,this formation is in many parts overlaid1 in the usual order of superposition1 by streams of basaltic lava1 which near the coast compose nearly the whole surface4 ,the course which these streams have followed from their craters1 can often be followed by the eye4 ,the town of ,angra is overlooked by a crateriform hill 7,mount ,brazil71 entirely built of thin strata of fine-grained1 harsh1 brown-coloured tuff4 ,the upper beds are seen to overlap the basaltic streams on which the town stands4 ,this hill is almost identical in structure and composition with numerous crateriformed hills in the ,galapagos ,archipelago4 ,effects of steam on the trachytic rocks4--,in the central part of the island there is a spot1 where steam is constantly issuing in jets from the bottom of a small ravine-like hollow1 which has no exit1 and which abuts against a range of trachytic mountains4 ,the steam is emitted from several irregular fissures3 it is scentless1 soon blackens iron1 and is of much too high temperature to be endured by the hand4 ,the manner in which the solid trachyte is changed on the borders of these orifices is curious3 first1 the base becomes earthy1 with red freckles evidently due to the oxidation of particles of iron2 then it becomes soft2 and lastly1 even the crystals of glassy feldspar yield to the dissolving agent4 ,after the mass is converted into clay1 the oxide of iron seems to be entirely removed from some parts1 which are left perfectly white1 whilst in other neighbouring parts1 which are of the brightest red colour1 it seems to be deposited in greater quantity2 some other masses are marbled with two distinct colours4 ,portions of the white clay1 now that they are dry1 cannot be distinguished by the eye from the finest prepared chalk2 and when placed between the teeth they feel equally soft-grained2 the inhabitants use this substance for white-washing their houses4 ,the cause of the iron being dissolved in one part1 and close by being again deposited1 is obscure2 but the fact has been observed in several other places4#a ,in some half-decayed specimens1 ,i found small1 globular aggregations of yellow hyalite1 resembling gum-arabic1 which no doubt had been deposited by the steam4 #a ,spallanzani1 ,dolomieu1 and ,hoffman have described similar cases in the ,italian volcanic islands4 ,dolomieu says the iron at the ,panza ,islands is redeposited in the form of veins 7p4 #hf 8,mmoire sur les ,isles ,ponces074 ,these authors likewise believe that the steam deposits silica3 it is now experimentally known that vapour of a high temperature is able to dissolve silica4 ,as there is no escape for the rain-water1 which trickles down the sides of the ravine-like hollow1 whence the steam issues1 it must all percolate downwards through the fissures at its bottom4 ,some of the inhabitants informed me that it was on record that flames 7some luminous appearance87 had originally proceeded from these cracks1 and that the flames had been succeeded by the steam2 but ,i was not able to ascertain how long this was ago1 or anything certain on the subject4 ,when viewing the spot1 ,i imagined that the injection of a large mass of rock4 like the cone of phonolite at ,fernando ,noronha1 in a semi-fluid state1 by arching the surface might have caused a wedge-shaped hollow with cracks at the bottom1 and that the rain-water percolating to the neighbourhood of the heated mass1 would during many succeeding years be driven back in the form of steam4 ,tahiti 7,otaheite74--,i visited only a part of the north-western side of this island1 and this part is entirely composed of volcanic rocks4 ,near the coast there are several varieties of basalt1 some abounding with large crystals of augite and tarnished olivine1 others compact and earthy1--some slightly vesicular1 and others occasionally amygdaloidal4 ,these rocks are generally much decomposed1 and to my surprise1 ,i found in several sections that it was impossible to distinguish1 even approximately1 the line of separation between the decayed lava and the alternating beds of tuff4 ,since the specimens have become dry1 it is rather more easy to distinguish the decomposed igneous rocks from the sedimentary tuffs4 ,this gradation in character between rocks having such widely different origins1 may ,i think be explained by the yielding under pressure of the softened sides of the vesicular cavities1 which in many volcanic rocks occupy a large proportion of their bulk4 ,as the vesicles generally increase in size and number in the upper parts of a stream of lava1 so would the effects of their compression increase2 the yielding1 moreover1 of each lower vesicle must tend to disturb all the softened matter above it4 ,hence we might expect to trace a perfect gradation from an unaltered crystalline rock to one in which all the particles 7although originally forming part of the same solid mass7 had undergone mechanical displacement2 and such particles could hardly be distinguished from others of similar composition1 which had been deposited as sediment4 ,as lavas are sometimes laminated in their upper parts even horizontal lines1 appearing like those of aqueous deposition1 could not in all cases be relied on as a criterion of sedimentary origin4 ,from these considerations it is not surprising that formerly many geologists believed in real transitions from aqueous deposits1 through wacke1 into igneous traps4 ,in the valley of ,tia-auru1 the commonest rocks are basalts with much olivine1 and in some cases almost composed of large crystals of augite4 ,i picked up some specimens1 with much glassy feldspar1 approaching in character to trachyte4 ,there were also many large blocks of vesicular basalt1 with the cavities beautifully lined with chabasie 7871 and radiating bundles of mesotype4 ,some of these specimens presented a curious appearance1 owing to a number of the vesicles being half filled up with a white1 soft1 earthy mesotypic mineral1 which intumesced under the blowpipe in a remarkable manner4 ,as the upper surfaces in all the half-filled cells are exactly parallel1 it is evident that this substance has sunk to the bottom of each cell from its weight4 ,sometimes1 however1 it entirely fills the cells4 ,other cells are either quite filled1 or lined1 with small crystals1 apparently of chabasie2 these crystals1 also1 frequently line the upper half of the cells partly filled with the earthy mineral1 as well as the upper surface of this substance itself1 in which case the two minerals appear to blend into each other4 ,i have never seen any other amygdaloid#b with the cells half filled in the manner here described2 and it is difficult to imagine the causes which determined the earthy mineral to sink from its gravity to the bottom of the cells1 and the crystalline mineral to adhere in a coating of equal thickness round the sides of the cells4 #b ,mac,culloch1 however1 has described and given a plate of 78,geolog4 ,trans40 #ast series1 vol4 iv1 p4 #bbe7 a trap rock1 with cavities filled up horizontally with quartz and chalcedony4 ,the upper halves of these cavities are often filled by layers1 which follow each irregularity of the surface1 and by little depending stalactites of the same siliceous substances4 ,the basic strata on the sides of the valley are gently inclined seaward1 and ,i nowhere observed any sign of disturbance2 the strata are separated from each other by thick1 compact beds of conglomerate1 in which the fragments are large1 some being rounded1 but most angular4 ,from the character of these beds1 from the compact and crystalline condition of most of the lavas1 and from the nature of the infiltrated minerals1 ,i was led to conjecture that they had originally flowed beneath the sea4 ,this conclusion agrees with the fact that the ,rev4 ,w4 ,ellis found marine remains at a considerable height1 which he believes were interstratified with volcanic matter2 as is likewise described to be the case by ,messrs4 ,tyerman and ,bennett at ,huaheine1 an island in this same archipelago4 ,mr4 ,stutchbury also discovered near the summit of one of the loftiest mountains of ,tahiti1 at the height of several thousand feet1 a stratum of semi-fossil coral4 ,none of these remains have been specifically examined4 ,on the coast1 where masses of coral-rock would have afforded the clearest evidence1 ,i looked in vain for any signs of recent elevation4 ,for references to the above authorities1 and for more detailed reasons for not believing that ,tahiti has been recently elevated1 ,i must refer to the 8,structure and ,distribution of ,coral-,reefs40 ,mauritius4--,approaching this island on the northern or north-western side1 a curved chain of bold mountains1 surmounted by rugged pinnacles1 is seen to rise from a smooth border of cultivated land1 which gently slopes down to the coast4 ,at the first glance1 one is tempted to believe that the sea lately reached the base of these mountains1 and upon examination1 this view1 at least with respect to the inferior parts of the border1 is found to be perfectly correct4 ,several authors#c have described masses of upraised coral-rock round the greater part of the circumference of the island4 ,between ,tamarin ,bay and the ,great ,black ,river ,i observed1 in company with ,captain ,lloyd1 two hillocks of coral-rock1 formed in their lower part of hard calcareous sandstone1 and in their upper of great blocks1 slightly aggregated1 of ,astra and ,madrepora1 and of fragments of basalt2 they were divided into beds dipping seaward1 in one case at an angle of #h1 and in the other at #ah2 they had a water-worn appearance1 and they rose abruptly from a smooth surface1 strewed with rolled dbris of organic remains1 to a height of about twenty feet4 ,the ,officier du ,roi1 in his most interesting tour in #agfh round the island1 has described masses of upraised coral-rocks1 still retaining that moat-like structure 7see my 8,coral ,reefs07 which is characteristic of the living reefs4 ,on the coast northward of ,port ,louis1 ,i found the lava concealed for a considerable space inland by a conglomerate of corals and shells1 like those on the beach1 but in parts consolidated by red ferruginous matter4 ,m4 ,bory ,st4 ,vincent has described similar calcareous beds over nearly the whole of the plain of ,pamplemousses4 ,near ,port ,louis1 when turning over some large stones1 which lay in the bed of a stream at the head of a protected creek1 and at the height of some yards above the level of spring tides1 ,i found several shells of serpula still adhering to their under sides4 #c ,captain ,carmichael1 in ,hooker's 8,bot4 ,misc410 vol4 ii1 p4 #cja4 ,captain ,lloyd has lately1 in the 8,proceedings of the ,geological ,society0 7vol4 iii1 p4 #cag71 described carefully some of these masses4 ,in the 8,voyage l',isle de ,france1 par un ,officier du ,roi10 many interesting facts are given on this subject4 ,consult also 8,voyage aux ,quatre ,isles d',afrique1 par ,m4 ,bory ,st4 ,vincent40 ,the jagged mountains near ,port ,louis rise to a height of between two and three thousand feet2 they consist of strata of basalt1 obscurely separated from each other by firmly aggregated beds of fragmentary matter2 and they are intersected by a few vertical dikes4 ,the basalt in some parts abounds with large crystals of augite and olivine1 and is generally compact4 ,the interior of the island forms a plain1 raised probably about a thousand feet above the level of the sea1 and composed of streams of lava which have flowed round and between the rugged basaltic mountains4 ,these more recent lavas are also basaltic1 but less compact1 and some of them abound with feldspar1 so that they even fuse into a pale coloured glass4 ,on the banks of the ,great ,river1 a section is exposed nearly five hundred feet deep1 worn through numerous thin sheets of the lava of this series1 which are separated from each other by beds of scori4 ,they seem to have been of subaerial formation1 and to have flowed from several points of eruption on the central platform1 of which the ,piton du ,milieu is said to be the principal one4 ,there are also several volcanic cones1 apparently of this modern period1 round the circumference of the island1 especially at the northern end1 where they form separate islets4 ,the mountains composed of the more compact and crystalline basalt1 form the main skeleton of the island4 ,m4 ,bailly#d states that they all 8se dveloppent autour d'elle comme une ceinture d'immenses remparts1 toutes affectant une pente plus ou moins encline vers le rivage de la mer2 tandis1 au contraire1 que vers le centre de l'ile elles presentent une coupe abrupte1 et souvent taille pic4 ,toutes ces montagnes sont formes de couches parallles inclines du centre de l'ile vers la mer40 ,these statements have been disputed1 though not in detail1 by ,m4 ,quoy1 in the voyage of ,freycinet4 ,as far as my limited means of observation went1 ,i found them perfectly correct4#e ,the mountains on the ,n4,w4 side of the island1 which ,i examined1 namely1 ,la ,pouce1 ,peter ,botts1 ,corps de ,garde1 ,les ,mamelles1 and apparently another farther southward1 have precisely the external shape and stratification described by ,m4 ,bailly4 ,they form about a quarter of his girdle of ramparts4 ,although these mountains now stand quite detached1 being separated from each other by breaches1 even several miles in width1 through which deluges of lava have flowed from the interior of the island2 nevertheless1 seeing their close general similarity1 one must feel convinced that they originally formed parts of one continuous mass4 ,judging from the beautiful map of the ,mauritius1 published by the ,admiralty from a ,french ,,ms41 there is a range of mountains 7,m4 ,bamboo7 on the opposite side of the island1 which correspond in height1 relative position1 and external form1 with those just described4 ,whether the girdle was ever complete may well be doubted2 but from ,m4 ,bailly's statements1 and my own observations1 it may be safely concluded that mountains with precipitous inland flanks1 and composed of strata dipping outwards1 once extended round a considerable portion of the circumference of the island4 ,the ring appears to have been oval and of vast size2 its shorter axis1 measured across from the inner sides of the mountains near ,port ,louis and those near ,grand ,port1 being no less than thirteen geographical miles in length4 ,m4 ,bailly boldly supposes that this enormous gulf1 which has since been filled up to a great extent by streams of modern lava1 was formed by the sinking in of the whole upper part of one great volcano4 #d 8,voyage aux ,terres ,australes10 tome i1 p4 #ed4 #e ,m4 ,lesson1 in his account of this island1 in the 8,voyage of the ,coquille10 seems to follow ,m4 ,bailly's views4 ,it is singular in how many respects those portions of ,st4 ,jago and of ,mauritius which ,i visited agree in their geological history4 ,at both islands1 mountains of similar external form1 stratification1 and 7at least in their upper beds7 composition1 follow in a curved chain the coast-line4 ,these mountains in each case appear originally to have formed parts of one continuous mass4 ,the basaltic strata of which they are composed1 from their compact and crystalline structure1 seem1 when contrasted with the neighbouring basaltic streams of subaerial formation1 to have flowed beneath the pressure of the sea1 and to have been subsequently elevated4 ,we may suppose that the wide breaches between the mountains were in both cases worn by the waves1 during their gradual elevation--of which process1 within recent times1 there is abundant evidence on the coast-land of both islands4 ,at both1 vast streams of more recent basaltic lavas have flowed from the interior of the island1 round and between the ancient basaltic hills2 at both1 moreover1 recent cones of eruption are scattered around the circumference of the island2 but at neither have eruptions taken place within the period of history4 ,as remarked in the last chapter1 it is probable that these ancient basaltic mountains1 which resemble 7at least in many respects7 the basal and disturbed remnants of two gigantic volcanoes1 owe their present form1 structure1 and position1 to the action of similar causes4 ,st4 ,paul's ,rocks4--,this small island is situated in the ,atlantic ,ocean1 nearly one degree north of the equator1 and #edj miles distant from ,south ,america1 in #bi #ae west longitude4 ,its highest point is scarcely fifty feet above the level of the sea2 its outline is irregular1 and its entire circumference barely three-quarters of a mile4 ,this little point of rock rises abruptly out of the ocean2 and1 except on its western side1 soundings were not obtained1 even at the short distance of a quarter of a mile from its shore4 ,it is not of volcanic origin2 and this circumstance1 which is the most remarkable point in its history 7as will hereafter be referred to71 properly ought to exclude it from the present volume4 ,it is composed of rocks1 unlike any which ,i have met with1 and which ,i cannot characterise by any name1 and must therefore describe4 ,the simplest1 and one of the most abundant kinds1 is a very compact1 heavy1 greenish-black rock1 having an angular1 irregular fracture1 with some points just hard enough to scratch glass1 and infusible4 ,this variety passes into others of paler green tints1 less hard1 but with a more crystalline fracture1 and translucent on their edges2 and these are fusible into a green enamel4 ,several other varieties are chiefly characterised by containing innumerable threads of dark-green serpentine1 and by having calcareous matter in their interstices4 ,these rocks have an obscure1 concretionary structure1 and are full of variously coloured angular pseudo fragments4 ,these angular pseudo fragments consist of the first-described dark green rock1 of a brown softer kind1 of serpentine1 and of a yellowish harsh stone1 which1 perhaps1 is related to serpentine rock4 ,there are other vesicular1 calcareo-ferruginous1 soft stones4 ,there is no distinct stratification1 but parts are imperfectly laminated2 and the whole abounds with innumerable veins1 and vein-like masses1 both small and large4 ,of these vein-like masses1 some calcareous ones1 which contain minute fragments of shells1 are clearly of subsequent origin to the others4 ,a glossy incrustation4--,extensive portions of these rocks are coated by a layer of a glossy polished substance1 with a pearly lustre and of a greyish white colour2 it follows all the inequalities of the surface1 to which it is firmly attached4 ,when examined with a lens1 it is found to consist of numerous exceedingly thin layers1 their aggregate thickness being about the tenth of an inch4 ,it is considerably harder than calcareous spar1 but can be scratched with a knife2 under the blowpipe it scales off1 decrepitates1 slightly blackens1 emits a fetid odour1 and becomes strongly alkaline3 it does not effervesce in acids4#f ,i presume this substance has been deposited by water draining from the birds' dung1 with which the rocks are covered4 ,at ,ascension1 near a cavity in the rocks which was filled with a laminated mass of infiltrated birds' dung1 ,i found some irregularly formed1 stalactitical masses of apparently the same nature4 ,these masses1 when broken1 had an earthy texture2 but on their outsides1 and especially at their extremities1 they were formed of a pearly substance1 generally in little globules1 like the enamel of teeth1 but more translucent1 and so hard as just to scratch plate-glass4 ,this substance slightly blackens under the blowpipe1 emits a bad smell1 then becomes quite white1 swelling a little1 and fuses into a dull white enamel2 it does not become alkaline2 nor does it effervesce in acids4 ,the whole mass had a collapsed appearance1 as if in the formation of the hard glossy crust the whole had shrunk much4 ,at the ,abrolhos ,islands on the coast of ,brazil1 where also there is much birds' dung1 ,i found a great quantity of a brown1 arborescent substance adhering to some trap-rock4 ,in its arborescent form1 this substance singularly resembles some of the branched species of ,nullipora4 ,under the blowpipe1 it behaves like the specimens from ,ascension2 but it is less hard and glossy1 and the surface has not the shrunk appearance4 #f ,in my 8,journal0 ,i have described this substance2 ,i then believed that it was an impure phosphate of lime4 ,chapter ,,iii ,,ascension4 ,basaltic lavas4--,numerous craters truncated on the same side4--,singular structure of volcanic bombs4--,aeriform explosions4--,ejected granitic fragments4--,trachytic rocks4--,singular veins4--,jasper1 its manner of formation4--,concretions in pumiceous tuff4--,calcareous deposits and frondescent incrustations on the coast4--,remarkable laminated beds1 alternating with1 and passing into1 obsidian4--,origin of obsidian4--,lamination of volcanic rocks4 ,this island is situated in the ,atlantic ,ocean1 in lat4 #h ,s41 long4 #ad ,w4 ,it has the form of an irregular triangle 7see map below71 each side being about six miles in length4 ,its highest point is #b1hgj feet#a above the level of the sea4 ,the whole is volcanic1 and1 from the absence of proofs to the contrary1 ,i believe of subaerial origin4 ,the fundamental rock is everywhere of a pale colour1 generally compact1 and of a feldspathic nature4 ,in the ,s4,e4 portion of the island1 where the highest land is situated1 well characterised trachyte1 and other congenerous rocks of that varying here and there a hill or single point of rock 7one of which near the sea-coast1 north of the ,fort1 is only two or three yards across7 of the trachyte still remaining exposed4 #a ,geographical ,journal1 vol4 v1 p4 #bdc4 ,illustration3 ,island of ,ascension ,,plate ,,iv4 ,basaltic rocks4--,the overlying basaltic lava is in some parts extremely vesicular1 in others little so2 it is of a black colour1 but sometimes contains crystals of glassy feldspar1 and seldom much olivine4 ,these streams appear to have possessed singularly little fluidity2 their side walls and lower ends being very steep1 and even as much as between twenty and thirty feet in height4 ,their surface is extraordinarily rugged1 and from a short distance appears as if studded with small craters4 ,these projections consist of broad1 irregularly conical1 hillocks1 traversed by fissures1 and composed of the same unequally scoriaceous basalt with the surrounding streams1 but having an obscure tendency to a columnar structure2 they rise to a height between ten and thirty feet above the general surface1 and have been formed1 as ,i presume1 by the heaping up of the viscid lava at points of greater resistance4 ,at the base of several of these hillocks1 and occasionally likewise on more level parts1 solid ribs1 composed of angulo-globular masses of basalt1 resembling in size and outline arched sewers or gutters of brickwork1 but not being hollow1 project between two or three feet above the surface of the streams2 what their origin may have been1 ,i do not know4 ,many of the superficial fragments from these basaltic streams present singularly convoluted forms2 and some specimens could hardly be distinguished from logs of dark-coloured wood without their bark4 ,many of the basaltic streams can be traced1 either to points of eruption at the base of the great central mass of trachyte1 or to separate1 conical1 red-coloured hills1 which are scattered over the northern and western borders of the island4 ,standing on the central eminence1 ,i counted between twenty and thirty of these cones of eruption4 ,the greater number of them had their truncated summits cut off obliquely1 and they all sloped towards the ,s4,e41 whence the trade-wind blows4#b ,this structure no doubt has been caused by the ejected fragments and ashes being always blown1 during eruptions1 in greater quantity towards one side than towards the other4 ,m4 ,moreau de ,jonnes has made a similar observation with respect to the volcanic orifices in the ,west ,indian ,islands4 #b ,m4 ,lesson in the 8,zoology of the ,voyage of the ,coquille10 p4 #dij has observed this fact4 ,mr4 ,hennah 78,geolog4 ,proceedings10 #ahce1 p4 #ahi7 further remarks that the most extensive beds of ashes at ,ascension invariably occur on the leeward side of the island4 ,volcanic bombs4--,these occur in great numbers strewed on the ground1 and some of them lie at considerable distances from any points of eruption4 ,they vary in size from that of an apple to that of a man's body2 they are either spherical or pear-shaped1 or with the hinder part 7corresponding to the tail of a comet7 irregular1 studded with projecting points1 and even concave4 ,their surfaces are rough1 and fissured with branching cracks2 their internal structure is either irregularly scoriaceous and compact1 or it presents a symmetrical and very curious appearance4 ,an irregular segment of a bomb of this latter kind1 of which ,i found several1 is accurately represented in figure ,no4 #c4 ,its size was about that of a man's head4 ,the whole interior is coarsely cellular2 the cells averaging in diameter about the tenth of an inch2 but nearer the outside they gradually decrease in size4 ,this part is succeeded by a well-defined shell of compact lava1 having a nearly uniform thickness of about the third of an inch2 and the shell is overlaid by a somewhat thicker coating of finely cellular lava 7the cells varying from the fiftieth to the hundredth of an inch in diameter71 which forms the external surface3 the line separating the shell of compact lava from the outer scoriaceous crust is distinctly defined4 ,this structure is very simply explained1 if we suppose a mass of viscid1 scoriaceous matter1 to be projected with a rapid1 rotatory motion through the air2 for whilst the external crust1 from cooling1 became solidified 7in the state we now see it71 the centrifugal force1 by relieving the pressure in the interior parts of the bomb1 would allow the heated vapours to expand their cells2 but these being driven by the same force against the already-hardened crust1 would become1 the nearer they were to this part1 smaller and smaller or less expanded1 until they became packed into a solid1 concentric shell4 ,as we know that chips from a grindstone#c can be flirted off1 when made to revolve with sufficient velocity1 we need not doubt that the centrifugal force would have power to modify the structure of a softened bomb1 in the manner here supposed4 ,geologists have remarked1 that the external form of a bomb at once bespeaks the history of its aerial course1 and few now see that the internal structure can speak1 with almost equal plainness1 of its rotatory movement4 #c ,nichol's 8,architecture of the ,heavens40 ,no4 #c ,illustration3 ,fragment of a spherical volcanic bomb4 ,fragment of a spherical volcanic bomb1 with the inferior parts coarsely cellular1 coated by a concentric layer of compact lava1 and this again by a crust of finely cellular rock4 ,no4 #d ,illustration3 ,volcanic bomb of obsidian from ,australia4 ,volcanic bomb of obsidian from ,australia4 ,the figure at left gives a front view2 the figure at right a side view of the same object4 ,m4 ,bory ,st4 ,vincent#d has described some balls of lava from the ,isle of ,bourbon1 which have a closely similar structure4 ,his explanation1 however 7if ,i understand it rightly71 is very different from that which ,i have given2 for he supposes that they have rolled1 like snowballs1 down the sides of the crater4 ,m4 ,beudant1#e also1 has described some singular little balls of obsidian1 never more than six or eight inches in diameter1 which he found strewed on the surface of the ground3 their form is always oval2 sometimes they are much swollen in the middle1 and even spindle-shaped3 their surface is regularly marked with concentric ridges and furrows1 all of which on the same ball are at right angles to one axis3 their interior is compact and glassy4 ,m4 ,beudant supposes that masses of lava1 when soft1 were shot into the air1 with a rotatory movement round the same axis1 and that the form and superficial ridges of the bombs were thus produced4 ,sir ,thomas ,mitchell has given me what at first appears to be the half of a much flattened oval ball of obsidian2 it has a singular artificial-like appearance1 which is well represented 7of the natural size7 in figure ,no4 #d4 ,it was found in its present state1 on a great sandy plain between the rivers ,darling and ,murray1 in ,australia1 and at the distance of several hundred miles from any known volcanic region4 ,it seems to have been embedded in some reddish tufaceous matter2 and may have been transported either by the aborigines or by natural means4 ,the external saucer consists of compact obsidian1 of a bottle-green colour1 and is filled with finely cellular black lava1 much less transparent and glassy than the obsidian4 ,the external surface is marked with four or five not quite perfect ridges1 which are represented rather too distinctly in figure ,no4 #d4 ,here1 then1 we have the external structure described by ,m4 ,beudant1 and the internal cellular condition of the bombs from ,ascension4 ,the lip of the saucer is slightly concave1 exactly like the margin of a soup-plate1 and its inner edge overlaps a little the central cellular lava4 ,this structure is so symmetrical round the entire circumference1 that one is forced to suppose that the bomb burst during its rotatory course1 before being quite solidified1 and that the lip and edges were thus slightly modified and turned inwards4 ,it may be remarked that the superficial ridges are in planes1 at right angles to an axis1 transverse to the longer axis of the flattened oval3 to explain this circumstance1 we may suppose that when the bomb burst1 the axis of rotation changed4 #d 8,voyage aux ,quatre ,isles d',afrique0 tome i1 p4 #bbb4 #e 8,voyage en ,hongrie10 tome ii1 p4 #bad4 ,aeriform explosions4--,the flanks of ,green ,mountain and the surrounding country are covered by a great mass1 some hundred feet in thickness1 of loose fragments4 ,the lower beds generally consist of fine-grained1 slightly consolidated tuffs1#f and the upper beds of great loose fragments1 with alternating finer beds4#g ,one white ribbon-like layer of decomposed1 pumiceous breccia1 was curiously bent into deep unbroken curves1 beneath each of the large fragments in the superincumbent stratum4 ,from the relative position of these beds1 ,i presume that a narrow-mouthed crater1 standing nearly in the position of ,green ,mountain1 like a great air-gun1 shot forth1 before its final extinction1 this vast accumulation of loose matter4 ,subsequently to this event1 considerable dislocations have taken place1 and an oval circus has been formed by subsidence4 ,this sunken space lies at the north-eastern foot of ,green ,mountain1 and is well represented in ,map #b4 ,its longer axis1 which is connected with a ,n4,e4 and ,s4,w4 line of fissure1 is three-fifths of a nautical mile in length2 its sides are nearly perpendicular1 except in one spot1 and about four hundred feet in height2 they consist1 in the lower part1 of a pale basalt with feldspar1 and in the upper part1 of the tuff and loose ejected fragments2 the bottom is smooth and level1 and under almost any other climate a deep lake would have been formed here4 ,from the thickness of the bed of loose fragments1 with which the surrounding country is covered1 the amount of aeriform matter necessary for their projection must have been enormous2 hence we may suppose it probable that after the explosions vast subterranean caverns were left1 and that the falling in of the roof of one of these produced the hollow here described4 ,at the ,galapagos ,archipelago1 pits of a similar character1 but of a much smaller size1 frequently occur at the bases of small cones of eruption4 #f ,some of this peperino1 or tuff1 is sufficiently hard not to be broken by the greatest force of the fingers4 #g ,on the northern side of the ,green ,mountain a thin seam1 about an inch in thickness1 of compact oxide of iron1 extends over a considerable area2 it lies conformably in the lower part of the stratified mass of ashes and fragments4 ,this substance is of a reddish-brown colour1 with an almost metallic lustre2 it is not magnetic1 but becomes so after having been heated under the blowpipe1 by which it is blackened and partly fused4 ,this seam of compact stone1 by intercepting the little rain-water which falls on the island1 gives rise to a small dripping spring1 first discovered by ,dampier4 ,it is the only fresh water on the island1 so that the possibility of its being inhabited has entirely depended on the occurrence of this ferruginous layer4 ,ejected granitic fragments4--,in the neighbourhood of ,green ,mountain1 fragments of extraneous rock are not unfrequently found embedded in the midst of masses of scori4 ,lieutenant ,evans1 to whose kindness ,i am indebted for much information1 gave me several specimens1 and ,i found others myself4 ,they nearly all have a granitic structure1 are brittle1 harsh to the touch1 and apparently of altered colours4 ,first1 a white syenite1 streaked and mottled with red2 it consists of well-crystallised feldspar1 numerous grains of quartz1 and brilliant1 though small1 crystals of hornblende4 ,the feldspar and hornblende in this and the succeeding cases have been determined by the reflecting goniometer1 and the quartz by its action under the blowpipe4 ,the feldspar in these ejected fragments1 like the glassy kind in the trachyte1 is from its cleavage a potash-feldspar4 ,secondly1 a brick-red mass of feldspar1 quartz1 and small dark patches of a decayed mineral2 one minute particle of which ,i was able to ascertain1 by its cleavage1 to be hornblende4 ,thirdly1 a mass of confusedly crystallised white feldspar1 with little nests of a dark-coloured mineral1 often carious1 externally rounded1 having a glossy fracture1 but no distinct cleavage3 from comparison with the second specimen1 ,i have no doubt that it is fused hornblende4 ,fourthly1 a rock1 which at first appears a simple aggregation of distinct and large-sized crystals of dusty-coloured ,labrador feldspar2#h but in their interstices there is some white granular feldspar1 abundant scales of mica1 a little altered hornblende1 and1 as ,i believe1 no quartz4 ,i have described these fragments in detail1 because it is rare#i to find granitic rocks ejected from volcanoes with their minerals unchanged1 as is the case with the first specimen1 and partially with the second4 ,one other large fragment1 found in another spot1 is deserving of notice2 it is a conglomerate1 containing small fragments of granitic1 cellular1 and jaspery rocks1 and of hornstone porphyries1 embedded in a base of wacke1 threaded by numerous thin layers of a concretionary pitchstone passing into obsidian4 ,these layers are parallel1 slightly tortuous1 and short2 they thin out at their ends1 and resemble in form the layers of quartz in gneiss4 ,it is probable that these small embedded fragments were not separately ejected1 but were entangled in a fluid volcanic rock1 allied to obsidian2 and we shall presently see that several varieties of this latter series of rock assume a laminated structure4 #h ,professor ,miller has been so kind as to examine this mineral4 ,he obtained two good cleavages of #hf #cj and #hf #ej4 ,the mean of several1 which ,i made1 was #hf #cj4 ,professor ,miller states that these crystals1 when reduced to a fine powder1 are soluble in hydrochloric acid1 leaving some undissolved silex behind2 the addition of oxalate of ammonia gives a copious precipitate of lime4 ,he further remarks1 that according to ,von ,kobell1 anorthite 7a mineral occurring in the ejected fragments at ,mount ,somma7 is always white and transparent1 so that if this be the case1 these crystals from ,ascension must be considered as ,labrador feldspar4 ,professor ,miller adds1 that he has seen an account1 in ,erdmann's 8,journal fr tecnische ,chemie10 of a mineral ejected from a volcano which had the external characters of ,labrador feldspar1 but differed in the analysis from that given by mineralogists of this mineral3 the author attributed this difference to an error in the analysis of ,labrador feldspar1 which is very old4 #i ,daubeny1 in his work on ,volcanoes 7p4 #chf71 remarks that this is the case2 and ,humboldt1 in his 8,personal ,narrative10 vol4 i1 p4 #bcf1 says 8,in general1 the masses of known primitive rocks1 ,i mean those which perfectly resemble our granites1 gneiss1 and mica-slate1 are very rare in lavas3 the substances we generally denote by the name of granite1 thrown out by ,vesuvius1 are mixtures of nepheline1 mica1 and pyroxene40 ,trachytic series of rocks4--,those occupy the more elevated and central1 and likewise the south-eastern1 parts of the island4 ,the trachyte is generally of a pale brown colour1 stained with small darker patches2 it contains broken and bent crystals of glassy feldspar1 grains of specular iron1 and black microscopical points1 which latter1 from being easily fused1 and then becoming magnetic1 ,i presume are hornblende4 ,the greater number of the hills1 however1 are composed of a quite white1 friable stone1 appearing like a trachytic tuff4 ,obsidian1 hornstone1 and several kinds of laminated feldspathic rocks1 are associated with the trachyte4 ,there is no distinct stratification2 nor could ,i distinguish a crateriform structure in any of the hills of this series4 ,considerable dislocations have taken place2 and many fissures in these rocks are yet left open1 or are only partially filled with loose fragments4 ,within the space1#aj mainly formed of trachyte1 some basaltic streams have burst forth2 and not far from the summit of ,green ,mountain1 there is one stream of quite black1 vesicular basalt1 containing minute crystals of glassy feldspar1 which have a rounded appearance4 #aj ,this space is nearly included by a line sweeping round ,green ,mountain1 and joining the hills1 called the ,weather ,port ,signal1 ,holyhead1 and that denominated 7improperly in a geological sense7 8the ,crater of an old volcano40 ,the soft white stone above mentioned is remarkable from its singular resemblance1 when viewed in mass1 to a sedimentary tuff3 it was long before ,i could persuade myself that such was not its origin2 and other geologists have been perplexed by closely similar formations in trachytic regions4 ,in two cases1 this white earthy stone formed isolated hills2 in a third1 it was associated with columnar and laminated trachyte2 but ,i was unable to trace an actual junction4 ,it contains numerous crystals of glassy feldspar and black microscopical specks1 and is marked with small darker patches1 exactly as in the surrounding trachyte4 ,its basis1 however1 when viewed under the microscope1 is generally quite earthy2 but sometimes it exhibits a decidedly crystalline structure4 ,on the hill marked 8,crater of an old volcano10 it passes into a pale greenish-grey variety1 differing only in its colour1 and in not being so earthy2 the passage was in one case effected insensibly2 in another1 it was formed by numerous1 rounded and angular1 masses of the greenish variety1 being embedded in the white variety2--in this latter case1 the appearance was very much like that of a sedimentary deposit1 torn up and abraded during the deposition of a subsequent stratum4 ,both these varieties are traversed by innumerable tortuous veins 7presently to be described71 which are totally unlike injected dikes1 or indeed any other veins which ,i have ever seen4 ,both varieties include a few scattered fragments1 large and small1 of dark-coloured scoriaceous rocks1 the cells of some of which are partially filled with the white earthy stone2 they likewise include some huge blocks of a cellular porphyry4#aa ,these fragments project from the weathered surface1 and perfectly resemble fragments embedded in a true sedimentary tuff4 ,but as it is known that extraneous fragments of cellular rock are sometimes included in columnar trachyte1 in phonolite1#ab and in other compact lavas1 this circumstance is not any real argument for the sedimentary origin of the white earthy stone4#ac ,the insensible passage of the greenish variety into the white one1 and likewise the more abrupt passage by fragments of the former being embedded in the latter1 might result from slight differences in the composition of the same mass of molten stone1 and from the abrading action of one such part still fluid on another part already solidified4 ,the curiously formed veins have1 ,i believe1 been formed by siliceous matter being subsequently segregated4 ,but my chief reason for believing that these soft earthy stones1 with their extraneous fragments1 are not of sedimentary origin1 is the extreme improbability of crystals of feldspar1 black microscopical specks1 and small stains of a darker colour occurring in the same proportional numbers in an aqueous deposit1 and in masses of solid trachyte4 ,moreover1 as ,i have remarked1 the microscope occasionally reveals a crystalline structure in the apparently earthy basis4 ,on the other hand1 the partial decomposition of such great masses of trachyte1 forming whole mountains1 is undoubtedly a circumstance of not easy explanation4 #aa ,the porphyry is dark coloured2 it contains numerous1 often fractured1 crystals of white opaque feldspar1 also decomposing crystals of oxide of iron2 its vesicles include masses of delicate1 hair-like1 crystals1 apparently of analcime4 #ab ,d',aubuisson1 8,trait de ,gognosie10 tome ii1 p4 #edh4 #ac ,dr4 ,daubeny 7on ,volcanoes1 p4 #ahj7 seems to have been led to believe that certain trachytic formations of ,ischia and of the ,puy de ,dme1 which closely resemble these of ,ascension1 were of sedimentary origin1 chiefly from the frequent presence in them 8of scoriform portions1 different in colour from the matrix40 ,dr4 ,daubeny adds1 that on the other hand1 ,brocchi1 and other eminent geologists1 have considered these beds as earthy varieties of trachyte2 he considers the subject deserving of further attention4 ,veins in the earthy trachytic masses4--,these veins are extraordinarily numerous1 intersecting in the most complicated manner both coloured varieties of the earthy trachyte3 they are best seen on the flanks of the 8,crater of the old volcano40 ,they contain crystals of glassy feldspar1 black microscopical specks and little dark stains1 precisely as in the surrounding rock2 but the basis is very different1 being exceedingly hard1 compact1 somewhat brittle1 and of rather less easy fusibility4 ,the veins vary much1 and suddenly1 from the tenth of an inch to one inch in thickness2 they often thin out1 not only on their edges1 but in their central parts1 thus leaving round1 irregular apertures2 their surfaces are rugged4 ,they are inclined at every possible angle with the horizon1 or are horizontal2 they are generally curvilinear1 and often interbranch one with another4 ,from their hardness they withstand weathering1 and projecting two or three feet above the ground1 they occasionally extend some yards in length2 these plate-like veins1 when struck1 emit a sound1 almost like that of a drum1 and they may be distinctly seen to vibrate2 their fragments1 which are strewed on the ground1 clatter like pieces of iron when knocked against each other4 ,they often assume the most singular forms2 ,i saw a pedestal of the earthy trachyte1 covered by a hemispherical portion of a vein1 like a great umbrella1 sufficiently large to shelter two persons4 ,i have never met with1 or seen described1 any veins like these2 but in form they resemble the ferruginous seams1 due to some process of segregation1 occurring not uncommonly in sandstones1--for instance1 in the ,new ,red sandstone of ,england4 ,numerous veins of jasper and of siliceous sinter1 occurring on the summit of this same hill1 show that there has been some abundant source of silica1 and as these plate-like veins differ from the trachyte only in their greater hardness1 brittleness1 and less easy fusibility1 it appears probable that their origin is due to the segregation or infiltration of siliceous matter1 in the same manner as happens with the oxides of iron in many sedimentary rocks4 ,siliceous sinter and jasper4--,the siliceous sinter is either quite white1 of little specific gravity1 and with a somewhat pearly fracture1 passing into pinkish pearl quartz2 or it is yellowish white1 with a harsh fracture1 and it then contains an earthy powder in small cavities4 ,both varieties occur1 either in large irregular masses in the altered trachyte1 or in seams included in broad1 vertical1 tortuous1 irregular veins of a compact1 harsh stone of a dull red colour1 appearing like a sandstone4 ,this stone1 however1 is only altered trachyte2 and a nearly similar variety1 but often honeycombed1 sometimes adheres to the projecting plate-like veins1 described in the last paragraph4 ,the jasper is of an ochre yellow or red colour2 it occurs in large irregular masses1 and sometimes in veins1 both in the altered trachyte and in an associated mass of scoriaceous basalt4 ,the cells of the scoriaceous basalt are lined or filled with fine1 concentric layers of chalcedony1 coated and studded with bright-red oxide of iron4 ,in this rock1 especially in the rather more compact parts1 irregular angular patches of the red jasper are included1 the edges of which insensibly blend into the surrounding mass2 other patches occur having an intermediate character between perfect jasper and the ferruginous1 decomposed1 basaltic base4 ,in these patches1 and likewise in the large vein-like masses of jasper1 there occur little rounded cavities1 of exactly the same size and form with the air-cells1 which in the scoriaceous basalt are filled and lined with layers of chalcedony4 ,small fragments of the jasper1 examined under the microscope1 seem to resemble the chalcedony with its colouring matter not separated into layers1 but mingled in the siliceous paste1 together with some impurities4 ,i can understand these facts1--namely1 the blending of the jasper into the semi-decomposed basalt1--its occurrence in angular patches1 which clearly do not occupy pre-existing hollows in the rock1--and its containing little vesicles filled with chalcedony1 like those in the scoriaceous lava1--only on the supposition that a fluid1 probably the same fluid which deposited the chalcedony in the air-cells1 removed in those parts where there were no cavities1 the ingredients of the basaltic rock1 and left in their place silica and iron1 and thus produced the jasper4 ,in some specimens of silicified wood1 ,i have observed1 that in the same manner as in the basalt1 the solid parts were converted into a dark-coloured homogeneous stone1 whereas the cavities formed by the larger sap-vessels 7which may be compared with the air-vesicles in the basaltic lava7 and other irregular hollows1 apparently produced by decay1 were filled with concentric layers of chalcedony2 in this case1 there can be little doubt that the same fluid deposited the homogeneous base and the chalcedonic layers4 ,after these considerations1 ,i cannot doubt but that the jasper of ,ascension may be viewed as a volcanic rock silicified1 in precisely the same sense as this term is applied to wood1 when silicified2 we are equally ignorant of the means by which every atom of wood1 whilst in a perfect state1 is removed and replaced by atoms of silica1 as we are of the means by which the constituent parts of a volcanic rock could be thus acted on4#ad ,i was led to the careful examination of these rocks1 and to the conclusion here given1 from having heard the ,rev4 ,professor ,henslow express a similar opinion1 regarding the origin in trap-rocks of many chalcedonies and agates4 ,siliceous deposits seem to be very general1 if not of universal occurrence1 in partially decomposed trachytic tuffs2#ae and as these hills1 according to the view above given1 consist of trachyte softened and altered in situ1 the presence of free silica in this case may be added as one more instance to the list4 #ad ,beudant 78,voyage en ,hongrie10 tome iii1 pp4 #ejb1 #ejd7 describes kidney-shaped masses of jasper-opal1 which either blend into the surrounding trachytic conglomerate1 or are embedded in it like chalk-flints2 and he compares them with the fragments of opalised wood1 which are abundant in this same formation4 ,beudant1 however1 appears to have viewed the process of their formation rather as one of simple infiltration than of molecular exchange2 but the presence of a concretion1 wholly different from the surrounding matter1 if not formed in a pre-existing hollow1 clearly seems to me to require1 either a molecular or mechanical displacement of the atoms1 which occupied the space afterwards filled by it4 ,the jasper-opal of ,hungary passes into chalcedony1 and therefore in this case1 as in that of ,ascension1 jasper seems to be intimately related in origin with chalcedony4 #ae ,beudant 78,voyage ,min410 tome iii1 p4 #ejg7 enumerates cases in ,hungary1 ,germany1 ,central ,france1 ,italy1 ,greece1 and ,mexico4 ,concretions in pumiceous tuff4--,the hill1 marked in ,map #b 8,crater of an old volcano10 has no claims to this appellation1 which ,i could discover1 except in being surmounted by a circular1 very shallow1 saucer-like summit1 nearly half a mile in diameter4 ,this hollow has been nearly filled up with many successive sheets of ashes and scori1 of different colours1 and slightly consolidated4 ,each successive saucer-shaped layer crops out all round the margin1 forming so many rings of various colours1 and giving to the hill a fantastic appearance4 ,the outer ring is broad1 and of a white colour2 hence it resembles a course round which horses have been exercised1 and has received the name of the ,devil's ,riding ,school1 by which it is most generally known4 ,these successive layers of ashes must have fallen over the whole surrounding country1 but they have all been blown away except in this one hollow1 in which probably moisture accumulated1 either during an extraordinary year when rain fell1 or during the storms often accompanying volcanic eruptions4 ,one of the layers of a pinkish colour1 and chiefly derived from small1 decomposed fragments of pumice1 is remarkable1 from containing numerous concretions4 ,these are generally spherical1 from half an inch to three inches in diameter2 but they are occasionally cylindrical1 like those of iron-pyrites in the chalk of ,europe4 ,they consist of a very tough1 compact1 pale-brown stone1 with a smooth and even fracture4 ,they are divided into concentric layers by thin white partitions1 resembling the external superficies2 six or eight of such layers are distinctly defined near the outside2 but those towards the inside generally become indistinct1 and blend into a homogeneous mass4 ,i presume that these concentric layers were formed by the shrinking of the concretion1 as it became compact4 ,the interior part is generally fissured by minute cracks or septaria1 which are lined1 both by black1 metallic1 and by other white and crystalline specks1 the nature of which ,i was unable to ascertain4 ,some of the larger concretions consist of a mere spherical shell1 filled with slightly consolidated ashes4 ,the concretions contain a small proportion of carbonate of lime3 a fragment placed under the blowpipe decrepitates1 then whitens and fuses into a blebby enamel1 but does not become caustic4 ,the surrounding ashes do not contain any carbonate of lime2 hence the concretions have probably been formed1 as is so often the case1 by the aggregation of this substance4 ,i have not met with any account of similar concretions2 and considering their great toughness and compactness1 their occurrence in a bed1 which probably has been subjected only to atmospheric moisture1 is remarkable4 ,formation of calcareous rocks on the sea-coast4--,on several of the sea-beaches1 there are immense accumulations of small1 well-rounded particles of shells and corals1 of white1 yellowish1 and pink colours1 interspersed with a few volcanic particles4 ,at the depth of a few feet1 these are found cemented together into stone1 of which the softer varieties are used for building2 there are other varieties1 both coarse and fine-grained1 too hard for this purpose3 and ,i saw one mass divided into even layers half an inch in thickness1 which were so compact that when struck with a hammer they rang like flint4 ,it is believed by the inhabitants1 that the particles become united in the course of a single year4 ,the union is effected by calcareous matter2 and in the most compact varieties1 each rounded particle of shell and volcanic rock can be distinctly seen to be enveloped in a husk of pellucid carbonate of lime4 ,extremely few perfect shells are embedded in these agglutinated masses2 and ,i have examined even a large fragment under a microscope1 without being able to discover the least vestige of stri or other marks of external form3 this shows how long each particle must have been rolled about1 before its turn came to be embedded and cemented4#af ,one of the most compact varieties1 when placed in acid1 was entirely dissolved1 with the exception of some flocculent animal matter2 its specific gravity was #bfc4 ,the specific gravity of ordinary limestone varies from #bf to #bge2 pure ,carrara marble was found by ,sir ,h4 ,de la ,beche#ag to be #bg4 ,it is remarkable that these rocks of ,ascension1 formed close to the surface1 should be nearly as compact as marble1 which has undergone the action of heat and pressure in the plutonic regions4 #af ,the eggs of the turtle being buried by the parent1 sometimes become enclosed in the solid rock4 ,mr4 ,lyell has given a figure 78,principles of ,geology10 book iii1 ch4 #ag7 of some eggs1 containing the bones of young turtles1 found thus entombed4 #ag 8,researches in ,theoretical ,geology10 p4 #ab4 ,the great accumulation of loose calcareous particles1 lying on the beach near the ,settlement1 commences in the month of ,october1 moving towards the ,s4,w41 which1 as ,i was informed by ,lieutenant ,evans1 is caused by a change in the prevailing direction of the currents4 ,at this period the tidal rocks1 at the ,s4,w4 end of the beach1 where the calcareous sand is accumulating1 and round which the currents sweep1 become gradually coated with a calcareous incrustation1 half an inch in thickness4 ,it is quite white1 compact1 with some parts slightly spathose1 and is firmly attached to the rock4 ,after a short time it gradually disappears1 being either redissolved1 when the water is less charged with lime1 or more probably is mechanically abraded4 ,lieutenant ,evans has observed these facts1 during the six years he has resided at ,ascension4 ,the incrustation varies in thickness in different years3 in #ahca it was unusually thick4 ,when ,i was there in ,july1 there was no remnant of the incrustation2 but on a point of basalt1 from which the quarrymen had lately removed a mass of the calcareous freestone1 the incrustation was perfectly preserved4 ,considering the position of the tidal-rocks1 and the period at which they become coated1 there can be no doubt that the movement and disturbance of the vast accumulation of calcareous particles1 many of them being partially agglutinated together1 cause the waves of the sea to be so highly charged with carbonate of lime1 that they deposit it on the first objects against which they impinge4 ,i have been informed by ,lieutenant ,holland1 ,r4,n41 that this incrustation is formed on many parts of the coast1 on most of which1 ,i believe1 there are likewise great masses of comminuted shells4 ,a frondescent calcareous incrustation4--,in many respects this is a singular deposit2 it coats throughout the year the tidal volcanic rocks1 that project from the beaches composed of broken shells4 ,its general appearance is well represented in ,figure #e2 but the fronds or discs1 of which it is composed1 are generally so closely crowded together as to touch4 ,these fronds have their sinuous edges finely crenulated1 and they project over their pedestals or supports2 their upper surfaces are either slightly concave1 or slightly convex2 they are highly polished1 and of a dark grey or jet black colour2 their form is irregular1 generally circular1 and from the tenth of an inch to one inch and a half in diameter2 their thickness1 or amount of their projection from the rock on which they stand1 varies much1 about a quarter of an inch being perhaps most usual4 ,the fronds occasionally become more and more convex1 until they pass into botryoidal masses with their summits fissured2 when in this state1 they are glossy and of an intense black1 so as to resemble some fused metallic substance4 ,i have shown the incrustation1 both in this latter and in its ordinary state to several geologists1 but not one could conjecture its origin1 except that perhaps it was of volcanic nature6 ,no4 #e ,illustration3 ,an incrustration of calcareous and animal matter4 ,an incrustation of calcareous and animal matter1 coating the tidal-rocks at ,ascension4 ,the substance forming the fronds has a very compact and often almost crystalline fracture2 the edges being translucent1 and hard enough easily to scratch calcareous spar4 ,under the blowpipe it immediately becomes white1 and emits a strong animal odour1 like that from fresh shells4 ,it is chiefly composed of carbonate of lime2 when placed in muriatic acid it froths much1 leaving a residue of sulphate of lime1 and of an oxide of iron1 together with a black powder1 which is not soluble in heated acids4 ,this latter substance seems to be carbonaceous1 and is evidently the colouring matter4 ,the sulphate of lime is extraneous1 and occurs in distinct1 excessively minute1 lamellar plates1 studded on the surface of the fronds1 and embedded between the fine layers of which they are composed2 when a fragment is heated in the blowpipe1 these lamell are immediately rendered visible4 ,the original outline of the fronds may often be traced1 either to a minute particle of shell fixed in a crevice of the rock1 or to several cemented together2 these first become deeply corroded1 by the dissolving power of the waves1 into sharp ridges1 and then are coated with successive layers of the glossy1 grey1 calcareous incrustation4 ,the inequalities of the primary support affect the outline of every successive layer1 in the same manner as may often be seen in bezoar-stones1 when an object like a nail forms the centre of aggregation4 ,the crenulated edges1 however1 of the frond appear to be due to the corroding power of the surf on its own deposit1 alternating with fresh depositions4 ,on some smooth basaltic rocks on the coast of ,st4 ,jago1 ,i found an exceedingly thin layer of brown calcareous matter1 which under a lens presented a miniature likeness of the crenulated and polished fronds of ,ascension2 in this case a basis was not afforded by any projecting extraneous particles4 ,although the incrustation at ,ascension is persistent throughout the year2 yet from the abraded appearance of some parts1 and from the fresh appearance of other parts1 the whole seems to undergo a round of decay and renovation1 due probably to changes in the form of the shifting beach1 and consequently in the action of the breakers3 hence probably it is1 that the incrustation never acquires a great thickness4 ,considering the position of the encrusted rocks in the midst of the calcareous beach1 together with its composition1 ,i think there can be no doubt that its origin is due to the dissolution and subsequent deposition of the matter composing the rounded particles of shells and corals4#ah ,from this source it derives its animal matter1 which is evidently the colouring principle4 ,the nature of the deposit1 in its incipient stage1 can often be well seen upon a fragment of white shell1 when jammed between two of the fronds2 it then appears exactly like the thinnest wash of a pale grey varnish4 ,its darkness varies a little1 but the jet blackness of some of the fronds and of the botryoidal masses seems due to the translucency of the successive grey layers4 ,there is1 however1 this singular circumstance1 that when deposited on the under side of ledges of rock or in fissures1 it appears always to be of a pale1 pearly grey colour1 even when of considerable thickness3 hence one is led to suppose1 that an abundance of light is necessary to the development of the dark colour1 in the same manner as seems to be the case with the upper and exposed surfaces of the shells of living mollusca1 which are always dark1 compared with their under surfaces and with the parts habitually covered by the mantle of the animal4 ,in this circumstance1--in the immediate loss of colour and in the odour emitted under the blowpipe1--in the degree of hardness and translucency of the edges1--and in the beautiful polish of the surface1#ai rivalling when in a fresh state that of the finest ,oliva1 there is a striking analogy between this inorganic incrustation and the shells of living molluscous animals4#bj ,this appears to me to be an interesting physiological fact4#ba #ah ,the selenite1 as ,i have remarked is extraneous1 and must have been derived from the sea-water4 ,it is an interesting circumstance thus to find the waves of the ocean1 sufficiently charged with sulphate of lime1 to deposit it on the rocks1 against which they dash every tide4 ,dr4 ,webster has described 78,voyage of the ,chanticleer10 vol4 ii1 p4 #cai7 beds of gypsum and salt1 as much as two feet in thickness1 left by the evaporation of the spray on the rocks on the windward coast4 ,beautiful stalactites of selenite1 resembling in form those of carbonate of lime1 are formed near these beds4 ,amorphous masses of gypsum1 also1 occur in caverns in the interior of the island2 and at ,cross ,hill 7an old crater7 ,i saw a considerable quantity of salt oozing from a pile of scori4 ,in these latter cases1 the salt and gypsum appear to be volcanic products47 #ai ,from the fact described in my 8,journal of ,researches0 of a coating of oxide of iron1 deposited by a streamlet on the rocks in its bed 7like a nearly similar coating at the great cataracts of the ,orinoco and ,nile71 becoming finely polished where the surf acts1 ,i presume that the surf in this instance1 also1 is the polishing agent47 #bj ,in the section descriptive of ,st4 ,paul's ,rocks1 ,i have described a glossy1 pearly substance1 which coats the rocks1 and an allied stalactitical incrustation from ,ascension1 the crust of which resembles the enamel of teeth1 but is hard enough to scratch plate-glass4 ,both these substances contain animal matter1 and seem to have been derived from water in filtering through birds' dung4 #ba ,mr4 ,horner and ,sir ,david ,brewster have described 78,philosophical ,transactions10 #ahcf1 p4 #fe7 a singular 8artificial substance1 resembling shell40 ,it is deposited in fine1 transparent1 highly polished1 brown-coloured lamin1 possessing peculiar optical properties1 on the inside of a vessel1 in which cloth1 first prepared with glue and then with lime1 is made to revolve rapidly in water4 ,it is much softer1 more transparent1 and contains more animal matter1 than the natural incrustation at ,ascension2 but we here again see the strong tendency which carbonate of lime and animal matter evince to form a solid substance allied to shell4 ,singular laminated beds alternating with and passing into obsidian4--,these beds occur within the trachytic district1 at the western base of ,green ,mountain1 under which they dip at a high inclination4 ,they are only partially exposed1 being covered up by modern ejections2 from this cause1 ,i was unable to trace their junction with the trachyte1 or to discover whether they had flowed as a stream of lava1 or had been injected amidst the overlying strata4 ,there are three principal beds of obsidian1 of which the thickest forms the base of the section4 ,the alternating stony layers appear to me eminently curious1 and shall be first described1 and afterwards their passage into the obsidian4 ,they have an extremely diversified appearance2 five principal varieties may be noticed1 but these insensibly blend into each other by endless gradations4 ,first4--,a pale grey1 irregularly and coarsely laminated1#bb harsh-feeling rock1 resembling clay-slate which has been in contact with a trap-dike1 and with a fracture of about the same degree of crystalline structure4 ,this rock1 as well as the following varieties1 easily fuses into a pale glass4 ,the greater part is honeycombed with irregular1 angular1 cavities1 so that the whole has a curious appearance1 and some fragments resemble in a remarkable manner silicified logs of decayed wood4 ,this variety1 especially where more compact1 is often marked with thin whitish streaks1 which are either straight or wrap round1 one behind the other1 the elongated carious hollows4 #bb ,this term is open to some misinterpretation1 as it may be applied both to rocks divided into lamin of exactly the same composition1 and to layers firmly attached to each other1 with no fissile tendency1 but composed of different minerals1 or of different shades of colour4 ,the term 8laminated10 in this chapter1 is applied in these latter senses2 where a homogeneous rock splits1 as in the former sense1 in a given direction1 like clay-slate1 ,i have used the term 8fissile40 ,secondly4--,a bluish grey or pale brown1 compact1 heavy1 homogeneous stone1 with an angular1 uneven1 earthy fracture2 viewed1 however1 under a lens of high power1 the fracture is seen to be distinctly crystalline1 and even separate minerals can be distinguished4 ,thirdly4--,a stone of the same kind with the last1 but streaked with numerous1 parallel1 slightly tortuous1 white lines of the thickness of hairs4 ,these white lines are more crystalline than the parts between them2 and the stone splits along them3 they frequently expand into exceedingly thin cavities1 which are often only just perceptible with a lens4 ,the matter forming the white lines becomes better crystallised in these cavities1 and ,professor ,miller was fortunate enough1 after several trials1 to ascertain that the white crystals1 which are the largest1 were of quartz1#bc and that the minute green transparent needles were augite1 or1 as they would more generally be called1 diopside3 besides these crystals1 there are some minute1 dark specks without a trace of crystalline1 and some fine1 white1 granular1 crystalline matter which is probably feldspar4 ,minute fragments of this rock are easily fusible4 #bc ,professor ,miller informs me that the crystals which he measured had the faces ,p1 z1 m of the figure 7#adg7 given by ,haidinger in his ,translation of ,mohs2 and he adds1 that it is remarkable1 that none of them had the slightest trace of faces r of the regular six-sided prism4 ,fourthly4--,a compact crystalline rock1 banded in straight lines with innumerable layers of white and grey shades of colour1 varying in width from the thirtieth to the two-hundredth of an inch2 these layers seem to be composed chiefly of feldspar1 and they contain numerous perfect crystals of glassy feldspar1 which are placed lengthways2 they are also thickly studded with microscopically minute1 amorphous1 black specks1 which are placed in rows1 either standing separately1 or more frequently united1 two or three or several together1 into black lines1 thinner than a hair4 ,when a small fragment is heated in the blowpipe1 the black specks are easily fused into black brilliant beads1 which become magnetic1--characters that apply to no common mineral except hornblende or augite4 ,with the black specks there are mingled some others of a red colour1 which are magnetic before being heated1 and no doubt are oxide of iron4 ,round two little cavities1 in a specimen of this variety1 ,i found the black specks aggregated into minute crystals1 appearing like those of augite or hornblende1 but too dull and small to be measured by the goniometer2 in the specimen1 also1 ,i could distinguish amidst the crystalline feldspar1 grains1 which had the aspect of quartz4 ,by trying with a parallel ruler1 ,i found that the thin grey layers and the black hair-like lines were absolutely straight and parallel to each other4 ,it is impossible to trace the gradation from the homogeneous grey rocks to these striped varieties1 or indeed the character of the different layers in the same specimen1 without feeling convinced that the more or less perfect whiteness of the crystalline feldspathic matter depends on the more or less perfect aggregation of diffused matter1 into the black and red specks of hornblende and oxide of iron4 ,fifthly4--,a compact heavy rock1 not laminated1 with an irregular1 angular1 highly crystalline1 fracture2 it abounds with distinct crystals of glassy feldspar1 and the crystalline feldspathic base is mottled with a black mineral1 which on the weathered surface is seen to be aggregated into small crystals1 some perfect1 but the greater number imperfect4 ,i showed this specimen to an experienced geologist1 and asked him what it was2 he answered1 as ,i think every one else would have done1 that it was a primitive greenstone4 ,the weathered surface1 also1 of the banded variety in figure ,no4 #d1 strikingly resembles a worn fragment of finely laminated gneiss4 ,these five varieties1 with many intermediate ones1 pass and repass into each other4 ,as the compact varieties are quite subordinate to the others1 the whole may be considered as laminated or striped4 ,the lamin1 to sum up their characteristics1 are either quite straight1 or slightly tortuous1 or convoluted2 they are all parallel to each other1 and to the intercalating strata of obsidian2 they are generally of extreme thinness2 they consist either of an apparently homogeneous1 compact rock1 striped with different shades of grey and brown colours1 or of crystalline feldspathic layers in a more or less perfect state of purity1 and of different thicknesses1 with distinct crystals of glassy feldspar placed lengthways1 or of very thin layers chiefly composed of minute crystals of quartz and augite1 or composed of black and red specks of an augitic mineral and of an oxide of iron1 either not crystallised or imperfectly so4 ,after having fully described the obsidian1 ,i shall return to the subject of the lamination of rocks of the trachytic series4 ,the passage of the foregoing beds into the strata of glassy obsidian is effected in several ways3 first1 angulo-modular masses of obsidian1 both large and small1 abruptly appear disseminated in a slaty1 or in an amorphous1 pale-coloured1 feldspathic rock1 with a somewhat pearly fracture4 ,secondly1 small irregular nodules of the obsidian1 either standing separately1 or united into thin layers1 seldom more than the tenth of an inch in thickness1 alternate repeatedly with very thin layers of a feldspathic rock1 which is striped with the finest parallel zones of colour1 like an agate1 and which sometimes passes into the nature of pitchstone2 the interstices between the nodules of obsidian are generally filled by soft white matter1 resembling pumiceous ashes4 ,thirdly1 the whole substance of the bounding rock suddenly passes into an angulo-concretionary mass of obsidian4 ,such masses 7as well as the small nodules7 of obsidian are of a pale green colour1 and are generally streaked with different shades of colour1 parallel to the lamin of the surrounding rock2 they likewise generally contain minute white sphrulites1 of which half is sometimes embedded in a zone of one shade of colour1 and half in a zone of another shade4 ,the obsidian assumes its jet black colour and perfectly conchoidal fracture1 only when in large masses2 but even in these1 on careful examination and on holding the specimens in different lights1 ,i could generally distinguish parallel streaks of different shades of darkness4 ,no4 #f ,illustration3 ,opaque brown sphrulites1 drawn on an enlarged scale4 ,opaque brown sphrulites1 drawn on an enlarged scale4 ,the upper ones are externally marked with parallel ridges4 ,the internal radiating structure of the lower ones1 is much too plainly represented4 ,no4 #g ,illustration3 ,a layer1 formed by the union of minute brown sphrulites4 ,a layer1 formed by the union of minute brown sphrulites1 intersecting two other similar layers3 the whole represented of nearly the natural size4 ,one of the commonest transitional rocks deserves in several respects a further description4 ,it is of a very complicated nature1 and consists of numerous thin1 slightly tortuous layers of a pale-coloured feldspathic stone1 often passing into an imperfect pitchstone1 alternating with layers formed of numberless little globules of two varieties of obsidian1 and of two kinds of sphrulites1 embedded in a soft or in a hard pearly base4 ,the sphrulites are either white and translucent1 or dark brown and opaque2 the former are quite spherical1 of small size1 and distinctly radiated from their centre4 ,the dark brown sphrulites are less perfectly round1 and vary in diameter from the twentieth to the thirtieth of an inch2 when broken they exhibit towards their centres1 which are whitish1 an obscure radiating structure2 two of them when united sometimes have only one central point of radiation2 there is occasionally a trace of or a hollow crevice in their centres4 ,they stand either separately1 or are united two or three or many together into irregular groups1 or more commonly into layers1 parallel to the stratification of the mass4 ,this union in many cases is so perfect1 that the two sides of the layer thus formed1 are quite even2 and these layers1 as they become less brown and opaque1 cannot be distinguished from the alternating layers of the pale-coloured feldspathic stone4 ,the sphrulites1 when not united1 are generally compressed in the plane of the lamination of the mass2 and in this same plane1 they are often marked internally1 by zones of different shades of colour1 and externally by small ridges and furrows4 ,in the upper part of figure ,no4 #f1 the sphrulites with the parallel ridges and furrows are represented on an enlarged scale1 but they are not well executed2 and in the lower part1 their usual manner of grouping is shown4 ,in another specimen1 a thin layer formed of the brown sphrulites closely united together1 intersects1 as represented in figure ,no4 #g1 a layer of similar composition2 and after running for a short space in a slightly curved line1 again intersects it1 and likewise a second layer lying a little way beneath that first intersected4 ,the small nodules also of obsidian are sometimes externally marked with ridges and furrows1 parallel to the lamination of the mass1 but always less plainly than the sphrulites4 ,these obsidian nodules are generally angular1 with their edges blunted3 they are often impressed with the form of the adjoining sphrulites1 than which they are always larger2 the separate nodules seldom appear to have drawn each other out by exerting a mutually attractive force4 ,had ,i not found in some cases1 a distinct centre of attraction in these nodules of obsidian1 ,i should have been led to have considered them as residuary matter1 left during the formation of the pearlstone1 in which they are embedded1 and of the sphrulitic globules4 ,the sphrulites and the little nodules of obsidian in these rocks so closely resemble1 in general form and structure1 concretions in sedimentary deposits1 that one is at once tempted to attribute to them an analogous origin4 ,they resemble ordinary concretions in the following respects3 in their external form1--in the union of two or three1 or of several1 into an irregular mass1 or into an even-sided layer1--in the occasional intersection of one such layer by another1 as in the case of chalk-flints1--in the presence of two or three kinds of nodules1 often close together1 in the same basis1--in their fibrous1 radiating structure1 with occasional hollows in their centres1--in the co-existence of a laminary1 concretionary1 and radiating structure1 as is so well developed in the concretions of magnesian limestone1 described by ,professor ,sedgwick4#bd ,concretions in sedimentary deposits1 it is known1 are due to the separation from the surrounding mass of the whole or part of some mineral substance1 and its aggregation round certain points of attraction4 ,guided by this fact1 ,i have endeavoured to discover whether obsidian and the sphrulites 7to which may be added marekanite and pearlstone1 both of them occurring in nodular concretions in the trachytic series7 differ in their constituent parts1 from the minerals generally composing trachytic rocks4 ,it appears from three analyses1 that obsidian contains on an average #gf per cent of silica2 from one analysis1 that sphrulites contain #giab2 from two1 that marekanite contains #gibe2 and from two other analyses1 that pearlstone contains #gefb of silica4#be ,now1 the constituent parts of trachyte1 as far as they can be distinguished consist of feldspar1 containing #feba of silica2 or of albite1 containing #fiji2 of hornblende1 containing #eebg1bf and of oxide of iron3 so that the foregoing glassy concretionary substances all contain a larger proportion of silica than that occurring in ordinary feldspathic or trachytic rocks4 ,d',aubuisson1#bg also1 has remarked on the large proportion of silica compared with alumina1 in six analyses of obsidian and pearlstone given in ,brongniart's 8,mineralogy40 ,hence ,i conclude1 that the foregoing concretions have been formed by a process of aggregation1 strictly analogous to that which takes place in aqueous deposits1 acting chiefly on the silica1 but likewise on some of the other elements of the surrounding mass1 and thus producing the different concretionary varieties4 ,from the well-known effects of rapid cooling#bh in giving glassiness of texture1 it is probably necessary that the entire mass1 in cases like that of ,ascension1 should have cooled at a certain rate2 but considering the repeated and complicated alterations of nodules and thin layers of a glassy texture with other layers quite stony or crystalline1 all within the space of a few feet or even inches1 it is hardly possible that they could have cooled at different rates1 and thus have acquired their different textures4 #bd 8,geological ,transactions10 vol4 #c1 part i1 p4 #cg4 #be ,the foregoing analyses are taken from ,beudant 8,trait de ,minralogie10 tome ii1 p4 #aac2 and one analysis of obsidian from ,phillips's 8,mineralogy40 #bf ,these analyses are taken from ,von ,kobell's 8,grundzge der ,mineralogie10 #ahch4 #bg 8,trait de ,gogn410 tome ii1 p4 #ece4 #bh ,this is seen in the manufacture of common glass1 and in ,gregory ,watts's experiments on molten trap2 also on the natural surfaces of lava-streams1 and on the side-walls of dikes4 ,the natural sphrulites in these rocks#bi very closely resemble those produced in glass1 when slowly cooled4 ,in some fine specimens of partially devitrified glass1 in the possession of ,mr4 ,stokes1 the sphrulites are united into straight layers with even sides1 parallel to each other1 and to one of the outer surfaces1 exactly as in the obsidian4 ,these layers sometimes interbranch and form loops2 but ,i did not see any case of actual intersection4 ,they form the passage from the perfectly glassy portions1 to those nearly homogeneous and stony1 with only an obscure concretionary structure4 ,in the same specimen1 also1 sphrulites differing slightly in colour and in structure1 occur embedded close together4 ,considering these facts1 it is some confirmation of the view above given of the concretionary origin of the obsidian and natural sphrulites1 to find that ,m4 ,dartigues1#cj in his curious paper on this subject1 attributes the production of sphrulites in glass1 to the different ingredients obeying their own laws of attraction and becoming aggregated4 ,he is led to believe that this takes place1 from the difficulty in remelting sphrulitic glass1 without the whole be first thoroughly pounded and mixed together2 and likewise from the fact1 that the change takes place most readily in glass composed of many ingredients4 ,in confirmation of ,m4 ,dartigues' view1 ,i may remark1 that ,m4 ,fleuriau de ,bellevue#ca found that the sphrulitic portions of devitrified glass were acted on both by nitric acid and under the blowpipe1 in a different manner from the compact paste in which they were embedded4 #bi ,i do not know whether it is generally known1 that bodies having exactly the same appearance as sphrulites1 sometimes occur in agates4 ,mr4 ,robert ,brown showed me in an agate1 formed within a cavity in a piece of silicified wood1 some little specks1 which were only just visible to the naked eye3 these specks1 when placed by him under a lens of high power1 presented a beautiful appearance3 they were perfectly circular1 and consisted of the finest fibres of a brown colour1 radiating with great exactness from a common centre4 ,these little radiating stars are occasionally intersected1 and portions are quite cut off by the fine1 ribbon-like zones of colour in the agate4 ,in the obsidian of ,ascension1 the halves of a sphrulite often lie in different zones of colour1 but they are not cut off by them1 as in the agate4 #cj ,journal de ,physique1 tome #ei 7#ahjd71 pp4 #aj1 #ab4 #ca ,idem1 tome #fj 7#ahje71 p4 #dah4 ,comparison of the obsidian beds and alternating strata of ascension1 with those of other countries4--,i have been struck with much surprise1 how closely the excellent description of the obsidian rocks of ,hungary1 given by ,beudant1#cb and that by ,humboldt1 of the same formation in ,mexico and ,peru1#cc and likewise the descriptions given by several authors#cd of the trachytic regions in the ,italian islands1 agree with my observations at ,ascension4 ,many passages might have been transferred without alteration from the works of the above authors1 and would have been applicable to this island4 ,they all agree in the laminated and stratified character of the whole series2 and ,humboldt speaks of some of the beds of obsidian being ribboned like jasper4#ce ,they all agree in the nodular or concretionary character of the obsidian1 and of the passage of these nodules into layers4 ,they all refer to the repeated alterations1 often in undulatory planes1 of glassy1 pearly1 stony1 and crystalline layers3 the crystalline layers1 however1 seem to be much more perfectly developed at ,ascension1 than in the above-named countries4 ,humboldt compares some of the stony beds1 when viewed from a distance1 to strata of a schistose sandstone4 ,sphrulites are described as occurring abundantly in all cases2 and they everywhere seem to mark the passage1 from the perfectly glassy to the stony and crystalline beds4 ,beudant's account#cf of his 8perlite lithoide globulaire0 in every1 even the most trifling particular1 might have been written for the little brown sphrulitic globules of the rocks of ,ascension4 #cb 8,voyage en ,hongrie10 tome i1 p4 #ccj2 tome ii1 pp4 #bba and #cae2 tome iii1 pp4 #cfi1 #cga1 #cgg1 #cha4 #cc 8,essai ,gognostique10 pp4 #agf1 #cbf1 #cbh4 #cd ,p4 ,scrope1 in 8,geological ,transactions10 vol4 ii 7second series7 p4 #aie4 ,consult1 also1 ,dolomieu's 8,voyage aux ,isles ,lipari10 and ,d',aubuisson1 8,trait de ,gogn410 tome ii1 p4 #ecd4 #ce ,in ,mr4 ,stokes' fine collection of obsidians from ,mexico1 ,i observe that the sphrulites are generally much larger than those of ,ascension2 they are generally white1 opaque1 and are united into distinct layers3 there are many singular varieties1 different from any at ,ascension4 ,the obsidians are finely zoned1 in quite straight or curved lines1 with exceedingly slight differences of tint1 of cellularity1 and of more or less perfect degrees of glassiness4 ,tracing some of the less perfectly glassy zones1 they are seen to become studded with minute white sphrulites1 which become more and more numerous1 until at last they unite and form a distinct layer3 on the other hand1 at ,ascension1 only the brown sphrulites unite and form layers2 the white ones always being irregularly disseminated4 ,some specimens at the ,geological ,society1 said to belong to an obsidian formation from ,mexico1 have an earthy fracture1 and are divided in the finest parallel lamin1 by specks of a black mineral1 like the augitic or hornblendic specks in the rocks at ,ascension4 #cf ,beudant's 8,voyage10 tome iii1 p4 #cgc4 ,from the close similarity in so many respects1 between the obsidian formations of ,hungary1 ,mexico1 ,peru1 and of some of the ,italian islands1 with that of ,ascension1 ,i can hardly doubt that in all these cases1 the obsidian and the sphrulites owe their origin to a concretionary aggregation of the silica1 and of some of the other constituent elements1 taking place whilst the liquified mass cooled at a certain required rate4 ,it is1 however1 well-known1 that in several places1 obsidian has flowed in streams like lava2 for instance1 at ,teneriffe1 at the ,lipari ,islands1 and at ,iceland4#cg ,in these cases1 the superficial parts are the most perfectly glassy1 the obsidian passing at the depth of a few feet into an opaque stone4 ,in an analysis by ,vauquelin of a specimen of obsidian from ,hecla1 which probably flowed as lava1 the proportion of silica is nearly the same as in the nodular or concretionary obsidian from ,mexico4 ,it would be interesting to ascertain1 whether the opaque interior portions and the superficial glassy coating contained the same proportional constituent parts3 we know from ,m4 ,dufrnoy#ch that the exterior and interior parts of the same stream of lava sometimes differ considerably in their composition4 ,even should the whole body of the stream of obsidian turn out to be similarly composed with nodular obsidian1 it would only be necessary1 in accordance with the foregoing facts1 to suppose that lava in these instances had been erupted with its ingredients mixed in the same proportion1 as in the concretionary obsidian4 #cg ,for ,teneriffe1 see von ,buch1 8,descript4 des ,isles ,canaries10 pp4 #ahd and #aij2 for the ,lipari ,islands1 see ,dolomieu's 8,voyage10 p4 #cd2 for ,iceland1 see ,mackenzie's 8,travels10 p4 #cfi4 #ch 8,mmoires pour servir a une ,descript4 ,golog4 de la ,france10 tome iv1 p4 #cga4 ,lamination of volcanic rocks of the trachytic series4 ,we have seen that1 in several and widely distant countries1 the strata alternating with beds of obsidian1 are highly laminated4 ,the nodules1 also1 both large and small1 of the obsidian1 are zoned with different shades of colour2 and ,i have seen a specimen from ,mexico in ,mr4 ,stokes' collection1 with its external surface weathered#ci into ridges and furrows1 corresponding with the zones of different degrees of glassiness3 ,humboldt1#dj moreover1 found on the ,peak of ,teneriffe1 a stream of obsidian divided by very thin1 alternating1 layers of pumice4 ,many other lavas of the feldspathic series are laminated2 thus1 masses of common trachyte at ,ascension are divided by fine earthy lines1 along which the rock splits1 separating thin layers of slightly different shades of colour2 the greater number1 also1 of the embedded crystals of glassy feldspar are placed lengthways in the same direction4 ,mr4 ,p4 ,scrope#da has described a remarkable columnar trachyte in the ,panza ,islands1 which seems to have been injected into an overlying mass of trachytic conglomerate3 it is striped with zones1 often of extreme tenuity1 of different textures and colours2 the harder and darker zones appearing to contain a larger proportion of silica4 ,in another part of the island1 there are layers of pearlstone and pitchstone1 which in many respects resemble those of ,ascension4 ,the zones in the columnar trachyte are generally contorted2 they extend uninterruptedly for a great length in a vertical direction1 and apparently parallel to the walls of the dike-like mass4 ,von ,buch#db has described at ,teneriffe1 a stream of lava containing innumerable thin1 plate-like crystals of feldspar1 which are arranged like white threads1 one behind the other1 and which mostly follow the same direction4 ,dolomieu#dc also states1 that the grey lavas of the modern cone of ,vulcano1 which have a vitreous texture1 are streaked with parallel white lines3 he further describes a solid pumice-stone which possesses a fissile structure1 like that of certain micaceous schists4 ,phonolite1 which ,i may observe is often1 if not always1 an injected rock1 also1 often has a fissile structure2 this is generally due to the parallel position of the embedded crystals of feldspar1 but sometimes1 as at ,fernando ,noronha1 seems to be nearly independent of their presence4#dd ,from these facts we see1 that various rocks of the feldspathic series have either a laminated or fissile structure1 and that it occurs both in masses which have injected into overlying strata1 and in others which have flowed as streams of lava4 #ci ,mac,culloch states 78,classification of ,rocks10 p4 #eca71 that the exposed surfaces of the pitchstone dikes in ,arran are furrowed 8with undulating lines1 resembling certain varieties of marbled paper1 and which evidently result from some corresponding difference of laminar structure40 #dj 8,personal ,narrative10 vol4 i1 p4 #bbb4 #da 8,geological ,transactions10 vol4 ii 7second series71 p4 #aie4 #db 8,description des ,iles ,canaries10 p4 #ahd4 #dc 8,voyage aux ,isles de ,lipari10 pp4 #ce and #he4 #dd ,in this case1 and in that of the fissile pumice-stone1 the structure is very different from that in the foregoing cases1 where the lamin consist of alternate layers of different composition or texture4 ,in some sedimentary formations1 however1 which apparently are homogeneous and fissile1 as in glossy clay-slate1 there is reason to believe1 according to ,d',aubuisson1 that the lamin are really due to excessively thin1 alternating1 layers of mica4 ,the lamin of the beds1 alternating with the obsidian at ,ascension1 dip at a high angle under the mountain1 at the base of which they are situated2 and they do not appear as if they had been inclined by violence4 ,a high inclination is common to these beds in ,mexico1 ,peru1 and in some of the ,italian islands3#de on the other hand1 in ,hungary1 the layers are horizontal2 the lamin1 also1 of some of the lava-streams above referred to1 as far as ,i can understand the descriptions given of them1 appear to be highly inclined or vertical4 ,i doubt whether in any of these cases1 the lamin have been tilted into their present position2 and in some instances1 as in that of the trachyte described by ,mr4 ,scrope1 it is almost certain that they have been originally formed with a high inclination4 ,in many of these cases1 there is evidence that the mass of liquified rock has moved in the direction of the lamin4 ,at ,ascension1 many of the air-cells have a drawn out appearance1 and are crossed by coarse semi-glassy fibres1 in the direction of the lamin2 and some of the layers1 separating the sphrulitic globules1 have a scored appearance1 as if produced by the grating of the globules4 ,i have seen a specimen of zoned obsidian from ,mexico1 in ,mr4 ,stokes' collection1 with the surfaces of the best-defined layers streaked or furrowed with parallel lines2 and these lines or streaks precisely resembled those1 produced on the surface of a mass of artificial glass by its having been poured out of a vessel4 ,humboldt1 also1 has described little cavities1 which he compares to the tails of comets1 behind sphrulites in laminated obsidian rocks from ,mexico1 and ,mr4 ,scrope has described other cavities behind fragments embedded in his laminated trachyte1 and which he supposes to have been produced during the movement of the mass4#df ,from such facts1 most authors have attributed the lamination of these volcanic rocks to their movement whilst liquified4 ,although it is easy to perceive1 why each separate air-cell1 or each fibre in pumice-stone1#dg should be drawn out in the direction of the moving mass2 it is by no means at first obvious why such air-cells and fibres should be arranged by the movement1 in the same planes1 in lamin absolutely straight and parallel to each other1 and often of extreme tenuity2 and still less obvious is it1 why such layers should come to be of slightly different composition and of different textures4 #de ,see ,phillips' 8,mineralogy10 for the ,italian ,islands1 p4 #acf4 ,for ,mexico and ,peru1 see ,humboldt's 8,essai ,gognostique40 ,mr4 ,edwards also describes the high inclination of the obsidian rocks of the ,cerro del ,navaja in ,mexico in the ,proc4 of the ,geolog4 ,soc4 ,june #ahch4 #df 8,geological ,transactions10 vol4 ii 7second series71 p4 #bjj etc4 ,these embedded fragments1 in some instances1 consist of the laminated trachyte broken off and 8enveloped in those parts1 which still remained liquid40 ,beudant1 also1 frequently refers in his great work on 8,hungary0 7tome iii1 p4 #chf71 to trachytic rocks1 irregularly spotted with fragments of the same varieties1 which in other parts form the parallel ribbons4 ,in these cases1 we must suppose1 that after part of the molten mass had assumed a laminated structure1 a fresh irruption of lava broke up the mass1 and involved fragments1 and that subsequently the whole became relaminated4 #dg ,dolomieu's 8,voyage10 p4 #fd4 ,in endeavouring to make out the cause of the lamination of these igneous feldspathic rocks1 let us return to the facts so minutely described at ,ascension4 ,we there see1 that some of the thinnest layers are chiefly formed by numerous1 exceedingly minute1 though perfect1 crystals of different minerals2 that other layers are formed by the union of different kinds of concretionary globules1 and that the layers thus formed1 often cannot be distinguished from the ordinary feldspathic and pitchstone layers1 composing a large portion of the entire mass4 ,the fibrous radiating structure of the sphrulites seems1 judging from many analogous cases1 to connect the concretionary and crystalline forces3 the separate crystals1 also1 of feldspar all lie in the same parallel planes4#dh ,these allied forces1 therefore1 have played an important part in the lamination of the mass1 but they cannot be considered the primary force2 for the several kinds of nodules1 both the smallest and largest1 are internally zoned with excessively fine shades of colour1 parallel to the lamination of the whole2 and many of them are1 also1 externally marked in the same direction with parallel ridges and furrows1 which have not been produced by weathering4 #dh ,the formation1 indeed1 of a large crystal of any mineral in a rock of mixed composition implies an aggregation of the requisite atoms1 allied to concretionary action4 ,the cause of the crystals of feldspar in these rocks of ,ascension1 being all placed lengthways1 is probably the same with that which elongates and flattens all the brown sphrulitic globules 7which behave like feldspar under the blowpipe7 in this same direction4 ,some of the finest streaks of colour in the stony layers1 alternating with the obsidian1 can be distinctly seen to be due to an incipient crystallisation of the constituent minerals4 ,the extent to which the minerals have crystallised can1 also1 be distinctly seen to be connected with the greater or less size1 and with the number1 of the minute1 flattened1 crenulated air-cavities or fissures4 ,numerous facts1 as in the case of geodes1 and of cavities in silicified wood1 in primary rocks1 and in veins1 show that crystallisation is much favoured by space4 ,hence1 ,i conclude1 that1 if in a mass of cooling volcanic rock1 any cause produced in parallel planes a number of minute fissures or zones of less tension 7which from the pent-up vapours would often be expanded into crenulated air-cavities71 the crystallisation of the constituent parts1 and probably the formation of concretions1 would be superinduced or much favoured in such planes2 and thus1 a laminated structure of the kind we are here considering would be generated4 ,that some cause does produce parallel zones of less tension in volcanic rocks1 during their consolidation1 we must admit in the case of the thin alternate layers of obsidian and pumice described by ,humboldt1 and of the small1 flattened1 crenulated air-cells in the laminated rocks of ,ascension2 for on no other principle can we conceive why the confined vapours should through their expansion form air-cells or fibres in separate1 parallel planes1 instead of irregularly throughout the mass4 ,in ,mr4 ,stokes' collection1 ,i have seen a beautiful example of this structure1 in a specimen of obsidian from ,mexico1 which is shaded and zoned1 like the finest agate1 with numerous1 straight1 parallel layers1 more or less opaque and white1 or almost perfectly glassy2 the degree of opacity and glassiness depending on the number of microscopically minute1 flattened air-cells2 in this case1 it is scarcely possible to doubt but that the mass1 to which the fragment belonged1 must have been subjected to some1 probably prolonged1 action1 causing the tension slightly to vary in the successive planes4 ,several causes appear capable of producing zones of different tension1 in masses semi-liquified by heat4 ,in a fragment of devitrified glass1 ,i have observed layers of sphrulites which appeared1 from the manner in which they were abruptly bent1 to have been produced by the simple contraction of the mass in the vessel1 in which it cooled4 ,in certain dikes on ,mount ,etna1 described by ,m4 ,elie de ,beaumont1#di as bordered by alternating bands of scoriaceous and compact rock1 one is led to suppose that the stretching movement of the surrounding strata1 which originally produced the fissures1 continued whilst the injected rock remained fluid4 ,guided1 however1 by ,professor ,forbes'#ej clear description of the zoned structure of glacier-ice1 far the most probable explanation of the laminated structure of these feldspathic rocks appears to be1 that they have been stretched whilst slowly flowing onwards in a pasty condition1#ea in precisely the same manner as ,professor ,forbes believes1 that the ice of moving glaciers is stretched and fissured4 ,in both cases1 the zones may be compared to those in the finest agates2 in both1 they extend in the direction in which the mass has flowed1 and those exposed on the surface are generally vertical3 in the ice1 the porous lamin are rendered distinct by the subsequent congelation of infiltrated water1 in the stony feldspathic lavas1 by subsequent crystalline and concretionary action4 ,the fragment of glassy obsidian in ,mr4 ,stokes' collection1 which is zoned with minute air-cells must strikingly resemble1 judging from ,professor ,forbes' descriptions1 a fragment of the zoned ice2 and if the rate of cooling and nature of the mass had been favourable to its crystallisation or to concretionary action1 we should here have had the finest parallel zones of different composition and texture4 ,in glaciers1 the lines of porous ice and of minute crevices seem to be due to an incipient stretching1 caused by the central parts of the frozen stream moving faster than the sides and bottom1 which are retarded by friction3 hence in glaciers of certain forms and towards the lower end of most glaciers1 the zones become horizontal4 ,may we venture to suppose that in the feldspathic lavas with horizontal lamin1 we see an analogous case8 ,all geologists1 who have examined trachytic regions1 have come to the conclusion1 that the lavas of this series have possessed an exceedingly imperfect fluidity2 and as it is evident that only matter thus characterised would be subject to become fissured and to be formed into zones of different tensions1 in the manner here supposed1 we probably see the reason why augitic lavas1 which appear generally to have possessed a high degree of fluidity1 are not1#eb like the feldspathic lavas1 divided into lamin of different composition and texture4 ,moreover1 in the augitic series1 there never appears to be any tendency to concretionary action1 which we have seen plays an important part in the lamination of rocks1 of the trachytic series1 or at least in rendering that structure apparent4 #di 8,mm4 pour servir10 etc41 tome iv1 p4 #aca4 #ej ,edinburgh ,new ,phil4 ,journal1 #ahdb1 p4 #cej4 #ea ,i presume that this is nearly the same explanation which ,mr4 ,scrope had in his mind1 when he speaks 78,geolog4 ,transact410 vol4 ii1 second series1 p4 #bbh7 of the ribboned structure of his trachytic rocks1 having arisen1 from 8a linear extension of the mass1 while in a state of imperfect liquidity1 coupled with a concretionary process40 #eb ,basaltic lavas1 and many other rocks1 are not unfrequently divided into thick lamin or plates1 of the same composition1 which are either straight or curved2 these being crossed by vertical lines of fissure1 sometimes become united into columns4 ,this structure seems related1 in its origin1 to that by which many rocks1 both igneous and sedimentary1 become traversed by parallel systems of fissures4 ,whatever may be thought of the explanation here advanced of the laminated structure of the rocks of the trachytic series1 ,i venture to call the attention of geologists to the simple fact1 that in a body of rock at ,ascension1 undoubtedly of volcanic origin1 layers often of extreme tenuity1 quite straight1 and parallel to each other1 have been produced2--some composed of distinct crystals of quartz and diopside1 mingled with amorphous augitic specks and granular feldspar1--others entirely composed of these black augitic specks1 with granules of oxide of iron1--and lastly1 others formed of crystalline feldspar1 in a more or less perfect state of purity1 together with numerous crystals of feldspar1 placed lengthways4 ,at this island1 there is reason to believe1 and in some analogous cases1 it is certainly known1 that the lamin have originally been formed with their present high inclination4 ,facts of this nature are manifestly of importance1 with relation to the structural origin of that grand series of plutonic rocks1 which like the volcanic have undergone the action of heat1 and which consist of alternate layers of quartz1 feldspar1 mica and other minerals4 ,chapter ,,iv ,,st4 ,,helena ,lavas of the feldspathic1 basaltic1 and submarine series4--,section of ,flagstaff ,hill and of the ,barn4--,dikes4--,turk's ,cap and ,prosperous ,bays4--,basaltic ring4--,central crateriform ridge1 with an internal ledge and a parapet4 ,cones of phonolite4 ,superficial beds of calcareous sandstone4--,extinct land-shells4--,beds of detritus4--,elevation of the land4--,denudation4--,craters of elevation4 ,the whole island is of volcanic origin2 its circumference1 according to ,beatson1#a is about twenty-eight miles4 ,the central and largest part consists of rocks of a feldspathic nature1 generally decomposed to an extraordinary degree2 and when in this state1 presenting a singular assemblage of alternating1 red1 purple1 brown1 yellow1 and white1 soft1 argillaceous beds4 ,from the shortness of our visit1 ,i did not examine these beds with care2 some of them1 especially those of the white1 yellow1 and brown shades1 originally existed as streams of lava1 but the greater number were probably ejected in the form of scori and ashes3 other beds of a purple tint1 porphyritic with crystal-shaped patches of a white1 soft substance1 which are now unctuous1 and yield1 like wax1 a polished streak to the nail1 seem once to have existed as solid claystone-porphyries3 the red argillaceous beds generally have a brecciated structure1 and no doubt have been formed by the decomposition of scori4 ,several extensive streams1 however1 belonging to this series1 retain their stony character2 these are either of a blackish-green colour1 with minute acicular crystals of feldspar1 or of a very pale tint1 and almost composed of minute1 often scaly1 crystals of feldspar1 abounding with microscopical black specks2 they are generally compact and laminated2 others1 however1 of similar composition1 are cellular and somewhat decomposed4 ,none of these rocks contain large crystals of feldspar1 or have the harsh fracture peculiar to trachyte4 ,these feldspathic lavas and tuffs are the uppermost or those last erupted2 innumerable dikes1 however1 and great masses of molten rock1 have subsequently been injected into them4 ,they converge1 as they rise1 towards the central curved ridge1 of which one point attains the elevation of #b1gjj feet4 ,this ridge is the highest land in the island2 and it once formed the northern rim of a great crater1 whence the lavas of this series flowed3 from its ruined condition1 from the southern half having been removed1 and from the violent dislocation which the whole island has undergone1 its structure is rendered very obscure4 #a ,governor ,beatson's 8,account of ,st4 ,helena40 ,basaltic series4--,the margin of the island is formed by a rude circle of great1 black1 stratified1 ramparts of basalt1 dipping seaward1 and worn into cliffs1 which are often nearly perpendicular1 and vary in height from a few hundred feet to two thousand4 ,this circle1 or rather horse-shoe shaped ring1 is open to the south1 and is breached by several other wide spaces4 ,its rim or summit generally projects little above the level of the adjoining inland country2 and the more recent feldspathic lavas1 sloping down from the central heights1 generally abut against and overlap its inner margin2 on the north-western side of the island1 however1 they appear 7judging from a distance7 to have flowed over and concealed portions of it4 ,in some parts1 where the basaltic ring has been breached1 and the black ramparts stand detached1 the feldspathic lavas have passed between them1 and now overhang the sea-coast in lofty cliffs4 ,the basaltic rocks are of a black colour and thinly stratified2 they are generally highly vesicular1 but occasionally compact2 some of them contain numerous crystals of glassy feldspar and octahedrons of titaniferous iron2 others abound with crystals of augite and grains of olivine4 ,the vesicles are frequently lined with minute crystals 7of chabasie87 and even become amygdaloidal with them4 ,the streams are separated from each other by cindery matter1 or by a bright red1 friable1 saliferous tuff1 which is marked by successive lines like those of aqueous deposition2 and sometimes it has an obscure1 concretionary structure4 ,the rocks of this basaltic series occur nowhere except near the coast4 ,in most volcanic districts the trachytic lavas are of anterior origin to the basaltic2 but here we see1 that a great pile of rock1 closely related in composition to the trachytic family1 has been erupted subsequently to the basaltic strata3 the number1 however1 of dikes1 abounding with large crystals of augite1 with which the feldspathic lavas have been injected1 shows perhaps some tendency to a return to the more usual order of superposition4 ,basal submarine lavas4--,the lavas of this basal series lie immediately beneath both the basaltic and feldspathic rocks4 ,according to ,mr4 ,seale1#b they may be seen at intervals on the sea-beach round the entire island4 ,in the sections which ,i examined1 their nature varied much2 some of the strata abound with crystals of augite2 others are of a brown colour1 either laminated or in a rubbly condition2 and many parts are highly amygdaloidal with calcareous matter4 ,the successive sheets are either closely united together1 or are separated from each other by beds of scoriaceous rock and of laminated tuff1 frequently containing well-rounded fragments4 ,the interstices of these beds are filled with gypsum and salt2 the gypsum also sometimes occurring in thin layers4 ,from the large quantity of these two substances1 from the presence of rounded pebbles in the tuffs1 and from the abundant amygdaloids1 ,i cannot doubt that these basal volcanic strata flowed beneath the sea4 ,this remark ought perhaps to be extended to a part of the superincumbent basaltic rocks2 but on this point1 ,i was not able to obtain clear evidence4 ,the strata of the basal series1 whenever ,i examined them1 were intersected by an extraordinary number of dikes4 #b 8,geognosy of the ,island of ,st4 ,helena40 ,mr4 ,seale has constructed a gigantic model of ,st4 ,helena1 well worth visiting1 which is now deposited at ,addiscombe ,college1 in ,surrey4 ,flagstaff ,hill and the ,barn4--,i will now describe some of the more remarkable sections1 and will commence with these two hills1 which form the principal external feature on the north-eastern side of the island4 ,the square1 angular outline1 and black colour of the ,barn1 at once show that it belongs to the basaltic series2 whilst the smooth1 conical figure1 and the varied bright tints of ,flagstaff ,hill1 render it equally clear1 that it is composed of the softened1 feldspathic rocks4 ,these two lofty hills are connected 7as is shown in figure ,no4 #h7 by a sharp ridge1 which is composed of the rubbly lavas of the basal series4 ,the strata of this ridge dip westward1 the inclination becoming less and less towards the ,flagstaff2 and the upper feldspathic strata of this hill can be seen1 though with some difficulty1 to dip conformably to the ,w4,s4,w4 ,close to the ,barn1 the strata of the ridge are nearly vertical1 but are much obscured by innumerable dikes2 under this hill1 they probably change from being vertical into being inclined into an opposite direction2 for the upper or basaltic strata1 which are about eight hundred or one thousand feet in thickness1 are inclined north-eastward1 at an angle between thirty and forty degrees4 ,no4 #h ,illustration3 ,flagstaff ,hill and the ,barn4 ,the double lines represent the basaltic strata2 the single1 the basal submarine strata2 the dotted1 the upper feldspathic strata2 the dikes are shaded transversely4 ,this ridge1 and likewise the ,barn and ,flagstaff ,hills1 are interlaced by dikes1 many of which preserve a remarkable parallelism in a ,n4,n4,w4 and ,s4,s4,e4 direction4 ,the dikes chiefly consist of a rock1 porphyritic with large crystals of augite2 others are formed of a fine-grained and brown-coloured trap4 ,most of these dikes are coated by a glossy layer1#c from one to two-tenths of an inch in thickness1 which1 unlike true pitchstone1 fuses into a black enamel2 this layer is evidently analogous to the glossy superficial coating of many lava streams4 ,the dikes can often be followed for great lengths both horizontally and vertically1 and they seem to preserve a nearly uniform thickness3#d ,mr4 ,seale states1 that one near the ,barn1 in a height of #a1bfj feet1 decreases in width only four inches1--from nine feet at the bottom1 to eight feet and eight inches at the top4 ,on the ridge1 the dikes appear to have been guided in their course1 to a considerable degree1 by the alternating soft and hard strata3 they are often firmly united to the harder strata1 and they preserve their parallelism for such great lengths1 that in very many instances it was impossible to conjecture1 which of the beds were dikes1 and which streams of lava4 ,the dikes1 though so numerous on this ridge1 are even more numerous in the valleys a little south of it1 and to a degree ,i never saw equalled anywhere else3 in these valleys they extend in less regular lines1 covering the ground with a network1 like a spider's web1 and with some parts of the surface even appearing to consist wholly of dikes1 interlaced by other dikes4 #c ,this circumstance has been observed 7,lyell1 8,principles of ,geology10 vol4 iv1 chap4 x1 p4 #i7 in the dikes of the ,atrio del ,cavallo1 but apparently it is not of very common occurrence4 ,sir ,g4 ,mackenzie1 however1 states 7p4 #cgb1 8,travels in ,iceland07 that all the veins in ,iceland have a 8black vitreous coating on their sides40 ,captain ,carmichael1 speaking of the dikes in ,tristan d',acunha1 a volcanic island in the ,southern ,atlantic1 says 78,linnan ,transactions10 vol4 xii1 p4 #dhe7 that their sides1 8where they come in contact with the rocks1 are invariably in a semi-vitrified state40 #d 8,geognosy of the ,island of ,st4 ,helena10 plate #e4 ,from the complexity produced by the dikes1 from the high inclination and anticlinal dip of the strata of the basal series1 which are overlaid1 at the opposite ends of the short ridge1 by two great masses of different ages and of different composition1 ,i am not surprised that this singular section has been misunderstood4 ,it has even been supposed to form part of a crater2 but so far is this from having been the case1 that the summit of ,flagstaff ,hill once formed the lower extremity of a sheet of lava and ashes1 which were erupted from the central1 crateriform ridge4 ,judging from the slope of the contemporaneous streams in an adjoining and undisturbed part of the island1 the strata of the ,flagstaff ,hill must have been upturned at least twelve hundred feet1 and probably much more1 for the great truncated dikes on its summit show that it has been largely denuded4 ,the summit of this hill now nearly equals in height the crateriform ridge2 and before having been denuded1 it was probably higher than this ridge1 from which it is separated by a broad and much lower tract of country2 we here1 therefore1 see that the lower extremities of a set of lava-streams have been tilted up to as great a height as1 or perhaps greater height than1 the crater1 down the flanks of which they originally flowed4 ,i believe that dislocations on so grand a scale are extremely rare#e in volcanic districts4 ,the formation of such numbers of dikes in this part of the island shows that the surface must here have been stretched to a quite extraordinary degree3 this stretching1 on the ridge between ,flagstaff and ,barn ,hills1 probably took place subsequently 7though perhaps immediately so7 to the strata being tilted2 for had the strata at that time extended horizontally1 they would in all probability have been fissured and injected transversely1 instead of in the planes of their stratification4 ,although the space between the ,barn and ,flagstaff ,hill presents a distinct anticlinal line extending north and south1 and though most of the dikes range with much regularity in the same line1 nevertheless1 at only a mile due south of the ridge the strata lie undisturbed4 ,hence the disturbing force seems to have acted under a point1 rather than along a line4 ,the manner in which it has acted1 is probably explained by the structure of ,little ,stony-top1 a mountain #b1jjj feet high1 situated a few miles southward of the ,barn2 we there see1 even from a distance1 a dark-coloured1 sharp1 wedge of compact columnar rock1 with the bright-coloured feldspathic strata1 sloping away on each side from its uncovered apex4 ,this wedge1 from which it derives its name of ,stony-top1 consists of a body of rock1 which has been injected whilst liquified into the overlying strata2 and if we may suppose that a similar body of rock lies injected1 beneath the ridge connecting the ,barn and ,flagstaff1 the structure there exhibited would be explained4 #e ,m4 ,constant ,prevost 78,mm4 de la ,soc4 ,golog410 tome ii7 observes that 8les produits volcaniques n'ont que localement et rarement mme drang le sol1 travers lequel ils se sont fait jour40 ,no4 #i ,illustration3 ,prosperous ,hill and ,the ,barn4 ,the double lines represent the basaltic strata2 the single1 the basal submarine strata2 the dotted1 the upper feldspathic strata4 ,turk's ,cap and ,prosperous ,bays4--,prosperous ,hill is a great1 black1 precipitous mountain1 situated two miles and a half south of the ,barn1 and composed1 like it1 of basaltic strata4 ,these rest1 in one part1 on the brown-coloured1 porphyritic beds of the basal series1 and in another part1 on a fissured mass of highly scoriaceous and amygdaloidal rock1 which seems to have formed a small point of eruption beneath the sea1 contemporaneously with the basal series4 ,prosperous ,hill1 like the ,barn1 is traversed by many dikes1 of which the greater number range north and south1 and its strata dip1 at an angle of about #bj degrees1 rather obliquely from the island towards the sea4 ,the space between ,prosperous ,hill and the ,barn1 as represented in figure ,no4 #i1 consists of lofty cliffs1 composed of the lavas of the upper or feldspathic series1 which rest1 though unconformably1 on the basal submarine strata1 as we have seen that they do at ,flagstaff ,hill4 ,differently1 however1 from in that hill1 these upper strata are nearly horizontal1 gently rising towards the interior of the island2 and they are composed of greenish-black1 or more commonly1 pale brown1 compact lavas1 instead of softened and highly coloured matter4 ,these brown-coloured1 compact lavas1 consist almost entirely of small glimmering scales1 or of minute acicular crystals1 of feldspar1 placed close by the side of each other1 and abounding with minute black specks1 apparently of hornblende4 ,the basaltic strata of ,prosperous ,hill project only a little above the level of the gently-sloping1 feldspathic streams1 which wind round and abut against their upturned edges4 ,the inclination of the basaltic strata seems to be too great to have been caused by their having flowed down a slope1 and they must have been tilted into their present position before the eruption of the feldspathic streams4 ,basaltic ring4--,proceeding round the ,island1 the lavas of the upper series1 southward of ,prosperous ,hill1 overhang the sea in lofty precipices4 ,further on1 the headland1 called ,great ,stony-top1 is composed1 as ,i believe1 of basalt2 as is ,long ,range ,point1 on the inland side of which the coloured beds abut4 ,on the southern side of the island1 we see the basaltic strata of the ,south ,barn1 dipping obliquely seaward at a considerable angle2 this headland1 also1 stands a little above the level of the more modern1 feldspathic lavas4 ,further on1 a large space of coast1 on each side of ,sandy ,bay1 has been much denuded1 and there seems to be left only the basal wreck of the great1 central crater4 ,the basaltic strata reappear1 with their seaward dip1 at the foot of the hill1 called ,man-and-,horse2 and thence they are continued along the whole north-western coast to ,sugar-,loaf ,hill1 situated near to the ,flagstaff2 and they everywhere have the same seaward inclination1 and rest1 in some parts at least1 on the lavas of the basal series4 ,we thus see that the circumference of the island is formed by a much-broken ring1 or rather1 a horse-shoe1 of basalt1 open to the south1 and interrupted on the eastern side by many wide breaches4 ,the breadth of this marginal fringe on the north-western side1 where alone it is at all perfect1 appears to vary from a mile to a mile and a half4 ,the basaltic strata1 as well as those of the subjacent basal series1 dip1 with a moderate inclination1 where they have not been subsequently disturbed1 towards the sea4 ,the more broken state of the basaltic ring round the eastern half1 compared with the western half of the island1 is evidently due to the much greater denuding power of the waves on the eastern or windward side1 as is shown by the greater height of the cliffs on that side1 than to leeward4 ,whether the margin of basalt was breached1 before or after the eruption of the lavas of the upper series1 is doubtful2 but as separate portions of the basaltic ring appear to have been tilted before that event1 and from other reasons1 it is more probable1 that some at least of the breaches were first formed4 ,reconstructing in imagination1 as far as is possible1 the ring of basalt1 the internal space or hollow1 which has since been filled up with the matter erupted from the great central crater1 appears to have been of an oval figure1 eight or nine miles in length by about four miles in breadth1 and with its axis directed in a ,n4,e4 and ,s4,w4 line1 coincident with the present longest axis of the island4 ,the central curved ridge4--,this ridge consists1 as before remarked1 of grey feldspathic lavas1 and of red1 brecciated1 argillaceous tuffs1 like the beds of the upper coloured series4 ,the grey lavas contain numerous1 minute1 black1 easily fusible specks2 and but very few large crystals of feldspar4 ,they are generally much softened2 with the exception of this character1 and of being in many parts highly cellular1 they are quite similar to those great sheets of lava which overhang the coast at ,prosperous ,bay4 ,considerable intervals of time appear to have elapsed1 judging from the marks of denudation1 between the formation of the successive beds1 of which this ridge is composed4 ,on the steep northern slope1 ,i observed in several sections a much worn undulating surface of red tuff1 covered by grey1 decomposed1 feldspathic lavas1 with only a thin earthy layer interposed between them4 ,in an adjoining part1 ,i noticed a trap-dike1 four feet wide1 cut off and covered up by the feldspathic lava1 as is represented in figure ,no4 #i4 ,the ridge ends on the eastern side in a hook1 which is not represented clearly enough in any map which ,i have seen2 towards the western end1 it gradually slopes down and divides into several subordinate ridges4 ,the best defined portion between ,diana's ,peak and ,nest ,lodge1 which supports the highest pinnacles in the island varying from #b1jjj to #b1gjj feet1 is rather less than three miles long in a straight line4 ,throughout this space the ridge has a uniform appearance and structure2 its curvature resembles that of the coast-line of a great bay1 being made up of many smaller curves1 all open to the south4 ,the northern and outer side is supported by narrow ridges or buttresses1 which slope down to the adjoining country4 ,the inside is much steeper1 and is almost precipitous2 it is formed of the basset edges of the strata1 which gently decline outwards4 ,along some parts of the inner side1 a little way beneath the summit1 a flat ledge extends1 which imitates in outline the smaller curvatures of the crest4 ,ledges of this kind occur not unfrequently within volcanic craters1 and their formation seems to be due to the sinking down of a level sheet of hardened lava1 the edges of which remain 7like the ice round a pool1 from which the water has been drained7 adhering to the sides4#f #f ,a most remarkable instance of this structure is described in ,ellis 8,polynesian ,researches0 7second edition71 where an admirable drawing is given of the successive ledges or terraces1 on the borders of the immense crater at ,hawaii1 in the ,sandwich ,islands4 ,no4 #aj ,illustration3 ,dike #a--;,grey feldspathic lava4 #b--;,a layer1 one inch in thickness1 of a reddish earthy matter4 #c--;,brecciated1 red1 argillaceous tuff4 ,in some parts1 the ridge is surmounted by a wall or parapet1 perpendicular on both sides4 ,near ,diana's ,peak this wall is extremely narrow4 ,at the ,galapagos ,archipelago ,i observed parapets1 having a quite similar structure and appearance1 surmounting several of the craters2 one1 which ,i more particularly examined1 was composed of glossy1 red scori firmly cemented together2 being externally perpendicular1 and extending round nearly the whole circumference of the crater1 it rendered it almost inaccessible4 ,the ,peak of ,teneriffe and ,cotopaxi1 according to ,humboldt1 are similarly constructed2 he states#g that 8at their summits a circular wall surrounds the crater1 which wall1 at a distance1 has the appearance of a small cylinder placed on a truncated cone4 ,on ,cotopaxi#h this peculiar structure is visible to the naked eye at more than two thousand toises' distance2 and no person has ever reached its crater4 ,on the ,peak of ,teneriffe1 the parapet is so high1 that it would be impossible to reach the caldera1 if on the eastern side there did not exist a breach40 ,the origin of these circular parapets is probably due to the heat or vapours from the crater1 penetrating and hardening the sides to a nearly equal depth1 and afterwards to the mountain being slowly acted on by the weather1 which would leave the hardened part1 projecting in the form of a cylinder or circular parapet4 #g 8,personal ,narrative10 vol4 i1 p4 #aga4 #h ,humboldt's 8,picturesque ,atlas10 folio1 pl4 #aj4 ,from the points of structure in the central ridge1 now enumerated1--namely1 from the convergence towards it of the beds of the upper series1--from the lavas there becoming highly cellular1--from the flat ledge1 extending along its inner and precipitous side1 like that within some still active craters1--from the parapet-like wall on its summit1--and lastly1 from its peculiar curvature1 unlike that of any common line of elevation1 ,i cannot doubt that this curved ridge forms the last remnant of a great crater4 ,in endeavouring1 however1 to trace its former outline1 one is soon baffled2 its western extremity gradually slopes down1 and1 branching into other ridges1 extends to the sea-coast2 the eastern end is more curved1 but it is only a little better defined4 ,some appearances lead me to suppose that the southern wall of the crater joined the present ridge near ,nest ,lodge2 in this case the crater must have been nearly three miles long1 and about a mile and a half in breadth4 ,had the denudation of the ridge and the decomposition of its constituent rocks proceeded a few steps further1 and had this ridge1 like several other parts of the island1 been broken up by great dikes and masses of injected matter1 we should in vain have endeavoured to discover its true nature4 ,even now we have seen that at ,flagstaff ,hill the lower extremity and most distant portion of one sheet of the erupted matter has been upheaved to as great a height as the crater down which it flowed1 and probably even to a greater height4 ,it is interesting thus to trace the steps by which the structure of a volcanic district becomes obscured1 and finally obliterated3 so near to this last stage is ,st4 ,helena1 that ,i believe no one has hitherto suspected that the central ridge or axis of the island is the last wreck of the crater1 whence the most modern volcanic streams were poured forth4 ,the great hollow space or valley southward of the central curved ridge1 across which the half of the crater must once have extended1 is formed of bare1 water-worn hillocks and ridges of red1 yellow1 and brown rocks1 mingled together in chaos-like confusion1 interlaced by dikes1 and without any regular stratification4 ,the chief part consists of red decomposing scori1 associated with various kinds of tuff and yellow argillaceous beds1 full of broken crystals1 those of augite being particularly large4 ,here and there masses of highly cellular and amygdaloidal lavas protrude4 ,from one of the ridges in the midst of the valley1 a conical precipitous hill1 called ,lot1 boldly stands up1 and forms a most singular and conspicuous object4 ,it is composed of phonolite1 divided in one part into great curved lamin1 in another1 into angular concretionary balls1 and in a third part into outwardly radiating columns4 ,at its base the strata of lava1 tuff1 and scori1 dip away on all sides2#i the uncovered portion is #aig feet#aj in height1 and its horizontal section gives an oval figure4 ,the phonolite is of a greenish-grey colour1 and is full of minute acicular crystals of feldspar2 in most parts it has a conchoidal fracture1 and is sonorous1 yet it is crenulated with minute air-cavities4 ,in a ,s4,w4 direction from ,lot1 there are some other remarkable columnar pinnacles1 but of a less regular shape1 namely1 ,lot's ,wife1 and the ,asses' ,ears1 composed of allied kinds of rock4 ,from their flattened shape1 and their relative position to each other1 they are evidently connected on the same line of fissure4 ,it is1 moreover1 remarkable that this same ,n4,e4 and ,s4,w4 line1 joining ,lot and ,lot's ,wife1 if prolonged would intersect ,flagstaff ,hill1 which1 as before stated1 is crossed by numerous dikes running in this direction1 and which has a disturbed structure1 rendering it probable that a great body of once fluid rock lies injected beneath it4 #i ,abich in his 8,views of ,vesuvius0 7plate vi71 has shown the manner in which beds1 under nearly similar circumstances1 are tilted up4 ,the upper beds are more turned up than the lower2 and he accounts for this1 by showing that the lava insinuates itself horizontally between the lower beds4 #aj ,this height is given by ,mr4 ,seale in his ,geognosy of the island4 ,the height of the summit above the level of the sea is said to be #a1ddd feet4 ,in this same great valley there are several other conical masses of injected rock 7one1 ,i observed1 was composed of compact greenstone71 some of which are not connected1 as far as is apparent1 with any line of dike2 whilst others are obviously thus connected4 ,of these dikes1 three or four great lines stretch across the valley in a ,n4,e4 and ,s4,w4 direction1 parallel to that one connecting the ,asses' ,ears1 ,lot's ,wife1 and probably ,lot4 ,the number of these masses of injected rock is a remarkable feature in the geology of ,st4 ,helena4 ,besides those just mentioned1 and the hypothetical one beneath ,flagstaff ,hill1 there is ,little ,stony-top and others1 as ,i have reason to believe1 at the ,man-and-,horse1 and at ,high ,hill4 ,most of these masses1 if not all of them1 have been injected subsequently to the last volcanic eruptions from the central crater4 ,the formation of conical bosses of rock on lines of fissure1 the walls of which are in most cases parallel1 may probably be attributed to inequalities in the tension1 causing small transverse fissures1 and at these points of intersection the edges of the strata would naturally yield1 and be easily turned upwards4 ,finally1 ,i may remark1 that hills of phonolite everywhere are apt#aa to assume singular and even grotesque shapes1 like that of ,lot3 the peak at ,fernando ,noronha offers an instance2 at ,st4 ,jago1 however1 the cones of phonolite1 though tapering1 have a regular form4 ,supposing1 as seems probable1 that all such hillocks or obelisks have originally been injected1 whilst liquified1 into a mould formed by yielding strata1 as certainly has been the case with ,lot1 how are we to account for the frequent abruptness and singularity of their outlines1 compared with similarly injected masses of greenstone and basalt8 ,can it be due to a less perfect degree of fluidity1 which is generally supposed to be characteristic of the allied trachytic lavas8 #aa ,d',aubuisson1 in his 8,trait de ,gognosie0 7tome ii1 p4 #edj7 particularly remarks that this is the case4 ,superficial deposits4--,soft calcareous sandstone occurs in extensive1 though thin1 superficial beds1 both on the northern and southern shores of the island4 ,it consists of very minute1 equal-sized1 rounded particles of shells1 and other organic bodies1 which partially retain their yellow1 brown1 and pink colours1 and occasionally1 though very rarely1 present an obscure trace of their original external forms4 ,i in vain endeavoured to find a single unrolled fragment of a shell4 ,the colour of the particles is the most obvious character by which their origin can be recognised1 the tints being affected 7and an odour produced7 by a moderate heat1 in the same manner as in fresh shells4 ,the particles are cemented together1 and are mingled with some earthy matter3 the purest masses1 according to ,beatson1 contain #gj per cent of carbonate of lime4 ,the beds1 varying in thickness from two or three feet to fifteen feet1 coat the surface of the ground2 they generally lie on that side of the valley which is protected from the wind1 and they occur at the height of several hundred feet above the level of the sea4 ,their position is the same which sand1 if now drifted by the trade-wind1 would occupy2 and no doubt they thus originated1 which explains the equal size and minuteness of the particles1 and likewise the entire absence of whole shells1 or even of moderately-sized fragments4 ,it is remarkable that at the present day there are no shelly beaches on any part of the coast1 whence calcareous dust could be drifted and winnowed2 we must1 therefore1 look back to a former period when before the land was worn into the present great precipices1 a shelving coast1 like that of ,ascension1 was favourable to the accumulation of shelly detritus4 ,some of the beds of this limestone are between six hundred and seven hundred feet above the sea2 but part of this height may possibly be due to an elevation of the land1 subsequent to the accumulation of the calcareous sand4 ,the percolation of rain-water has consolidated parts of these beds into a solid rock1 and has formed masses of dark brown1 stalagmitic limestone4 ,at the ,sugar-,loaf quarry1 fragments of rock on the adjoining slopes#ab have been thickly coated by successive fine layers of calcareous matter4 ,it is singular1 that many of these pebbles have their entire surfaces coated1 without any point of contact having been left uncovered2 hence1 these pebbles must have been lifted up by the slow deposition between them of the successive films of carbonate of lime4 ,masses of white1 finely oolitic rock are attached to the outside of some of these coated pebbles4 ,von ,buch has described a compact limestone at ,lanzarote1 which seems perfectly to resemble the stalagmitic deposition just mentioned3 it coats pebbles1 and in parts is finely oolitic3 it forms a far-extended layer1 from one inch to two or three feet in thickness1 and it occurs at the height of #hjj feet above the sea1 but only on that side of the island exposed to the violent north-western winds4 ,von ,buch remarks1#ac that it is not found in hollows1 but only on the unbroken and inclined surfaces of the mountain4 ,he believes1 that it has been deposited by the spray which is borne over the whole island by these violent winds4 ,it appears1 however1 to me much more probable that it has been formed1 as at ,st4 ,helena1 by the percolation of water through finely comminuted shells3 for when sand is blown on a much-exposed coast1 it always tends to accumulate on broad1 even surfaces1 which offer a uniform resistance to the winds4 ,at the neighbouring island1 moreover1 of ,feurteventura1#ad there is an earthy limestone1 which1 according to ,von ,buch1 is quite similar to specimens which he has seen from ,st4 ,helena1 and which he believes to have been formed by the drifting of shelly detritus4 #ab ,in the earthy detritus on several parts of this hill1 irregular masses of very impure1 crystallised sulphate of lime occur4 ,as this substance is now being abundantly deposited by the surf at ,ascension1 it is possible that these masses may thus have originated2 but if so1 it must have been at a period when the land stood at a much lower level4 ,this earthy selenite is now found at a height of between six hundred and seven hundred feet4 #ac 8,description des ,isles ,canaries10 p4 #bic4 #ad ,idem1 pp4 #cad and #cgd4 ,the upper beds of the limestone1 at the above-mentioned quarry on the ,sugar-,loaf ,hill1 are softer1 finer-grained and less pure1 than the lower beds4 ,they abound with fragments of land-shells1 and with some perfect ones2 they contain1 also1 the bones of birds1 and the large eggs1#ae apparently of water-fowl4 ,it is probable that these upper beds remained long in an unconsolidated form1 during which time1 these terrestrial productions were embedded4 ,mr4 ,g4 ,r4 ,sowerby has kindly examined three species of land-shells1 which ,i procured from this bed1 and has described them in detail4 ,one of them is a ,succinea1 identical with a species now living abundantly on the island2 the two others1 namely1 ,cochlogena fossilis and ,helix biplicata1 are not known in a recent state3 the latter species was also found in another and different locality1 associated with a species of ,cochlogena which is undoubtedly extinct4 #ae ,colonel ,wilkes1 in a catalogue presented with some specimens to the ,geological ,society1 states that as many as ten eggs were found by one person4 ,dr4 ,buckland has remarked 78,geolog4 ,trans410 vol4 v1 p4 #dgd7 on these eggs4 ,beds of extinct land-shells4--,land-shells1 all of which appear to be species now extinct1 occur embedded in earth1 in several parts of the island4 ,the greater number have been found at a considerable height on ,flagstaff ,hill4 ,on the ,n4,w4 side of this hill1 a rain-channel exposes a section of about twenty feet in thickness1 of which the upper part consists of black vegetable mould1 evidently washed down from the heights above1 and the lower part of less black earth1 abounding with young and old shells1 and with their fragments3 part of this earth is slightly consolidated by calcareous matter1 apparently due to the partial decomposition of some of the shells4 ,mr4 ,seale1 an intelligent resident1 who first called attention to these shells1 gave me a large collection from another locality1 where the shells appear to have been embedded in very black earth4 ,mr4 ,g4 ,r4 ,sowerby has examined these shells1 and has described them4 ,there are seven species1 namely1 one ,cochlogena1 two species of the genus ,cochlicopa1 and four of ,helix2 none of these are known in a recent state1 or have been found in any other country4 ,the smaller species were picked out of the inside of the large shells of the ,cochlogena aurisvulpina4 ,this last-mentioned species is in many respects a very singular one2 it was classed1 even by ,lamarck1 in a marine genus1 and having thus been mistaken for a sea-shell1 and the smaller accompanying species having been overlooked1 the exact localities where it was found have been measured1 and the elevation of this island thus deduced6 ,it is very remarkable that all the shells of this species found by me in one spot1 form a distinct variety1 as described by ,mr4 ,sowerby1 from those procured from another locality by ,mr4 ,seale4 ,as this ,cochlogena is a large and conspicuous shell1 ,i particularly inquired from several intelligent countrymen whether they had ever seen it alive2 they all assured me that they had not1 and they would not even believe that it was a land animal3 ,mr4 ,seale1 moreover1 who was a collector of shells all his life at ,st4 ,helena1 never met with it alive4 ,possibly some of the smaller species may turn out to be yet living kinds2 but1 on the other hand1 the two land-shells which are now living on the island in great numbers1 do not occur embedded1 as far as is yet known1 with the extinct species4 ,i have shown in my 8,journal10#af that the extinction of these land-shells possibly may not be an ancient event2 as a great change took place in the state of the island about one hundred and twenty years ago1 when the old trees died1 and were not replaced by young ones1 these being destroyed by the goats and hogs1 which had run wild in numbers1 from the year #aejb4 ,mr4 ,seale states1 that on ,flagstaff ,hill1 where we have seen that the embedded land-shells are especially numerous1 traces are everywhere discoverable1 which plainly indicate that it was once thickly clothed with trees2 at present not even a bush grows there4 ,the thick bed of black vegetable mould which covers the shell-bed1 on the flanks of this hill1 was probably washed down from the upper part1 as soon as the trees perished1 and the shelter afforded by them was lost4 #af 8,journal of ,researches10 p4 #ehb4 ,elevation of the land4--,seeing that the lavas of the basal series1 which are of submarine origin1 are raised above the level of the sea1 and at some places to the height of many hundred feet1 ,i looked out for superficial signs of the elevation of the land4 ,the bottoms of some of the gorges1 which descend to the coast1 are filled up to the depth of about a hundred feet1 by rudely divided layers of sand1 muddy clay1 and fragmentary masses2 in these beds1 ,mr4 ,seale has found the bones of the tropic-bird and of the albatross2 the former now rarely1 and the latter never visiting the island4 ,from the difference between these layers1 and the sloping piles of detritus which rest on them1 ,i suspect that they were deposited1 when the gorges stood beneath the sea4 ,mr4 ,seale1 moreover1 has shown that some of the fissure-like gorges#ag become1 with a concave outline1 gradually rather wider at the bottom than at the top2 and this peculiar structure was probably caused by the wearing action of the sea1 when it entered the lower part of these gorges4 ,at greater heights1 the evidence of the rise of the land is even less clear3 nevertheless1 in a bay-like depression on the table-land behind ,prosperous ,bay1 at the height of about a thousand feet1 there are flat-topped masses of rock1 which it is scarcely conceivable1 could have been insulated from the surrounding and similar strata1 by any other agency than the denuding action of a sea-beach4 ,much denudation1 indeed1 has been effected at great elevations1 which it would not be easy to explain by any other means3 thus1 the flat summit of the ,barn1 which is #b1jjj feet high1 presents1 according to ,mr4 ,seale1 a perfect network of truncated dikes2 on hills like the ,flagstaff1 formed of soft rock1 we might suppose that the dikes had been worn down and cut off by meteoric agency1 but we can hardly suppose this possible with the hard1 basaltic strata of the ,barn4 #ag ,a fissure-like gorge1 near ,stony-top1 is said by ,mr4 ,seale to be #hdj feet deep1 and only #aae feet in width4 ,coast denudation4--,the enormous cliffs1 in many parts between one and two thousand feet in height1 with which this prison-like island is surrounded1 with the exception of only a few places1 where narrow valleys descend to the coast1 is the most striking feature in its scenery4 ,we have seen that portions of the basaltic ring1 two or three miles in length by one or two miles in breadth1 and from one to two thousand feet in height1 have been wholly removed4 ,there are1 also1 ledges and banks of rock1 rising out of profoundly deep water1 and distant from the present coast between three and four miles1 which1 according to ,mr4 ,seale1 can be traced to the shore1 and are found to be the continuations of certain well-known great dikes4 ,the swell of the ,atlantic ,ocean has obviously been the active power in forming these cliffs2 and it is interesting to observe that the lesser1 though still great1 height of the cliffs on the leeward and partially protected side of the island 7extending from the ,sugar-,loaf ,hill to ,south ,west ,point71 corresponds with the lesser degree of exposure4 ,when reflecting on the comparatively low coasts of many volcanic islands1 which also stand exposed in the open ocean1 and are apparently of considerable antiquity1 the mind recoils from an attempt to grasp the number of centuries of exposure1 necessary to have ground into mud and to have dispersed the enormous cubic mass of hard rock which has been pared off the circumference of this island4 ,the contrast in the superficial state of ,st4 ,helena1 compared with the nearest island1 namely1 ,ascension1 is very striking4 ,at ,ascension1 the surfaces of the lava-streams are glossy1 as if just poured forth1 their boundaries are well defined1 and they can often be traced to perfect craters1 whence they were erupted2 in the course of many long walks1 ,i did not observe a single dike2 and the coast round nearly the entire circumference is low1 and has been eaten back 7though too much stress must not be placed on this fact1 as the island may have been subsiding7 into a little wall only from ten to thirty feet high4 ,yet during the #cdj years1 since ,ascension has been known1 not even the feeblest signs of volcanic action have been recorded4#ah ,on the other hand1 at ,st4 ,helena1 the course of no one stream of lava can be traced1 either by the state of its boundaries or of its superficies2 the mere wreck of one great crater is left2 not the valleys only1 but the surfaces of some of the highest hills1 are interlaced by worn-down dikes1 and1 in many places1 the denuded summits of great cones of injected rock stand exposed and naked2 lastly1 as we have seen1 the entire circuit of the island has been deeply worn back into the grandest precipices4 #ah ,in the ,nautical ,magazine for #ahce1 p4 #fdb1 and for #ahch1 p4 #cfa1 and in the 8,comptes ,rendus10 ,april #ahch1 accounts are given of a series of volcanic phenomena--earthquakes--troubled water--floating scori and columns of smoke--which have been observed at intervals since the middle of the last century1 in a space of open sea between longitudes #bj and #bb west1 about half a degree south of the equator4 ,these facts seem to show1 that an island or an archipelago is in process of formation in the middle of the ,atlantic3 a line joining ,st4 ,helena and ,ascension1 prolonged1 intersects this slowly nascent focus of volcanic action4 ,craters of ,elevation4 ,there is much resemblance in structure and in geological history between ,st4 ,helena1 ,st4 ,jago1 and ,mauritius4 ,all three islands are bounded 7at least in the parts which ,i was able to examine7 by a ring of basaltic mountains1 now much broken1 but evidently once continuous4 ,these mountains have1 or apparently once had1 their escarpments steep towards the interior of the island1 and their strata dip outwards4 ,i was able to ascertain1 only in a few cases1 the inclination of the beds2 nor was this easy1 for the stratification was generally obscure1 except when viewed from a distance4 ,i feel1 however1 little doubt that1 according to the researches of ,m4 ,elie de ,beaumont1 their average inclination is greater than that which they could have acquired1 considering their thickness and compactness1 by flowing down a sloping surface4 ,at ,st4 ,helena1 and at ,st4 ,jago1 the basaltic strata rest on older and probably submarine beds of different composition4 ,at all three islands1 deluges of more recent lavas have flowed from the centre of the island1 towards and between the basaltic mountains2 and at ,st4 ,helena the central platform has been filled up by them4 ,all three islands have been raised in mass4 ,at ,mauritius the sea1 within a late geological period1 must have reached to the foot of the basaltic mountains1 as it now does at ,st4 ,helena2 and at ,st4 ,jago it is cutting back the intermediate plain towards them4 ,in these three islands1 but especially at ,st4 ,jago and at ,mauritius1 when1 standing on the summit of one of the old basaltic mountains1 one looks in vain towards the centre of the island1--the point towards which the strata beneath one's feet1 and of the mountains on each side1 rudely converge1--for a source whence these strata could have been erupted2 but one sees only a vast hollow platform stretched beneath1 or piles of matter of more recent origin4 ,these basaltic mountains come1 ,i presume1 into the class of ,craters of elevation3 it is immaterial whether the rings were ever completely formed1 for the portions which now exist have so uniform a structure1 that1 if they do not form fragments of true craters1 they cannot be classed with ordinary lines of elevation4 ,with respect to their origin1 after having read the works of ,mr4 ,lyell1#ai and of ,,mm4 ,c4 ,prevost and ,virlet1 ,i cannot believe that the great central hollows have been formed by a simple dome-shaped elevation1 and the consequent arching of the strata4 ,on the other hand1 ,i have very great difficulty in admitting that these basaltic mountains are merely the basal fragments of great volcanoes1 of which the summits have either been blown off1 or more probably swallowed up by subsidence4 ,these rings are1 in some instances1 so immense1 as at ,st4 ,jago and at ,mauritius1 and their occurrence is so frequent1 that ,i can hardly persuade myself to adopt this explanation4 ,moreover1 ,i suspect that the following circumstances1 from their frequent concurrence1 are someway connected together1--a connection not implied in either of the above views3 namely1 first1 the broken state of the ring2 showing that the now detached portions have been exposed to great denudation1 and in some cases1 perhaps1 rendering it probable that the ring never was entire2 secondly1 the great amount of matter erupted from the central area after or during the formation of the ring2 and thirdly1 the elevation of the district in mass4 ,as far as relates to the inclination of the strata being greater than that which the basal fragments of ordinary volcanoes would naturally possess1 ,i can readily believe that this inclination might have been slowly acquired by that amount of elevation1 of which1 according to ,m4 ,elie de ,beaumont1 the numerous upfilled fissures or dikes are the evidence and the measure1--a view equally novel and important1 which we owe to the researches of that geologist on ,mount ,etna4 #ai 8,principles of ,geology0 7fifth edit471 vol4 ii1 p4 #aga4 ,a conjecture1 including the above circumstances1 occurred to me1 when1-- with my mind fully convinced1 from the phenomena of #ahce in ,south ,america1#bj that the forces which eject matter from volcanic orifices and raise continents in mass are identical1--,i viewed that part of the coast of ,st4 ,jago1 where the horizontally upraised1 calcareous stratum dips into the sea1 directly beneath a cone of subsequently erupted lava4 ,the conjecture is that1 during the slow elevation of a volcanic district or island1 in the centre of which one or more orifices continue open1 and thus relieve the subterranean forces1 the borders are elevated more than the central area2 and that the portions thus upraised do not slope gently into the central1 less elevated area1 as does the calcareous stratum under the cone at ,st4 ,jago1 and as does a large part of the circumference of ,iceland1#ba but that they are separated from it by curved faults4 ,we might expect1 from what we see along ordinary faults1 that the strata on the upraised side1 already dipping outwards from their original formation as lava-streams1 would be tilted from the line of fault1 and thus have their inclination increased4 ,according to this hypothesis1 which ,i am tempted to extend only to some few cases1 it is not probable that the ring would ever be formed quite perfect2 and from the elevation being slow1 the upraised portions would generally be exposed to much denudation1 and hence the ring become broken2 we might also expect to find occasional inequalities in the dip of the upraised masses1 as is the case at ,st4 ,jago4 ,by this hypothesis the elevation of the districts in mass1 and the flowing of deluges of lava from the central platforms1 are likewise connected together4 ,on this view the marginal basaltic mountains of the three foregoing islands might still be considered as forming 8,craters of elevation20 the kind of elevation implied having been slow1 and the central hollow or platform having been formed1 not by the arching of the surface1 but simply by that part having been upraised to a less height4 #bj ,i have given a detailed account of these phenomena1 in a paper read before the ,geological ,society in ,march #ahch4 ,at the instant of time1 when an immense area was convulsed and a large tract elevated1 the districts immediately surrounding several of the great vents in the ,cordillera remained quiescent2 the subterranean forces being apparently relieved by the eruptions1 which then recommenced with great violence4 ,an event of somewhat the same kind1 but on an infinitely smaller scale1 appears to have taken place1 according to ,abich 78,views of ,vesuvius10 plates i and ix71 within the great crater of ,vesuvius1 where a platform on one side of a fissure was raised in mass twenty feet1 whilst on the other side1 a train of small volcanoes burst forth in eruption4 #ba ,it appears1 from information communicated to me in the most obliging manner by ,m4 ,e4 ,robert1 that the circumferential parts of ,iceland1 which are composed of ancient basaltic strata alternating with tuff1 dip inland1 thus forming a gigantic saucer4 ,m4 ,robert found that this was the case1 with a few and quite local exceptions1 for a space of coast several hundred miles in length4 ,i find this statement corroborated1 as far as regards one place1 by ,mackenzie in his 8,travels0 7p4 #cgg71 and in another place by some ,,ms4 notes kindly lent me by ,dr4 ,holland4 ,the coast is deeply indented by creeks1 at the head of which the land is generally low4 ,m4 ,robert informs me1 that the inwardly dipping strata appear to extend as far as this line1 and that their inclination usually corresponds with the slope of the surface1 from the high coast-mountains to the low land at the head of these creeks4 ,in the section described by ,sir ,g4 ,mackenzie1 the dip is #abj4 ,the interior parts of the island chiefly consist1 as far as is known1 of recently erupted matter4 ,the great size1 however1 of ,iceland1 equalling the bulkiest part of ,england1 ought perhaps to exclude it from the class of islands we have been considering2 but ,i cannot avoid suspecting that if the coast-mountains1 instead of gently sloping into the less elevated central area1 had been separated from it by irregularly curved faults1 the strata would have been tilted seaward1 and a 8,crater of elevation10 like that of ,st4 ,jago or that of ,mauritius1 but of much vaster dimensions1 would have been formed4 ,i will only further remark1 that the frequent occurrence of extensive lakes at the foot of large volcanoes1 and the frequent association of volcanic and fresh-water strata1 seem to indicate that the areas around volcanoes are apt to be depressed beneath the level of the adjoining country1 either from having been less elevated1 or from the effects of subsidence4 ,chapter ,v ,,galapagos ,,archipelago4 ,chatham ,island4--,craters composed of a peculiar kind of tuff4--,small basaltic craters1 with hollows at their bases4--,albemarle ,island1 fluid lavas1 their composition4--,craters of tuff1 inclination of their exterior diverging strata1 and structure of their interior converging strata4--,james ,island1 segment of a small basaltic crater2 fluidity and composition of its lava-streams1 and of its ejected fragments4--,concluding remarks on the craters of tuff1 and on the breached condition of their southern sides4--,mineralogical composition of the rocks of the archipelago4--,elevation of the land4 ,direction of the fissures of eruption4 ,this archipelago is situated under the equator1 at a distance of between five and six hundred miles from the west coast of ,south ,america4 ,it consists of five principal islands1 and of several small ones1 which together are equal in area1#a but not in extent of land1 to ,sicily1 conjointly with the ,ionian ,islands4 ,they are all volcanic3 on two1 craters have been seen in eruption1 and on several of the other islands1 streams of lava have a recent appearance4 ,the larger islands are chiefly composed of solid rock1 and they rise with a tame outline to a height of between one and four thousand feet4 ,they are sometimes1 but not generally1 surmounted by one principal orifice4 ,the craters vary in size from mere spiracles to huge caldrons several miles in circumference2 they are extraordinarily numerous1 so that ,i should think1 if enumerated1 they would be found to exceed two thousand2 they are formed either of scori and lava1 or of a brown-coloured tuff2 and these latter craters are in several respects remarkable4 ,the whole group was surveyed by the officers of the ,beagle4 ,i visited myself four of the principal islands1 and received specimens from all the others4 ,under the head of the different islands ,i will describe only that which appears to me deserving of attention4 #a ,i exclude from this measurement1 the small volcanic islands of ,culpepper and ,wenman1 lying seventy miles northward of the group4 ,craters were visible on all the islands of the group1 except on ,towers ,island1 which is one of the lowest2 this island is1 however1 formed of volcanic rocks4 ,no4 #aa ,illustration3 ,galapagos ,archipelago4 ,galapagos ,archipelago ,,chatham ,,island4 ,craters composed of a singular kind of tuff4--,towards the eastern end of this island there occur two craters composed of two kinds of tuff2 one kind being friable1 like slightly consolidated ashes2 and the other compact1 and of a different nature from anything which ,i have met with described4 ,this latter substance1 where it is best characterised1 is of a yellowish-brown colour1 translucent1 and with a lustre somewhat resembling resin2 it is brittle1 with an angular1 rough1 and very irregular fracture1 sometimes1 however1 being slightly granular1 and even obscurely crystalline3 it can readily be scratched with a knife1 yet some points are hard enough just to mark common glass2 it fuses with ease into a blackish-green glass4 ,the mass contains numerous broken crystals of olivine and augite1 and small particles of black and brown scori2 it is often traversed by thin seams of calcareous matter4 ,it generally affects a nodular or concretionary structure4 ,in a hand specimen1 this substance would certainly be mistaken for a pale and peculiar variety of pitchstone2 but when seen in mass its stratification1 and the numerous layers of fragments of basalt1 both angular and rounded1 at once render its subaqueous origin evident4 ,an examination of a series of specimens shows that this resin-like substance results from a chemical change on small particles of pale and dark-coloured scoriaceous rocks2 and this change could be distinctly traced in different stages round the edges of even the same particle4 ,the position near the coast of all the craters composed of this kind of tuff or peperino1 and their breached condition1 renders it probable that they were all formed when standing immersed in the sea2 considering this circumstance1 together with the remarkable absence of large beds of ashes in the whole archipelago1 ,i think it highly probable that much the greater part of the tuff has originated from the trituration of fragments of the grey1 basaltic lavas in the mouths of craters standing in the sea4 ,it may be asked whether the heated water within these craters has produced this singular change in the small scoriaceous particles and given to them their translucent1 resin-like fracture4 ,or has the associated lime played any part in this change8 ,i ask these questions from having found at ,st4 ,jago1 in the ,cape de ,verde ,islands1 that where a great stream of molten lava has flowed over a calcareous bottom into the sea1 the outermost film1 which in other parts resembles pitchstone1 is changed1 apparently by its contact with the carbonate of lime1 into a resin-like substance1 precisely like the best characterised specimens of the tuff from this archipelago4#b #b ,the concretions containing lime1 which ,i have described at ,ascension1 as formed in a bed of ashes1 present some degree of resemblance to this substance1 but they have not a resinous fracture4 ,at ,st4 ,helena1 also1 ,i found veins of a somewhat similar1 compact1 but non-resinous substance1 occurring in a bed of pumiceous ashes1 apparently free from calcareous matter3 in neither of these cases could heat have acted4 ,to return to the two craters3 one of them stands at the distance of a league from the coast1 the intervening tract consisting of a calcareous tuff1 apparently of submarine origin4 ,this crater consists of a circle of hills some of which stand quite detached1 but all have a very regular1 qu-qu versal dip1 at an inclination of between thirty and forty degrees4 ,the lower beds1 to the thickness of several hundred feet1 consist of the resin-like stone1 with embedded fragments of lava4 ,the upper beds1 which are between thirty and forty feet in thickness1 are composed of a thinly stratified1 fine-grained1 harsh1 friable1 brown-coloured tuff1 or peperino4#c ,a central mass without any stratification1 which must formerly have occupied the hollow of the crater1 but is now attached only to a few of the circumferential hills1 consists of a tuff1 intermediate in character between that with a resin-like1 and that with an earthy fracture4 ,this mass contains white calcareous matter in small patches4 ,the second crater 7#ebj feet in height7 must have existed until the eruption of a recent1 great stream of lava1 as a separate islet2 a fine section1 worn by the sea1 shows a grand funnel-shaped mass of basalt1 surrounded by steep1 sloping flanks of tuff1 having in parts an earthy1 and in others a semi-resinous fracture4 ,the tuff is traversed by several broad1 vertical dikes1 with smooth and parallel sides1 which ,i did not doubt were formed of basalt1 until ,i actually broke off fragments4 ,these dikes1 however1 consist of tuff like that of the surrounding strata1 but more compact1 and with a smoother fracture2 hence we must conclude1 that fissures were formed and filled up with the finer mud or tuff from the crater1 before its interior was occupied1 as it now is1 by a solidified pool of basalt4 ,other fissures have been subsequently formed1 parallel to these singular dikes1 and are merely filled with loose rubbish4 ,the change from ordinary scoriaceous particles to the substance with a semi-resinous fracture1 could be clearly followed in portions of the compact tuff of these dikes4 #c ,those geologists who restrict the term of 8tuff0 to ashes of a white colour1 resulting from the attrition of feldspathic lavas1 would call these brown-coloured strata 8peperino40 ,no4 #ab ,illustration3 ,the ,kicker ,rock4 ,the ,kicker ,rock1 #djj feet high4 ,at the distance of a few miles from these two craters1 stands the ,kicker ,rock1 or islet1 remarkable from its singular form4 ,it is unstratified1 and is composed of compact tuff1 in parts having the resin-like fracture4 ,it is probable that this amorphous mass1 like that similar mass in the case first described1 once filled up the central hollow of a crater1 and that its flanks1 or sloping walls1 have since been worn quite away by the sea1 in which it stands exposed4 ,small basaltic craters4--,a bare1 undulating tract1 at the eastern end of ,chatham ,island1 is remarkable from the number1 proximity1 and form of the small basaltic craters with which it is studded4 ,they consist1 either of a mere conical pile1 or1 but less commonly1 of a circle1 of black and red1 glossy scori1 partially cemented together4 ,they vary in diameter from thirty to one hundred and fifty yards1 and rise from about fifty to one hundred feet above the level of the surrounding plain4 ,from one small eminence1 ,i counted sixty of these craters1 all of which were within a third of a mile from each other1 and many were much closer4 ,i measured the distance between two very small craters1 and found that it was only thirty yards from the summit-rim of one to the rim of the other4 ,small streams of black1 basaltic lava1 containing olivine and much glassy feldspar1 have flowed from many1 but not from all of these craters4 ,the surfaces of the more recent streams were exceedingly rugged1 and were crossed by great fissures2 the older streams were only a little less rugged2 and they were all blended and mingled together in complete confusion4 ,the different growth1 however1 of the trees on the streams1 often plainly marked their different ages4 ,had it not been for this latter character1 the streams could in few cases have been distinguished2 and1 consequently1 this wide undulatory tract might have 7as probably many tracts have7 been erroneously considered as formed by one great deluge of lava1 instead of by a multitude of small streams1 erupted from many small orifices4 ,in several parts of this tract1 and especially at the base of the small craters1 there are circular pits1 with perpendicular sides1 from twenty to forty feet deep4 ,at the foot of one small crater1 there were three of these pits4 ,they have probably been formed1 by the falling in of the roofs of small caverns4#d ,in other parts1 there are mammiform hillocks1 which resemble great bubbles of lava1 with their summits fissured by irregular cracks1 which appeared1 upon entering them1 to be very deep2 lava has not flowed from these hillocks4 ,there are1 also1 other very regular1 mammiform hillocks1 composed of stratified lava1 and surmounted by circular1 steep-sided hollows1 which1 ,i suppose have been formed by a body of gas1 first1 arching the strata into one of the bubble-like hillocks1 and then1 blowing off its summit4 ,these several kinds of hillocks and pits1 as well as the numerous1 small1 scoriaceous craters1 all show that this tract has been penetrated1 almost like a sieve1 by the passage of heated vapours4 ,the more regular hillocks could only have been heaved up1 whilst the lava was in a softened state4#e #d 7,m4 ,elie de ,beaumont has described 78,mm4 pour servir10 etc41 tome iv1 p4 #aac7 many 8petits cirques d'eboulement0 on ,etna1 of some of which the origin is historically known4 #e ,sir ,g4 ,mackenzie 78,travels in ,iceland10 pp4 #chi to #cib7 has described a plain of lava at the foot of ,hecla1 everywhere heaved up into great bubbles or blisters4 ,sir ,george states that this cavernous lava composes the uppermost stratum2 and the same fact is affirmed by ,von ,buch 78,descript4 des ,isles ,canaries10 p4 #aei71 with respect to the basaltic stream near ,rialejo1 in ,teneriffe4 ,it appears singular that it should be the upper streams that are chiefly cavernous1 for one sees no reason why the upper and lower should not have been equally affected at different times2--have the inferior streams flowed beneath the pressure of the sea1 and thus been flattened1 after the passage through them1 of bodies of gas8 ,,albemarle ,,island4--,this island consists of five1 great1 flat-topped craters1 which1 together with the one on the adjoining island of ,narborough1 singularly resemble each other1 in form and height4 ,the southern one is #d1gjj feet high1 two others are #c1gbj feet1 a third only #ej feet higher1 and the remaining ones apparently of nearly the same height4 ,three of these are situated on one line1 and their craters appear elongated in nearly the same direction4 ,the northern crater1 which is not the largest1 was found by the triangulation to measure1 externally1 no less than three miles and one-eighth of a mile in diameter4 ,over the lips of these great1 broad caldrons1 and from little orifices near their summits1 deluges of black lava have flowed down their naked sides4 ,fluidity of different lavas4--,near ,tagus or ,banks' ,cove1 ,i examined one of these great streams of lava1 which is remarkable from the evidence of its former high degree of fluidity1 especially when its composition is considered4 ,near the sea-coast this stream is several miles in width4 ,it consists of a black1 compact base1 easily fusible into a black bead1 with angular and not very numerous air-cells1 and thickly studded with large1 fractured crystals of glassy albite1#f varying from the tenth of an inch to half an inch in diameter4 ,this lava1 although at first sight appearing eminently porphyritic1 cannot properly be considered so1 for the crystals have evidently been enveloped1 rounded1 and penetrated by the lava1 like fragments of foreign rock in a trap-dike4 ,this was very clear in some specimens of a similar lava1 from ,abingdon ,island1 in which the only difference was1 that the vesicles were spherical and more numerous4 ,the albite in these lavas is in a similar condition with the leucite of ,vesuvius1 and with the olivine1 described by ,von ,buch1#g as projecting in great balls from the basalt of ,lanzarote4 ,besides the albite1 this lava contains scattered grains of a green mineral1 with no distinct cleavage1 and closely resembling olivine2#h but as it fuses easily into a green glass1 it belongs probably to the augitic family3 at ,james ,island1 however1 a similar lava contained true olivine4 ,i obtained specimens from the actual surface1 and from a depth of four feet1 but they differed in no respect4 ,the high degree of fluidity of this lava-stream was at once evident1 from its smooth and gently sloping surface1 from the manner in which the main stream was divided by small inequalities into little rills1 and especially from the manner in which its edges1 far below its source1 and where it must have been in some degree cooled1 thinned out to almost nothing2 the actual margin consisting of loose fragments1 few of which were larger than a man's head4 ,the contrast between this margin1 and the steep walls1 above twenty feet high1 bounding many of the basaltic streams at ,ascension1 is very remarkable4 ,it has generally been supposed that lavas abounding with large crystals1 and including angular vesicles1#i have possessed little fluidity2 but we see that the case has been very different at ,albemarle ,island4 ,the degree of fluidity in different lavas1 does not seem to correspond with any apparent corresponding amount of difference in their composition3 at ,chatham ,island1 some streams1 containing much glassy albite and some olivine1 are so rugged1 that they may be compared to a sea frozen during a storm2 whilst the great stream at ,albemarle ,island is almost as smooth as a lake when ruffled by a breeze4 ,at ,james ,island1 black basaltic lava1 abounding with small grains of olivine1 presents an intermediate degree of roughness2 its surface being glossy1 and the detached fragments resembling1 in a very singular manner1 folds of drapery1 cables1 and pieces of the bark of trees4#aj #f ,in the ,cordillera of ,chile1 ,i have seen lava very closely resembling this variety at the ,galapagos ,archipelago4 ,it contained1 however1 besides the albite1 well-formed crystals of augite1 and the base 7perhaps in consequence of the aggregation of the augitic particles7 was a shade lighter in colour4 ,i may here remark1 that in all these cases1 ,i call the feldspathic crystals1 albite1 from their cleavage-planes 7as measured by the reflecting goniometer7 corresponding with those of that mineral4 ,as1 however1 other species of this genus have lately been discovered to cleave in nearly the same planes with albite1 this determination must be considered as only provisional4 ,i examined the crystals in the lavas of many different parts of the ,galapagos group1 and ,i found that none of them1 with the exception of some crystals from one part of ,james ,island1 cleaved in the direction of orthite or potash-feldspar4 #g 8,description des ,isles ,canaries10 p4 #bie4 #h ,humboldt mentions that he mistook a green augitic mineral1 occurring in the volcanic rocks of the ,cordillera of ,quito1 for olivine4 #i ,the irregular and angular form of the vesicles is probably caused by the unequal yielding of a mass composed1 in almost equal proportion1 of solid crystals and of a viscid base4 ,it certainly seems a general circumstance1 as might have been expected1 that in lava1 which has possessed a high degree of fluidity1 as well as an even-sized grain1 the vesicles are internally smooth and spherical4 #aj ,a specimen of basaltic lava1 with a few small broken crystals of albite1 given me by one of the officers1 is perhaps worthy of description4 ,it consists of cylindrical ramifications1 some of which are only the twentieth of an inch in diameter1 and are drawn out into the sharpest points4 ,the mass has not been formed like a stalactite1 for the points terminate both upwards and downwards4 ,globules1 only the fortieth of an inch in diameter1 have dropped from some of the points1 and adhere to the adjoining branches4 ,the lava is vesicular1 but the vesicles never reach the surface of the branches1 which are smooth and glossy4 ,as it is generally supposed that vesicles are always elongated in the direction of the movement of the fluid mass1 ,i may observe1 that in these cylindrical branches1 which vary from a quarter to only the twentieth of an inch in diameter1 every air-cell is spherical4 ,craters of tuff4--,about a mile southward of ,banks' ,cove1 there is a fine elliptic crater1 about five hundred feet in depth1 and three-quarters of a mile in diameter4 ,its bottom is occupied by a lake of brine1 out of which some little crateriform hills of tuff rise4 ,the lower beds are formed of compact tuff1 appearing like a subaqueous deposit2 whilst the upper beds1 round the entire circumference1 consist of a harsh1 friable tuff1 of little specific gravity1 but often containing fragments of rock in layers4 ,this upper tuff contains numerous pisolitic balls1 about the size of small bullets1 which differ from the surrounding matter1 only in being slightly harder and finer grained4 ,the beds dip away very regularly on all sides1 at angles varying1 as ,i found by measurement1 from twenty-five to thirty degrees4 ,the external surface of the crater slopes at a nearly similar inclination1 and is formed by slightly convex ribs1 like those on the shell of a pecten or scallop1 which become broader as they extend from the mouth of the crater to its base4 ,these ribs are generally from eight to twenty feet in breadth1 but sometimes they are as much as forty feet broad2 and they resemble old1 plastered1 much flattened vaults1 with the plaster scaling off in plates3 they are separated from each other by gullies1 deepened by alluvial action4 ,at their upper and narrow ends1 near the mouth of the crater1 these ribs often consist of real hollow passages1 like1 but rather smaller than1 those often formed by the cooling of the crust of a lava-stream1 whilst the inner parts have flowed onward2--of which structure ,i saw many examples at ,chatham ,island4 ,there can be no doubt but that these hollow ribs or vaults have been formed in a similar manner1 namely1 by the setting or hardening of a superficial crust on streams of mud1 which have flowed down from the upper part of the crater4 ,in another part of this same crater1 ,i saw open concave gutters between one and two feet wide1 which appear to have been formed by the hardening of the lower surface of a mud stream1 instead of1 as in the former case1 of the upper surface4 ,from these facts ,i think it is certain that the tuff must have flowed as mud4#aa ,this mud may have been formed either within the crater1 or from ashes deposited on its upper parts1 and afterwards washed down by torrents of rain4 ,the former method1 in most of the cases1 appears the more probable one2 at ,james ,island1 however1 some beds of the friable kind of tuff extend so continuously over an uneven surface1 that probably they were formed by the falling of showers of ashes4 #aa ,this conclusion is of some interest1 because ,m4 ,dufrenoy 78,mm4 pour servir10 tome iv1 p4 #bgd7 has argued from strata of tuff1 apparently of similar composition with that here described1 being inclined at angles between #ah and #bj1 that ,monte ,nuevo and some other craters of ,southern ,italy have been formed by upheaval4 ,from the facts given above1 of the vaulted character of the separate rills1 and from the tuff not extending in horizontal sheets round these crateriform hills1 no one will suppose that the strata have here been produced by elevation2 and yet we see that their inclination is above #bj1 and often as much as #cj4 ,the consolidated strata also1 of the internal talus1 as will be immediately seen1 dips at an angle of above #cj4 ,within this same crater1 strata of coarse tuff1 chiefly composed of fragments of lava1 abut1 like a consolidated talus1 against the inside walls4 ,they rise to a height of between one hundred and one hundred and fifty feet above the surface of the internal brine-lake2 they dip inwards1 and are inclined at an angle varying from thirty to thirty-six degrees4 ,they appear to have been formed beneath water1 probably at a period when the sea occupied the hollow of the crater4 ,i was surprised to observe that beds having this great inclination did not1 as far as they could be followed1 thicken towards their lower extremities4 ,banks' ,cove4--,this harbour occupies part of the interior of a shattered crater of tuff larger than that last described4 ,all the tuff is compact1 and includes numerous fragments of lava2 it appears like a subaqueous deposit4 ,the most remarkable feature in this crater is the great development of strata converging inwards1 as in the last case1 at a considerable inclination1 and often deposited in irregular curved layers4 ,these interior converging beds1 as well as the proper1 diverging crateriform strata1 are represented in figure ,no4 #ac1 a rude1 sectional sketch of the headlands1 forming this ,cove4 ,the internal and external strata differ little in composition1 and the former have evidently resulted from the wear and tear1 and redeposition of the matter forming the external crateriform strata4 ,from the great development of these inner beds1 a person walking round the rim of this crater might fancy himself on a circular anticlinal ridge of stratified sandstone and conglomerate4 ,the sea is wearing away the inner and outer strata1 and especially the latter2 so that the inwardly converging strata will1 perhaps1 in some future age1 be left standing alone--a case which might at first perplex a geologist4#ab #ab ,i believe that this case actually occurs in the ,azores1 where ,dr4 ,webster 78,description10 p4 #ahe7 has described a basin-formed1 little island1 composed of strata of tuff1 dipping inwards and bounded externally by steep sea-worn cliffs4 ,dr4 ,daubeny supposes 7on ,volcanoes1 p4 #bff71 that this cavity must have been formed by a circular subsidence4 ,it appears to me far more probable1 that we here have strata which were originally deposited within the hollow of a crater1 of which the exterior walls have since been removed by the sea4 ,no4 #ac ,illustration3 ,sectional sketch of headlands forming ,banks' ,cove4 ,a sectional sketch of the headlands forming ,banks' ,cove1 showing the diverging craterform strata1 and the converging stratified talus4 ,the highest point of these hills is #hag feet above the sea4 ,,james ,,island4--,two craters of tuff on this island are the only remaining ones which require any notice4 ,one of them lies a mile and a half inland from ,puerto ,grande3 it is circular1 about the third of a mile in diameter1 and #djj feet in depth4 ,it differs from all the other tuff-craters which ,i examined1 in having the lower part of its cavity1 to the height of between one hundred and one hundred and fifty feet1 formed by a precipitous wall of basalt1 giving to the crater the appearance of having burst through a solid sheet of rock4 ,the upper part of this crater consists of strata of the altered tuff1 with a semi-resinous fracture4 ,its bottom is occupied by a shallow lake of brine1 covering layers of salt1 which rest on deep black mud4 ,the other crater lies at the distance of a few miles1 and is only remarkable from its size and perfect condition4 ,its summit is #a1bjj feet above the level of the sea1 and the interior hollow is #fjj feet deep4 ,its external sloping surface presented a curious appearance from the smoothness of the wide layers of tuff1 which resembled a vast plastered floor4 ,brattle ,island is1 ,i believe1 the largest crater in the ,archipelago composed of tuff2 its interior diameter is nearly a nautical mile4 ,at present it is in a ruined condition1 consisting of little more than half a circle open to the south2 its great size is probably due1 in part1 to internal degradation1 from the action of the sea4 ,no4 #ad ,illustration3 ,segment of very small orifice of eruption4 ,segment of a very small orifice of eruption1 on the beach of ,fresh-water ,bay4 ,segment of a basaltic crater4--,one side of ,fresh-water ,bay1 in ,james ,island1 is bounded by a promontory1 which forms the last wreck of a great crater4 ,on the beach of this promontory1 a quadrant-shaped segment of a small subordinate point of eruption stands exposed4 ,it consists of nine separate little streams of lava piled upon each other2 and of an irregular pinnacle1 about fifteen feet high1 of reddish-brown1 vesicular basalt1 abounding with large crystals of glassy albite1 and with fused augite4 ,this pinnacle1 and some adjoining paps of rock on the beach1 represent the axis of the crater4 ,the streams of lava can be followed up a little ravine1 at right angles to the coast1 for between ten and fifteen yards1 where they are hidden by detritus3 along the beach they are visible for nearly eighty yards1 and ,i do not believe that they extend much further4 ,the three lower streams are united to the pinnacle2 and at the point of junction 7as shown in figure ,no4 #ad1 a rude sketch made on the spot71 they are slightly arched1 as if in the act of flowing over the lip of the crater4 ,the six upper streams no doubt were originally united to this same column before it was worn down by the sea4 ,the lava of these streams is of similar composition with that of the pinnacle1 excepting that the crystals of albite appear to be more comminuted1 and the grains of fused augite are absent4 ,each stream is separated from the one above it by a few inches1 or at most by one or two feet in thickness1 of loose fragmentary scori1 apparently derived from the abrasion of the streams in passing over each other4 ,all these streams are very remarkable from their thinness4 ,i carefully measured several of them2 one was eight inches thick1 but was firmly coated with three inches above1 and three inches below1 of red scoriaceous rock 7which is the case with all the streams71 making altogether a thickness of fourteen inches3 this thickness was preserved quite uniformly along the entire length of the section4 ,a second stream was only eight inches thick1 including both the upper and lower scoriaceous surfaces4 ,until examining this section1 ,i had not thought it possible that lava could have flowed in such uniformly thin sheets over a surface far from smooth4 ,these little streams closely resemble in composition that great deluge of lava at ,albemarle ,island1 which likewise must have possessed a high degree of fluidity4 ,pseudo-extraneous1 ejected fragments4--,in the lava and in the scori of this little crater1 ,i found several fragments1 which1 from their angular form1 their granular structure1 their freedom from air-cells1 their brittle and burnt condition1 closely resembled those fragments of primary rocks which are occasionally ejected1 as at ,ascension1 from volcanoes4 ,these fragments consist of glassy albite1 much mackled1 and with very imperfect cleavages1 mingled with semi-rounded grains1 having tarnished1 glossy surfaces1 of a steel-blue mineral4 ,the crystals of albite are coated by a red oxide of iron1 appearing like a residual substance2 and their cleavage-planes also are sometimes separated by excessively fine layers of this oxide1 giving to the crystals the appearance of being ruled like a glass micrometer4 ,there was no quartz4 ,the steel-blue mineral1 which is abundant in the pinnacle1 but which disappears in the streams derived from the pinnacle1 has a fused appearance1 and rarely presents even a trace of cleavage2 ,i obtained1 however1 one measurement1 which proved that it was augite2 and in one other fragment1 which differed from the others1 in being slightly cellular1 and in gradually blending into the surrounding matrix the small grains of this mineral were tolerably well crystallised4 ,although there is so wide a difference in appearance between the lava of the little streams1 and especially of their red scoriaceous crusts1 and one of these angular ejected fragments1 which at first sight might readily be mistaken for syenite1 yet ,i believe that the lava has originated from the melting and movement of a mass of rock of absolutely similar composition with the fragments4 ,besides the specimen above alluded to1 in which we see a fragment becoming slightly cellular1 and blending into the surrounding matrix1 some of the grains of the steel-blue augite also have their surfaces becoming very finely vesicular1 and passing into the nature of the surrounding paste2 other grains are throughout1 in an intermediate condition4 ,the paste seems to consist of the augite more perfectly fused1 or1 more probably1 merely disturbed in its softened state by the movement of the mass1 and mingled with the oxide of iron and with finely comminuted1 glassy albite4 ,hence probably it is that the fused albite1 which is abundant in the pinnacle1 disappears in the streams4 ,the albite is in exactly the same state1 with the exception of most of the crystals being smaller in the lava and in the embedded fragments2 but in the fragments they appear to be less abundant3 this1 however1 would naturally happen from the intumescence of the augitic base1 and its consequent apparent increase in bulk4 ,it is interesting thus to trace the steps by which a compact granular rock becomes converted into a vesicular1 pseudo-porphyritic lava1 and finally into red scori4 ,the structure and composition of the embedded fragments show that they are parts either of a mass of primary rock which has undergone considerable change from volcanic action1 or more probably of the crust of a body of cooled and crystallised lava1 which has afterwards been broken up and re-liquified2 the crust being less acted on by the renewed heat and movement4 ,concluding remarks on the tuff-craters4--,these craters1 from the peculiarity of the resin-like substance which enters largely into their composition1 from their structure1 their size and number1 present the most striking feature in the geology of this ,archipelago4 ,the majority of them form either separate islets1 or promontories attached to the larger islands2 and those which now stand at some little distance from the coast are worn and breached1 as if by the action of the sea4 ,from this general circumstance of their position1 and from the small quantity of ejected ashes in any part of the ,archipelago1 ,i am led to conclude1 that the tuff has been chiefly produced1 by the grinding together of fragments of lava within active craters1 communicating with the sea4 ,in the origin and composition of the tuff1 and in the frequent presence of a central lake of brine and of layers of salt1 these craters resemble1 though on a gigantic scale1 the 8salses10 or hillocks of mud1 which are common in some parts of ,italy and in other countries4#ac ,their closer connection1 however1 in this ,archipelago1 with ordinary volcanic action1 is shown by the pools of solidified basalt1 with which they are sometimes filled up4 #ac ,d',aubuisson's 8,trait de ,gognosie10 tome i1 p4 #ahi4 ,i may remark1 that ,i saw at ,terceira1 in the ,azores1 a crater of tuff or peperino1 very similar to these of the ,galapagos ,archipelago4 ,from the description given in ,freycinet 8,voyage10 similar ones occur at the ,sandwich ,islands2 and probably they are present in many other places4 ,it at first appears very singular1 that all the craters formed of tuff have their southern sides1 either quite broken down and wholly removed1 or much lower than the other sides4 ,i saw and received accounts of twenty-eight of these craters2 of these1 twelve form separate islets1#ad and now exist as mere crescents quite open to the south1 with occasionally a few points of rock marking their former circumference3 of the remaining sixteen1 some form promontories1 and others stand at a little distance inland from the shore2 but all have their southern sides either the lowest1 or quite broken down4 ,two1 however1 of the sixteen had their northern sides also low1 whilst their eastern and western sides were perfect4 ,i did not see1 or hear of1 a single exception to the rule1 of these craters being broken down or low on the side1 which faces a point of the horizon between ,s4,e4 and ,s4,w4 ,this rule does not apply to craters composed of lava and scori4 ,the explanation is simple3 at this ,archipelago1 the waves from the trade-wind1 and the swell propagated from the distant parts of the open ocean1 coincide in direction 7which is not the case in many parts of the ,pacific71 and with their united forces attack the southern sides of all the islands2 and consequently the southern slope1 even when entirely formed of hard basaltic rock1 is invariably steeper than the northern slope4 ,as the tuff-craters are composed of a soft material1 and as probably all1 or nearly all1 have at some period stood immersed in the sea1 we need not wonder that they should invariably exhibit on their exposed sides the effects of this great denuding power4 ,judging from the worn condition of many of these craters1 it is probable that some have been entirely washed away4 ,as there is no reason to suppose1 that the craters formed of scori and lava were erupted whilst standing in the sea1 we can see why the rule does not apply to them4 ,at ,ascension1 it was shown that the mouths of the craters1 which are there all of terrestrial origin1 have been affected by the trade-wind2 and this same power might here1 also1 aid in making the windward and exposed sides of some of the craters originally the lowest4 #ad ,these consist of the three ,crossman ,islets1 the largest of which is #fjj feet in height2 ,enchanted ,island2 ,gardner ,island 7#gfj feet high72 ,champion ,island 7#cca feet high72 ,enderby ,island2 ,brattle ,island2 two islets near ,indefatigable ,island2 and one near ,james ,island4 ,a second crater near ,james ,island 7with a salt lake in its centre7 has its southern side only about twenty feet high1 whilst the other parts of the circumference are about three hundred feet in height4 ,mineralogical composition of the rocks4--,in the northern islands1 the basaltic lavas seem generally to contain more albite than they do in the southern half of the ,archipelago2 but almost all the streams contain some4 ,the albite is not unfrequently associated with olivine4 ,i did not observe in any specimen distinguishable crystals of hornblende or augite2 ,i except the fused grains in the ejected fragments1 and in the pinnacle of the little crater1 above described4 ,i did not meet with a single specimen of true trachyte2 though some of the paler lavas1 when abounding with large crystals of the harsh and glassy albite1 resemble in some degree this rock2 but in every case the basis fuses into a black enamel4 ,beds of ashes and far-ejected scori1 as previously stated1 are almost absent2 nor did ,i see a fragment of obsidian or of pumice4 ,von ,buch#ae believes that the absence of pumice on ,mount ,etna is consequent on the feldspar being of the ,labrador variety2 if the presence of pumice depends on the constitution of the feldspar1 it is remarkable1 that it should be absent in this archipelago1 and abundant in the ,cordillera of ,south ,america1 in both of which regions the feldspar is of the albitic variety4 ,owing to the absence of ashes1 and the general indecomposable character of the lava in this ,archipelago1 the islands are slowly clothed with a poor vegetation1 and the scenery has a desolate and frightful aspect4 #ae 8,description des ,isles ,canaries10 p4 #cbh4 ,elevation of the land4--,proofs of the rising of the land are scanty and imperfect4 ,at ,chatham ,island1 ,i noticed some great blocks of lava1 cemented by calcareous matter1 containing recent shells2 but they occurred at the height of only a few feet above high-water mark4 ,one of the officers gave me some fragments of shells1 which he found embedded several hundred feet above the sea1 in the tuff of two craters1 distant from each other4 ,it is possible1 that these fragments may have been carried up to their present height in an eruption of mud2 but as1 in one instance1 they were associated with broken oyster-shells1 almost forming a layer1 it is more probable that the tuff was uplifted with the shells in mass4 ,the specimens are so imperfect that they can be recognised only as belonging to recent marine genera4 ,on ,charles ,island1 ,i observed a line of great rounded blocks1 piled on the summit of a vertical cliff1 at the height of fifteen feet above the line1 where the sea now acts during the heaviest gales4 ,this appeared1 at first1 good evidence in favour of the elevation of the land2 but it was quite deceptive1 for ,i afterwards saw on an adjoining part of this same coast1 and heard from eye-witnesses1 that wherever a recent stream of lava forms a smooth inclined plane1 entering the sea1 the waves during gales have the power of rolling up rounded blocks to a great height1 above the line of their ordinary action4 ,as the little cliff in the foregoing case is formed by a stream of lava1 which1 before being worn back1 must have entered the sea with a gently sloping surface1 it is possible or rather it is probable1 that the rounded boulders1 now lying on its summit1 are merely the remnants of those which had been rolled up during storms to their present height4 ,direction of the fissures of eruption4--,the volcanic orifices in this group cannot be considered as indiscriminately scattered4 ,three great craters on ,albermarle ,island form a well-marked line1 extending ,n4,w4 by ,n4 and ,s4,e4 by ,s4 ,narborough ,island1 and the great crater on the rectangular projection of ,albemarle ,island1 form a second parallel line4 ,to the east1 ,hood's ,island1 and the islands and rocks between it and ,james ,island1 form another nearly parallel line1 which1 when prolonged1 includes ,culpepper and ,wenman ,islands1 lying seventy miles to the north4 ,the other islands lying further eastward1 form a less regular fourth line4 ,several of these islands1 and the vents on ,albemarle ,island1 are so placed1 that they likewise fall on a set of rudely parallel lines1 intersecting the former lines at right angles2 so that the principal craters appear to lie on the points where two sets of fissures cross each other4 ,the islands themselves1 with the exception of ,albemarle ,island1 are not elongated in the same direction with the lines on which they stand4 ,the direction of these islands is nearly the same with that which prevails in so remarkable a manner in the numerous archipelagoes of the great ,pacific ,ocean4 ,finally1 ,i may remark1 that amongst the ,galapagos ,islands there is no one dominant vent much higher than all the others1 as may be observed in many volcanic archipelagoes3 the highest is the great mound on the south-western extremity of ,albemarle ,island1 which exceeds by barely a thousand feet several other neighbouring craters4 ,chapter ,,vi ,,trachyte ,,and ,,basalt4--,,distribution ,,of ,,volcanic ,,isles4 ,the sinking of crystals in fluid lava4--,specific gravity of the constituent parts of trachyte and of basalt1 and their consequent separation4--,obsidian4--,apparent non-separation of the elements of plutonic rocks4--,origin of trap-dikes in the plutonic series4--,distribution of volcanic islands2 their prevalence in the great oceans4--,they are generally arranged in lines4--,the central volcanoes of ,von ,buch doubtful4--,volcanic islands bordering continents4--,antiquity of volcanic islands1 and their elevation in mass4--,eruptions on parallel lines of fissure within the same geological period4 ,on the separation of the constituent minerals of lava1 according to their specific gravities4--,one side of ,fresh-water ,bay1 in ,james ,island1 is formed by the wreck of a large crater1 mentioned in the last chapter1 of which the interior has been filled up by a pool of basalt1 about two hundred feet in thickness4 ,this basalt is of a grey colour1 and contains many crystals of glassy albite1 which become much more numerous in the lower1 scoriaceous part4 ,this is contrary to what might have been expected1 for if the crystals had been originally disseminated in equal numbers1 the greater intumescence of this lower scoriaceous part would have made them appear fewer in number4 ,von ,buch#a has described a stream of obsidian on the ,peak of ,teneriffe1 in which the crystals of feldspar become more and more numerous1 as the depth or thickness increases1 so that near the lower surface of the stream the lava even resembles a primary rock4 ,von ,buch further states1 that ,m4 ,dree1 in his experiments in melting lava1 found that the crystals of feldspar always tended to precipitate themselves to the bottom of the crucible4 ,in these cases1 ,i presume there can be no doubt#b that the crystals sink from their weight4 ,the specific gravity of feldspar varies#c from #bd to #beh1 whilst obsidian seems commonly to be from #bc to #bd2 and in a fluidified state its specific gravity would probably be less1 which would facilitate the sinking of the crystals of feldspar4 ,at ,james ,island1 the crystals of albite1 though no doubt of less weight than the grey basalt1 in the parts where compact1 might easily be of greater specific gravity than the scoriaceous mass1 formed of melted lava and bubbles of heated gas4 #a 8,description des ,isles ,canaries10 pp4 #aij and #aia4 #b ,in a mass of molten iron1 it is found 7,edinburgh ,new ,philosophical ,journal1 vol4 xxiv1 p4 #ff7 that the substances1 which have a closer affinity for oxygen than iron has1 rise from the interior of the mass to the surface4 ,but a similar cause can hardly apply to the separation of the crystals of these lava-streams4 ,the cooling of the surface of lava seems1 in some cases1 to have affected its composition2 for ,dufrenoy 78,mm4 pour servir10 tome iv1 p4 #bga7 found that the interior parts of a stream near ,naples contained two-thirds of a mineral which was acted on by acids1 whilst the surface consisted chiefly of a mineral unattackable by acids4 #c ,i have taken the specific gravities of the simple minerals from ,von ,kobell1 one of the latest and best authorities1 and of the rocks from various authorities4 ,obsidian1 according to ,phillips1 is #bce2 and ,jameson says it never exceeds #bd2 but a specimen from ,ascension1 weighed by myself1 was #bdb4 ,the sinking of crystals through a viscid substance like molten rock1 as is unequivocally shown to have been the case in the experiments of ,m4 ,dre1 is worthy of further consideration1 as throwing light on the separation of the trachytic and basaltic series of lavas4 ,mr4 ,p4 ,scrope has speculated on this subject2 but he does not seem to have been aware of any positive facts1 such as those above given2 and he has overlooked one very necessary element1 as it appears to me1 in the phenomenon--namely1 the existence of either the lighter or heavier mineral in globules or in crystals4 ,in a substance of imperfect fluidity1 like molten rock1 it is hardly credible1 that the separate1 infinitely small atoms1 whether of feldspar1 augite1 or of any other mineral1 would have power from their slightly different gravities to overcome the friction caused by their movement2 but if the atoms of any one of these minerals became1 whilst the others remained fluid1 united into crystals or granules1 it is easy to perceive that from the lessened friction1 their sinking or floating power would be greatly increased4 ,on the other hand1 if all the minerals became granulated at the same time1 it is scarcely possible1 from their mutual resistance1 that any separation could take place4 ,a valuable1 practical discovery1 illustrating the effect of the granulation of one element in a fluid mass1 in aiding its separation1 has lately been made3 when lead containing a small proportion of silver1 is constantly stirred whilst cooling1 it becomes granulated1 and the grains of imperfect crystals of nearly pure lead sink to the bottom1 leaving a residue of melted metal much richer in silver2 whereas if the mixture be left undisturbed1 although kept fluid for a length of time1 the two metals show no signs of separating4#d ,the sole use of the stirring seems to be1 the formation of detached granules4 ,the specific gravity of silver is #ajd1 and of lead #aace3 the granulated lead1 which sinks1 is never absolutely pure1 and the residual fluid metal contains1 when richest1 only #a_/aai part of silver4 ,as the difference in specific gravity1 caused by the different proportions of the two metals1 is so exceedingly small1 the separation is probably aided in a great degree by the difference in gravity between the lead1 when granular though still hot1 and when fluid4 #d ,a full and interesting account of this discovery1 by ,mr4 ,pattinson1 was read before the ,british ,association in ,september #ahch4 ,in some alloys1 according to ,turner 78,chemistry10 p4 #baj71 the heaviest metal sinks1 and it appears that this takes place whilst both metals are fluid4 ,where there is a considerable difference in gravity1 as between iron and the slag formed during the fusion of the ore1 we need not be surprised at the atoms separating1 without either substance being granulated4 ,in a body of liquified volcanic rock1 left for some time without any violent disturbance1 we might expect1 in accordance with the above facts1 that if one of the constituent minerals became aggregated into crystals or granules1 or had been enveloped in this state from some previously existing mass1 such crystals or granules would rise or sink1 according to their specific gravity4 ,now we have plain evidence of crystals being embedded in many lavas1 whilst the paste or basis has continued fluid4 ,i need only refer1 as instances1 to the several1 great1 pseudo-porphyritic streams at the ,galapagos ,islands1 and to the trachytic streams in many parts of the world1 in which we find crystals of feldspar bent and broken by the movement of the surrounding1 semi-fluid matter4 ,lavas are chiefly composed of three varieties of feldspar1 varying in specific gravity from #bd to #bgd2 of hornblende and augite1 varying from #cj to #cd2 of olivine1 varying from #cc to #cd2 and lastly1 of oxides of iron1 with specific gravities from #dh to #eb4 ,hence crystals of feldspar1 enveloped in a mass of liquified1 but not highly vesicular lava1 would tend to rise to the upper parts2 and crystals or granules of the other minerals1 thus enveloped1 would tend to sink4 ,we ought not1 however1 to expect any perfect degree of separation in such viscid materials4 ,trachyte1 which consists chiefly of feldspar1 with some hornblende and oxide of iron1 has a specific gravity of about #bde2e whilst basalt1 composed chiefly of augite and feldspar1 often with much iron and olivine1 has a gravity of about #cj4 ,accordingly we find1 that where both trachytic and basaltic streams have proceeded from the same orifice1 the trachytic streams have generally been first erupted owing1 as we must suppose1 to the molten lava of this series having accumulated in the upper parts of the volcanic focus4 ,this order of eruption has been observed by ,beudant1 ,scrope1 and by other authors2 three instances1 also1 have been given in this volume4 ,as the later eruptions1 however1 from most volcanic mountains1 burst through their basal parts1 owing to the increased height and weight of the internal column of molten rock1 we see why1 in most cases1 only the lower flanks of the central1 trachytic masses1 are enveloped by basaltic streams4 ,the separation of the ingredients of a mass of lava1 would1 perhaps1 sometimes take place within the body of a volcanic mountain1 if lofty and of great dimensions1 instead of within the underground focus2 in which case1 trachytic streams might be poured forth1 almost contemporaneously1 or at short recurrent intervals1 from its summit1 and basaltic streams from its base3 this seems to have taken place at ,teneriffe4#f ,i need only further remark1 that from violent disturbances the separation of the two series1 even under otherwise favourable conditions1 would naturally often be prevented1 and likewise their usual order of eruption be inverted4 ,from the high degree of fluidity of most basaltic lavas1 these perhaps1 alone1 would in many cases reach the surface4 #e ,trachyte from ,java was found by ,von ,buch to be #bdg2 from ,auvergne1 by ,de la ,beche1 it was #bdb2 from ,ascension1 by myself1 it was #bdb4 ,jameson and other authors give to basalt a specific gravity of #cj2 but specimens from ,auvergne were found1 by ,de la ,beche1 to be only #bgh2 and from the ,giant's ,causeway1 to be #bia4 #f ,consult ,von ,buch's well-known and admirable 8,description ,physique0 of this island1 which might serve as a model of descriptive geology4 ,as we have seen that crystals of feldspar1 in the instance described by ,von ,buch1 sink in obsidian1 in accordance with their known greater specific gravity1 we might expect to find in every trachytic district1 where obsidian has flowed as lava1 that it had proceeded from the upper or highest orifices4 ,this1 according to ,von ,buch1 holds good in a remarkable manner both at the ,lipari ,islands and on the ,peak of ,teneriffe2 at this latter place obsidian has never flowed from a less height than #i1bjj feet4 ,obsidian1 also1 appears to have been erupted from the loftiest peaks of the ,peruvian ,cordillera4 ,i will only further observe1 that the specific gravity of quartz varies from #bf to #bh2 and therefore1 that when present in a volcanic focus1 it would not tend to sink with the basaltic bases2 and this1 perhaps1 explains the frequent presence1 and the abundance of this mineral1 in the lavas of the trachytic series1 as observed in previous parts of this volume4 ,an objection to the foregoing theory will1 perhaps1 be drawn from the plutonic rocks not being separated into two evidently distinct series1 of different specific gravities2 although1 like the volcanic1 they have been liquified4 ,in answer1 it may first be remarked1 that we have no evidence of the atoms of any one of the constituent minerals in the plutonic series having been aggregated1 whilst the others remained fluid1 which we have endeavoured to show is an almost necessary condition of their separation2 on the contrary1 the crystals have generally impressed each other with their forms4#g #g ,the crystalline paste of phonolite is frequently penetrated by long needles of hornblende2 from which it appears that the hornblende1 though the more fusible mineral1 has crystallised before1 or at the same time with a more refractory substance4 ,phonolite1 as far as my observations serve1 in every instance appears to be an injected rock1 like those of the plutonic series2 hence probably1 like these latter1 it has generally been cooled without repeated and violent disturbances4 ,those geologists who have doubted whether granite could have been formed by igneous liquefaction1 because minerals of different degrees of fusibility impress each other with their forms1 could not have been aware of the fact of crystallised hornblende penetrating phonolite1 a rock undoubtedly of igneous origin4 ,the viscidity1 which it is now known1 that both feldspar and quartz retain at a temperature much below their points of fusion1 easily explains their mutual impressment4 ,consult on this subject ,mr4 ,horner's paper on ,bonn1 8,geolog4 ,transact410 vol4 iv1 p4 #dci2 and 8,l',institut10 with respect to quartz1 #ahci1 p4 #afa4 ,in the second place1 the perfect tranquillity1 under which it is probable that the plutonic masses1 buried at profound depths1 have cooled1 would1 most likely1 be highly unfavourable to the separation of their constituent minerals2 for1 if the attractive force1 which during the progressive cooling draws together the molecules of the different minerals1 has power sufficient to keep them together1 the friction between such half-formed crystals or pasty globules would effectually prevent the heavier ones from sinking1 or the lighter ones from rising4 ,on the other hand1 a small amount of disturbance1 which would probably occur in most volcanic foci1 and which we have seen does not prevent the separation of granules of lead from a mixture of molten lead and silver1 or crystals of feldspar from streams of lava1 by breaking and dissolving the less perfectly formed globules1 would permit the more perfect and therefore unbroken crystals1 to sink or rise1 according to their specific gravity4 ,although in plutonic rocks two distinct species1 corresponding to the trachytic and basaltic series1 do not exist1 ,i much suspect that a certain amount of separation of their constituent parts has often taken place4 ,i suspect this from having observed how frequently dikes of greenstone and basalt intersect widely extended formations of granite and the allied metamorphic rocks4 ,i have never examined a district in an extensive granitic region without discovering dikes2 ,i may instance the numerous trap-dikes1 in several districts of ,brazil1 ,chile1 and ,australia1 and at the ,cape of ,good ,hope3 many dikes likewise occur in the great granitic tracts of ,india1 in the north of ,europe1 and in other countries4 ,whence1 then1 has the greenstone and basalt1 forming these dikes1 come8 ,are we to suppose1 like some of the elder geologists1 that a zone of trap is uniformly spread out beneath the granitic series1 which composes1 as far as we know1 the foundations of the earth's crust8 ,is it not more probable1 that these dikes have been formed by fissures penetrating into partially cooled rocks of the granitic and metamorphic series1 and by their more fluid parts1 consisting chiefly of hornblende1 oozing out1 and being sucked into such fissures8 ,at ,bahia1 in ,brazil1 in a district composed of gneiss and primitive greenstone1 ,i saw many dikes1 of a dark augitic 7for one crystal certainly was of this mineral7 or hornblendic rock1 which1 as several appearances clearly proved1 either had been formed before the surrounding mass had become solid1 or had together with it been afterwards thoroughly softened4#h ,on both sides of one of these dikes1 the gneiss was penetrated1 to the distance of several yards1 by numerous1 curvilinear threads or streaks of dark matter1 which resembled in form clouds of the class called cirrhi-com2 some few of these threads could be traced to their junction with the dike4 ,when examining them1 ,i doubted whether such hair-like and curvilinear veins could have been injected1 and ,i now suspect1 that instead of having been injected from the dike1 they were its feeders4 ,if the foregoing views of the origin of trap-dikes in widely extended granitic regions far from rocks of any other formation1 be admitted as probable1 we may further admit1 in the case of a great body of plutonic rock1 being impelled by repeated movements into the axis of a mountain-chain1 that its more liquid constituent parts might drain into deep and unseen abysses2 afterwards1 perhaps1 to be brought to the surface under the form1 either of injected masses of greenstone and augitic porphyry1#i or of basaltic eruptions4 ,much of the difficulty which geologists have experienced when they have compared the composition of volcanic with plutonic formations1 will1 ,i think1 be removed1 if we may believe that most plutonic masses have been1 to a certain extent1 drained of those comparatively weighty and easily liquified elements1 which compose the trappean and basaltic series of rocks4 #h ,portions of these dikes have been broken off1 and are now surrounded by the primary rocks1 with their lamin conformably winding round them4 ,dr4 ,hubbard also 7,silliman's ,journal1 vol4 xxxiv1 p4 #aai71 has described an interlacement of trap-veins in the granite of the ,white ,mountains1 which he thinks must have been formed when both rocks were soft4 #i ,mr4 ,phillips 78,lardner's ,encyclop410 vol4 ii1 p4 #aae7 quotes ,von ,buch's statement1 that augitic porphyry ranges parallel to1 and is found constantly at the base of1 great chains of mountains4 ,humboldt1 also1 has remarked the frequent occurrence of trap-rock1 in a similar position2 of which fact ,i have observed many examples at the foot of the ,chilian ,cordillera4 ,the existence of granite in the axes of great mountain chains is always probable1 and ,i am tempted to suppose1 that the laterally injected masses of augitic porphyry and of trap1 bear nearly the same relation to the granitic axes which basaltic lavas bear to the central trachytic masses1 round the flanks of which they have so frequently been erupted4 ,on the distribution of volcanic islands4--,during my investigations on coral-reefs1 ,i had occasion to consult the works of many voyagers1 and ,i was invariably struck with the fact1 that with rare exceptions1 the innumerable islands scattered throughout the ,pacific1 ,indian1 and ,atlantic ,oceans1 were composed either of volcanic1 or of modern coral-rocks4 ,it would be tedious to give a long catalogue of all the volcanic islands2 but the exceptions which ,i have found are easily enumerated3 in the ,atlantic1 we have ,st4 ,paul's ,rock1 described in this volume1 and the ,falkland ,islands1 composed of quartz and clay-slate2 but these latter islands are of considerable size1 and lie not very far from the ,south ,american coast3#aj in the ,indian ,ocean1 the ,seychelles 7situated in a line prolonged from ,madagascar7 consist of granite and quartz3 in the ,pacific ,ocean1 ,new ,caledonia1 an island of large size1 belongs 7as far as is known7 to the primary class4 ,new ,zealand1 which contains much volcanic rock and some active volcanoes1 from its size cannot be classed with the small islands1 which we are now considering4 ,the presence of a small quantity of non-volcanic rock1 as of clay-slate on three of the ,azores1#aa or of tertiary limestone at ,madeira1 or of clay-slate at ,chatham ,island in the ,pacific1 or of lignite at ,kerguelen ,land1 ought not to exclude such islands or archipelagoes1 if formed chiefly of erupted matter1 from the volcanic class4 #aj ,judging from ,forster's imperfect observation1 perhaps ,georgia is not volcanic4 ,dr4 ,allan is my informant with regard to the ,seychelles4 ,i do not know of what formation ,rodriguez1 in the ,indian ,ocean1 is composed4 #aa ,this is stated on the authority of ,count ,v4 de ,bedemar1 with respect to ,flores and ,graciosa 7,charlsworth1 8,magazine of ,nat4 ,hist410 vol4 i1 p4 #eeg74 ,st4 ,maria has no volcanic rock1 according to ,captain ,boyd 7,von ,buch 8,descript410 p4 #cfe74 ,chatham ,island has been described by ,dr4 ,dieffenbach in the 8,geographical ,journal10 #ahda1 p4 #bja4 ,as yet we have received only imperfect notices on ,kerguelen ,land1 from the ,antarctic ,expedition4 ,the composition of the numerous islands scattered through the great oceans being with such rare exceptions volcanic1 is evidently an extension of that law1 and the effect of those same causes1 whether chemical or mechanical1 from which it results1 that a vast majority of the volcanoes now in action stand either as islands in the sea1 or near its shores4 ,this fact of the ocean-islands being so generally volcanic is also interesting in relation to the nature of the mountain-chains on our continents1 which are comparatively seldom volcanic2 and yet we are led to suppose that where our continents now stand an ocean once extended4 ,do volcanic eruptions1 we may ask1 reach the surface more readily through fissures formed during the first stages of the conversion of the bed of the ocean into a tract of land8 ,looking at the charts of the numerous volcanic archipelagoes1 we see that the islands are generally arranged either in single1 double1 or triple rows1 in lines which are frequently curved in a slight degree4#ab ,each separate island is either rounded1 or more generally elongated in the same direction with the group in which it stands1 but sometimes transversely to it4 ,some of the groups which are not much elongated present little symmetry in their forms2 ,m4 ,virlet#ac states that this is the case with the ,grecian ,archipelago3 in such groups ,i suspect 7for ,i am aware how easy it is to deceive oneself on these points71 that the vents are generally arranged on one line1 or on a set of short parallel lines1 intersecting at nearly right angles another line1 or set of lines4 ,the ,galapagos ,archipelago offers an example of this structure1 for most of the islands and the chief orifices on the largest island are so grouped as to fall on a set of lines ranging about ,n4,w4 by ,n41 and on another set ranging about ,w4,s4,w43 in the ,canary ,archipelago we have a simpler structure of the same kind3 in the ,cape de ,verde group1 which appears to be the least symmetrical of any oceanic volcanic archipelago1 a ,n4,w4 and ,s4,e4 line formed by several islands1 if prolonged1 would intersect at right angles a curved line1 on which the remaining islands are placed4 ,von ,buch#ad has classed all volcanoes under two heads1 namely1 central volcanoes1 round which numerous eruptions have taken place on all sides1 in a manner almost regular1 and volcanic chains4 ,in the examples given of the first class1 as far as position is concerned1 ,i can see no grounds for their being called 8central20 and the evidence of any difference in mineralogical nature between central volcanoes and volcanic chains appears slight4 ,no doubt some one island in most small volcanic archipelagoes is apt to be considerably higher than the others2 and in a similar manner1 whatever the cause may be1 that on the same island one vent is generally higher than all the others4 ,von ,buch does not include in his class of volcanic chains small archipelagoes1 in which the islands are admitted by him1 as at the ,azores1 to be arranged in lines2 but when viewing on a map of the world how perfect a series exists from a few volcanic islands placed in a row to a train of linear archipelagoes following each other in a straight line1 and so on to a great wall like the ,cordillera of ,america1 it is difficult to believe that there exists any essential difference between short and long volcanic chains4 ,von ,buch#ae states that his volcanic chains surmount1 or are closely connected with1 mountain-ranges of primary formation3 but if trains of linear archipelagoes are1 in the course of time1 by the long-continued action of the elevatory and volcanic forces1 converted into mountain-ranges1 it would naturally result that the inferior primary rocks would often be uplifted and brought into view4 #ab ,professors ,william and ,henry ,darwin ,rogers have lately insisted much1 in a memoir read before the ,american ,association1 on the regularly curved lines of elevation in parts of the ,appalachian range4 #ac 8,bulletin de la ,soc4 ,golog410 tome iii1 p4 #aaj4 #ad 8,description des ,isles ,canaries10 p4 #cbd4 #ae ,idem1 p4 #cic4 ,some authors have remarked that volcanic islands occur scattered1 though at very unequal distances1 along the shores of the great continents1 as if in some measure connected with them4 ,in the case of ,juan ,fernandez1 situated #ccj miles from the coast of ,chile1 there was undoubtedly a connection between the volcanic forces acting under this island and under the continent1 as was shown during the earthquake of #ahce4 ,the islands1 moreover1 of some of the small volcanic groups which thus border continents1 are placed in lines1 related to those along which the adjoining shores of the continents trend2 ,i may instance the lines of intersection at the ,galapagos1 and at the ,cape de ,verde ,archipelagoes1 and the best marked line of the ,canary ,islands4 ,if these facts be not merely accidental1 we see that many scattered volcanic islands and small groups are related not only by proximity1 but in the direction of the fissures of eruption to the neighbouring continents--a relation1 which ,von ,buch considers1 characteristic of his great volcanic chains4 ,in volcanic archipelagoes1 the orifices are seldom in activity on more than one island at a time2 and the greater eruptions usually recur only after long intervals4 ,observing the number of craters1 that are usually found on each island of a group1 and the vast amount of matter which has been erupted from them1 one is led to attribute a high antiquity even to those groups1 which appear1 like the ,galapagos1 to be of comparatively recent origin4 ,this conclusion accords with the prodigious amount of degradation1 by the slow action of the sea1 which their originally sloping coasts must have suffered1 when they are worn back1 as is so often the case1 into grand precipices4 ,we ought not1 however1 to suppose1 in hardly any instance1 that the whole body of matter1 forming a volcanic island1 has been erupted at the level1 on which it now stands3 the number of dikes1 which seem invariably to intersect the interior parts of every volcano1 show1 on the principles explained by ,m4 ,elie de ,beaumont1 that the whole mass has been uplifted and fissured4 ,a connection1 moreover1 between volcanic eruptions and contemporaneous elevations in mass#af has1 ,i think1 been shown to exist in my work on ,coral-,reefs1 both from the frequent presence of upraised organic remains1 and from the structure of the accompanying coral-reefs4 ,finally1 ,i may remark1 that in the same ,archipelago1 eruptions have taken place within the historical period on more than one of the parallel lines of fissure3 thus1 at the ,galapagos ,archipelago1 eruptions have taken place from a vent on ,narborough ,island1 and from one on ,albemarle ,island1 which vents do not fall on the same line2 at the ,canary ,islands1 eruptions have taken place in ,teneriffe and ,lanzarote2 and at the ,azores1 on the three parallel lines of ,pico1 ,st4 ,jorge1 and ,terceira4 ,believing that a mountain-axis differs essentially from a volcano1 only in plutonic rocks having been injected1 instead of volcanic matter having been ejected1 this appears to me an interesting circumstance2 for we may infer from it as probable1 that in the elevation of a mountain-chain1 two or more of the parallel lines forming it may be upraised and injected within the same geological period4 #af ,a similar conclusion is forced on us1 by the phenomena1 which accompanied the earthquake of #ahce1 at ,concepcion1 and which are detailed in my paper 7vol4 v1 p4 #fja7 in the 8,geological ,transactions40 ,chapter ,,vii ,,australia2 ,,new ,,zealand2 ,,cape ,,of ,,good ,,hope4 ,new ,south ,wales4--,sandstone formation4--,embedded pseudo-fragments of shale4--,stratification4--,current-cleav age4--,great valleys4--,van ,diemen's ,land4--,palozoic formation4--,newer formation with volcanic rocks4--,travertin with leaves of extinct plants4--,elevation of the land4--,new ,zealand4--,king ,george's ,sound4--,superficial ferruginous beds4--,superficial calcareous deposits1 with casts of branches4--,their origin from drifted particles of shells and corals4--,their extent4--,cape of ,good ,hope4--,junction of the granite and clay-slate4--,sandstone formation4 ,the ,beagle1 in her homeward voyage1 touched at ,new ,zealand1 ,australia1 ,van ,diemen's ,land1 and the ,cape of ,good ,hope4 ,in order to confine the ,third ,part of these ,geological ,observations to ,south ,america1 ,i will here briefly describe all that ,i observed at these places worthy of the attention of geologists4 ,new ,south ,wales4--,my opportunities of observation consisted of a ride of ninety geographical miles to ,bathurst1 in a ,w4,n4,w4 direction from ,sydney4 ,the first thirty miles from the coast passes over a sandstone country1 broken up in many places by trap-rocks1 and separated by a bold escarpment overhanging the river ,nepean1 from the great sandstone platform of the ,blue ,mountains4 ,this upper platform is #a1jjj feet high at the edge of the escarpment1 and rises in a distance of twenty-five miles to between three and four thousand feet above the level of the sea4 ,at this distance the road descends to a country rather less elevated1 and composed in chief part of primary rocks4 ,there is much granite1 in one part passing into a red porphyry with octagonal crystals of quartz1 and intersected in some places by trap-dikes4 ,near the ,downs of ,bathurst ,i passed over much pale-brown1 glossy clay-slate1 with the shattered lamin running north and south2 ,i mention this fact1 because ,captain ,king informs me that1 in the country a hundred miles southward1 near ,lake ,george1 the mica-slate ranges so invariably north and south that the inhabitants take advantage of it in finding their way through the forests4 ,the sandstone of the ,blue ,mountains is at least #a1bjj feet thick1 and in some parts is apparently of greater thickness2 it consists of small grains of quartz1 cemented by white earthy matter1 and it abounds with ferruginous veins4 ,the lower beds sometimes alternate with shales and coal3 at ,wolgan ,i found in carbonaceous shale leaves of the ,glossopteris ,brownii1 a fern which so frequently accompanies the coal of ,australia4 ,the sandstone contains pebbles of quartz2 and these generally increase in number and size 7seldom1 however1 exceeding an inch or two in diameter7 in the upper beds3 ,i observed a similar circumstance in the grand sandstone formation at the ,cape of ,good ,hope4 ,on the ,south ,american coast1 where tertiary and supra-tertiary beds have been extensively elevated1 ,i repeatedly noticed that the uppermost beds were formed of coarser materials than the lower3 this appears to indicate that1 as the sea became shallower1 the force of the waves or currents increased4 ,on the lower platform1 however1 between the ,blue ,mountains and the coast1 ,i observed that the upper beds of the sandstone frequently passed into argillaceous shale1--the effect1 probably1 of this lower space having been protected from strong currents during its elevation4 ,the sandstone of the ,blue ,mountains evidently having been of mechanical origin1 and not having suffered any metamorphic action1 ,i was surprised at observing that1 in some specimens1 nearly all the grains of quartz were so perfectly crystallised with brilliant facets that they evidently had not in their present form been aggregated in any previously existing rock4#a ,it is difficult to imagine how these crystals could have been formed2 one can hardly believe that they were separately precipitated in their present crystallised state4 ,is it possible that rounded grains of quartz may have been acted on by a fluid corroding their surfaces1 and depositing on them fresh silica8 ,i may remark that1 in the sandstone formation of the ,cape of ,good ,hope1 it is evident that silica has been profusely deposited from aqueous solution4 #a ,i have lately seen1 in a paper by ,smith 7the father of ,english geologists71 in the ,magazine of ,natural ,history1 that the grains of quartz in the millstone grit of ,england are often crystallised4 ,sir ,david ,brewster1 in a paper read before the ,british ,association1 #ahdj1 states1 that in old decomposed glass1 the silex and metals separate into concentric rings1 and that the silex regains its crystalline structure1 as is shown by its action on light4 ,in several parts of the sandstone ,i noticed patches of shale which might at the first glance have been mistaken for extraneous fragments2 their horizontal lamin1 however1 being parallel with those of the sandstone1 showed that they were the remnants of thin1 continuous beds4 ,one such fragment 7probably the section of a long narrow strip7 seen in the face of a cliff1 was of greater vertical thickness than breadth1 which proves that this bed of shale must have been in some slight degree consolidated1 after having been deposited1 and before being worn away by the currents4 ,each patch of the shale shows1 also1 how slowly many of the successive layers of sandstone were deposited4 ,these pseudo-fragments of shale will perhaps explain1 in some cases1 the origin of apparently extraneous fragments in crystalline metamorphic rocks4 ,i mention this1 because ,i found near ,rio de ,janeiro a well-defined angular fragment1 seven yards long by two yards in breadth1 of gneiss containing garnets and mica in layers1 enclosed in the ordinary1 stratified1 porphyritic gneiss of the country4 ,the lamin of the fragment and of the surrounding matrix ran in exactly the same direction1 but they dipped at different angles4 ,i do not wish to affirm that this singular fragment 7a solitary case1 as far as ,i know7 was originally deposited in a layer1 like the shale in the ,blue ,mountains1 between the strata of the porphyritic gneiss1 before they were metamorphosed2 but there is sufficient analogy between the two cases to render such an explanation possible4 ,stratification of the escarpment4--,the strata of the ,blue ,mountains appear to the eye horizontal2 but they probably have a similar inclination with the surface of the platform1 which slopes from the west towards the escarpment over the ,nepean1 at an angle of one degree1 or of one hundred feet in a mile4#b ,the strata of the escarpment dip almost conformably with its steeply inclined face1 and with so much regularity1 that they appear as if thrown into their present position2 but on a more careful examination1 they are seen to thicken and to thin out1 and in the upper part to be succeeded and almost capped by horizontal beds4 ,these appearances render it probable1 that we here see an original escarpment1 not formed by the sea having eaten back into the strata1 but by the strata having originally extended only thus far4 ,those who have been in the habit of examining accurate charts of sea-coasts1 where sediment is accumulating1 will be aware1 that the surfaces of the banks thus formed1 generally slope from the coast very gently towards a certain line in the offing1 beyond which the depth in most cases suddenly becomes great4 ,i may instance the great banks of sediment within the ,west ,indian ,archipelago1#c which terminate in submarine slopes1 inclined at angles of between thirty and forty degrees1 and sometimes even at more than forty degrees3 every one knows how steep such a slope would appear on the land4 ,banks of this nature1 if uplifted1 would probably have nearly the same external form as the platform of the ,blue ,mountains1 where it abruptly terminates over the ,nepean4 #b ,this is stated on the authority of ,sir ,t4 ,mitchell1 in his 8,travels10 vol4 ii1 p4 #ceg4 #c ,i have described these very curious banks in the ,appendix to my volume on the structure of ,coral-,reefs4 ,i have ascertained the inclination of the edges of the banks1 from information given me by ,captain ,b4 ,allen1 one of the surveyors1 and by carefully measuring the horizontal distances between the last sounding on the bank and the first in the deep water4 ,widely extended banks in all parts of the ,west ,indies have the same general form of surface4 ,current-cleavage4--,the strata of sandstone in the low coast country1 and likewise on the ,blue ,mountains1 are often divided by cross or current lamin1 which dip in different directions1 and frequently at an angle of forty-five degrees4 ,most authors have attributed these cross layers to successive small accumulations on an inclined surface2 but from a careful examination in some parts of the ,new ,red ,sandstone of ,england1 ,i believe that such layers generally form parts of a series of curves1 like gigantic tidal ripples1 the tops of which have since been cut off1 either by nearly horizontal layers1 or by another set of great ripples1 the folds of which do not exactly coincide with those below them4 ,it is well-known to surveyors that mud and sand are disturbed during storms at considerable depths1 at least from three hundred to four hundred and fifty feet1#d so that the nature of the bottom even becomes temporarily changed2 the bottom1 also1 at a depth between sixty and seventy feet1 has been observed#e to be broadly rippled4 ,one may1 therefore1 be allowed to suspect1 from the appearance just mentioned in the ,new ,red ,sandstone1 that at greater depths1 the bed of the ocean is heaped up during gales into great ripple-like furrows and depressions1 which are afterwards cut off by the currents during more tranquil weather1 and again furrowed during gales4 #d ,see ,martin ,white1 on 8,soundings in the ,british ,channel10 pp4 #d and #aff4 #e ,m4 ,siau on the 8,action of ,waves10 ,edin4 ,new ,phil4 ,journ41 vol4 xxxi1 p4 #bde4 ,valleys in the sandstone platforms4--,the grand valleys1 by which the ,blue ,mountains and the other sandstone platforms of this part of ,australia are penetrated1 and which long offered an insuperable obstacle to the attempts of the most enterprising colonist to reach the interior country1 form the most striking feature in the geology of ,new ,south ,wales4 ,they are of grand dimensions1 and are bordered by continuous links of lofty cliffs4 ,it is not easy to conceive a more magnificent spectacle1 than is presented to a person walking on the summit-plains1 when without any notice he arrives at the brink of one of these cliffs1 which are so perpendicular1 that he can strike with a stone 7as ,i have tried7 the trees growing1 at the depth of between one thousand and one thousand five hundred feet below him2 on both hands he sees headland beyond headland of the receding line of cliff1 and on the opposite side of the valley1 often at the distance of several miles1 he beholds another line rising up to the same height with that on which he stands1 and formed of the same horizontal strata of pale sandstone4 ,the bottoms of these valleys are moderately level1 and the fall of the rivers flowing in them1 according to ,sir ,t4 ,mitchell1 is gentle4 ,the main valleys often send into the platform great baylike arms1 which expand at their upper ends2 and on the other hand1 the platform often sends promontories into the valley1 and even leaves in them great1 almost insulated1 masses4 ,so continuous are the bounding lines of cliff1 that to descend into some of these valleys1 it is necessary to go round twenty miles2 and into others1 the surveyors have only lately penetrated1 and the colonists have not yet been able to drive in their cattle4 ,but the most remarkable point of structure in these valleys1 is1 that although several miles wide in their upper parts1 they generally contract towards their mouths to such a degree as to become impassable4 ,the ,surveyor-,general1 ,sir ,t4 ,mitchell1#f in vain endeavoured1 first on foot and then by crawling between the great fallen fragments of sandstone1 to ascend through the gorge by which the river ,grose joins the ,nepean2 yet the valley of the ,grose in its upper part1 as ,i saw1 forms a magnificent basin some miles in width1 and is on all sides surrounded by cliffs1 the summits of which are believed to be nowhere less than #c1jjj feet above the level of the sea4 ,when cattle are driven into the valley of the ,wolgan by a path 7which ,i descended7 partly cut by the colonists1 they cannot escape2 for this valley is in every other part surrounded by perpendicular cliffs1 and eight miles lower down1 it contracts1 from an average width of half a mile1 to a mere chasm impassable to man or beast4 ,sir ,t4 ,mitchell#g states1 that the great valley of the ,cox river with all its branches contracts1 where it unites with the ,nepean1 into a gorge #b1bjj yards wide1 and about one thousand feet in depth4 ,other similar cases might have been added4 #f 8,travels in ,australia10 vol4 i1 p4 #aed4--;,i must express my obligation to ,sir ,t4 ,mitchell for several interesting personal communications on the subject of these great valleys of ,new ,south ,wales4 #g ,idem1 vol4 ii1 p4 #ceh4 ,the first impression1 from seeing the correspondence of the horizontal strata1 on each side of these valleys and great amphitheatre-like depressions1 is that they have been in chief part hollowed out1 like other valleys1 by aqueous erosion2 but when one reflects on the enormous amount of stone1 which on this view must have been removed1 in most of the above cases through mere gorges or chasms1 one is led to ask whether these spaces may not have subsided4 ,but considering the form of the irregularly branching valleys1 and of the narrow promontories1 projecting into them from the platforms1 we are compelled to abandon this notion4 ,to attribute these hollows to alluvial action1 would be preposterous2 nor does the drainage from the summit-level always fall1 as ,i remarked near the ,weatherboard1 into the head of these valleys1 but into one side of their bay-like recesses4 ,some of the inhabitants remarked to me1 that they never viewed one of these baylike recesses1 with the headlands receding on both hands1 without being struck with their resemblance to a bold sea-coast4 ,this is certainly the case2 moreover1 the numerous fine harbours1 with their widely branching arms1 on the present coast of ,new ,south ,wales1 which are generally connected with the sea by a narrow mouth1 from one mile to a quarter of a mile in width1 passing through the sandstone coast-cliffs1 present a likeness1 though on a miniature scale1 to the great valleys of the interior4 ,but then immediately occurs the startling difficulty1 why has the sea worn out these great1 though circumscribed1 depressions on a wide platform1 and left mere gorges1 through which the whole vast amount of triturated matter must have been carried away8 ,the only light ,i can throw on this enigma1 is by showing that banks appear to be forming in some seas of the most irregular forms1 and that the sides of such banks are so steep 7as before stated7 that a comparatively small amount of subsequent erosion would form them into cliffs3 that the waves have power to form high and precipitous cliffs1 even in landlocked harbours1 ,i have observed in many parts of ,south ,america4 ,in the ,red ,sea1 banks with an extremely irregular outline and composed of sediment1 are penetrated by the most singularly shaped creeks with narrow mouths3 this is likewise the case1 though on a larger scale1 with the ,bahama ,banks4 ,such banks1 ,i have been led to suppose1#h have been formed by currents heaping sediment on an irregular bottom4 ,that in some cases1 the sea1 instead of spreading out sediment in a uniform sheet1 heaps it round submarine rocks and islands1 it is hardly possible to doubt1 after having examined the charts of the ,west ,indies4 ,to apply these ideas to the sandstone platforms of ,new ,south ,wales1 ,i imagine that the strata might have been heaped on an irregular bottom by the action of strong currents1 and of the undulations of an open sea2 and that the valley-like spaces thus left unfilled might1 during a slow elevation of the land1 have had their steeply sloping flanks worn into cliffs2 the worn-down sandstone being removed1 either at the time when the narrow gorges were cut by the retreating sea1 or subsequently by alluvial action4 #h ,see the 8,appendix0 to the ,part on ,coral-,reefs4 ,the fact of the sea heaping up mud round a submarine nucleus1 is worthy of the notice of geologists3 for outlyers of the same composition with the coast banks are thus formed2 and these1 if upheaved and worn into cliffs1 would naturally be thought to have been once connected together4 ,van ,diemen's ,land4 ,the southern part of this island is mainly formed of mountains of greenstone1 which often assumes a syenitic character1 and contains much hypersthene4 ,these mountains1 in their lower half1 are generally encased by strata containing numerous small corals and some shells4 ,these shells have been examined by ,mr4 ,g4 ,b4 ,sowerby1 and have been described by him3 they consist of two species of ,producta1 and of six of ,spirifera2 two of these1 namely1 ,p4 rugata and ,s4 rotundata1 resemble1 as far as their imperfect condition allows of comparison1 ,british mountain-limestone shells4 ,mr4 ,lonsdale has had the kindness to examine the corals2 they consist of six undescribed species1 belonging to three genera4 ,species of these genera occur in the ,silurian1 ,devonian1 and ,carboniferous strata of ,europe4 ,mr4 ,lonsdale remarks1 that all these fossils have undoubtedly a ,palozoic character1 and that probably they correspond in age to a division of the system above the ,silurian formations4 ,the strata containing these remains are singular from the extreme variability of their mineralogical composition4 ,every intermediate form is present1 between flinty-slate1 clay-slate passing into grey wacke1 pure limestone1 sandstone1 and porcellanic rock2 and some of the beds can only be described as composed of a siliceo-calcareo-clay-slate4 ,the formation1 as far as ,i could judge1 is at least a thousand feet in thickness3 the upper few hundred feet usually consist of a siliceous sandstone1 containing pebbles and no organic remains2 the inferior strata1 of which a pale flinty slate is perhaps the most abundant1 are the most variable2 and these chiefly abound with the remains4 ,between two beds of hard crystalline limestone1 near ,newtown1 a layer of white soft calcareous matter is quarried1 and is used for whitewashing houses4 ,from information given to me by ,mr4 ,frankland1 the ,surveyor-,general1 it appears that this ,palozoic formation is found in different parts of the whole island2 from the same authority1 ,i may add1 that on the north-eastern coast and in ,bass' ,straits primary rocks extensively occur4 ,the shores of ,storm ,bay are skirted1 to the height of a few hundred feet1 by strata of sandstone1 containing pebbles of the formation just described1 with its characteristic fossils1 and therefore belonging to a subsequent age4 ,these strata of sandstone often pass into shale1 and alternate with layers of impure coal2 they have in many places been violently disturbed4 ,near ,hobart ,town1 ,i observed one dike1 nearly a hundred yards in width1 on one side of which the strata were tilted at an angle of #fj degrees1 and on the other they were in some parts vertical1 and had been altered by the effects of the heat4 ,on the west side of ,storm ,bay1 ,i found these strata capped by streams of basaltic lava with olivine2 and close by there was a mass of brecciated scori1 containing pebbles of lava1 which probably marks the place of an ancient submarine crater4 ,two of these streams of basalt were separated from each other by a layer of argillaceous wacke1 which could be traced passing into partially altered scori4 ,the wacke contained numerous rounded grains of a soft1 grass-green mineral1 with a waxy lustre1 and translucent on its edges3 under the blowpipe it instantly blackened1 and the points fused into a strongly magnetic1 black enamel4 ,in these characters1 it resembles those masses of decomposed olivine1 described at ,st4 ,jago in the ,cape de ,verde group2 and ,i should have thought that it had thus originated1 had ,i not found a similar substance1 in cylindrical threads1 within the cells of the vesicular basalt1--a state under which olivine never appears2 this substance1#i ,i believe1 would be classed as bole by mineralogists4 #i ,chlorophite1 described by ,dr4 ,mac,culloch 78,western ,islands10 vol4 i1 p4 #ejd7 as occurring in a basaltic amygdaloid1 differs from this substance1 in remaining unchanged before the blowpipe1 and in blackening from exposure to the air4 ,may we suppose that olivine1 in undergoing the remarkable change described at ,st4 ,jago1 passes through several states8 ,travertin with extinct plants4--,behind ,hobart ,town there is a small quarry of a hard travertin1 the lower strata of which abound with distinct impressions of leaves4 ,mr4 ,robert ,brown has had the kindness to look at my specimens1 and he informed me that there are four or five kinds1 none of which he recognises as belonging to existing species4 ,the most remarkable leaf is palmate1 like that of a fan-palm1 and no plant having leaves of this structure has hitherto been discovered in ,van ,diemen's ,land4 ,the other leaves do not resemble the most usual form of the ,eucalyptus 7of which tribe the existing forests are chiefly composed71 nor do they resemble that class of exceptions to the common form of the leaves of the ,eucalyptus1 which occur in this island4 ,the travertin containing this remnant of a lost vegetation1 is of a pale yellow colour1 hard1 and in parts even crystalline2 but not compact1 and is everywhere penetrated by minute1 tortuous1 cylindrical pores4 ,it contains a very few pebbles of quartz1 and occasionally layers of chalcedonic nodules1 like those of chert in our ,greensand4 ,from the pureness of this calcareous rock1 it has been searched for in other places1 but has never been found4 ,from this circumstance1 and from the character of the deposit1 it was probably formed by a calcareous spring entering a small pool or narrow creek4 ,the strata have subsequently been tilted and fissured2 and the surface has been covered by a singular mass1 with which1 also1 a large fissure has been filled up1 formed of balls of trap embedded in a mixture of wacke and a white1 earthy1 alumino-calcareous substance4 ,hence it would appear1 as if a volcanic eruption had taken place on the borders of the pool1 in which the calcareous matter was depositing1 and had broken it up and drained it4 ,elevation of the land4--,both the eastern and western shores of the bay1 in the neighbourhood of ,hobart ,town1 are in most parts covered to the height of thirty feet above the level of high-water mark1 with broken shells1 mingled with pebbles4 ,the colonists attribute these shells to the aborigines having carried them up for food3 undoubtedly1 there are many large mounds1 as was pointed out to me by ,mr4 ,frankland1 which have been thus formed2 but ,i think from the numbers of the shells1 from their frequent small size1 from the manner in which they are thinly scattered1 and from some appearances in the form of the land1 that we must attribute the presence of the greater number to a small elevation of the land4 ,on the shore of ,ralph ,bay 7opening into ,storm ,bay7 ,i observed a continuous beach about fifteen feet above high-water mark1 clothed with vegetation1 and by digging into it1 pebbles encrusted with ,serpul were found3 along the banks1 also1 of the river ,derwent1 ,i found a bed of broken sea-shells above the surface of the river1 and at a point where the water is now much too fresh for sea-shells to live2 but in both these cases1 it is just possible1 that before certain spits of sand and banks of mud in ,storm ,bay were accumulated1 the tides might have risen to the height where we now find the shells4#aj #aj ,it would appear that some changes are now in progress in ,ralph ,bay1 for ,i was assured by an intelligent farmer1 that oysters were formerly abundant in it1 but that about the year #ahcd they had1 without any apparent cause1 disappeared4 ,in the 8,transactions of the ,maryland ,academy0 7vol4 i1 part i1 p4 #bh7 there is an account by ,mr4 ,ducatel of vast beds of oysters and clams having been destroyed by the gradual filling up of the shallow lagoons and channels1 on the shores of the southern ,united ,states4 ,at ,chiloe1 in ,south ,america1 ,i heard of a similar loss1 sustained by the inhabitants1 in the disappearance from one part of the coast of an edible species of ,ascidia4 ,evidence more or less distinct of a change of level between the land and water1 has been detected on almost all the land on this side of the globe4 ,captain ,grey1 and other travellers1 have found in ,southern ,australia upraised shells1 belonging either to the recent1 or to a late tertiary period4 ,the ,french naturalists in ,baudin's expedition1 found shells similarly circumstanced on the ,s4,w4 coast of ,australia4 ,the ,rev4 ,w4 ,b4 ,clarke#aa finds proofs of the elevation of the land1 to the amount of #djj feet1 at the ,cape of ,good ,hope4 ,in the neighbourhood of the ,bay of ,islands in ,new ,zealand1#ab ,i observed that the shores were scattered to some height1 as at ,van ,diemen's ,land1 with sea-shells1 which the colonists attribute to the natives4 ,whatever may have been the origin of these shells1 ,i cannot doubt1 after having seen a section of the valley of the ,thames ,river 7#cg degrees ,s471 drawn by the ,rev4 ,w4 ,williams1 that the land has been there elevated3 on the opposite sides of this great valley1 three step-like terraces1 composed of an enormous accumulation of rounded pebbles1 exactly correspond with each other3 the escarpment of each terrace is about fifty feet in height4 ,no one after having examined the terraces in the valleys on the western shores of ,south ,america1 which are strewed with sea-shells1 and have been formed during intervals of rest in the slow elevation of the land1 could doubt that the ,new ,zealand terraces have been similarly formed4 ,i may add1 that ,dr4 ,dieffenbach1 in his description of the ,chatham ,islands#ac 7,s4,w4 of ,new ,zealand71 states that it is manifest 8that the sea has left many places bare which were once covered by its waters40 #aa 8,proceedings of the ,geological ,society10 vol4 iii1 p4 #dbj4 #ab ,i will here give a catalogue of the rocks which ,i met with near the ,bay of ,islands1 in ,new ,zealand3--#ast1 ,much basaltic lava1 and scoriform rocks1 forming distinct craters2--#bnd1 ,a castellated hill of horizontal strata of flesh-coloured limestone1 showing when fractured distinct crystalline facets3 the rain has acted on this rock in a remarkable manner1 corroding its surface into a miniature model of an ,alpine country3 ,i observed here layers of chert and clay ironstone2 and in the bed of a stream1 pebbles of clay-slate2--#crd1 ,the shores of the ,bay of ,islands are formed of a feldspathic rock1 of a bluish-grey colour1 often much decomposed1 with an angular fracture1 and crossed by numerous ferruginous seams1 but without any distinct stratification or cleavage4 ,some varieties are highly crystalline1 and would at once be pronounced to be trap2 others strikingly resembled clay-slate1 slightly altered by heat3 ,i was unable to form any decided opinion on this formation4 #ac ,geographical ,journal1 vol4 xi1 pp4 #bjb1 #bje4 ,king ,george's ,sound4 ,this settlement is situated at the south-western angle of the ,australian continent3 the whole country is granitic1 with the constituent minerals sometimes obscurely arranged in straight or curved lamin4 ,in these cases1 the rock would be called by ,humboldt1 gneiss-granite1 and it is remarkable that the form of the bare conical hills1 appearing to be composed of great folding layers1 strikingly resembles1 on a small scale1 those composed of gneiss-granite at ,rio de ,janeiro1 and those described by ,humboldt at ,venezuela4 ,these plutonic rocks are1 in many places1 intersected by trappean-dikes2 in one place1 ,i found ten parallel dikes ranging in an ,e4 and ,w4 line2 and not far off another set of eight dikes1 composed of a different variety of trap1 ranging at right angles to the former ones4 ,i have observed in several primary districts1 the occurrence of systems of dikes parallel and close to each other4 ,superficial ferruginous beds4--,the lower parts of the country are everywhere covered by a bed1 following the inequalities of the surface1 of a honeycombed sandstone1 abounding with oxides of iron4 ,beds of nearly similar composition are common1 ,i believe1 along the whole western coast of ,australia1 and on many of the ,east ,indian islands4 ,at the ,cape of ,good ,hope1 at the base of the mountains formed of granite and capped with sandstone1 the ground is everywhere coated either by a fine-grained1 rubbly1 ochraceous mass1 like that at ,king ,george's ,sound1 or by a coarser sandstone with fragments of quartz1 and rendered hard and heavy by an abundance of the hydrate of iron1 which presents1 when freshly broken1 a metallic lustre4 ,both these varieties have a very irregular texture1 including spaces either rounded or angular1 full of loose sand3 from this cause the surface is always honeycombed4 ,the oxide of iron is most abundant on the edges of the cavities1 where alone it affords a metallic fracture4 ,in these formations1 as well as in many true sedimentary deposits1 it is evident that iron tends to become aggregated1 either in the form of a shell1 or of a network4 ,the origin of these superficial beds1 though sufficiently obscure1 seems to be due to alluvial action on detritus abounding with iron4 ,superficial calcareous deposit4--,a calcareous deposit on the summit of ,bald ,head1 containing branched bodies1 supposed by some authors to have been corals1 has been celebrated by the descriptions of many distinguished voyagers4#ad ,it folds round and conceals irregular hummocks of granite1 at the height of #fjj feet above the level of the sea4 ,it varies much in thickness2 where stratified1 the beds are often inclined at high angles1 even as much as at thirty degrees1 and they dip in all directions4 ,these beds are sometimes crossed by oblique and even-sided lamin4 ,the deposit consists either of a fine1 white calcareous powder1 in which not a trace of structure can be discovered1 or of exceedingly minute1 rounded grains1 of brown1 yellowish1 and purplish colours2 both varieties being generally1 but not always1 mixed with small particles of quartz1 and being cemented into a more or less perfect stone4 ,the rounded calcareous grains1 when heated in a slight degree1 instantly lose their colours2 in this and in every other respect1 closely resembling those minute1 equal-sized particles of shells and corals1 which at ,st4 ,helena have been drifted up the side of the mountains1 and have thus been winnowed of all coarser fragments4 ,i cannot doubt that the coloured calcareous particles here have had a similar origin4 ,the impalpable powder has probably been derived from the decay of the rounded particles2 this certainly is possible1 for on the coast of ,peru1 ,i have traced large unbroken shells gradually falling into a substance as fine as powdered chalk4 ,both of the above-mentioned varieties of calcareous sandstone frequently alternate with1 and blend into1 thin layers of a hard substalagmitic#ae rock1 which1 even when the stone on each side contains particles of quartz1 is entirely free from them3 hence we must suppose that these layers1 as well as certain vein-like masses1 have been formed by rain dissolving the calcareous matter and re-precipitating it1 as has happened at ,st4 ,helena4 ,each layer probably marks a fresh surface1 when the1 now firmly cemented1 particles existed as loose sand4 ,these layers are sometimes brecciated and re-cemented1 as if they had been broken by the slipping of the sand when soft4 ,i did not find a single fragment of a sea-shell2 but bleached shells of the ,helix melo1 an existing land species1 abound in all the strata2 and ,i likewise found another ,helix1 and the case of an ,oniscus4 #ad ,i visited this hill1 in company with ,captain ,fitzroy1 and we came to a similar conclusion regarding these branching bodies4 #ae ,i adopt this term from ,lieutenant ,nelson's excellent paper on the ,bermuda ,islands 78,geolog4 ,trans410 vol4 v1 p4 #ajf71 for the hard1 compact1 cream- or brown-coloured stone1 without any crystalline structure1 which so often accompanies superficial calcareous accumulations4 ,i have observed such superficial beds1 coated with substalagmitic rock1 at the ,cape of ,good ,hope1 in several parts of ,chile1 and over wide spaces in ,la ,plata and ,patagonia4 ,some of these beds have been formed from decayed shells1 but the origin of the greater number is sufficiently obscure4 ,the causes which determine water to dissolve lime1 and then soon to redeposit it1 are not1 ,i think1 known4 ,the surface of the substalagmitic layers appears always to be corroded by the rain-water4 ,as all the above-mentioned countries have a long dry season1 compared with the rainy one1 ,i should have thought that the presence of the substalagmitic was connected with the climate1 had not ,lieutenant ,nelson found this substance forming under sea-water4 ,disintegrated shell seems to be extremely soluble2 of which ,i found good evidence1 in a curious rock at ,coquimbo in ,chile1 which consisted of small1 pellucid1 empty husks1 cemented together4 ,a series of specimens clearly showed that these husks had originally contained small rounded particles of shells1 which had been enveloped and cemented together by calcareous matter 7as often happens on sea-beaches71 and which subsequently had decayed1 and been dissolved by water1 that must have penetrated through the calcareous husks1 without corroding them1--of which processes every stage could be seen4 ,the branches are absolutely undistinguishable in shape from the broken and upright stumps of a thicket2 their roots are often uncovered1 and are seen to diverge on all sides2 here and there a branch lies prostrate4 ,the branches generally consist of the sandstone1 rather firmer than the surrounding matter1 with the central parts filled1 either with friable1 calcareous matter1 or with a substalagmitic variety2 this central part is also frequently penetrated by linear crevices1 sometimes1 though rarely1 containing a trace of woody matter4 ,these calcareous1 branching bodies1 appear to have been formed by fine calcareous matter being washed into the casts or cavities1 left by the decay of branches and roots of thickets1 buried under drifted sand4 ,the whole surface of the hill is now undergoing disintegration1 and hence the casts1 which are compact and hard1 are left projecting4 ,in calcareous sand at the ,cape of ,good ,hope1 ,i found the casts1 described by ,abel1 quite similar to these at ,bald ,head2 but their centres are often filled with black carbonaceous matter not yet removed4 ,it is not surprising1 that the woody matter should have been almost entirely removed from the casts on ,bald ,head2 for it is certain1 that many centuries must have elapsed since the thickets were buried2 at present1 owing to the form and height of the narrow promontory1 no sand is drifted up1 and the whole surface1 as ,i have remarked1 is wearing away4 ,we must1 therefore1 look back to a period when the land stood lower1 of which the ,french naturalists#af found evidence in upraised shells of recent species1 for the drifting on ,bald ,head of the calcareous and quartzose sand1 and the consequent embedment of the vegetable remains4 ,there was only one appearance which at first made me doubt concerning the origin of the cast1--namely1 that the finer roots from different stems sometimes became united together into upright plates or veins2 but when the manner is borne in mind in which fine roots often fill up cracks in hard earth1 and that these roots would decay and leave hollows1 as well as the stems1 there is no real difficulty in this case4 ,besides the calcareous branches from the ,cape of ,good ,hope1 ,i have seen casts1 of exactly the same forms1 from ,madeira#ag and from ,bermuda2 at this latter place1 the surrounding calcareous rocks1 judging from the specimens collected by ,lieutenant ,nelson1 are likewise similar1 as is their subaerial formation4 ,reflecting on the stratification of the deposit on ,bald ,head1--on the irregularly alternating layers of substalagmitic rock1--on the uniformly sized1 and rounded particles1 apparently of sea-shells and corals1--on the abundance of land-shells throughout the mass1--and finally1 on the absolute resemblance of the calcareous casts1 to the stumps1 roots1 and branches of that kind of vegetation1 which would grow on sand-hillocks1 ,i think there can be no reasonable doubt1 notwithstanding the different opinion of some authors1 that a true view of their origin has been here given4 #af ,see ,m4 ,pron's 8,voyage10 tome i1 p4 #bjd4 #ag ,dr4 ,j4 ,macaulay has fully described 7,edinb4 ,new ,phil4 ,journ41 vol4 xxix1 p4 #cej7 the casts from ,madeira4 ,he considers 7differently from ,mr4 ,smith of ,jordan ,hill7 these bodies to be corals1 and the calcareous deposit to be of subaqueous origin4 ,his arguments chiefly rest 7for his remarks on their structure are vague7 on the great quantity of the calcareous matter1 and on the casts containing animal matter1 as shown by their evolving ammonia4 ,had ,dr4 ,macaulay seen the enormous masses of rolled particles of shells and corals on the beach of ,ascension1 and especially on coral-reefs2 and had he reflected on the effects of long-continued1 gentle winds1 in drifting up the finer particles1 he would hardly have advanced the argument of quantity1 which is seldom trustworthy in geology4 ,if the calcareous matter has originated from disintegrated shells and corals1 the presence of animal matter is what might have been expected4 ,mr4 ,anderson analysed for ,dr4 ,macaulay part of a cast1 and he found it composed of-- ,carbonate of lime #gcae ,silica #aaij ,phosphate of lime #hha ,animal matter #dbe ,sulphate of lime a trace ------ #ihaa ,calcareous deposits1 like these of ,king ,george's ,sound1 are of vast extent on the ,australian shores4 ,dr4 ,fitton remarks1 that 8recent calcareous breccia 7by which term all these deposits are included7 was found during ,baudin's voyage1 over a space of no less than twenty-five degrees of latitude and an equal extent of longitude1 on the southern1 western1 and north-western coasts40#ah ,it appears also from ,m4 ,peron1 with whose observations and opinions on the origin of the calcareous matter and branching casts mine entirely accord1 that the deposit is generally much more continuous than near ,king ,george's ,sound4 ,at ,swan ,river1 ,archdeacon ,scott#ai states that in one part it extends ten miles inland4 ,captain ,wickham1 moreover1 informs me that during his late survey of the western coast1 the bottom of the sea1 wherever the vessel anchored1 was ascertained1 by crowbars being let down1 to consist of white calcareous matter4 ,hence it seems that along this coast1 as at ,bermuda and at ,keeling ,atoll1 submarine and subaerial deposits are contemporaneously in process of formation1 from the disintegration of marine organic bodies4 ,the extent of these deposits1 considering their origin1 is very striking2 and they can be compared in this respect only with the great coral-reefs of the ,indian and ,pacific ,oceans4 ,in other parts of the world1 for instance in ,south ,america1 there are superficial calcareous deposits of great extent1 in which not a trace of organic structure is discoverable2 these observations would lead to the inquiry1 whether such deposits may not1 also1 have been formed from disintegrated shells and corals4 #ah ,for ample details on this formation consult ,dr4 ,fitton's 8,appendix to ,captain ,king's ,voyage40 ,dr4 ,fitton is inclined to attribute a concretionary origin to the branching bodies3 ,i may remark1 that ,i have seen in beds of sand in ,la ,plata cylindrical stems which no doubt thus originated2 but they differed much in appearance from these at ,bald ,head1 and the other places above specified4 #ai 8,proceedings of the ,geolog4 ,soc410 vol4 i1 p4 #cbj4 ,cape of ,good ,hope4 ,after the accounts given by ,barrow1 ,carmichael1 ,basil ,hall1 and ,w4 ,b4 ,clarke of the geology of this district1 ,i shall confine myself to a few observations on the junction of the three principal formations4 ,the fundamental rock is granite1#bj overlaid by clay-slate3 the latter is generally hard1 and glossy from containing minute scales of mica2 it alternates with1 and passes into1 beds of slightly crystalline1 feldspathic1 slaty rock4 ,this clay-slate is remarkable from being in some places 7as on the ,lion's ,rump7 decomposed1 even to the depth of twenty feet1 into a pale-coloured1 sandstone-like rock1 which has been mistaken1 ,i believe1 by some observers1 for a separate formation4 ,i was guided by ,dr4 ,andrew ,smith to a fine junction at ,green ,point between the granite and clay-slate3 the latter at the distance of a quarter of a mile from the spot1 where the granite appears on the beach 7though1 probably1 the granite is much nearer underground71 becomes slightly more compact and crystalline4 ,at a less distance1 some of the beds of clay-slate are of a homogeneous texture1 and obscurely striped with different zones of colour1 whilst others are obscurely spotted4 ,within a hundred yards of the first vein of granite1 the clay-slate consists of several varieties2 some compact with a tinge of purple1 others glistening with numerous minute scales of mica and imperfectly crystallised feldspar2 some obscurely granular1 others porphyritic with small1 elongated spots of a soft white mineral1 which being easily corroded1 gives to this variety a vesicular appearance4 ,close to the granite1 the clay-slate is changed into a dark-coloured1 laminated rock1 having a granular fracture1 which is due to imperfect crystals of feldspar1 coated by minute1 brilliant scales of mica4 #bj ,in several places ,i observed in the granite1 small dark-coloured balls1 composed of minute scales of black mica in a tough basis4 ,in another place1 ,i found crystals of black schorl radiating from a common centre4 ,dr4 ,andrew ,smith found1 in the interior parts of the country1 some beautiful specimens of granite1 with silvery mica radiating or rather branching1 like moss1 from central points4 ,at the ,geological ,society1 there are specimens of granite with crystallised feldspar branching and radiating in like manner4 ,the actual junction between the granitic and clay-slate districts extends over a width of about two hundred yards1 and consists of irregular masses and of numerous dikes of granite1 entangled and surrounded by the clay-slate3 most of the dikes range in a ,n4,w4 and ,s4,e4 line1 parallel to the cleavage of the slate4 ,as we leave the junction1 thin beds1 and lastly1 mere films of the altered clay-slate are seen1 quite isolated1 as if floating1 in the coarsely crystallised granite2 but although completely detached1 they all retain traces of the uniform ,n4,w4 and ,s4,e4 cleavage4 ,this fact has been observed in other similar cases1 and has been advanced by some eminent geologists1#ba as a great difficulty on the ordinary theory1 of granite having been injected whilst liquified2 but if we reflect on the probable state of the lower surface of a laminated mass1 like clay-slate1 after having been violently arched by a body of molten granite1 we may conclude that it would be full of fissures parallel to the planes of cleavage2 and that these would be filled with granite1 so that wherever the fissures were close to each other1 mere parting layers or wedges of the slate would depend into the granite4 ,should1 therefore1 the whole body of rock afterwards become worn down and denuded1 the lower ends of these dependent masses or wedges of slate would be left quite isolated in the granite2 yet they would retain their proper lines of cleavage1 from having been united1 whilst the granite was fluid1 with a continuous covering of clay-slate4 #ba ,see ,m4 ,keilhau's 8,theory on ,granite0 translated in the ,edinburgh ,new ,philosophical ,journal1 vol4xxiv1 p4 #djb4 ,following1 in company with ,dr4 ,a4 ,smith1 the line of junction between the granite and the slate1 as it stretched inland1 in a ,s4,e4 direction1 we came to a place1 where the slate was converted into a fine-grained1 perfectly characterised gneiss1 composed of yellow-brown granular feldspar1 of abundant black brilliant mica1 and of few and thin lamin of quartz4 ,from the abundance of the mica in this gneiss1 compared with the small quantity and excessively minute scales1 in which it exists in the glossy clay-slate1 we must conclude1 that it has been here formed by the metamorphic action--a circumstance doubted1 under nearly similar circumstances1 by some authors4 ,the lamin of the clay-slate are straight2 and it was interesting to observe1 that as they assumed the character of gneiss1 they became undulatory with some of the smaller flexures angular1 like the lamin of many true metamorphic schists4 ,sandstone formation4--,this formation makes the most imposing feature in the geology of ,southern ,africa4 ,the strata are in many parts horizontal1 and attain a thickness of about two thousand feet4 ,the sandstone varies in character2 it contains little earthy matter1 but is often stained with iron2 some of the beds are very fine-grained and quite white2 others are as compact and homogeneous as quartz rock4 ,in some places ,i observed a breccia of quartz1 with the fragments almost dissolved in a siliceous paste4 ,broad veins of quartz1 often including large and perfect crystals1 are very numerous2 and it is evident in nearly all the strata1 that silica has been deposited from solution in remarkable quantity4 ,many of the varieties of quartzite appeared quite like metamorphic rocks2 but from the upper strata being as siliceous as the lower1 and from the undisturbed junctions with the granite1 which in many places can be examined1 ,i can hardly believe that these sandstone-strata have been exposed to heat4#bb ,on the lines of junction between these two great formations1 ,i found in several places the granite decayed to the depth of a few inches1 and succeeded1 either by a thin layer of ferruginous shale1 or by four or five inches in thickness of the re-cemented crystals of the granite1 on which the great pile of sandstone immediately rested4 #bb ,the ,rev4 ,w4 ,b4 ,clarke1 however1 states1 to my surprise 78,geolog4 ,proceedings10 vol4 iii1 p4 #dbb71 that the sandstone in some parts is penetrated by granitic dikes3 such dikes must belong to an epoch altogether subsequent to that when the molten granite acted on the clay-slate4 ,mr4 ,schomburgk has described#bc a great sandstone formation in ,northern ,brazil1 resting on granite1 and resembling to a remarkable degree1 in composition and in the external form of the land1 this formation of the ,cape of ,good ,hope4 ,the sandstones of the great platforms of ,eastern ,australia1 which also rest on granite1 differ in containing more earthy and less siliceous matter4 ,no fossil remains have been discovered in these three vast deposits4 ,finally1 ,i may add that ,i did not see any boulders of far-transported rocks at the ,cape of ,good ,hope1 or on the eastern and western shores of ,australia1 or at ,van ,diemen's ,land4 ,in the northern island of ,new ,zealand1 ,i noticed some large blocks of greenstone1 but whether their parent rock was far distant1 ,i had no opportunity of determining4 #bc ,geographical ,journal1 vol4 x1 p4 #bdf4 ,,geological ,,observations ,,on ,,south ,,america4 ,,critical ,,introduction4 ,of the remarkable 8trilogy0 constituted by ,darwin's writings which deal with the geology of the ,beagle1 the member which has perhaps attracted least attention1 up to the present time is that which treats of the geology of ,south ,america4 ,the actual writing of this book appears to have occupied ,darwin a shorter period than either of the other volumes of the series2 his diary records that the work was accomplished within ten months1 namely1 between ,july #ahdd and ,april #ahde2 but the book was not actually issued till late in the year following1 the preface bearing the date 8,september #ahdf40 ,altogether1 as ,darwin informs us in his 8,autobiography10 the geological books 8consumed four and a half years' steady work10 most of the remainder of the ten years that elapsed between the return of the ,beagle1 and the completion of his geological books being1 it is sad to relate1 8lost through illness60 ,concerning the 8,geological ,observations on ,south ,america10 ,darwin wrote to his friend ,lyell1 as follows3--8,my volume will be about #bdj pages1 dreadfully dull1 yet much condensed4 ,i think whenever you have time to look through it1 you will think the collection of facts on the elevation of the land and on the formation of terraces pretty good40 8,much condensed0 is the verdict that everyone must endorse1 on rising from the perusal of this remarkable book2 but by no means 8dull40 ,the three and a half years from ,april #ahcb to ,september #ahce1 were spent by ,darwin in ,south ,america1 and were devoted to continuous scientific work2 the problems he dealt with were either purely geological or those which constitute the borderland between the geological and biological sciences4 ,it is impossible to read the journal which he kept during this time without being impressed by the conviction that it contains all the germs of thought which afterwards developed into the 8,origin of ,species40 ,but it is equally evident that after his return to ,england1 biological speculations gradually began to exercise a more exclusive sway over ,darwin's mind1 and tended to dispossess geology1 which during the actual period of the voyage certainly engrossed most of his time and attention4 ,the wonderful series of observations made during those three and a half years in ,south ,america could scarcely be done justice to1 in the #bdj pages devoted to their exposition4 ,that he executed the work of preparing the book on ,south ,america in somewhat the manner of a task1 is shown by many references in his letters4 ,writing to ,sir ,joseph ,hooker in #ahde1 he says1 8,i hope this next summer to finish my ,south ,american ,geology1 then to get out a little ,zoology1 and hurrah for my species work60 ,it would seem that the feeling of disappointment1 which ,darwin so often experienced in comparing a book when completed1 with the observations and speculations which had inspired it1 was more keenly felt in the case of his volume on ,south ,america than any other4 ,to one friend he writes1 8,i have of late been slaving extra hard1 to the great discomfiture of wretched digestive organs1 at ,south ,america1 and thank all the fates1 ,i have done three-fourths of it4 ,writing plain ,english grows with me more and more difficult1 and never attainable4 ,as for your pretending that you will read anything so dull as my pure geological descriptions1 lay not such a flattering unction on my soul1 for it is incredible40 ,to another friend he writes1 8,you do not know what you threaten when you propose to read it--it is purely geological4 ,i said to my brother1 #8,you will of course read it1' and his answer was1 #8,upon my life1 ,i would sooner even buy it4'0 ,in spite of these disparaging remarks1 however1 we are strongly inclined to believe that this book1 despised by its author1 and neglected by his contemporaries1 will in the end be admitted to be one of ,darwin's chief titles to fame4 ,it is1 perhaps1 an unfortunate circumstance that the great success which he attained in biology by the publication of the 8,origin of ,species0 has1 to some extent1 overshadowed the fact that ,darwin's claims as a geologist1 are of the very highest order4 ,it is not too much to say that1 had ,darwin not been a geologist1 the 8,origin of ,species0 could never have been written by him4 ,but apart from those geological questions1 which have an important bearing on biological thought and speculation1 such as the proofs of imperfection in the geological record1 the relations of the later tertiary faunas to the recent ones in the same areas1 and the apparent intermingling of types belonging to distant geological epochs1 when we study the palontology of remote districts1--there are other purely geological problems1 upon which the contributions made by ,darwin are of the very highest value4 ,i believe that the verdict of the historians of science will be that if ,darwin had not taken a foremost place among the biologists of this century1 his position as a geologist would have been an almost equally commanding one4 ,but in the case of ,darwin's principal geological work--that relating to the origin of the crystalline schists1--geologists were not at the time prepared to receive his revolutionary teachings4 ,the influence of powerful authority was long exercised1 indeed1 to stifle his teaching1 and only now1 when this unfortunate opposition has disappeared1 is the true nature and importance of ,darwin's purely geological work beginning to be recognised4 ,the two first chapters of the 8,geological ,observations on ,south ,america10 deal with the proofs which exist of great1 but frequently interrupted1 movements of elevation during very recent geological times4 ,in connection with this subject1 ,darwin's particular attention was directed to the relations between the great earthquakes of ,south ,america--of some of which he had impressive experience--and the permanent changes of elevation which were taking place4 ,he was much struck by the rapidity with which the evidence of such great earth movements is frequently obliterated2 and especially with the remarkable way in which the action of rain-water1 percolating through deposits on the earth's surface1 removes all traces of shells and other calcareous organisms4 ,it was these considerations which were the parents of the generalisation that a palontological record can only be preserved during those periods in which long-continued slow subsidence is going on4 ,this in turn1 led to the still wider and more suggestive conclusion that the geological record as a whole is1 and never can be more than1 a series of more or less isolated fragments4 ,the recognition of this important fact constitutes the keystone to any theory of evolution which seeks to find a basis in the actual study of the types of life that have formerly inhabited our globe4 ,in his third chapter1 ,darwin gives a number of interesting facts1 collected during his visits to the plains and valleys of ,chili1 which bear on the question of the origin of saliferous deposits--the accumulation of salt1 gypsum1 and nitrate of soda4 ,this is a problem that has excited much discussion among geologists1 and which1 in spite of many valuable observations1 still remains to a great extent very obscure4 ,among the important considerations insisted upon by ,darwin is that relating to the absence of marine shells in beds associated with such deposits4 ,he justly argues that if the strata were formed in shallow waters1 and then exposed by upheaval to subaerial action1 all shells and other calcareous organisms would be removed by solution4 ,following ,lyell's method1 ,darwin proceeds from the study of deposits now being accumulated on the earth's surface1 to those which have been formed during the more recent periods of the geological history4 ,his account of the great ,pampean formation1 with its wonderful mammalian remains--,mastodon1 ,toxodon1 ,scelidotherium1 ,macrauchenia1 ,megatherium1 ,megalonyx1 ,mylodon1 and ,glyptodon--this full of interest4 ,his discovery of the remains of a true ,equus afforded a remarkable confirmation of the fact--already made out in ,north ,america--that species of horse had existed and become extinct in the ,new ,world1 before their introduction by the ,spaniards in the sixteenth century4 ,fully perceiving the importance of the microscope in studying the nature and origin of such deposits as those of the ,pampas1 ,darwin submitted many of his specimens both to ,dr4 ,carpenter in this country1 and to ,professor ,ehrenberg in ,berlin4 ,many very important notes on the microscopic organisms contained in the formation will be found scattered through the chapter4 ,darwin's study of the older tertiary formations1 with their abundant shells1 and their relics of vegetable life buried under great sheets of basalt1 led him to consider carefully the question of climate during these earlier periods4 ,in opposition to prevalent views on this subject1 ,darwin points out that his observations are opposed to the conclusion that a higher temperature prevailed universally over the globe during early geological periods4 ,he argues that 8the causes which gave to the older tertiary productions of the quite temperate zones of ,europe a tropical character1 were of a local character and did not affect the whole globe40 ,in this1 as in many similar instances1 we see the beneficial influence of extensive travel in freeing ,darwin's mind from prevailing prejudices4 ,it was this widening of experience which rendered him so especially qualified to deal with the great problem of the origin of species1 and in doing so to emancipate himself from ideas which were received with unquestioning faith by geologists whose studies had been circumscribed within the limits of ,western ,europe4 ,in the ,cordilleras of ,northern and ,central ,chili1 ,darwin1 when studying still older formations1 clearly recognised that they contain an admixture of the forms of life1 which in ,europe are distinctive of the ,cretaceous and ,jurassic periods respectively4 ,he was thus led to conclude that the classification of geological periods1 which fairly well expresses the facts that had been discovered in the areas where the science was first studied1 is no longer capable of being applied when we come to the study of widely distant regions4 ,this important conclusion led up to the further generalisation that each great geological period has exhibited a geographical distribution of the forms of animal and vegetable life1 comparable to that which prevails in the existing fauna and flora4 ,to those who are familiar with the extent to which the doctrine of universal formations has affected geological thought and speculation1 both long before and since the time that ,darwin wrote1 the importance of this new standpoint to which he was able to attain will be sufficiently apparent4 ,like the idea of the extreme imperfection of the ,geological ,record1 the doctrine of local geological formations is found permeating and moulding all the palontological reasonings of his great work4 ,in one of ,darwin's letters1 written while he was in ,south ,america1 there is a passage we have already quoted1 in which he expresses his inability to decide between the rival claims upon his attention of 8the old crystalline group of rocks10 and 8the softer fossiliferous beds0 respectively4 ,the sixth chapter of the work before us1 entitled 8,plutonic and ,metamorphic ,rocks--,cleavage and ,foliation10 contains a brief summary of a series of observations and reasonings upon these crystalline rocks1 which are1 we believe1 calculated to effect a revolution in geological science1 and--though their value and importance have long been overlooked--are likely to entitle ,darwin in the future to a position among geologists1 scarcely1 if at all1 inferior to that which he already occupies among biologists4 ,darwin's studies of the great rock-masses of the ,andes convinced him of the close relations between the granitic or ,plutonic rocks1 and those which were undoubtedly poured forth as lavas4 ,upon his return1 he set to work1 with the aid of ,professor ,miller1 to make a careful study of the minerals composing the granites and those which occur in the lavas1 and he was able to show that in all essential respects they are identical4 ,he was further able to prove that there is a complete gradation between the highly crystalline or granitic rock-masses1 and those containing more or less glassy matter between their crystals1 which constitute ordinary lavas4 ,the importance of this conclusion will be realised when we remember that it was then the common creed of geologists--and still continues to be so on the ,continent--that all highly crystalline rocks are of great geological antiquity1 and that the igneous ejections which have taken place since the beginning of the tertiary periods differ essentially1 in their composition1 their structure1 and their mode of occurrence1 from those which have made their appearance at earlier periods of the world's history4 ,very completely have the conclusions of ,darwin upon these subjects been justified by recent researches4 ,in ,england1 the ,united ,states1 and ,italy1 examples of the gradual passage of rocks of truly granitic structure into ordinary lavas have been described1 and the reality of the transition has been demonstrated by the most careful studies with the microscope4 ,recent researches carried on in ,south ,america by ,professor ,stelzner1 have also shown the existence of a class of highly crystalline rocks--the 8,andengranites0--which combine in themselves many of the characteristics which were once thought to be distinctive of the so-called ,plutonic and volcanic rocks4 ,no one familiar with recent geological literature--even in ,germany and ,france1 where the old views concerning the distinction of igneous products of different ages have been most stoutly maintained--can fail to recognise the fact that the principles contended for by ,darwin bid fair at no distant period to win universal acceptance among geologists all over the globe4 ,still more important are the conclusions at which ,darwin arrived with respect to the origin of the schists and gneisses which cover so large an area in ,south ,america4 ,carefully noting1 by the aid of his compass and clinometer1 at every point which he visited1 the direction and amount of inclination of the parallel divisions in these rocks1 he was led to a very important generalisation--namely1 that over very wide areas the direction 7strike7 of the planes of cleavage in slates1 and of foliation in schists and gneisses1 remained constant1 though the amount of their inclination 7dip7 often varied within wide limits4 ,further than this it appeared that there was always a close correspondence between the strike of the cleavage and foliation and the direction of the great axes along which elevation had taken place in the district4 ,in ,tierra del ,fuego1 ,darwin found striking evidence that the cleavage intersecting great masses of slate-rocks was quite independent of their original stratification1 and could often1 indeed1 be seen cutting across it at right angles4 ,he was also able to verify ,sedgwick's observation that1 in some slates1 glossy surfaces on the planes of cleavage arise from the development of new minerals1 chlorite1 epidote or mica1 and that in this way a complete graduation from slates to true schists may be traced4 ,darwin further showed that in highly schistose rocks1 the folia bend around and encircle any foreign bodies in the mass1 and that in some cases they exhibit the most tortuous forms and complicated puckerings4 ,he clearly saw that in all cases the forces by which these striking phenomena must have been produced were persistent over wide areas1 and were connected with the great movements by which the rocks had been upheaved and folded4 ,that the distinct folia of quartz1 feldspar1 mica1 and other minerals composing the metamorphic schists could not have been separately deposited as sediment was strongly insisted upon by ,darwin2 and in doing so he opposed the view generally prevalent among geologists at that time4 ,he was thus driven to the conclusion that foliation1 like cleavage1 is not an original1 but a superinduced structure in rock-masses1 and that it is the result of re-crystallisation1 under the controlling influence of great pressure1 of the materials of which the rock was composed4 ,in studying the lavas of ,ascension1 as we have already seen1 ,darwin was led to recognise the circumstance that1 when igneous rocks are subjected to great differential movements during the period of their consolidation1 they acquire a foliated structure1 closely analogous to that of the crystalline schists4 ,like his predecessor in this field of inquiry1 ,mr4 ,poulett ,scrope1 ,charles ,darwin seems to have been greatly impressed by these facts1 and he argued from them that the rocks exhibiting the foliated structure must have been in a state of plasticity1 like that of a cooling mass of lava4 ,at that time the suggestive experiments of ,tresca1 ,daubree1 and others1 showing that solid masses under the influence of enormous pressure become actually plastic1 had not been published4 ,had ,darwin been aware of these facts he would have seen that it was not necessary to assume a state of imperfect solidity in rock-masses in order to account for their having yielded to pressure and tension1 and1 in doing so1 acquiring the new characters which distinguish the crystalline schists4 ,the views put forward by ,darwin on the origin of the crystalline schists found an able advocate in ,mr4 ,daniel ,sharpe1 who in #aheb and #ahed published two papers1 dealing with the geology of the ,scottish ,highlands and of the ,alps respectively1 in which he showed that the principles arrived at by ,darwin when studying the ,south ,american rocks afford a complete explanation of the structure of the two districts in question4 ,but1 on the other hand1 the conclusions of ,darwin and ,sharpe were met with the strongest opposition by ,sir ,roderick ,murchison and ,dr4 ,a4 ,geikie1 who in #ahfa read a paper before the ,geological ,society 8,on the ,coincidence between ,stratification and ,foliation in the ,crystalline ,rocks of the ,scottish ,highlands10 in which they insisted that their observations in ,scotland tended to entirely disprove the conclusions of ,darwin that foliation in rocks is a secondary structure1 and entirely independent of the original stratification of the rock-masses4 ,now it is a most significant circumstance that1 no sooner did the officers of the ,geological ,survey commence the careful and detailed study of the ,scottish ,highlands than they found themselves compelled to make a formal retraction of the views which had been put forward by ,murchison and ,geikie in opposition to the conclusions of ,darwin4 ,the officers of the ,geological ,survey have completely abandoned the view that the foliation of the ,highland rocks has been determined by their original stratification1 and admit that the structure is the result of the profound movements to which the rocks have been subjected4 ,the same conclusions have recently been supported by observations made in many different districts--among which we may especially refer to those of ,dr4 ,h4 ,reusch in ,norway1 and those of ,dr4 ,j4 ,lehmann in ,saxony4 ,at the present time the arguments so clearly stated by ,darwin in the work before us1 have1 after enduring opposition or neglect for a whole generation1 begun to 8triumph all along the line10 and we may look forward confidently to the near future1 when his claim to be regarded as one of the greatest of geological discoverers shall be fully vindicated4 ,,john ,w4 ,,judd4 ,chapter ,i ,,on ,,the ,,elevation ,,of ,,the ,,eastern ,,coast ,,of ,,south ,,america4 ,upraised shells of ,la ,plata4--,bahia ,blanca1 ,sand-dunes and ,pumice-pebbles4--,step-formed plains of ,patagonia1 with upraised ,shells4--,terrace-bounded ,valley of ,santa ,cruz1 formerly a ,sea-strait4--,upraised shells of ,tierra del ,fuego4--,length and breadth of the elevated area4--,equability of the movements1 as shown by the similar heights of the plains4--,slowness of the elevatory process4--,mode of formation of the step-formed plains4--,summary4--,great ,shingle ,formation of ,patagonia2 its extent1 origin1 and distribution4--,formation of sea-cliffs4 ,in the following ,volume1 which treats of the geology of ,south ,america1 and almost exclusively of the parts southward of the ,tropic of ,capricorn1 ,i have arranged the chapters according to the age of the deposits1 occasionally departing from this order1 for the sake of geographical simplicity4 ,the elevation of the land within the recent period1 and the modifications of its surface through the action of the sea 7to which subjects ,i paid particular attention7 will be first discussed2 ,i will then pass on to the tertiary deposits1 and afterwards to the older rocks4 ,only those districts and sections will be described in detail which appear to me to deserve some particular attention2 and ,i will1 at the end of each chapter1 give a summary of the results4 ,we will commence with the proofs of the upheaval of the eastern coast of the continent1 from the ,rio ,plata southward2 and1 in the ,second ,chapter1 follow up the same subject along the shores of ,chile and ,peru4 ,on the northern bank of the great estuary of the ,rio ,plata1 near ,maldonado1 ,i found at the head of a lake1 sometimes brackish but generally containing fresh water1 a bed of muddy clay1 six feet in thickness1 with numerous shells of species still existing in the ,plata1 namely1 the ,azara labiata1 d',orbigny1 fragments of ,mytilus eduliformis1 d',orbigny1 ,paludestrina ,isabellei1 d',orbigny1 and the ,solen ,caribus1 ,lam41 which last was embedded vertically in the position in which it had lived4 ,these shells lie at the height of only two feet above the lake1 nor would they have been worth mentioning1 except in connection with analogous facts4 ,at ,monte ,video1 ,i noticed near the town1 and along the base of the mount1 beds of a living ,mytilus1 raised some feet above the surface of the ,plata3 in a similar bed1 at a height from thirteen to sixteen feet1 ,m4 ,isabelle collected eight species1 which1#a according to ,m4 d',orbigny1 now live at the mouth of the estuary4 ,at ,colonia del ,sacramiento1 further westward1 ,i observed at the height of about fifteen feet above the river1 there of quite fresh water1 a small bed of the same ,mytilus1 which lives in brackish water at ,monte ,video4 ,near the mouth of ,uruguay1 and for at least thirty-five miles northward1 there are at intervals large sandy tracts1 extending several miles from the banks of the river1 but not raised much above its level1 abounding with small bivalves1 which occur in such numbers that at the ,agraciado they are sifted and burnt for lime4 ,those which ,i examined near the ,a4 ,s4 ,juan were much worn3 they consisted of ,mactra ,isabellei1 d',orbigny1 mingled with few of ,venus sinuosa1 ,lam41 both inhabiting1 as ,i am informed by ,m4 d',orbigny1 brackish water at the mouth of the ,plata1 nearly or quite as salt as the open sea4 ,the loose sand1 in which these shells are packed1 is heaped into low1 straight1 long lines of dunes1 like those left by the sea at the head of many bays4 ,m4 d',orbigny has described#b an analogous phenomenon on a greater scale1 near ,san ,pedro on the river ,parana1 where he found widely extended beds and hillocks of sand1 with vast numbers of the ,azara labiata1 at the height of nearly #ajj feet 7,english7 above the surface of that river4 ,the ,azara inhabits brackish water1 and is not known to be found nearer to ,san ,pedro than ,buenos ,ayres1 distant above a hundred miles in a straight line4 ,nearer ,buenos ,ayres1 on the road from that place to ,san ,isidro1 there are extensive beds1 as ,i am informed by ,sir ,woodbine ,parish1#c of the ,azara labiata1 lying at about forty feet above the level of the river1 and distant between two and three miles from it4 ,these shells are always found on the highest banks in the district3 they are embedded in a stratified earthy mass1 precisely like that of the great ,pampean deposit hereafter to be described4 ,in one collection of these shells1 there were some valves of the ,venus sinuosa1 ,lam41 the same species found with the ,mactra on the banks of the ,uruguay4 ,south of ,buenos ,ayres1 near ,ensenada1 there are other beds of the ,azara1 some of which seem to have been embedded in yellowish1 calcareous1 semi-crystalline matter2 and ,sir ,w4 ,parish has given me from the banks of the ,arroyo del ,tristan1 situated in this same neighbourhood1 at the distance of about a league from the ,plata1 a specimen of a pale-reddish1 calcereo-argillaceous stone 7precisely like parts of the ,pampean deposit the importance of which fact will be referred to in a succeeding chapter71 abounding with shells of an ,azara1 much worn1 but which in general form and appearance closely resemble1 and are probably identical with1 the ,a4 labiata4 ,besides these shells1 cellular1 highly crystalline rock1 formed of the casts of small bivalves1 is found near ,ensenada2 and likewise beds of sea-shells1 which from their appearance appear to have lain on the surface4 ,sir ,w4 ,parish has given me some of these shells1 and ,m4 d',orbigny pronounces them to be3-- ,buccinanops globulosum1 d',orbigny4 ,olivancillaria auricularia1 d',orbigny4 ,venus flexuosa1 ,lam4 ,cythera 7imperfect74 ,mactra ,isabellei1 d',orbigny4 ,ostrea pulchella1 d',orbigny4 ,besides these1 ,sir ,w4 ,parish procured#d 7as named by ,mr4 ,g4 ,b4 ,sowerby7 the following shells3-- ,voluta colocynthis4 ,voluta angulata4 ,buccinum 7not spec4874 #a 8,voyage dans l',amrique ,mrid43 ,part4 ,golog410 p4 #ba4 #b ,ibid41 p4 #dc4 #c 8,buenos ,ayres10 etc41 by ,sir ,woodbine ,parish1 p4 #afh4 #d 8,buenos ,ayres10 etc41 by ,sir ,w4 ,parish1 p4 #afh4 ,all these species 7with1 perhaps1 the exception of the last7 are recent1 and live on the ,south ,american coast4 ,these shell-beds extend from one league to six leagues from the ,plata1 and must lie many feet above its level4 ,i heard1 also1 of beds of shells on the ,somborombon1 and on the ,rio ,salado1 at which latter place1 as ,m4 d',orbigny informs me1 the ,mactra ,isabellei and ,venus sinuosa are found4 ,during the elevation of the ,provinces of ,la ,plata1 the waters of the ancient estuary have but little affected 7with the exception of the sand-hills on the banks of the ,parana and ,uruguay7 the outline of the land4 ,m4 ,parchappe1#e however1 has described groups of sand dunes scattered over the wide extent of the ,pampas southward of ,buenos ,ayres1 which ,m4 d',orbigny attributes with much probability to the action of the sea1 before the plains were raised above its level4#f #e ,d',orbigny's 8,voyage ,golog410 p4 #dd4 #f ,before proceeding to the districts southward of ,la ,plata1 it may be worth while just to state1 that there is some evidence that the coast of ,brazil has participated in a small amount of elevation4 ,mr4 ,burchell informs me1 that he collected at ,santos 7lat4 #bd ,s47 oyster-shells1 apparently recent1 some miles from the shore1 and quite above the tidal action4 ,westward of ,rio de ,janeiro1 ,captain ,elliot is asserted 7see ,harlan1 8,med4 and ,phys4 ,res410 p4 #ce1 and ,dr4 ,meigs1 in 8,trans4 ,amer4 ,phil4 ,soc4071 to have found human bones1 encrusted with sea-shells1 between fifteen and twenty feet above the level of the sea4 ,between ,rio de ,janeiro and ,cape ,frio ,i crossed sandy tracts abounding with sea-shells1 at a distance of a league from the coast2 but whether these tracts have been formed by upheaval1 or through the mere accumulation of drift sand1 ,i am not prepared to assert4 ,at ,bahia 7lat4 #ac ,s471 in some parts near the coast1 there are traces of sea-action at the height of about twenty feet above its present level2 there are also1 in many parts1 remnants of beds of sandstone and conglomerate with numerous recent shells1 raised a little above the sea-level4 ,i may add1 that at the head of ,bahia ,bay there is a formation1 about forty feet in thickness1 containing tertiary shells apparently of fresh-water origin1 now washed by the sea and encrusted with ,balini2 this appears to indicate a small amount of subsidence subsequent to its deposition4 ,at ,pernambuco 7lat4 #h ,s471 in the alluvial or tertiary cliffs1 surrounding the low land on which the city stands1 ,i looked in vain for organic remains1 or other evidence of changes in level4 ,southward of the ,plata4--,the coast as far as ,bahia ,blanca 7in lat4 #ci ,s47 is formed either of a horizontal range of cliffs1 or of immense accumulations of sand-dunes4 ,within ,bahia ,blanca1 a small piece of tableland1 about twenty feet above high-water mark1 called ,punta ,alta1 is formed of strata of cemented gravel and of red earthy mud1 abounding with shells 7with others lying loose on the surface71 and the bones of extinct mammifers4 ,these shells1 twenty in number1 together with a ,balanus and two corals1 are all recent species1 still inhabiting the neighbouring seas4 ,they will be enumerated in the ,fourth ,chapter1 when describing the ,pampean formation2 five of them are identical with the upraised ones from near ,buenos ,ayres4 ,the northern shore of ,bahia ,blanca is1 in main part1 formed of immense sand-dunes1 resting on gravel with recent shells1 and ranging in lines parallel to the shore4 ,these ranges are separated from each other by flat spaces1 composed of stiff impure red clay1 in which1 at the distance of about two miles from the coast1 ,i found by digging a few minute fragments of sea-shells4 ,the sand-dunes extend several miles inland1 and stand on a plain1 which slopes up to a height of between one hundred and two hundred feet4 ,numerous1 small1 well-rounded pebbles of pumice lie scattered both on the plain and sand-hillocks3 at ,monte ,hermoso1 on the flat summit of a cliff1 ,i found many of them at a height of #abj feet 7angular measurement7 above the level of the sea4 ,these pumice pebbles1 no doubt1 were originally brought down from the ,cordillera by the rivers which cross the continent1 in the same way as the river ,negro anciently brought down1 and still brings down1 pumice1 and as the river ,chupat brings down scori3 when once delivered at the mouth of a river1 they would naturally have travelled along the coasts1 and been cast up during the elevation of the land1 at different heights4 ,the origin of the argillaceous flats1 which separate the parallel ranges of sand-dunes1 seems due to the tides here having a tendency 7as ,i believe they have on most shoal1 protected coasts7 to throw up a bar parallel to the shore1 and at some distance from it2 this bar gradually becomes larger1 affording a base for the accumulation of sand-dunes1 and the shallow space within then becomes silted up with mud4 ,the repetition of this process1 without any elevation of the land1 would form a level plain traversed by parallel lines of sand-hillocks2 during a slow elevation of the land1 the hillocks would rest on a gently inclined surface1 like that on the northern shore of ,bahia ,blanca4 ,i did not observe any shells in this neighbourhood at a greater height than twenty feet2 and therefore the age of the sea-drifted pebbles of pumice1 now standing at the height of #abj feet1 must remain uncertain4 ,the main plain surrounding ,bahia ,blanca ,i estimated at from two hundred to three hundred feet2 it insensibly rises towards the distant ,sierra ,ventana4 ,there are in this neighbourhood some other and lower plains1 but they do not abut one at the foot of the other1 in the manner hereafter to be described1 so characteristic of ,patagonia4 ,the plain on which the settlement stands is crossed by many low sand-dunes1 abounding with the minute shells of the ,paludestrina australis1 d',orbigny1 which now lives in the bay4 ,this low plain is bounded to the south1 at the ,cabeza del ,buey1 by the cliff-formed margin of a wide plain of the ,pampean formation1 which ,i estimated at sixty feet in height4 ,on the summit of this cliff there is a range of high sand-dunes extending several miles in an east and west line4 ,southward of ,bahia ,blanca1 the river ,colorado flows between two plains1 apparently from thirty to forty feet in height4 ,of these plains1 the southern one slopes up to the foot of the great sandstone plateau of the ,rio ,negro2 and the northern one against an escarpment of the ,pampean deposit2 so that the ,colorado flows in a valley fifty miles in width1 between the upper escarpments4 ,i state this1 because on the low plain at the foot of the northern escarpment1 ,i crossed an immense accumulation of high sand-dunes1 estimated by the ,gauchos at no less than eight miles in breadth4 ,these dunes range westward from the coast1 which is twenty miles distant1 to far inland1 in lines parallel to the valley2 they are separated from each other by argillaceous flats1 precisely like those on the northern shore of ,bahia ,blanca4 ,at present there is no source whence this immense accumulation of sand could proceed2 but if1 as ,i believe1 the upper escarpments once formed the shores of an estuary1 in that case the sandstone formation of the river ,negro would have afforded an inexhaustible supply of sand1 which would naturally have accumulated on the northern shore1 as on every part of the coast open to the south winds between ,bahia ,blanca and ,buenos ,ayres4 ,at ,san ,blas 7#dj #dj ,s47 a little south of the mouth of the ,colorado1 ,m4 d',orbigny#g found fourteen species of existing shells 7six of them identical with those from ,bahia ,blanca71 embedded in their natural positions4 ,from the zone of depth which these shells are known to inhabit1 they must have been uplifted thirty-two feet4 ,he also found1 at from fifteen to twenty feet above this bed1 the remains of an ancient beach4 #g 8,voyage10 etc41 p4 #ed4 ,ten miles southward1 but #abj miles to the west1 at ,port ,s4 ,antonio1 the ,officers employed on the ,survey assured me that they saw many old sea-shells strewed on the surface of the ground1 similar to those found on other parts of the coast of ,patagonia4 ,at ,san ,josef1 ninety miles south in nearly the same longitude1 ,i found1 above the gravel1 which caps an old tertiary formation1 an irregular bed and hillock of sand1 several feet in thickness1 abounding with shells of ,patella deaurita1 ,mytilus ,magellanicus1 the latter retaining much of its colour2 ,fusus ,magellanicus 7and a variety of the same71 and a large ,balanus 7probably ,b4 ,tulipa71 all now found on this coast3 ,i estimated this bed at from eighty to one hundred feet above the level of the sea4 ,to the westward of this bay1 there is a plain estimated at between two hundred and three hundred feet in height3 this plain seems1 from many measurements1 to be a continuation of the sandstone platform of the river ,negro4 ,the next place southward1 where ,i landed1 was at ,port ,desire1 #cdj miles distant2 but from the intermediate districts ,i received1 through the kindness of the ,officers of the ,survey1 especially from ,lieutenant ,stokes and ,mr4 ,king1 many specimens and sketches1 quite sufficient to show the general uniformity of the whole line of coast4 ,i may here state1 that the whole of ,patagonia consists of a tertiary formation1 resting on and sometimes surrounding hills of porphyry and quartz3 the surface is worn into many wide valleys and into level step-formed plains1 rising one above another1 all capped by irregular beds of gravel1 chiefly composed of porphyritic rocks4 ,this gravel formation will be separately described at the end of the chapter4 ,in the following diagrams3 ,baseline is ,level of sea4 ,scale is #a_/bj of inch to #ajj feet vertical4 ,height is shown in feet thus3 ,an4 ,m4 always stands for angular or trigonometrical measurement4 ,ba4 ,m4 always stands for barometrical measurement4 ,est4 always stands for estimation by the ,officers of the ,survey4 ,no4 #a ,section of step-formed plains south of ,nuevo ,gulf4 ,illustration3 ,section of step-formed plains south of ,nuevo ,gulf4 ,my object in giving the following measurements of the plains1 as taken by the ,officers of the ,survey1 is1 as will hereafter be seen1 to show the remarkable equability of the recent elevatory movements4 ,round the southern parts of ,nuevo ,gulf1 as far as the ,river ,chupat 7seventy miles southward of ,san ,josef71 there appear to be several plains1 of which the best defined are here represented4 ,the upper plain is here well defined 7called ,table ,hills72 its edge forms a cliff or line of escarpment many miles in length1 projecting over a lower plain4 ,the lowest plain corresponds with that at ,san ,josef with the recent shells on its surface4 ,between this lowest and the uppermost plain1 there is probably more than one step-formed terrace3 several measurements show the existence of the intermediate one of the height given in diagram ,no4 #a4 ,no4 #b ,section of plains in the ,bay of ,st4 ,george4 ,illustration3 ,section of plains in the ,bay of ,st4 ,george4 ,near the north headland of the great ,bay of ,st4 ,george 7#ajj miles south of the ,chupat71 two well-marked plains of #bej and #ccj feet were measured3 these are said to sweep round a great part of the ,bay4 ,at its south headland1 #abj miles distant from the north headland1 the #bej feet plain was again measured4 ,in the middle of the bay1 a higher plain was found at two neighbouring places 7,tilli ,roads and ,c4 ,marques7 to be #ehj feet in height4 ,above this plain1 towards the interior1 ,mr4 ,stokes informs me that there were several other step-formed plains1 the highest of which was estimated at #a1bjj feet1 and was seen ranging at apparently the same height for #aej miles northward4 ,all these plains have been worn into great valleys and much denuded4 ,the section in diagram ,no4 #c is illustrative of the general structure of the great ,bay of ,st4 ,george4 ,at the south headland of the ,bay of ,st4 ,george 7near ,c4 ,three ,points7 the #bej plain is very extensive4 ,at ,port ,desire 7forty miles southward7 ,i made several measurements with the barometer of a plain1 which extends along the north side of the port and along the open coast1 and which varies from #bde to #bee feet in height3 this plain abuts against the foot of a higher plain of #ccj feet1 which extends also far northward along the coast1 and likewise into the interior4 ,in the distance a higher inland platform was seen1 of which ,i do not know the height4 ,in three separate places1 ,i observed the cliff of the #bde-bee feet plain1 fringed by a terrace or narrow plain estimated at about one hundred feet in height4 ,these plains are represented in the following section3-- ,no4 #c ,section of plains at ,port ,desire4 ,illustration3 ,section of plains at ,port ,desire4 ,in many places1 even at the distance of three and four miles from the coast1 ,i found on the gravel-capped surface of the #bde-bee feet1 and of the #ccj feet plain1 shells of ,mytilus ,magellanicus1 ,m4 edulis1 ,patella deaurita1 and another ,patella1 too much worn to be identified1 but apparently similar to one found abundantly adhering to the leaves of the kelp4 ,these species are the commonest now living on this coast4 ,the shells all appeared very old2 the blue of the mussels was much faded2 and only traces of colour could be perceived in the ,patellas1 of which the outer surfaces were scaling off4 ,they lay scattered on the smooth surface of the gravel1 but abounded most in certain patches1 especially at the heads of the smaller valleys3 they generally contained sand in their insides2 and ,i presume that they have been washed by alluvial action out of thin sandy layers1 traces of which may sometimes be seen covering the gravel4 ,the several plains have very level surfaces2 but all are scooped out by numerous broad1 winding1 flat-bottomed valleys1 in which1 judging from the bushes1 streams never flow4 ,these remarks on the state of the shells1 and on the nature of the plains1 apply to the following cases1 so need not be repeated4 ,southward of ,port ,desire1 the plains have been greatly denuded1 with only small pieces of tableland marking their former extension4 ,but opposite ,bird ,island1 two considerable step-formed plains were measured1 and found respectively to be #cej and #eij feet in height4 ,this latter plain extends along the coast close to ,port ,st4 ,julian 7#aaj miles south of ,port ,desire72 where we have the following section3-- ,no4 #d ,section of plains at ,port ,st4 ,julian4 ,illustration3 ,section of plains at ,port ,st4 ,julian4 ,the lowest plain was estimated at ninety feet3 it is remarkable from the usual gravel-bed being deeply worn into hollows1 which are filled up with1 as well as the general surface covered by1 sandy and reddish earthy matter3 in one of the hollows thus filled up1 the skeleton of the ,macrauchenia ,patachonica1 as will hereafter be described1 was embedded4 ,on the surface and in the upper parts of this earthy mass1 there were numerous shells of ,mytilus ,magellanicus and ,m4 edulis1 ,patella deaurita1 and fragments of other species4 ,this plain is tolerably level1 but not extensive2 it forms a promontory seven or eight miles long1 and three or four wide4 ,the upper plains in ,diagram #d were measured by the ,officers of the ,survey2 they were all capped by thick beds of gravel1 and were all more or less denuded2 the #iej plain consists merely of separate1 truncated1 gravel-capped hills1 two of which1 by measurement1 were found to differ only three feet4 ,the #dcj feet plain extends1 apparently with hardly a break1 to near the northern entrance of the ,rio ,santa ,cruz 7fifty miles to the south72 but it was there found to be only #ccj feet in height4 ,on the southern side of the mouth of the ,santa ,cruz we have ,diagram #e1 which ,i am able to give with more detail than in the foregoing cases3-- ,no4 #e ,section of plains at the mouth of the ,rio ,santa ,cruz4 ,illustration3 ,section of plains at the mouth of the ,rio ,santa ,cruz4 ,the plain marked #cee feet 7as ascertained by the barometer and by angular measurement7 is a continuation of the above-mentioned #ccj feet plain3 it extends in a ,n4,w4 direction along the southern shores of the estuary4 ,it is capped by gravel1 which in most parts is covered by a thin bed of sandy earth1 and is scooped out by many flat-bottomed valleys4 ,it appears to the eye quite level1 but in proceeding in a ,s4,s4,w4 course1 towards an escarpment distant about six miles1 and likewise ranging across the country in a ,n4,w4 line1 it was found to rise at first insensibly1 and then for the last half-mile1 sensibly1 close up to the base of the escarpment3 at this point it was #dfc feet in height1 showing a rise of #ajh feet in the six miles4 ,on this #cee-dfc feet plain1 ,i found several shells of ,mytilus ,magellanicus and of a ,mytilus1 which ,mr4 ,sowerby informs me is yet unnamed1 though well-known as recent on this coast2 ,patella deaurita2 ,fusus1 ,i believe1 ,magellanicus1 but the specimen has been lost2 and at the distance of four miles from the coast1 at the height of about four hundred feet1 there were fragments of the same ,patella and of a ,voluta 7apparently ,v4 ancilla7 partially embedded in the superficial sandy earth4 ,all these shells had the same ancient appearance with those from the foregoing localities4 ,as the tides along this part of the coast rise at the ,syzygal period forty feet1 and therefore form a well-marked beach-line1 ,i particularly looked out for ridges in crossing this plain1 which1 as we have seen1 rises #ajh feet in about six miles1 but ,i could not see any traces of such4 ,the next highest plain is #gaj feet above the sea2 it is very narrow1 but level1 and is capped with gravel2 it abuts to the foot of the #hdj feet plain4 ,this summit-plain extends as far as the eye can range1 both inland along the southern side of the valley of the ,santa ,cruz1 and southward along the ,atlantic4 ,the ,valley of the ,r4 ,santa ,cruz4--,this valley runs in an east and west direction to the ,cordillera1 a distance of about one hundred and sixty miles4 ,it cuts through the great ,patagonian tertiary formation1 including1 in the upper half of the valley1 immense streams of basaltic lava1 which as well as the softer beds1 are capped by gravel2 and this gravel1 high up the river1 is associated with a vast boulder formation4#h ,in ascending the valley1 the plain which at the mouth on the southern side is #cee feet high1 is seen to trend towards the corresponding plain on the northern side1 so that their escarpments appear like the shores of a former estuary1 larger than the existing one3 the escarpments1 also1 of the #hdj feet summit-plain 7with a corresponding northern one1 which is met with some way up the valley71 appear like the shores of a still larger estuary4 ,farther up the valley1 the sides are bounded throughout its entire length by level1 gravel-capped terraces1 rising above each other in steps4 ,the width between the upper escarpments is on an average between seven and ten miles2 in one spot1 however1 where cutting through the basaltic lava1 it was only one mile and a half4 ,between the escarpments of the second highest terrace the average width is about four or five miles4 ,the bottom of the valley1 at the distance of #aaj miles from its mouth1 begins sensibly to expand1 and soon forms a considerable plain1 #ddj feet above the level of the sea1 through which the river flows in a gut from twenty to forty feet in depth4 ,i here found1 at a point #adj miles from the ,atlantic1 and seventy miles from the nearest creek of the ,pacific1 at the height of #daj feet1 a very old and worn shell of ,patella deaurita4 ,lower down the valley1 #aje miles from the ,atlantic 7long4 #ga ,w471 and at an elevation of about #cjj feet1 ,i also found1 in the bed of the river1 two much worn and broken shells of the ,voluta ancilla1 still retaining traces of their colours2 and one of the ,patella deaurita4 ,it appeared that these shells had been washed from the banks into the river2 considering the distance from the sea1 the desert and absolutely unfrequented character of the country1 and the very ancient appearance of the shells 7exactly like those found on the plains nearer the coast71 there is1 ,i think1 no cause to suspect that they could have been brought here by ,indians4 #h ,i have described this formation in a paper in the 8,geological ,transactions10 vol4 vi1 p4 #dae4 ,the plain at the head of the valley is tolerably level1 but water-worn1 and with many sand-dunes on it like those on a sea-coast4 ,at the highest point to which we ascended1 it was sixteen miles wide in a north and south line2 and forty-five miles in length in an east and west line4 ,it is bordered by the escarpments1 one above the other1 of two plains1 which diverge as they approach the ,cordillera1 and consequently resemble1 at two levels1 the shores of great bays facing the mountains2 and these mountains are breached in front of the lower plain by a remarkable gap4 ,the valley1 therefore1 of the ,santa ,cruz consists of a straight broad cut1 about ninety miles in length1 bordered by gravel-capped terraces and plains1 the escarpments of which at both ends diverge or expand1 one over the other1 after the manner of the shores of great bays4 ,bearing in mind this peculiar form of the land--the sand-dunes on the plain at the head of the valley--the gap in the ,cordillera1 in front of it--the presence in two places of very ancient shells of existing species--and lastly1 the circumstance of the #cee-dec feet plain1 with the numerous marine remains on its surface1 sweeping from the ,atlantic coast1 far up the valley1 ,i think we must admit1 that within the recent period1 the course of the ,santa ,cruz formed a sea-strait intersecting the continent4 ,at this period1 the southern part of ,south ,america consisted of an archipelago of islands #cfj miles in a north and south line4 ,we shall presently see1 that two other straits also1 since closed1 then cut through ,tierra del ,fuego2 ,i may add1 that one of them must at that time have expanded at the foot of the ,cordillera into a great bay 7now ,otway ,water7 like that which formerly covered the #ddj feet plain at the head of the ,santa ,cruz4 ,i have said that the valley in its whole course is bordered by gravel-capped plains4 ,the section 7diagram ,no4 #f71 supposed to be drawn in a north and south line across the valley1 can scarcely be considered as more than illustrative2 for during our hurried ascent it was impossible to measure all the plains at any one place4 ,at a point nearly midway between the ,cordillera and the ,atlantic1 ,i found the plain 7,a north7 #a1abb feet above the river2 all the lower plains on this side were here united into one great broken cliff3 at a point sixteen miles lower down the stream1 ,i found by measurement and estimation that ,b 7n7 was #hfi above the river3 very near to where ,a 7n7 was measured1 ,c 7n7 was #fci above the same level3 the terrace ,d 7n7 was nowhere measured3 the lowest ,e 7n7 was in many places about twenty feet above the river4 ,these plains or terraces were best developed where the valley was widest2 the whole five1 like gigantic steps1 occurred together only at a few points4 ,the lower terraces are less continuous than the higher ones1 and appear to be entirely lost in the upper third of the valley4 ,terrace ,c 7s71 however was traced continuously for a great distance4 ,the terrace ,b 7n71 at a point fifty-five miles from the mouth of the river1 was four miles in width2 higher up the valley this terrace 7or at least the second highest one1 for ,i could not always trace it continuously7 was about eight miles wide4 ,this second plain was generally wider than the lower ones--as indeed follows from the valley from ,a 7n7 to ,a 7s7 being generally nearly double the width of from ,b 7n7 to ,b 7s74 ,no4 #f ,north and ,south ,section across the terraces bounding the valley of the ,river ,santa ,cruz1 high up its course4 ,illustration3 ,north and ,south ,section across the terraces bounding the ,river ,santa ,cruz4 ,low down the valley1 the summit-plain ,a 7s7 is continuous with the #hdj feet plain on the coast1 but it is soon lost or unites with the escarpment of ,b 7s74 ,the corresponding plain ,a 7n71 on the north side of the valley1 appears to range continuously from the ,cordillera to the head of the present estuary of the ,santa ,cruz1 where it trends northward towards ,port ,st4 ,julian4 ,near the ,cordillera the summit-plain on both sides of the valley is between #c1bjj and #c1cjj feet in height2 at #ajj miles from the ,atlantic1 it is #a1daf feet1 and on the coast #hdj feet1 all above the sea-beach2 so that in a distance of #ajj miles the plain rises #egf feet1 and much more rapidly near to the ,cordillera4 ,the lower terraces ,b and ,c also appear to rise as they run up the valley2 thus ,d 7n71 measured at two points twenty-four miles apart1 was found to have risen #ahe feet4 ,from several reasons ,i suspect1 that this gradual inclination of the plains up the valley1 has been chiefly caused by the elevation of the continent in mass1 having been the greater the nearer to the ,cordillera4 ,all the terraces are capped with well-rounded gravel1 which rests either on the denuded and sometimes furrowed surface of the soft tertiary deposits1 or on the basaltic lava4 ,the difference in height between some of the lower steps or terraces seems to be entirely owing to a difference in the thickness of the capping gravel4 ,furrows and inequalities in the gravel1 where such occur1 are filled up and smoothed over with sandy earth4 ,the pebbles1 especially on the higher plains1 are often whitewashed1 and even cemented together by a white aluminous substance1 and ,i occasionally found this to be the case with the gravel on the terrace ,d4 ,i could not perceive any trace of a similar deposition on the pebbles now thrown up by the river1 and therefore ,i do not think that terrace ,d was river-formed4 ,as the terrace ,e generally stands about twenty feet above the bed of the river1 my first impression was to doubt whether even this lowest one could have been so formed2 but it should always be borne in mind1 that the horizontal upheaval of a district1 by increasing the total descent of the streams1 will always tend to increase1 first near the sea-coast and then further and further up the valley1 their corroding and deepening powers3 so that an alluvial plain1 formed almost on a level with a stream1 will1 after an elevation of this kind1 in time be cut through1 and left standing at a height never again to be reached by the water4 ,with respect to the three upper terraces of the ,santa ,cruz1 ,i think there can be no doubt1 that they were modelled by the sea1 when the valley was occupied by a strait1 in the same manner 7hereafter to be discussed7 as the greater step-formed1 shell-strewed plains along the coast of ,patagonia4 ,to return to the shores of the ,atlantic3 the #hdj feet plain1 at the mouth of the ,santa ,cruz1 is seen extending horizontally far to the south2 and ,i am informed by the ,officers of the ,survey1 that bending round the head of ,coy ,inlet 7sixty-five miles southward71 it trends inland4 ,outliers of apparently the same height are seen forty miles farther south1 inland of the river ,gallegos2 and a plain comes down to ,cape ,gregory 7thirty-five miles southward71 in the ,strait of ,magellan1 which was estimated at between eight hundred and one thousand feet in height1 and which1 rising towards the interior1 is capped by the boulder formation4 ,south of the ,strait of ,magellan1 there are large outlying masses of apparently the same great tableland1 extending at intervals along the eastern coast of ,tierra del ,fuego3 at two places here1 #aaj miles a part1 this plain was found to be #iej and #igj feet in height4 ,from ,coy ,inlet1 where the high summit-plain trends inland1 a plain estimated at #cej feet in height1 extends for forty miles to the river ,gallegos4 ,from this point to the ,strait of ,magellan1 and on each side of that ,strait1 the country has been much denuded and is less level4 ,it consists chiefly of the boulder formation1 which rises to a height of between one hundred and fifty and two hundred and fifty feet1 and is often capped by beds of gravel4 ,at ,n4,s4 ,gracia1 on the north side of the ,inner ,narrows of the ,strait of ,magellan1 ,i found on the summit of a cliff1 #afj feet in height1 shells of existing ,patell and ,mytili1 scattered on the surface and partially embedded in earth4 ,on the eastern coast1 also1 of ,tierra del ,fuego1 in latitude #ec #bj south1 ,i found many ,mytili on some level land1 estimated at #bjj feet in height4 ,anterior to the elevation attested by these shells1 it is evident by the present form of the land1 and by the distribution of the great erratic boulders#i on the surface1 that two sea-channels connected the ,strait of ,magellan both with ,sebastian ,bay and with ,otway ,water4 #i 8,geolog4 ,transactions10 vol4 vi1 p4 #dai4 ,concluding remarks on the recent elevation of the south-eastern coasts of ,america1 and on the action of the sea on the land4--,upraised shells of species1 still existing as the commonest kinds in the adjoining sea1 occur1 as we have seen1 at heights of between a few feet and #daj feet1 at intervals from latitude #cc #dj to #ec #bj south4 ,this is a distance of #a1ahj geographical miles--about equal from ,london to the ,north ,cape of ,sweden4 ,as the boulder formation extends with nearly the same height #aej miles south of #ec #bj1 the most southern point where ,i landed and found upraised shells2 and as the level ,pampas ranges many hundred miles northward of the point1 where ,m4 d',orbigny found at the height of #ajj feet beds of the ,azara1 the space in a north and south line1 which has been uplifted within the recent period1 must have been much above the #a1ahj miles4 ,by the term 8recent10 ,i refer only to that period within which the now living mollusca were called into existence2 for it will be seen in the ,fourth ,chapter1 that both at ,bahia ,blanca and ,p4 ,s4 ,julian1 the mammiferous quadrupeds which co-existed with these shells belong to extinct species4 ,i have said that the upraised shells were found only at intervals on this line of coast1 but this in all probability may be attributed to my not having landed at the intermediate points2 for wherever ,i did land1 with the exception of the river ,negro1 shells were found3 moreover1 the shells are strewed on plains or terraces1 which1 as we shall immediately see1 extend for great distances with a uniform height4 ,i ascended the higher plains only in a few places1 owing to the distance at which their escarpments generally range from the coast1 so that ,i am far from knowing that #daj feet is the maximum of elevation of these upraised remains4 ,the shells are those now most abundant in a living state in the adjoining sea4#aj ,all of them have an ancient appearance2 but some1 especially the mussels1 although lying fully exposed to the weather1 retain to a considerable extent their colours3 this circumstance appears at first surprising1 but it is now known that the colouring principle of the ,mytilus is so enduring1 that it is preserved when the shell itself is completely disintegrated4#aa ,most of the shells are broken2 ,i nowhere found two valves united2 the fragments are not rounded1 at least in none of the specimens which ,i brought home4 #aj ,captain ,king1 8,voyages of ,adventure and ,beagle10 vol4 i1 #a pp4 #f and #acc4 #aa ,see ,mr4 ,lyell 8,proofs of a ,gradual ,rising in ,sweden10 in the 8,philosoph4 ,transact410 #ahce1 p4 #a4 ,see also ,mr4 ,smith of ,jordan ,hill in the ,edin4 ,new ,phil4 ,journal1 vol4 xxv1 p4 #cic4 ,with respect to the breadth of the upraised area in an east and west line1 we know from the shells found at the ,inner ,narrows of the ,strait of ,magellan1 that the entire width of the plain1 although there very narrow1 has been elevated4 ,it is probable that in this southernmost part of the continent1 the movement has extended under the sea far eastward2 for at the ,falkland ,islands1 though ,i could not find any shells1 the bones of whales have been noticed by several competent observers1 lying on the land at a considerable distance from the sea1 and at the height of some hundred feet above it4#ab ,moreover1 we know that in ,tierra del ,fuego the boulder formation has been uplifted within the recent period1 and a similar formation occurs#ac on the north-western shores 7,byron ,sound7 of these islands4 ,the distance from this point to the ,cordillera of ,tierra del ,fuego1 is #cfj miles1 which we may take as the probable width of the recently upraised area4 ,in the latitude of the ,r4 ,santa ,cruz1 we know from the shells found at the mouth and head1 and in the middle of the valley1 that the entire width 7about #afj miles7 of the surface eastward of the ,cordillera has been upraised4 ,from the slope of the plains1 as shown by the course of the rivers1 for several degrees northward of the ,santa ,cruz1 it is probable that the elevation attested by the shells on the coast has likewise extended to the ,cordillera4 ,when1 however1 we look as far northward as the provinces of ,la ,plata1 this conclusion would be very hazardous2 not only is the distance from ,maldonado 7where ,i found upraised shells7 to the ,cordillera great1 namely1 #gfj miles1 but at the head of the estuary of the ,plata1 a ,n4,n4,e4 and ,s4,s4,w4 range of tertiary volcanic rocks has been observed1#ad which may well indicate an axis of elevation quite distinct from that of the ,andes4 ,moreover1 in the centre of the ,pampas in the chain of ,cordova1 severe earthquakes have been felt2#ae whereas at ,mendoza1 at the eastern foot of the ,cordillera1 only gentle oscillations1 transmitted from the shores of the ,pacific1 have ever been experienced4 ,hence the elevation of the ,pampas may be due to several distinct axes of movement2 and we cannot judge1 from the upraised shells round the estuary of the ,plata1 of the breadth of the area uplifted within the recent period4 #ab 8,voyages of the ,adventure and ,beagle10 vol4 ii1 p4 #bbg4 ,and ,bougainville's 8,voyage10 tome i1 p4 #aab4 #ac ,i owe this fact to the kindness of ,captain ,sulivan1 ,r4,n41 a highly competent observer4 ,i mention it more especially1 as in my ,paper 7p4 #dbg7 on the ,boulder ,formation1 ,i have1 after having examined the northern and middle parts of the eastern island1 said that the formation was here wholly absent4 #ad ,this volcanic formation will be described in ,chapter ,,iv4 ,it is not improbable that the height of the upraised shells at the head of the estuary of the ,plata1 being greater than at ,bahia ,blanca or at ,san ,blas1 may be owing to the upheaval of these latter places having been connected with the distant line of the ,cordillera1 whilst that of the provinces of ,la ,plata was in connection with the adjoining tertiary volcanic axis4 #ae ,see ,sir ,w4 ,parish's work on 8,la ,plata10 p4 #bdb4 ,for a notice of an earthquake which drained a lake near ,cordova1 see also ,temple's 8,travels in ,peru40 ,sir ,w4 ,parish informs me1 that a town between ,salta and ,tucuman 7north of ,cordova7 was formerly utterly overthrown by an earthquake4 ,not only has the above specified long range of coast been elevated within the recent period1 but ,i think it may be safely inferred from the similarity in height of the gravel-capped plains at distant points1 that there has been a remarkable degree of equability in the elevatory process4 ,i may premise1 that when ,i measured the plains1 it was simply to ascertain the heights at which shells occurred2 afterwards1 comparing these measurements with some of those made during the ,survey1 ,i was struck with their uniformity1 and accordingly tabulated all those which represented the summit-edges of plains4 ,the extension of the #ccj to #cee feet plain is very striking1 being found over a space of #ejj geographical miles in a north and south line4 ,a table 7,table #a7 of the measurements is given below4 ,the angular measurements and all the estimations 7in feet7 are by the ,officers of the ,survey2 the barometrical ones by myself3-- ,feet ,gallegos ,river to ,coy ,inlet 7partly angular partly estimation7 #cej ,south ,side of ,santa ,cruz 7angular and barometric7 #cee ,north ,side of ,santa ,cruz 7angular and barometric7 #ccj ,bird ,island1 plain opposite to 7angular7 #cej ,port ,desire1 plain extending far along coast 7barometric7 #ccj ,st4 ,george's ,bay1 north promontory 7angular7 #ccj ,table ,land1 south of ,new ,bay 7angular7 #cej ,a plain1 varying from #bde to #bee feet1 seems to extend with much uniformity from ,port ,desire to the north of ,st4 ,george's ,bay1 a distance of #agj miles2 and some approximate measurements 7in feet71 also given in the table below1 indicate the much greater extension of #ghj miles3-- ,feet ,coy ,inlet1 south of 7partly angular and partly estimation7 #bjj to #cjj ,port ,desire 7barometric7 #bde to #bee ,c4 ,blanco 7angular7 #bej ,north ,promontory of ,st4 ,george's ,bay 7angular7 #bej ,south of ,new ,bay 7angular7 #bjj to #bbj ,north of ,s4 ,josef 7estimation7 #bjj to #cjj ,plain of ,rio ,negro 7angular7 #bjj to #bbj ,bahia ,blanca 7estimation7 #bjj to #cjj ,the extension1 moreover1 of the #efj to #ehj1 and of the #hj to #ajj feet1 plains is remarkable1 though somewhat less obvious than in the former cases4 ,bearing in mind that ,i have not picked these measurements out of a series1 but have used all those which represented the edges of plains1 ,i think it scarcely possible that these coincidences in height should be accidental4 ,we must therefore conclude that the action1 whatever it may have been1 by which these plains have been modelled into their present forms1 has been singularly uniform4 ,these plains or great terraces1 of which three and four often rise like steps one behind the other1 are formed by the denudation of the old ,patagonian tertiary beds1 and by the deposition on their surfaces of a mass of well-rounded gravel1 varying1 near the coast1 from ten to thirty-five feet in thickness1 but increasing in thickness towards the interior4 ,the gravel is often capped by a thin irregular bed of sandy earth4 ,the plains slope up1 though seldom sensibly to the eye1 from the summit edge of one escarpment to the foot of the next highest one4 ,within a distance of #aej miles1 between ,santa ,cruz to ,port ,desire1 where the plains are particularly well developed1 there are at least seven stages or steps1 one above the other4 ,on the three lower ones1 namely1 those of #ajj feet1 #bej feet1 and #cej feet in height1 existing littoral shells are abundantly strewed1 either on the surface1 or partially embedded in the superficial sandy earth4 ,by whatever action these three lower plains have been modelled1 so undoubtedly have all the higher ones1 up to a height of #iej feet at ,s4 ,julian1 and of #a1bjj feet 7by estimation7 along ,st4 ,george's ,bay4 ,i think it will not be disputed1 considering the presence of the upraised marine shells1 that the sea has been the active power during stages of some kind in the elevatory process4 ,we will now briefly consider this subject3 if we look at the existing coast-line1 the evidence of the great denuding power of the sea is very distinct2 for1 from ,cape ,st4 ,diego1 in lat4 #ed #cj to the mouth of the ,rio ,negro1 in lat4 #ca 7a length of more than eight hundred miles71 the shore is formed1 with singularly few exceptions1 of bold and naked cliffs3 in many places the cliffs are high2 thus1 south of the ,santa ,cruz1 they are between eight and nine hundred feet in height1 with their horizontal strata abruptly cut off1 showing the immense mass of matter which has been removed4 ,nearly this whole line of coast consists of a series of greater or lesser curves1 the horns of which1 and likewise certain straight projecting portions1 are formed of hard rocks2 hence the concave parts are evidently the effect and the measure of the denuding action on the softer strata4 ,at the foot of all the cliffs1 the sea shoals very gradually far outwards2 and the bottom1 for a space of some miles1 everywhere consists of gravel4 ,i carefully examined the bed of the sea off the ,santa ,cruz1 and found that its inclination was exactly the same1 both in amount and in its peculiar curvature1 with that of the #cee feet plain at this same place4 ,if1 therefore1 the coast1 with the bed of the adjoining sea1 were now suddenly elevated one or two hundred feet1 an inland line of cliffs1 that is an escarpment1 would be formed1 with a gravel-capped plain at its foot gently sloping to the sea1 and having an inclination like that of the existing #cee feet plain4 ,from the denuding tendency of the sea1 this newly formed plain would in time be eaten back into a cliff3 and repetitions of this elevatory and denuding process would produce a series of gravel-capped sloping terraces1 rising one above another1 like those fronting the shores of ,patagonia4 ,the chief difficulty 7for there are other inconsiderable ones7 on this view1 is the fact1--as far as ,i can trust two continuous lines of soundings carefully taken between ,santa ,cruz and the ,falkland ,islands1 and several scattered observations on this and other coasts1--that the pebbles at the bottom of the sea quickly and regularly decrease in size with the increasing depth and distance from the shore1 whereas in the gravel on the sloping plains1 no such decrease in size was perceptible4 ,the following table gives the average result of many soundings off the ,santa ,cruz3-- ,under two miles from the shore1 many of the pebbles were of large size1 mingled with some small ones4 ,distance in miles from shore ,depth in fathoms ,size of ,pebbles #c to #d #aa to #ab ,as large as walnuts2 mingled in every case with some smaller ones4 #f to #g #ag to #ai ,as large as hazel-nuts4 #aj to #aa #bc to #be ,from three- to four-tenths of an inch in diameter4 #ab #cj to #dj ,two-tenths of an inch4 #bb to #aej #de to #fe ,one-tenth of an inch1 to the finest sand4 ,i particularly attended to the size of the pebbles on the #cee feet ,santa ,cruz plain1 and ,i noticed that on the summit-edge of the present sea cliffs many were as large as half a man's head2 and in crossing from these cliffs to the foot of the next highest escarpment1 a distance of six miles1 ,i could not observe any increase in their size4 ,we shall presently see that the theory of a slow and almost insensible rise of the land1 will explain all the facts connected with the gravel-capped terraces1 better than the theory of sudden elevations of from one to two hundred feet4 ,m4 d',orbigny has argued1 from the upraised shells at ,san ,blas being embedded in the positions in which they lived1 and from the valves of the ,azara labiata high on the banks of the ,parana being united and unrolled1 that the elevation of ,northern ,patagonia and of ,la ,plata must have been sudden2 for he thinks1 if it had been gradual1 these shells would all have been rolled on successive beach-lines4 ,but in protected bays1 such as in that of ,bahia ,blanca1 wherever the sea is accumulating extensive mud-banks1 or where the winds quietly heap up sand-dunes1 beds of shells might assuredly be preserved buried in the positions in which they had lived1 even whilst the land retained the same level2 any1 the smallest1 amount of elevation would directly aid in their preservation4 ,i saw a multitude of spots in ,bahia ,blanca where this might have been effected2 and at ,maldonado it almost certainly has been effected4 ,in speaking of the elevation of the land having been slow1 ,i do not wish to exclude the small starts which accompany earthquakes1 as on the coast of ,chile2 and by such movements beds of shells might easily be uplifted1 even in positions exposed to a heavy surf1 without undergoing any attrition3 for instance1 in #ahce1 a rocky flat off the island of ,santa ,maria was at one blow upheaved above high-water mark1 and was left covered with gaping and putrefying mussel-shells1 still attached to the bed on which they had lived4 ,if ,m4 d',orbigny had been aware of the many long parallel lines of sand-hillocks1 with infinitely numerous shells of the ,mactra and ,venus1 at a low level near the ,uruguay2 if he had seen at ,bahia ,blanca the immense sand-dunes1 with water-worn pebbles of pumice1 ranging in parallel lines1 one behind the other1 up a height of at least #abj feet2 if he had seen the sand-dunes1 with the countless ,paludestrinas1 on the low plain near the ,fort at this place1 and that long line on the edge of the cliff1 sixty feet higher up2 if he had crossed that long and great belt of parallel sand-dunes1 eight miles in width1 standing at the height of from forty to fifty feet above the ,colorado1 where sand could not now collect1--,i cannot believe he would have thought that the elevation of this great district had been sudden4 ,certainly the sand-dunes 7especially when abounding with shells71 which stand in ranges at so many different levels1 must all have required long time for their accumulation2 and hence ,i do not doubt that the last #ajj feet of elevation of ,la ,plata and ,northern ,patagonia has been exceedingly slow4 ,if we extend this conclusion to ,central and ,southern ,patagonia1 the inclination of the successively rising gravel-capped plains can be explained quite as well1 as by the more obvious view already given of a few comparatively great and sudden elevations2 in either case we must admit long periods of rest1 during which the sea ate deeply into the land4 ,let us suppose the present coast to rise at a nearly equable1 slow rate1 yet sufficiently quick to prevent the waves quite removing each part as soon as brought up2 in this case every portion of the present bed of the sea will successively form a beach-line1 and from being exposed to a like action will be similarly affected4 ,it cannot matter to what height the tides rise1 even if to forty feet as at ,santa ,cruz1 for they will act with equal force and in like manner on each successive line4 ,hence there is no difficulty in the fact of the #cee feet plain at ,santa ,cruz sloping up #ajh feet to the foot of the next highest escarpment1 and yet having no marks of any one particular beach-line on it2 for the whole surface on this view has been a beach4 ,i cannot pretend to follow out the precise action of the tidal-waves during a rise of the land1 slow1 yet sufficiently quick to prevent or check denudation3 but if it be analogous to what takes place on protected parts of the present coast1 where gravel is now accumulating in large quantities1#af an inclined surface1 thickly capped by well-rounded pebbles of about the same size1 would be ultimately left4 ,on the gravel now accumulating1 the waves1 aided by the wind1 sometimes throw up a thin covering of sand1 together with the common coast-shells4 ,shells thus cast up by gales1 would1 during an elevatory period1 never again be touched by the sea4 ,hence1 on this view of a slow and gradual rising of the land1 interrupted by periods of rest and denudation1 we can understand the pebbles being of about the same size over the entire width of the step-like plains1--the occasional thin covering of sandy earth1--and the presence of broken1 unrolled fragments of those shells1 which now live exclusively near the coast4 #af ,on the eastern side of ,chiloe1 which island we shall see in the next chapter is now rising1 ,i observed that all the beaches and extensive tidal-flats were formed of shingle4 ,summary of results4--,it may be concluded that the coast on this side of the continent1 for a space of at least #a1ahj miles1 has been elevated to a height of #ajj feet in ,la ,plata1 and of #djj feet in ,southern ,patagonia1 within the period of existing shells1 but not of existing mammifers4 ,that in ,la ,plata the elevation has been very slowly effected3 that in ,patagonia the movement may have been by considerable starts1 but much more probably slow and quiet4 ,in either case1 there have been long intervening periods of comparative rest1#ag during which the sea corroded deeply1 as it is still corroding1 into the land4 ,that the periods of denudation and elevation were contemporaneous and equable over great spaces of coast1 as shown by the equable heights of the plains2 that there have been at least eight periods of denudation1 and that the land1 up to a height of from #iej to #a1bjj feet1 has been similarly modelled and affected3 that the area elevated1 in the southernmost part of the continent1 extended in breadth to the ,cordillera1 and probably seaward to the ,falkland ,islands2 that northward1 in ,la ,plata1 the breadth is unknown1 there having been probably more than one axis of elevation2 and finally1 that1 anterior to the elevation attested by these upraised shells1 the land was divided by a ,strait where the ,river ,santa ,cruz now flows1 and that further southward there were other sea-straits1 since closed4 ,i may add1 that at ,santa ,cruz1 in lat4 #ej ,s41 the plains have been uplifted at least #a1djj feet1 since the period when gigantic boulders were transported between sixty and seventy miles from their parent rock1 on floating icebergs4 #ag ,i say comparative and not absolute rest1 because the sea acts1 as we have seen1 with great denuding power on this whole line of coast2 and therefore1 during an elevation of the land1 if excessively slow 7and of course during a subsidence of the land71 it is quite possible that lines of cliff might be formed47 ,lastly1 considering the great upward movements which this long line of coast has undergone1 and the proximity of its southern half to the volcanic axis of the ,cordillera1 it is highly remarkable that in the many fine sections exposed in the ,pampean1 ,patagonian tertiary1 and ,boulder formations1 ,i nowhere observed the smallest fault or abrupt curvature in the strata4 ,gravel ,formation of ,patagonia4 ,i will here describe in more detail than has been as yet incidentally done1 the nature1 origin1 and extent of the great shingle covering of ,patagonia3 but ,i do not mean to affirm that all of this shingle1 especially that on the higher plains1 belongs to the recent period4 ,a thin bed of sandy earth1 with small pebbles of various porphyries and of quartz1 covering a low plain on the north side of the ,rio ,colorado1 is the extreme northern limit of this formation4 ,these little pebbles have probably been derived from the denudation of a more regular bed of gravel1 capping the old tertiary sandstone plateau of the ,rio ,negro4 ,the gravel-bed near the ,rio ,negro is1 on an average1 about ten or twelve feet in thickness2 and the pebbles are larger than on the northern side of the ,colorado1 being from one or two inches in diameter1 and composed chiefly of rather dark-tinted porphyries4 ,amongst them ,i here first noticed a variety often to be referred to1 namely1 a peculiar gallstone-yellow siliceous porphyry1 frequently1 but not invariably1 containing grains of quartz4 ,the pebbles are embedded in a white1 gritty1 calcareous matrix1 very like mortar1 sometimes merely coating with a whitewash the separate stones1 and sometimes forming the greater part of the mass4 ,in one place ,i saw in the gravel concretionary nodules 7not rounded7 of crystallised gypsum1 some as large as a man's head4 ,i traced this bed for forty-five miles inland1 and was assured that it extended far into the interior4 ,as the surface of the calcareo-argillaceous plain of ,pampean formation1 on the northern side of the wide valley of the ,colorado1 stands at about the same height with the mortar-like cemented gravel capping the sandstone on the southern side1 it is probable1 considering the apparent equability of the subterranean movements along this side of ,america1 that this gravel of the ,rio ,negro and the upper beds of the ,pampean formation northward of the ,colorado1 are of nearly contemporaneous origin1 and that the calcareous matter has been derived from the same source4 ,southward of the ,rio ,negro1 the cliffs along the great bay of ,s4 ,antonio are capped with gravel3 at ,san ,josef1 ,i found that the pebbles closely resembled those on the plain of the ,rio ,negro1 but that they were not cemented by calcareous matter4 ,between ,san ,josef and ,port ,desire1 ,i was assured by the ,officers of the ,survey that the whole face of the country is coated with gravel4 ,at ,port ,desire and over a space of twenty-five miles inland1 on the three step-formed plains and in the valleys1 ,i everywhere passed over gravel which1 where thickest1 was between thirty and forty feet4 ,here1 as in other parts of ,patagonia1 the gravel1 or its sandy covering1 was1 as we have seen1 often strewed with recent marine shells4 ,the sandy covering sometimes fills up furrows in the gravel1 as does the gravel in the underlying tertiary formations4 ,the pebbles are frequently whitewashed and even cemented together by a peculiar1 white1 friable1 aluminous1 fusible substance1 which ,i believe is decomposed feldspar4 ,at ,port ,desire1 the gravel rested sometimes on the basal formation of porphyry1 and sometimes on the upper or the lower denuded tertiary strata4 ,it is remarkable that most of the porphyritic pebbles differ from those varieties of porphyry which occur here abundantly in situ4 ,the peculiar gallstone-yellow variety was common1 but less numerous than at ,port ,s4 ,julian1 where it formed nearly one-third of the mass of the gravel2 the remaining part there consisting of pale grey and greenish porphyries with many crystals of feldspar4 ,at ,port ,s4 ,julian1 ,i ascended one of the flat-topped hills1 the denuded remnant of the highest plain1 and found it1 at the height of #iej feet1 capped with the usual bed of gravel4 ,near the mouth of the ,santa ,cruz1 the bed of gravel on the #cee feet plain is from twenty to about thirty-five feet in thickness4 ,the pebbles vary from minute ones to the size of a hen's egg1 and even to that of half a man's head2 they consist of paler varieties of porphyry than those found further northward1 and there are fewer of the gallstone-yellow kind2 pebbles of compact black clay-slate were here first observed4 ,the gravel1 as we have seen1 covers the step-formed plains at the mouth1 head1 and on the sides of the great valley of the ,santa ,cruz4 ,at a distance of #aaj miles from the coast1 the plain has risen to the height of #a1daf feet above the sea2 and the gravel1 with the associated great boulder formation1 has attained a thickness of #bab feet4 ,the plain1 apparently with its usual gravel covering1 slopes up to the foot of the ,cordillera to the height of between #c1bjj and #c1cjj feet4 ,in ascending the valley1 the gravel gradually becomes entirely altered in character3 high up1 we have pebbles of crystalline feldspathic rocks1 compact clay-slate1 quartzose schists1 and pale-coloured porphyries2 these rocks1 judging both from the gigantic boulders in the surface and from some small pebbles embedded beneath #gjj feet in thickness of the old tertiary strata1 are the prevailing kinds in this part of the ,cordillera2 pebbles of basalt from the neighbouring streams of basaltic lava are also numerous2 there are few or none of the reddish or of the gallstone-yellow porphyries so common near the coast4 ,hence the pebbles on the #cej feet plain at the mouth of the ,santa ,cruz cannot have been derived 7with the exception of those of compact clay-slate1 which1 however1 may equally well have come from the south7 from the ,cordillera in this latitude2 but probably1 in chief part1 from farther north4 ,southward of the ,santa ,cruz1 the gravel may be seen continuously capping the great #hdj feet plain3 at the ,rio ,gallegos1 where this plain is succeeded by a lower one1 there is1 as ,i am informed by ,captain ,sulivan1 an irregular covering of gravel from ten to twelve feet in thickness over the whole country4 ,the district on each side of the ,strait of ,magellan is covered up either with gravel or the boulder formation3 it was interesting to observe the marked difference between the perfectly rounded state of the pebbles in the great shingle formation of ,patagonia1 and the more or less angular fragments in the boulder formation4 ,the pebbles and fragments near the ,strait of ,magellan nearly all belong to rocks known to occur in ,fuegia4 ,i was therefore much surprised in dredging south of the ,strait to find1 in lat4 #ed #aj south1 many pebbles of the gallstone-yellow siliceous porphyry2 ,i procured others from a great depth off ,staten ,island1 and others were brought me from the western extremity of the ,falkland ,islands4#ah ,the distribution of the pebbles of this peculiar porphyry1 which ,i venture to affirm is not found in situ either in ,fuegia1 the ,falkland ,islands1 or on the coast of ,patagonia1 is very remarkable1 for they are found over a space of #hdj miles in a north and south line1 and at the ,falklands1 #cjj miles eastward of the coast of ,patagonia4 ,their occurrence in ,fuegia and the ,falklands may1 however1 perhaps be due to the same ice-agency by which the boulders have been there transported4 #ah ,at my request1 ,mr4 ,kent collected for me a bag of pebbles from the beach of ,white ,rock harbour1 in the northern part of the sound1 between the two ,falkland ,islands4 ,out of these well-rounded pebbles1 varying in size from a walnut to a hen's egg1 with some larger1 thirty-eight evidently belonged to the rocks of these islands2 twenty-six were similar to the pebbles of porphyry found on the ,patagonian plains1 which rocks do not exist in situ in the ,falklands2 one pebble belonged to the peculiar yellow siliceous porphyry2 thirty were of doubtful origin4 ,we have seen that porphyritic pebbles of a small size are first met with on the northern side of the ,rio ,colorado1 the bed becoming well developed near the ,rio ,negro3 from this latter point ,i have every reason to believe that the gravel extends uninterruptedly over the plains and valleys of ,patagonia for at least #fcj nautical miles southward to the ,rio ,gallegos4 ,from the slope of the plains1 from the nature of the pebbles1 from their extension at the ,rio ,negro far into the interior1 and at the ,santa ,cruz close up to the ,cordillera1 ,i think it highly probable that the whole breadth of ,patagonia is thus covered4 ,if so1 the average width of the bed must be about two hundred miles4 ,near the coast the gravel is generally from ten to thirty feet in thickness2 and as in the valley of ,santa ,cruz it attains1 at some distance from the ,cordillera1 a thickness of #bad feet1 we may1 ,i think1 safely assume its average thickness over the whole area of #fcj by #bjj miles1 at fifty feet6 ,the transportal and origin of this vast bed of pebbles is an interesting problem4 ,from the manner in which they cap the step-formed plains1 worn by the sea within the period of existing shells1 their deposition1 at least on the plains up to a height of #djj feet1 must have been a recent geological event4 ,from the form of the continent1 we may feel sure that they have come from the westward1 probably1 in chief part from the ,cordillera1 but1 perhaps1 partly from unknown rocky ridges in the central districts of ,patagonia4 ,that the pebbles have not been transported by rivers1 from the interior towards the coast1 we may conclude from the fewness and smallness of the streams of ,patagonia3 moreover1 in the case of the one great and rapid river of ,santa ,cruz1 we have good evidence that its transporting power is very trifling4 ,this river is from two to three hundred yards in width1 about seventeen feet deep in its middle1 and runs with a singular degree of uniformity five knots an hour1 with no lakes and scarcely any still reaches3 nevertheless1 to give one instance of its small transporting power1 upon careful examination1 pebbles of compact basalt could not be found in the bed of the river at a greater distance than ten miles below the point where the stream rushes over the debris of the great basaltic cliffs forming its shore3 fragments of the cellular varieties have been washed down twice or thrice as far4 ,that the pebbles in ,central and ,northern ,patagonia have not been transported by ice-agency1 as seems to have been the case to a considerable extent farther south1 and likewise in the northern hemisphere1 we may conclude1 from the absence of all angular fragments in the gravel1 and from the complete contrast in many other respects between the shingle and neighbouring boulder formation4 ,looking to the gravel on any one of the step-formed plains1 ,i cannot doubt1 from the several reasons assigned in this chapter1 that it has been spread out and leveled by the long-continued action of the sea1 probably during the slow rise of the land4 ,the smooth and perfectly rounded condition of the innumerable pebbles alone would prove long-continued action4 ,but how the whole mass of shingle on the coast-plains has been transported from the mountains of the interior1 is another and more difficult question4 ,the following considerations1 however1 show that the sea by its ordinary action has considerable power in distributing pebbles4 ,a table has already been given1 showing how very uniformly and gradually#ai the pebbles decrease in size with the gradually seaward increasing depth and distance4 ,a series of this kind irresistibly leads to the conclusion1 that the sea has the power of sifting and distributing the loose matter on its bottom4 ,according to ,martin ,white1#bj the bed of the ,british ,channel is disturbed during gales at depths of sixty-three and sixty-seven fathoms1 and at thirty fathoms1 shingle and fragments of shells are often deposited1 afterwards to be carried away again4 ,groundswells1 which are believed to be caused by distant gales1 seem especially to affect the bottom3 at such times1 according to ,sir ,r4 ,schomburgk1#ba the sea to a great distance round the ,west ,indian ,islands1 at depths from five to fifteen fathoms1 becomes discoloured1 and even the anchors of vessels have been moved4 ,there are1 however1 some difficulties in understanding how the sea can transport pebbles lying at the bottom1 for1 from experiments instituted on the power of running water1 it would appear that the currents of the sea have not sufficient velocity to move stones of even moderate size3 moreover1 ,i have repeatedly found in the most exposed situations that the pebbles which lie at the bottom are encrusted with full-grown living corallines1 furnished with the most delicate1 yet unbroken spines3 for instance1 in ten fathoms water off the mouth of the ,santa ,cruz1 many pebbles1 under half an inch in diameter1 were thus coated with ,flustracean zoophytes4#bb ,hence we must conclude that these pebbles are not often violently disturbed3 it should1 however1 be borne in mind that the growth of corallines is rapid4 ,the view1 propounded by ,professor ,playfair1 will1 ,i believe1 explain this apparent difficulty1--namely1 that from the undulations of the sea tending to lift up and down pebbles or other loose bodies at the bottom1 such are liable1 when thus quite or partially raised1 to be moved even by a very small force1 a little onwards4 ,we can thus understand how oceanic or tidal currents of no great strength1 or that recoil movement of the bottom-water near the land1 called by sailors the 8undertow0 7which ,i presume must extend out seaward as far as the breaking waves impel the surface-water towards the beach71 may gain the power during storms of sifting and distributing pebbles even of considerable size1 and yet without so violently disturbing them as to injure the encrusting corallines4#bc #ai ,i may mention1 that at the distance of #aej miles from the ,patagonian shore ,i carefully examined the minute rounded particles in the sand1 and found them to be fusible like the porphyries of the great shingle bed4 ,i could even distinguish particles of the gallstone-yellow porphyry4 ,it was interesting to notice how gradually the particles of white quartz increased1 as we approached the ,falkland ,islands1 which are thus constituted4 ,in the whole line of soundings between these islands and the coast of ,patagonia dead or living organic remains were most rare4 ,on the relations between the depth of water and the nature of the bottom1 see ,martin ,white on 8,soundings in the ,channel10 pp4 #d1 #f1 #age2 also ,captain ,beechey's 8,voyage to the ,pacific10 chap4 xviii4 #bj 8,soundings in the ,channel10 pp4 #d1 #aff4 ,m4 ,siau states 7,edin4 ,new ,phil4 ,jour41 vol4 xxxi1 p4 #bdf71 that he found the sediment1 at a depth of #ahh metres1 arranged in ripples of different degrees of fineness4 ,there are some excellent discussions on this and allied subjects in ,sir ,h4 ,de la ,beche's 8,theoretical ,researches40 #ba ,journal of ,royal ,geograph4 ,soc41 vol4 v1 p4 #be4 ,it appears from ,mr4 ,scott ,russell's investigations 7see ,mr4 ,murchison's 8,anniver4 ,address ,geolog4 ,soc410 #ahdc1 p4 #dj71 that in waves of translation the motion of the particles of water is nearly as great at the bottom as at the top4 #bb 7,a pebble1 one and a half inch square and half an inch thick1 was given me1 dredged up from twenty-seven fathoms depth off the western end of the ,falkland ,islands1 where the sea is remarkably stormy1 and subject to violent tides4 ,this pebble was encrusted on all sides by a delicate living coralline4 ,i have seen many pebbles from depths between forty and seventy fathoms thus encrusted2 one from the latter depth off ,cape ,horn4 #bc ,i may take this opportunity of remarking on a singular1 but very common character in the form of the bottom1 in the creeks which deeply penetrate the western shores of ,tierra del ,fuego2 namely1 that they are almost invariably much shallower close to the open sea at their mouths than inland4 ,thus1 ,cook1 in entering ,christmas ,sound1 first had soundings in thirty-seven fathoms1 then in fifty1 then in sixty1 and a little farther in no bottom with #agj fathoms4 ,the sealers are so familiar with this fact1 that they always look out for anchorage near the entrances of the creeks4 ,see1 also1 on this subject1 the 8,voyages of the ,adventure and ,beagle10 vol4 i1 p4 #cge and 8,appendix10 p4 #cac4 ,this ,shoalness of the sea-channels near their entrances probably results from the quantity of sediment formed by the wear and tear of the outer rocks exposed to the full force of the open sea4 ,i have no doubt that many lakes1 for instance in ,scotland1 which are very deep within1 and are separated from the sea apparently only by a tract of detritus1 were originally sea-channels with banks of this nature near their mouths1 which have since been upheaved4 ,the sea acts in another and distinct manner in the distribution of pebbles1 namely by the waves on the beach4 ,mr4 ,palmer1#bd in his excellent memoir on this subject1 has shown that vast masses of shingle travel with surprising quickness along lines of coast1 according to the direction with which the waves break on the beach and that this is determined by the prevailing direction of the winds4 ,this agency must be powerful in mingling together and disseminating pebbles derived from different sources3 we may1 perhaps1 thus understand the wide distribution of the gallstone-yellow porphyry2 and likewise1 perhaps1 the great difference in the nature of the pebbles at the mouth of the ,santa ,cruz from those in the same latitude at the head of the valley4 #bd 8,philosophical ,transactions10 #ahcd1 p4 #egf4 ,i will not pretend to assign to these several and complicated agencies their shares in the distribution of the ,patagonian shingle3 but from the several considerations given in this chapter1 and ,i may add1 from the frequency of a capping of gravel on tertiary deposits in all parts of the world1 as ,i have myself observed and seen stated in the works of various authors1 ,i cannot doubt that the power of widely dispersing gravel is an ordinary contingent on the action of the sea2 and that even in the case of the great ,patagonian shingle-bed we have no occasion to call in the aid of debacles4 ,i at one time imagined that perhaps an immense accumulation of shingle had originally been collected at the foot of the ,cordillera2 and that this accumulation1 when upraised above the level of the sea1 had been eaten into and partially spread out 7as off the present line of coast72 and that the newly-spread out bed had in its turn been upraised1 eaten into1 and re-spread out2 and so onwards1 until the shingle1 which was first accumulated in great thickness at the foot of the ,cordillera1 had reached in thinner beds its present extension4 ,by whatever means the gravel formation of ,patagonia may have been distributed1 the vastness of its area1 its thickness1 its superficial position1 its recent origin1 and the great degree of similarity in the nature of its pebbles1 all appear to me well deserving the attention of geologists1 in relation to the origin of the widely-spread beds of conglomerate belonging to past epochs4 ,no4 #g ,section of coast-cliffs and bottom of sea1 off the island of ,st4 ,helena4 ,illustration3 ,section of clast-cliffs and bottom of sea1 off the island of ,st4 ,helena4 ,formation of ,cliffs4--,when viewing the sea-worn cliffs of ,patagonia1 in some parts between eight hundred and nine hundred feet in height1 and formed of horizontal tertiary strata1 which must once have extended far seaward--or again1 when viewing the lofty cliffs round many volcanic islands1 in which the gentle inclination of the lava-streams indicates the former extension of the land1 a difficulty often occurred to me1 namely1 how the strata could possibly have been removed by the action of the sea at a considerable depth beneath its surface4 ,the section in diagram ,no4 #g1 which represents the general form of the land on the northern and leeward side of ,st4 ,helena 7taken from ,mr4 ,seale's large model and various measurements71 and of the bottom of the adjoining sea 7taken chiefly from ,captain ,austin's survey and some old charts71 will show the nature of this difficulty4 ,if1 as seems probable1 the basaltic streams were originally prolonged with nearly their present inclination1 they must1 as shown by the dotted line in the section1 once have extended at least to a point1 now covered by the sea to a depth of nearly thirty fathoms3 but ,i have every reason to believe they extended considerably further1 for the inclination of the streams is less near the coast than further inland4 ,it should also be observed1 that other sections on the coast of this island would have given far more striking results1 but ,i had not the exact measurements2 thus1 on the windward side1 the cliffs are about two thousand feet in height and the cut-off lava streams very gently inclined1 and the bottom of the sea has nearly a similar slope all round the island4 ,how1 then1 has all the hard basaltic rock1 which once extended beneath the surface of the sea1 been worn away8 ,according to ,captain ,austin1 the bottom is uneven and rocky only to that very small distance from the beach within which the depth is from five to six fathoms2 outside this line1 to a depth of about one hundred fathoms1 the bottom is smooth1 gently inclined1 and formed of mud and sand2 outside the one hundred fathoms1 it plunges suddenly into unfathomable depths1 as is so very commonly the case on all coasts where sediment is accumulating4 ,at greater depths than the five or six fathoms1 it seems impossible1 under existing circumstances1 that the sea can both have worn away hard rock1 in parts to a thickness of at least #aej feet1 and have deposited a smooth bed of fine sediment4 ,now1 if we had any reason to suppose that ,st4 ,helena had1 during a long period1 gone on slowly subsiding1 every difficulty would be removed3 for looking at the diagram1 and imagining a fresh amount of subsidence1 we can see that the waves would then act on the coast-cliffs with fresh and unimpaired vigour1 whilst the rocky ledge near the beach would be carried down to that depth1 at which sand and mud would be deposited on its bare and uneven surface3 after the formation near the shore of a new rocky shoal1 fresh subsidence would carry it down and allow it to be smoothly covered up4 ,but in the case of the many cliff-bounded islands1 for instance in some of the ,canary ,islands and of ,madeira1 round which the inclination of the strata shows that the land once extended far into the depths of the sea1 where there is no apparent means of hard rock being worn away--are we to suppose that all these islands have slowly subsided8 ,madeira1 ,i may remark1 has1 according to ,mr4 ,smith of ,jordan ,hill1 subsided4 ,are we to extend this conclusion to the high1 cliff-bound1 horizontally stratified shores of ,patagonia1 off which1 though the water is not deep even at the distance of several miles1 yet the smooth bottom of pebbles gradually decreasing in size with the increasing depth1 and derived from a foreign source1 seem to declare that the sea is now a depositing and not a corroding agent8 ,i am much inclined to suspect1 that we shall hereafter find in all such cases1 that the land with the adjoining bed of the sea has in truth subsided3 the time will1 ,i believe1 come1 when geologists will consider it as improbable1 that the land should have retained the same level during a whole geological period1 as that the atmosphere should have remained absolutely calm during an entire season4 ,chapter ,,ii ,,on ,,the ,,elevation ,,of ,,the ,,western ,,coast ,,of ,,south ,,america4 ,chonos ,archipelago4--,chiloe1 recent and gradual elevation of1 traditions of the inhabitants on this subject4--,concepcion1 earthquake and elevation of4--,,valparaiso1 great elevation of1 upraised shells1 earth of marine origin1 gradual rise of the land within the historical period4--,,coquimbo1 elevation of1 in recent times2 terraces of marine origin1 their inclination1 their escarpments not horizontal4--,guasco1 gravel terraces of4--,copiapo4--,,peru4--,upraised shells of ,cobija1 ,iquique1 and ,arica4--,lima1 shell-beds and sea-beach on ,san ,lorenzo1 human remains1 fossil earthenware1 earthquake debacle1 recent subsidence4 ,on the decay of upraised shells4--,general summary4 ,commencing at the south and proceeding northward1 the first place at which ,i landed1 was at ,cape ,tres ,montes1 in lat4 #df #ce4 ,here1 on the shores of ,christmas ,cove1 ,i observed in several places a beach of pebbles with recent shells1 about twenty feet above high-water mark4 ,southward of ,tres ,montes 7between lat4 #dg and #dh71 ,byron#a remarks1 8,we thought it very strange1 that upon the summits of the highest hills were found beds of shells1 a foot or two thick40 ,in the ,chonos ,archipelago1 the island of ,lemus 7lat4 #dd #cj7 was1 according to ,m4 ,coste1#b suddenly elevated eight feet1 during the earthquake of #ahbi3 he adds1 8,des roches jadis toujours couvertes par la mer1 restant aujourd'hui constamment decouvertes40 ,in other parts of this archipelago1 ,i observed two terraces of gravel1 abutting to the foot of each other3 at ,lowe's ,harbour 7#dc #dh71 under a great mass of the boulder formation1 about three hundred feet in thickness1 ,i found a layer of sand1 with numerous comminuted fragments of sea-shells1 having a fresh aspect1 but too small to be identified4 #a 8,narrative of the ,loss of the ,wager40 #b 8,comptes ,rendus10 ,october #ahch1 p4 #gjf4 ,the ,island of ,chiloe4--,the evidence of recent elevation is here more satisfactory4 ,the bay of ,san ,carlos is in most parts bounded by precipitous cliffs from about ten to forty feet in height1 their bases being separated from the present line of tidal action by a talus1 a few feet in height1 covered with vegetation4 ,in one sheltered creek 7west of ,p4 ,arena71 instead of a loose talus1 there was a bare sloping bank of tertiary mudstone1 perforated1 above the line of the highest tides1 by numerous shells of a ,pholas now common in the harbour4 ,the upper extremities of these shells1 standing upright in their holes with grass growing out of them1 were abraded about a quarter of an inch1 to the same level with the surrounding worn strata4 ,in other parts1 ,i observed 7as at ,pudeto7 a great beach1 formed of comminuted shells1 twenty feet above the present shore4 ,in other parts again1 there were small caves worn into the foot of the low cliffs1 and protected from the waves by the talus with its vegetation3 one such cave1 which ,i examined1 had its mouth about twenty feet1 and its bottom1 which was filled with sand containing fragments of shells and legs of crabs1 from eight to ten feet above high-water mark4 ,from these several facts1 and from the appearance of the upraised shells1 ,i inferred that the elevation had been quite recent2 and on inquiring from ,mr4 ,williams1 the ,portmaster1 he told me he was convinced that the land had risen1 or the sea fallen1 four feet within the last four years4 ,during this period1 there had been one severe earthquake1 but no particular change of level was then observed2 from the habits of the people who all keep boats in the protected creeks1 it is absolutely impossible that a rise of four feet could have taken place suddenly and been unperceived4 ,mr4 ,williams believes that the change has been quite gradual4 ,without the elevatory movement continues at a quick rate1 there can be no doubt that the sea will soon destroy the talus of earth at the foot of the cliffs round the bay1 and will then reach its former lateral extension1 but not of course its former level3 some of the inhabitants assured me that one such talus1 with a footpath on it1 was even already sensibly decreasing in width4 ,i received several accounts of beds of shells1 existing at considerable heights in the inland parts of ,chiloe2 and to one of these1 near ,catiman1 ,i was guided by a countryman4 ,here1 on the south side of the peninsula of ,lacuy1 there was an immense bed of the ,venus costellata and of an oyster1 lying on the summit-edge of a piece of tableland1 #cej feet 7by the barometer7 above the level of the sea4 ,the shells were closely packed together1 embedded in and covered by a very black1 damp1 peaty mould1 two or three feet in thickness1 out of which a forest of great trees was growing4 ,considering the nature and dampness of this peaty soil1 it is surprising that the fine ridges on the outside of the ,venus are perfectly preserved1 though all the shells have a blackened appearance4 ,i did not doubt that the black soil1 which when dry1 cakes hard1 was entirely of terrestrial origin1 but on examining it under the microscope1 ,i found many very minute rounded fragments of shells1 amongst which ,i could distinguish bits of ,serpul and mussels4 ,the ,venus costellata1 and the ,ostrea 7,o4 edulis1 according to ,captain ,king7 are now the commonest shells in the adjoining bays4 ,in a bed of shells1 a few feet below the #cej feet bed1 ,i found a horn of the little ,cervus humilis1 which now inhabits ,chiloe4 ,the eastern or inland side of ,chiloe1 with its many adjacent islets1 consists of tertiary and boulder deposits1 worn into irregular plains capped by gravel4 ,near ,castro1 and for ten miles southward1 and on the islet of ,lemuy1 ,i found the surface of the ground to a height of between twenty and thirty feet above high-water mark1 and in several places apparently up to fifty feet1 thickly coated by much comminuted shells1 chiefly of the ,venus costellata and ,mytilus ,chiloensis2 the species now most abundant on this line of coast4 ,as the inhabitants carry immense numbers of these shells inland1 the continuity of the bed at the same height was often the only means of recognising its natural origin4 ,near ,castro1 on each side of the creek and rivulet of the ,gamboa1 three distinct terraces are seen3 the lowest was estimated at about one hundred and fifty feet in height1 and the highest at about five hundred feet1 with the country irregularly rising behind it2 obscure traces1 also1 of these same terraces could be seen along other parts of the coast4 ,there can be no doubt that their three escarpments record pauses in the elevation of the island4 ,i may remark that several promontories have the word ,huapi1 which signifies in the ,indian tongue1 island1 appended to them1 such as ,huapilinao1 ,huapilacuy1 ,caucahuapi1 etc42 and these1 according to ,indian traditions1 once existed as islands4 ,in the same manner the term ,pulo in ,sumatra is appended#c to the names of promontories1 traditionally said to have been islands2 in ,sumatra1 as in ,chiloe1 there are upraised recent shells4 ,the ,bay of ,carelmapu1 on the mainland north of ,chiloe1 according to ,agerros1#d was in #afdc a good harbour2 it is now quite useless1 except for boats4 #c ,marsden's 8,sumatra10 p4 #ca4 #d 8,descripcion ,hist4 de la ,provincia de ,chilo10 p4 #gh4 ,from the account given by the old ,spanish writers1 it would appear that several other harbours1 between this point and ,concepcion1 were formerly much deeper than they now are4 ,valdivia4--,i did not observe here any distinct proofs of recent elevation2 but in a bed of very soft sandstone1 forming a fringe-like plain1 about sixty feet in height1 round the hills of mica-slate1 there are shells of ,mytilus1 ,crepidula1 ,solen1 ,novaculina1 and ,cythera1 too imperfect to be specifically recognised4 ,at ,imperial1 seventy miles north of ,valdivia1 ,agerros#e states that there are large beds of shells1 at a considerable distance from the coast1 which are burnt for lime4 ,the island of ,mocha1 lying a little north of ,imperial1 was uplifted two feet1#f during the earthquake of #ahce4 #e ,ibid41 p4 #be4 #f 8,voyages of ,adventure and ,beagle10 vol4 ii1 p4 #dae4 ,concepcion4--,i cannot add anything to the excellent account by ,captain ,fitzroy#g of the elevation of the land at this place1 which accompanied the earthquake of #ahce4 ,i will only recall to the recollection of geologists1 that the southern end of the island of ,st4 ,mary was uplifted eight feet1 the central part nine1 and the northern end ten feet2 and the whole island more than the surrounding districts4 ,great beds of mussels1 patell1 and chitons still adhering to the rocks were upraised above high-water mark2 and some acres of a rocky flat1 which was formerly always covered by the sea1 was left standing dry1 and exhaled an offensive smell1 from the many attached and putrefying shells4 ,it appears from the researches of ,captain ,fitzroy that both the island of ,st4 ,mary and ,concepcion 7which was uplifted only four or five feet7 in the course of some weeks subsided1 and lost part of their first elevation4 ,i will only add as a lesson of caution1 that round the sandy shores of the great ,bay of ,concepcion1 it was most difficult1 owing to the obliterating effects of the great accompanying wave1 to recognise any distinct evidence of this considerable upheaval2 one spot must be excepted1 where there was a detached rock which before the earthquake had always been covered by the sea1 but afterwards was left uncovered4 #g ,ibid41 vol4 ii1 p4 #dab1 et seq4 ,in vol4 v1 p4 #fja of the 8,geological ,transactions0 ,i have given an account of the remarkable volcanic phenomena1 which accompanied this earthquake4 ,these phenomena appear to me to prove that the action1 by which large tracts of land are uplifted1 and by which volcanic eruptions are produced1 is in every respect identical4 ,on the island of ,quiriquina 7in the ,bay of ,concepcion71 ,i found1 at an estimated height of four hundred feet1 extensive layers of shells1 mostly comminuted1 but some perfectly preserved and closely packed in black vegetable mould2 they consisted of ,concholepas1 ,fissurella1 ,mytilus1 ,trochus1 and ,balanus4 ,some of these layers of shells rested on a thick bed of bright-red1 dry1 friable earth1 capping the surface of the tertiary sandstone1 and extending1 as ,i observed whilst sailing along the coast1 for #aej miles southward3 at ,valparaiso1 we shall presently see that a similar red earthy mass1 though quite like terrestrial mould1 is really in chief part of recent marine origin4 ,on the flanks of this island of ,quiriquina1 at a less height than the #djj feet1 there were spaces several feet square1 thickly strewed with fragments of similar shells4 ,during a subsequent visit of the ,beagle to ,concepcion1 ,mr4 ,kent1 the assistant-surgeon1 was so kind as to make for me some measurements with the barometer3 he found many marine remains along the shores of the whole bay1 at a height of about twenty feet2 and from the hill of ,sentinella behind ,talcahuano1 at the height of #afj feet1 he collected numerous shells1 packed together close beneath the surface in black earth1 consisting of two species of ,mytilus1 two of ,crepidula1 one of ,concholepas1 of ,fissurella1 ,venus1 ,mactra1 ,turbo1 ,monoceros1 and the ,balanus psittacus4 ,these shells were bleached1 and within some of the ,balani other ,balani were growing1 showing that they must have long lain dead in the sea4 ,the above species ,i compared with living ones from the bay1 and found them identical2 but having since lost the specimens1 ,i cannot give their names3 this is of little importance1 as ,mr4 ,broderip has examined a similar collection1 made during ,captain ,beechey's expedition1 and ascertained that they consisted of ten recent species1 associated with fragments of ,echini1 crabs1 and ,flustr2 some of these remains were estimated by ,lieutenant ,belcher to lie at the height of nearly a thousand feet above the level of the sea4#h ,in some places round the bay1 ,mr4 ,kent observed that there were beds formed exclusively of the ,mytilus ,chiloensis3 this species now lives in parts never uncovered by the tides4 ,at considerable heights1 ,mr4 ,kent found only a few shells2 but from the summit of one hill1 #fbe feet high1 he brought me specimens of the ,concholepas1 ,mytilus ,chiloensis1 and a ,turbo4 ,these shells were softer and more brittle than those from the height of #afd feet2 and these latter had obviously a much more ancient appearance than the same species from the height of only twenty feet4 #h 8,zoology of ,captain ,beechey's ,voyage10 p4 #afb4 ,coast north of ,concepcion4--,the first point examined was at the mouth of the ,rapel 7#afj miles north of ,concepcion and sixty miles south of ,valparaiso71 where ,i observed a few shells at the height of #ajj feet1 and some barnacles adhering to the rocks three or four feet above the highest tides3 ,m4 ,gay#i found here recent shells at the distance of two leagues from the shore4 ,inland there are some wide1 gravel-capped plains1 intersected by many broad1 flat-bottomed valleys 7now carrying insignificant streamlets71 with their sides cut into successive wall-like escarpments1 rising one above another1 and in many places1 according to ,m4 ,gay1 worn into caves4 ,the one cave 7,c4 del ,obispo7 which ,i examined1 resembled those formed on many sea-coasts1 with its bottom filled with shingle4 ,these inland plains1 instead of sloping towards the coast1 are inclined in an opposite direction towards the ,cordillera1 like the successively rising terraces on the inland or eastern side of ,chiloe3 some points of granite1 which project through the plains near the coast1 no doubt once formed a chain of outlying islands1 on the inland shores of which the plains were accumulated4 ,at ,bucalemu1 a few miles northward of the ,rapel1 ,i observed at the foot1 and on the summit-edge of a plain1 ten miles from the coast1 many recent shells1 mostly comminuted1 but some perfect4 ,there were1 also1 many at the bottom of the great valley of the ,maypu4 ,at ,san ,antonio1 shells are said to be collected and burnt for lime4 ,at the bottom of a great ravine 7,quebrada ,onda1 on the road to ,casa ,blanca71 at the distance of several miles from the coast1 ,i noticed a considerable bed1 composed exclusively of ,mesodesma donaciforme1 ,desh41 lying on a bed of muddy sand3 this shell now lives associated together in great numbers1 on tidal-flats on the coast of ,chile4 #i 8,annales des ,scienc4 ,nat410 ,avril #ahcc4 ,valparaiso4 ,during two successive years ,i carefully examined1 part of the time in company with ,mr4 ,alison1 into all the facts connected with the recent elevation of this neighbourhood4 ,in very many parts a beach of broken shells1 about fourteen or fifteen feet above high-water mark1 may be observed2 and at this level the coast-rocks1 where precipitous1 are corroded in a band4 ,at one spot1 ,mr4 ,alison1 by removing some birds' dung1 found at this same level barnacles adhering to the rocks4 ,for several miles southward of the bay1 almost every flat little headland1 between the heights of #fj and #bcj feet 7measured by the barometer71 is smoothly coated by a thick mass of comminuted shells1 of the same species1 and apparently in the same proportional numbers with those existing in the adjoining sea4 ,the ,concholepas is much the most abundant1 and the best preserved shell2 but ,i extracted perfectly preserved specimens of the ,fissurella biradiata1 a ,trochus and ,balanus 7both well-known1 but according to ,mr4 ,sowerby yet unnamed7 and parts of the ,mytilus ,chiloensis4 ,most of these shells1 as well as an encrusting ,nullipora1 partially retain their colour2 but they are brittle1 and often stained red from the underlying brecciated mass of primary rocks2 some are packed together1 either in black or reddish moulds2 some lie loose on the bare rocky surfaces4 ,the total number of these shells is immense2 they are less numerous1 though still far from rare1 up a height of #a1jjj feet above the sea4 ,on the summit of a hill1 measured #eeg feet1 there was a small horizontal band of comminuted shells1 of which many consisted 7and likewise from lesser heights7 of very young and small specimens of the still living ,concholepas1 ,trochus1 ,patell1 ,crepidul1 and of ,mytilus ,magellanicus 7873#aj several of these shells were under a quarter of an inch in their greatest diameter4 ,my attention was called to this circumstance by a native fisherman1 whom ,i took to look at these shell-beds2 and he ridiculed the notion of such small shells having been brought up for food2 nor could some of the species have adhered when alive to other larger shells4 ,on another hill1 some miles distant1 and #fdh feet high1 ,i found shells of the ,concholepas and ,trochus1 perfect1 though very old1 with fragments of ,mytilus ,chiloensis1 all embedded in reddish-brown mould3 ,i also found these same species1 with fragments of an ,echinus and of ,balanus psittacus1 on a hill #a1jjj feet high4 ,above this height1 shells became very rare1 though on a hill #a1cjj feet high1#aa ,i collected the ,concholepas1 ,trochus1 ,fissurella1 and a ,patella4 ,at these greater heights the shells are almost invariably embedded in mould1 and sometimes are exposed only by tearing up bushes4 ,these shells obviously had a very much more ancient appearance than those from the lesser heights2 the apices of the ,trochi were often worn down2 the little holes made by burrowing animals were greatly enlarged2 and the ,concholepas was often perforated quite through1 owing to the inner plates of shell having scaled off4 #aj ,mr4 ,cuming informs me that he does not think this species identical with1 though closely resembling1 the true ,m4 ,magellanicus of the southern and eastern coast of ,south ,america2 it lives abundantly on the coast of ,chile4 #aa ,measured by the barometer3 the highest point in the range behind ,valparaiso ,i found to be #a1fbf feet above the level of the sea4 ,many of these shells1 as ,i have said1 were packed in1 and were quite filled with1 blackish or reddish-brown earth1 resting on the granitic detritus4 ,i did not doubt until lately that this mould was of purely terrestrial origin1 when with a microscope examining some of it from the inside of a ,concholepas from the height of about one hundred feet1 ,i found that it was in considerable part composed of minute fragments of the spines1 mouth-bones1 and shells of ,echini1 and of minute fragments1 of chiefly very young ,patell1 ,mytili1 and other species4 ,i found similar microscopical fragments in earth filling up the central orifices of some large ,fissurell4 ,this earth when crushed emits a sickly smell1 precisely like that from garden-mould mixed with guano4 ,the earth accidentally preserved within the shells1 from the greater heights1 has the same general appearance1 but it is a little redder2 it emits the same smell when rubbed1 but ,i was unable to detect with certainty any marine remains in it4 ,this earth resembles in general appearance1 as before remarked1 that capping the rocks of ,quiriquina in the ,bay of ,concepcion1 on which beds of sea-shells lay4 ,i have1 also1 shown that the black1 peaty soil1 in which the shells at the height of #cej feet at ,chiloe were packed1 contained many minute fragments of marine animals4 ,these facts appear to me interesting1 as they show that soils1 which would naturally be considered of purely terrestrial nature1 may owe their origin in chief part to the sea4 ,being well aware from what ,i have seen at ,chiloe and in ,tierra del ,fuego1 that vast quantities of shells are carried1 during successive ages1 far inland1 where the inhabitants chiefly subsist on these productions1 ,i am bound to state that at greater heights than #eeg feet1 where the number of very young and small shells proved that they had not been carried up for food1 the only evidence of the shells having been naturally left by the sea1 consists in their invariable and uniform appearance of extreme antiquity--in the distance of some of the places from the coast1 in others being inaccessible from the nearest part of the beach1 and in the absence of fresh water for men to drink--in the shells not lying in heaps1--and1 lastly1 in the close similarity of the soil in which they are embedded1 to that which lower down can be unequivocally shown to be in great part formed from the debris of the sea animals4#ab #ab ,in the 8,proceedings of the ,geolog4 ,soc410 vol4 ii1 p4 #ddf1 ,i have given a brief account of the upraised shells on the coast of ,chile1 and have there stated that the proofs of elevation are not satisfactory above the height of #bcj feet4 ,i had at that time unfortunately overlooked a separate page written during my second visit to ,valparaiso1 describing the shells now in my possession from the #eeg feet hill2 ,i had not then unpacked my collections1 and had not reconsidered the obvious appearance of greater antiquity of the shells from the greater heights1 nor had ,i at that time discovered the marine origin of the earth in which many of the shells are packed4 ,considering these facts1 ,i do not now feel a shadow of doubt that the shells1 at the height of #a1cjj feet1 have been upraised by natural causes into their present position4 ,with respect to the position in which the shells lie1 ,i was repeatedly struck here1 at ,concepcion1 and at other places1 with the frequency of their occurrence on the summits and edges either of separate hills1 or of little flat headlands often terminating precipitously over the sea4 ,the several above-enumerated species of mollusca1 which are found strewed on the surface of the land from a few feet above the level of the sea up to the height of #a1cjj feet1 all now live either on the beach1 or at only a few fathoms' depth3 ,mr4 ,edmondston1 in a letter to ,professor ,e4 ,forbes1 states that in dredging in the ,bay of ,valparaiso1 he found the common species of ,concholepas1 ,fissurella1 ,trochus1 ,monoceros1 ,chitons1 etc41 living in abundance from the beach to a depth of seven fathoms2 and dead shells occurred only a few fathoms deeper4 ,the common ,turritella cingulata was dredged up living at even from ten to fifteen fathoms2 but this is a species which ,i did not find here amongst the upraised shells4 ,considering this fact of the species being all littoral or sub-littoral1 considering their occurrence at various heights1 their vast numbers1 and their generally comminuted state1 there can be little doubt that they were left on successive beach-lines during a gradual elevation of the land4 ,the presence1 however1 of so many whole and perfectly preserved shells appears at first a difficulty on this view1 considering that the coast is exposed to the full force of an open ocean3 but we may suppose1 either that these shells were thrown during gales on flat ledges of rock just above the level of high-water mark1 and that during the elevation of the land they are never again touched by the waves1 or1 that during earthquakes1 such as those of #ahbb1 #ahce1 and #ahcg1 rocky reefs covered with marine-animals were it one blow uplifted above the future reach of the sea4 ,this latter explanation is1 perhaps1 the most probable one with respect to the beds at ,concepcion entirely composed of the ,mytilus ,chiloensis1 a species which lives below the lowest tides2 and likewise with respect to the great beds occurring both north and south of ,valparaiso1 of the ,mesodesma donaciforme1--a shell which1 as ,i am informed by ,mr4 ,cuming1 inhabits sandbanks at the level of the lowest tides4 ,but even in the case of shells having the habits of this ,mytilus and ,mesodesma1 beds of them1 wherever the sea gently throws up sand or mud1 and thus protects its own accumulations1 might be upraised by the slowest movement1 and yet remain undisturbed by the waves of each new beach-line4 ,it is worthy of remark1 that nowhere near ,valparaiso above the height of twenty feet1 or rarely of fifty feet1 ,i saw any lines of erosion on the solid rocks1 or any beds of pebbles2 this1 ,i believe1 may be accounted for by the disintegrating tendency of most of the rocks in this neighbourhood4 ,nor is the land here modelled into terraces3 ,mr4 ,alison1 however1 informs me1 that on both sides of one narrow ravine1 at the height of #cjj feet above the sea1 he found a succession of rather indistinct step-formed beaches1 composed of broken shells1 which together covered a space of about eighty feet vertical4 ,i can add nothing to the accounts already published of the elevation of the land at ,valparaiso1#ac which accompanied the earthquake of #ahbb3 but ,i heard it confidently asserted1 that a sentinel on duty1 immediately after the shock1 saw a part of a fort1 which previously was not within the line of his vision1 and this would indicate that the uplifting was not horizontal3 it would even appear from some facts collected by ,mr4 ,alison1 that only the eastern half of the bay was then elevated4 ,through the kindness of this same gentleman1 ,i am able to give an interesting account of the changes of level1 which have supervened here within historical periods3 about the year #afhj a long sea-wall 7or ,prefil7 was built1 of which only a few fragments now remain2 up to the year #ahag1 the sea often broke over it1 and washed the houses on the opposite side of the road 7where the prison now stands72 and even in #ahai1 ,mr4 ,j4 ,martin remembers walking at the foot of this wall1 and being often obliged to climb over it to escape the waves4 ,there now stands 7#ahcd7 on the seaward side of this wall1 and between it and the beach1 in one part a single row of houses1 and in another part two rows with a street between them4 ,this great extension of the beach in so short a time cannot be attributed simply to the accumulation of detritus2 for a resident engineer measured for me the height between the lowest part of the wall visible1 and the present beach-line at spring-tides1 and the difference was eleven feet six inches4 ,the church of ,s4 ,augustin is believed to have been built in #afad1 and there is a tradition that the sea formerly flowed very near it2 by levelling1 its foundations were found to stand nineteen feet six inches above the highest beach-line2 so that we see in a period of #bbj years1 the elevation cannot have been as much as nineteen feet six inches4 ,from the facts given with respect to the sea-wall1 and from the testimony of the elder inhabitants1 it appears certain that the change in level began to be manifest about the year #ahag4 ,the only sudden elevation of which there is any record occurred in #ahbb1 and this seems to have been less than three feet4 ,since that year1 ,i was assured by several competent observers1 that part of an old wreck1 which is firmly embedded near the beach1 has sensibly emerged2 hence here1 as at ,chiloe1 a slow rise of the land appears to be now in progress4 ,it seems highly probable that the rocks which are corroded in a band at the height of fourteen feet above the sea were acted on during the period1 when by tradition the base of ,s4 ,augustin church1 now nineteen feet six inches above the highest water-mark1 was occasionally washed by the waves4 #ac ,dr4 ,meyen 78,reise um ,erde10 ,th4 ,i1 s4 #bba7 found in #ahca seaweed and other bodies still adhering to some rocks which during the shock of #ahbb were lifted above the sea4 ,valparaiso to ,coquimbo4--,for the first seventy-five miles north of ,valparaiso ,i followed the coast-road1 and throughout this space ,i observed innumerable masses of upraised shells4 ,about ,quintero there are immense accumulations 7worked for lime7 of the ,mesodesma donaciforme1 packed in sandy earth2 they abound chiefly about fifteen feet above high-water1 but shells are here found1 according to ,mr4 ,miers1#ad to a height of #ejj feet1 and at a distance of three leagues from the coast3 ,i here noticed barnacles adhering to the rocks three or four feet above the highest tides4 ,in the neighbourhood of ,plazilla and ,catapilco1 at heights of between two hundred and three hundred feet1 the number of comminuted shells1 with some perfect ones1 especially of the ,mesodesma1 packed in layers1 was truly immense3 the land at ,plazilla had evidently existed as a bay1 with abrupt rocky masses rising out of it1 precisely like the islets in the broken bays now indenting this coast4 ,on both sides of the rivers ,ligua1 ,longotomo1 ,guachen1 and ,quilimari1 there are plains of gravel about two hundred feet in height1 in many parts absolutely covered with shells4 ,close to ,conchalee1 a gravel-plain is fronted by a lower and similar plain about sixty feet in height1 and this again is separated from the beach by a wide tract of low land3 the surfaces of all three plains or terraces were strewed with vast numbers of the ,concholepas1 ,mesodesma1 an existing ,venus1 and other still existing littoral shells4 ,the two upper terraces closely resemble in miniature the plains of ,patagonia2 and like them are furrowed by dry1 flat-bottomed1 winding valleys4 ,northward of this place ,i turned inward2 and therefore found no more shells3 but the valleys of ,chuapa1 ,illapel1 and ,limari1 are bounded by gravel-capped plains1 often including a lower terrace within4 ,these plains send bay-like arms between and into the surrounding hills2 and they are continuously united with other extensive gravel-capped plains1 separating the coast mountain-ranges from the ,cordillera4 #ad 8,travels in ,chile10 vol4 i1 pp4 #cie1 #deh4 ,i received several similar accounts from the inhabitants1 and was assured that there are many shells on the plain of ,casa ,blanca1 between ,valparaiso and ,santiago1 at the height of #hjj feet4 ,coquimbo4 ,a narrow fringe-like plain1 gently inclined towards the sea1 here extends for eleven miles along the coast1 with arms stretching up between the coast-mountains1 and likewise up the valley of ,coquimbo3 at its southern extremity it is directly connected with the plain of ,limari1 out of which hills abruptly rise like islets1 and other hills project like headlands on a coast4 ,the surface of the fringe-like plain appears level1 but differs insensibly in height1 and greatly in composition1 in different parts4 ,at the mouth of the valley of ,coquimbo1 the surface consists wholly of gravel1 and stands from #cjj to #cej feet above the level of the sea1 being about one hundred feet higher than in other parts4 ,in these other and lower parts the superficial beds consist of calcareous matter1 and rest on ancient tertiary deposits hereafter to be described4 ,the uppermost calcareous layer is cream-coloured1 compact1 smooth-fractured1 sub-stalactiform1 and contains some sand1 earthy matter1 and recent shells4 ,it lies on1 and sends wedge-like veins into1#ae a much more friable1 calcareous1 tuff-like variety2 and both rest on a mass about twenty feet in thickness1 formed of fragments of recent shells1 with a few whole ones1 and with small pebbles firmly cemented together4 ,this latter rock is called by the inhabitants losa1 and is used for building3 in many parts it is divided into strata1 which dip at an angle of ten degrees seaward1 and appear as if they had originally been heaped in successive layers 7as may be seen on coral-reefs7 on a steep beach4 ,this stone is remarkable from being in parts entirely formed of empty1 pellucid capsules or cells of calcareous matter1 of the size of small seeds3 a series of specimens unequivocally showed that all these capsules once contained minute rounded fragments of shells which have since been gradually dissolved by water percolating through the mass4#af #ae ,in many respects this upper hard1 and the underlying more friable1 varieties1 resemble the great superficial beds at ,king ,george's ,sound in ,australia1 which ,i have described in my 8,geological ,observations on ,volcanic ,islands40 ,there could be little doubt that the upper layers there have been hardened by the action of rain on the friable1 calcareous matter1 and that the whole mass has originated in the decay of minutely comminuted sea-shells and corals4 #af ,i have incidentally described this rock in the above work on ,volcanic ,islands4 ,the shells embedded in the calcareous beds forming the surface of this fringe-like plain1 at the height of from #bjj to #bej feet above the sea1 consist of3-- ,venus opaca4 ,mulinia ,byronensis4 ,pecten purpuratus4 ,mesodesma donaciforme4 ,turritella cingulata4 ,monoceros costatum4 ,concholepas ,peruviana4 ,trochus 7common ,valparaiso species74 ,calyptra ,byronensis4 ,although these species are all recent1 and are all found in the neighbouring sea1 yet ,i was particularly struck with the difference in the proportional numbers of the several species1 and of those now cast up on the present beach4 ,i found only one specimen of the ,concholepas1 and the ,pecten was very rare1 though both these shells are now the commonest kinds1 with the exception1 perhaps1 of the ,calyptra radians1 of which ,i did not find one in the calcareous beds4 ,i will not pretend to determine how far this difference in the proportional numbers depends on the age of the deposit1 and how far on the difference in nature between the present sandy beaches and the calcareous bottom1 on which the embedded shells must have lived4 ,no4 #h ,section of plain of ,coquimbo4 ,illustration3 ,section of plain of ,coquimbo4 ,on the bare surface of the calcareous plain1 or in a thin covering of sand1 there were lying1 at a height from #bjj to #beb feet1 many recent shells1 which had a much fresher appearance than the embedded ones3 fragments of the ,concholepas1 and of the common ,mytilus1 still retaining a tinge of its colour1 were numerous1 and altogether there was manifestly a closer approach in proportional numbers to those now lying on the beach4 ,in a mass of stratified1 slightly agglutinated sand1 which in some places covers up the lower half of the seaward escarpment of the plain1 the included shells appeared to be in exactly the same proportional numbers with those on the beach4 ,on one side of a steep-sided ravine1 cutting through the plain behind ,herradura ,bay1 ,i observed a narrow strip of stratified sand1 containing similar shells in similar proportional numbers2 a section of the ravine is represented in ,diagram #h1 which serves also to show the general composition of the plain4 ,i mention this case of the ravine chiefly because without the evidence of the marine shells in the sand1 any one would have supposed that it had been hollowed out by simple alluvial action4 ,the escarpment of the fringe-like plain1 which stretches for eleven miles along the coast1 is in some parts fronted by two or three narrow1 step-formed terraces1 one of which at ,herradura ,bay expands into a small plain4 ,its surface was there formed of gravel1 cemented together by calcareous matter2 and out of it ,i extracted the following recent shells1 which are in a more perfect condition than those from the upper plain3-- ,calyptra radians4 ,turritella cingulata4 ,oliva ,peruviana4 ,murex labiosus1 var4 ,nassa 7identical with a living species74 ,solen ,dombeiana4 ,pecten purpuratus4 ,venus ,chilensis4 ,amphidesma rugulosum4 ,the small irregular wrinkles of the posterior part of this shell are rather stronger than in the recent specimens of this species from ,coquimbo4 7,g4 ,b4 ,sowerby47 ,balanus 7identical with living species74 ,on the syenitic ridge1 which forms the southern boundary of ,herradura ,bay and ,plain1 ,i found the ,concholepas and ,turritella cingulata 7mostly in fragments71 at the height of #bdb feet above the sea4 ,i could not have told that these shells had not formerly been brought up by man1 if ,i had not found one very small mass of them cemented together in a friable calcareous tuff4 ,i mention this fact more particularly1 because ,i carefully looked1 in many apparently favourable spots1 at lesser heights on the side of this ridge1 and could not find even the smallest fragment of a shell4 ,this is only one instance out of many1 proving that the absence of sea-shells on the surface1 though in many respects inexplicable1 is an argument of very little weight in opposition to other evidence on the recent elevation of the land4 ,the highest point in this neighbourhood at which ,i found upraised shells of existing species was on an inland calcareous plain1 at the height of #beb feet above the sea4 ,it would appear from ,mr4 ,caldcleugh's researches1#ag that a rise has taken place here within the last century and a half2 and as no sudden change of level has been observed during the not very severe earthquakes1 which have occasionally occurred here1 the rising has probably been slow1 like that now1 or quite lately1 in progress at ,chiloe and at ,valparaiso3 there are three well-known rocks1 called the ,pelicans1 which in #agaj1 according to ,feuille1 were fleur d'eau1 but now are said to stand twelve feet above low-water mark3 the spring-tides rise here only five feet4 ,there is another rock1 now nine feet above high-water mark1 which in the time of ,frezier and ,feuille rose only five or six feet out of water4 ,mr4 ,caldcleugh1 ,i may add1 also shows 7and ,i received similar accounts7 that there has been a considerable decrease in the soundings during the last twelve years in the ,bays of ,coquimbo1 ,concepcion1 ,valparaiso1 and ,guasco2 but as in these cases it is nearly impossible to distinguish between the accumulation of sediment and the upheavement of the bottom1 ,i have not entered into any details4 #ag 8,proceedings of the ,geological ,society10 vol4 ii1 p4 #ddf4 ,valley of ,coquimbo4--,the narrow coast-plain sends1 as before stated1 an arm1 or more correctly a fringe1 on both sides1 but chiefly on the southern side1 several miles up the valley4 ,these fringes are worn into steps or terraces1 which present a most remarkable appearance1 and have been compared 7though not very correctly7 by ,captain ,basil ,hall1 to the parallel roads of ,glen ,roy in ,scotland3 their origin has been ably discussed by ,mr4 ,lyell4#ah ,the first section which ,i will give 7,figure #i71 is not drawn across the valley1 but in an east and west line at its mouth1 where the step-formed terraces debouch and present their very gently inclined surfaces towards the ,pacific4 #ah 8,principles of ,geology0 7#ast edit471 vol4 iii1 p4 #aca4 ,no4 #i ,east and west section through the terraces at ,coquimbo1 where they debouch from the valley1 and front the sea4 ,illustration3 ,east and west section through terraces at ,coquimbo4 ,the bottom plain 7,a7 is about a mile in width1 and rises quite insensibly from the beach to a height of twenty-five feet at the foot of the next plain2 it is sandy1 and abundantly strewed with shells4 ,plain or terrace ,b is of small extent1 and is almost concealed by the houses of the town1 as is likewise the escarpment of terrace ,c4 ,on both sides of a ravine1 two miles south of the town1 there are two little terraces1 one above the other1 evidently corresponding with ,b and ,c2 and on them marine remains of the species already enumerated were plentiful4 ,terrace ,e is very narrow1 but quite distinct and level2 a little southward of the town there were traces of a terrace ,d intermediate between ,e and ,c4 ,terrace ,f is part of the fringe-like plain1 which stretches for the eleven miles along the coast2 it is here composed of shingle1 and is #ajj feet higher than where composed of calcareous matter4 ,this greater height is obviously due to the quantity of shingle1 which at some former period has been brought down the great valley of ,coquimbo4 ,considering the many shells strewed over the terraces ,a1 ,b1 and ,c1 and a few miles southward on the calcareous plain1 which is continuously united with the upper step-like plain ,f1 there cannot1 ,i apprehend1 be any doubt1 that these six terraces have been formed by the action of the sea2 and that their five escarpments mark so many periods of comparative rest in the elevatory movement1 during which the sea wore into the land4 ,the elevation between these periods may have been sudden and on an average not more than seventy-two feet each time1 or it may have been gradual and insensibly slow4 ,from the shells on the three lower terraces1 and on the upper one1 and ,i may add on the three gravel-capped terraces at ,conchalee1 being all littoral and sub-littoral species1 and from the analogical facts given at ,valparaiso1 and lastly from the evidence of a slow rising lately or still in progress here1 it appears to me far more probable that the movement has been slow4 ,the existence of these successive escarpments1 or old cliff-lines1 is in another respect highly instructive1 for they show periods of comparative rest in the elevatory movement1 and of denudation1 which would never even have been suspected from a close examination of many miles of coast southward of ,coquimbo4 ,we come now to the terraces on the opposite sides of the east and west valley of ,coquimbo3 the section in figure ,no4 #aj is taken in a north and south line across the valley at a point about three miles from the sea4 ,the valley measured from the edges of the escarpments of the upper plain ,,ff is about a mile in width2 but from the bases of the bounding mountains it is from three to four miles wide4 ,the terraces marked with an interrogative do not exist on that side of the valley1 but are introduced merely to render the diagram more intelligible4 ,no4 #aj ,north and south section across the valley of ,coquimbo4 ,illustration3 ,north and south section across the valley of ,coquimbo4 ,terraces marked with 8 do not occur on that side of the valley1 and are introduced only to make the diagram more intelligible4 ,a river and bottom-plain of valley ,c1 ,e1 and ,f1 on the south side of valley1 are respectively1 #aig1 #cgg1 and #dbj feet above the level of the sea4 ,,aa4 ,the bottom of the valley1 believed to be #ajj feet above the sea3 it is continuously united with the lowest plain ,a of figure ,no4 #i4 ,b4 ,this terrace higher up the valley expands considerably2 seaward it is soon lost1 its escarpment being united with that of ,c3 it is not developed at all on the south side of the valley4 ,c4 ,this terrace1 like the last1 is considerably expanded higher up the valley4 ,these two terraces apparently correspond with ,b and ,c of figure ,no4 #i4 ,d is not well developed in the line of this section2 but seaward it expands into a plain3 it is not present on the south side of the valley2 but it is met with1 as stated under the former section1 a little south of the town4 ,e is well developed on the south side1 but absent on the north side of the valley3 though not continuously united with ,e of figure ,no4 #i1 it apparently corresponds with it4 ,f4 ,this is the surface-plain1 and is continuously united with that which stretches like a fringe along the coast4 ,in ascending the valley it gradually becomes narrower1 and is at last1 at the distance of about ten miles from the sea1 reduced to a row of flat-topped patches on the sides of the mountains4 ,none of the lower terraces extend so far up the valley4 ,these five terraces are formed of shingle and sand2 three of them1 as marked by ,captain ,b4 ,hall 7namely1 ,b1 ,c1 and ,f71 are much more conspicuous than the others4 ,from the marine remains copiously strewed at the mouth of the valley on the lower terraces1 and southward of the town on the upper one1 they are1 as before remarked1 undoubtedly of marine origin2 but within the valley1 and this fact well deserves notice1 at a distance of from only a mile and a half to three or four miles from the sea1 ,i could not find even a fragment of a shell4 ,on the inclination of the terraces of ,coquimbo1 and on the upper and basal edges of their escarpments not being horizontal4--,the surfaces of these terraces slope in a slight degree1 as shown by the two last sections taken conjointly1 both towards the centre of the valley1 and seawards towards its mouth4 ,this double or diagonal inclination1 which is not the same in the several terraces1 is1 as we shall immediately see1 of simple explanation4 ,there are1 however1 some other points which at first appear by no means obvious1--namely1 first1 that each terrace1 taken in its whole breadth from the summit-edge of one escarpment to the base of that above it1 and followed up the valley1 is not horizontal2 nor have the several terraces1 when followed up the valley1 all the same inclination2 thus ,i found the terraces ,c1 ,e1 and ,f1 measured at a point about two miles from the mouth of the valley1 stood severally between fifty-six to seventy-seven feet higher than at the mouth4 ,again1 if we look to any one line of cliff or escarpment1 neither its summit-edge nor its base is horizontal4 ,on the theory of the terraces having been formed during a slow and equable rise of the land1 with as many intervals of rest as there are escarpments1 it appears at first very surprising that horizontal lines of some kind should not have been left on the land4 ,the direction of the diagonal inclination in the different terraces being different1--in some being directed more towards the middle of the valley1 in others more towards its mouth1--naturally follows on the view of each terrace1 being an accumulation of successive beach-lines round bays1 which must have been of different forms and sizes when the land stood at different levels3 for if we look to the actual beach of a narrow creek1 its slope is directed towards the middle2 whereas1 in an open bay1 or slight concavity on a coast1 the slope is towards the mouth1 that is1 almost directly seaward2 hence as a bay alters in form and size1 so will the direction of the inclination of its successive beaches become changed4 ,illustration3 ,a bay in the district which has begun slowly rising4 ,if it were possible to trace any one of the many beach-lines1 composing each sloping terrace1 it would of course be horizontal2 but the only lines of demarcation are the summit and basal edges of the escarpments4 ,now the summit-edge of one of these escarpments marks the furthest line or point to which the sea has cut into a mass of gravel sloping seaward2 and as the sea will generally have greater power at the mouth than at the protected head of the bay1 so will the escarpment at the mouth be cut deeper into the land1 and its summit-edge be higher2 consequently it will not be horizontal4 ,with respect to the basal or lower edges of the escarpments1 from picturing in one's mind ancient bays entirely surrounded at successive periods by cliff-formed shores1 one's first impression is that they at least necessarily must be horizontal1 if the elevation has been horizontal4 ,but here is a fallacy3 for after the sea has1 during a cessation of the elevation1 worn cliffs all round the shores of a bay1 when the movement recommences1 and especially if it recommences slowly1 it might well happen that1 at the exposed mouth of the bay1 the waves might continue for some time wearing into the land1 whilst in the protected and upper parts successive beach-lines might be accumulating in a sloping surface or terrace at the foot of the cliffs which had been lately reached3 hence1 supposing the whole line of escarpment to be finally uplifted above the reach of the sea1 its basal line or foot near the mouth will run at a lower level than in the upper and protected parts of the bay2 consequently this basal line will not be horizontal4 ,and it has already been shown that the summit-edges of each escarpment will generally be higher near the mouth 7from the seaward sloping land being there most exposed and cut into7 than near the head of the bay2 therefore the total height of the escarpments will be greatest near the mouth2 and further up the old bay or valley they will on both sides generally thin out and die away3 ,i have observed this thinning out of the successive escarpment at other places besides ,coquimbo2 and for a long time ,i was quite unable to understand its meaning4 ,the rude diagram in ,figure #aa will perhaps render what ,i mean more intelligible2 it represents a bay in a district which has begun slowly rising4 ,before the movement commenced1 it is supposed that the waves had been enabled to eat into the land and form cliffs1 as far up1 but with gradually diminishing power1 as the points ,,aa3 after the movement had commenced and gone on for a little time1 the sea is supposed still to have retained the power1 at the exposed mouth of the bay1 of cutting down and into the land as it slowly emerged2 but in the upper parts of the bay it is supposed soon to have lost this power1 owing to the more protected situation and to the quantity of detritus brought down by the river2 consequently low land was there accumulated4 ,as this low land was formed during a slow elevatory movement1 its surface will gently slope upwards from the beach on all sides4 ,now1 let us imagine the bay1 not to make the diagram more complicated1 suddenly converted into a valley3 the basal line of the cliffs will of course be horizontal1 as far as the beach is now seen extending in the diagram2 but in the upper part of the valley1 this line will be higher1 the level of the district having been raised whilst the low land was accumulating at the foot of the inland cliffs4 ,if1 instead of the bay in the diagram being suddenly converted into a valley1 we suppose with much more probability it to be upraised slowly1 then the waves in the upper parts of the bay will continue very gradually to fail to reach the cliffs1 which are now in the diagram represented as washed by the sea1 and which1 consequently1 will be left standing higher and higher above its level2 whilst at the still exposed mouth1 it might well happen that the waves might be enabled to cut deeper and deeper1 both down and into the cliffs1 as the land slowly rose4 ,the greater or lesser destroying power of the waves at the mouths of successive bays1 comparatively with this same power in their upper and protected parts1 will vary as the bays become changed in form and size1 and therefore at different levels1 at their mouths and heads1 more or less of the surfaces between the escarpments 7that is1 the accumulated beach-lines or terraces7 will be left undestroyed3 from what has gone before we can see that1 according as the elevatory movements after each cessation recommence more or less slowly1 according to the amount of detritus delivered by the river at the heads of the successive bays1 and according to the degree of protection afforded by their altered forms1 so will a greater or less extent of terrace be accumulated in the upper part1 to which there will be no surface at a corresponding level at the mouth3 hence we can perceive why no one terrace1 taken in its whole breadth and followed up the valley1 is horizontal1 though each separate beach-line must have been so2 and why the inclination of the several terraces1 both transversely1 and longitudinally up the valley1 is not alike4 ,i have entered into this case in some detail1 for ,i was long perplexed 7and others have felt the same difficulty7 in understanding how1 on the idea of an equable elevation with the sea at intervals eating into the land1 it came that neither the terraces nor the upper nor lower edges of the escarpments were horizontal4 ,along lines of coast1 even of great lengths1 such as that of ,patagonia1 if they are nearly uniformly exposed1 the corroding power of the waves will be checked and conquered by the elevatory movement1 as often as it recommences1 at about the same period2 and hence the terraces1 or accumulated beach-lines1 will commence being formed at nearly the same levels3 at each succeeding period of rest1 they will1 also1 be eaten into at nearly the same rate1 and consequently there will be a much closer coincidence in their levels and inclinations1 than in the terraces and escarpments formed round bays with their different parts very differently exposed to the action of the sea4 ,it is only where the waves are enabled1 after a long lapse of time1 slowly to corrode hard rocks1 or to throw up1 owing to the supply of sediment being small and to the surface being steeply inclined1 a narrow beach or mound1 that we can expect1 as at ,glen ,roy in ,scotland1#ai a distinct line marking an old sea-level1 and which will be strictly horizontal1 if the subsequent elevatory movements have been so3 for in these cases no discernible effects will be produced1 except during the long intervening periods of rest2 whereas in the case of step-formed coasts1 such as those described in this and the preceding chapter1 the terraces themselves are accumulated during the slow elevatory process1 the accumulation commencing sooner in protected than in exposed situations1 and sooner where there is copious supply of detritus than where there is little2 on the other hand1 the steps or escarpments are formed during the stationary periods1 and are more deeply cut down and into the coast-land in exposed than in protected situations2--the cutting action1 moreover1 being prolonged in the most exposed parts1 both during the beginning and ending1 if slow1 of the upward movement4 #ai 8,philosophical ,transactions10 #ahci1 p4 #ci4 ,although in the foregoing discussion ,i have assumed the elevation to have been horizontal1 it may be suspected1 from the considerable seaward slope of the terraces1 both up the valley of ,s4 ,cruz and up that of ,coquimbo1 that the rising has been greater inland than nearer the coast4 ,there is reason to believe1#bj from the effects produced on the water-course of a mill during the earthquake of #ahbb in ,chile1 that the upheaval one mile inland was nearly double1 namely1 between five and seven feet1 to what it was on the ,pacific4 ,we know1 also1 from the admirable researches of ,m4 ,bravais1#ba that in ,scandinavia the ancient sea-beaches gently slope from the interior mountain-ranges towards the coast1 and that they are not parallel one to the other showing that the proportional difference in the amount of elevation on the coast and in the interior1 varied at different periods4 #bj ,mr4 ,place in the ,quarterly ,journal of ,science1 #ahbd1 vol4 xvii1 p4 #db4 #ba 8,voyages de la ,comm4 du ,nord10 etc41 also 8,comptes ,rendus10 ,oct4 #ahdb4 ,coquimbo to ,guasco4--,in this distance of ninety miles1 ,i found in almost every part marine shells up to a height of apparently from two hundred to three hundred feet4 ,the desert plain near ,choros is thus covered2 it is bounded by the escarpment of a higher plain1 consisting of pale-coloured1 earthy1 calcareous stone1 like that of ,coquimbo1 with the same recent shells embedded in it4 ,in the valley of ,chaneral1 a similar bed occurs in which1 differently from that of ,coquimbo1 ,i observed many shells of the ,concholepas3 near ,guasco the same calcareous bed is likewise met with4 ,in the valley of ,guasco1 the step-formed terraces of gravel are displaced in a more striking manner than at any other point4 ,i followed the valley for thirty-seven miles 7as reckoned by the inhabitants7 from the coast to ,ballenar2 in nearly the whole of this distance1 five grand terraces1 running at corresponding heights on both sides of the broad valley1 are more conspicuous than the three best-developed ones at ,coquimbo4 ,they give to the landscape the most singular and formal aspect2 and when the clouds hung low1 hiding the neighbouring mountains1 the valley resembled in the most striking manner that of ,santa ,cruz4 ,the whole thickness of these terraces or plains seems composed of gravel1 rather firmly aggregated together1 with occasional parting seams of clay3 the pebbles on the upper plain are often whitewashed with an aluminous substance1 as in ,patagonia4 ,near the coast ,i observed many sea-shells on the lower plains4 ,at ,freyrina 7twelve miles up the valley71 there are six terraces beside the bottom-surface of the valley3 the two lower ones are here only from two hundred to three hundred yards in width1 but higher up the valley they expand into plains2 the third terrace is generally narrow2 the fourth ,i saw only in one place1 but there it was distinct for the length of a mile2 the fifth is very broad2 the sixth is the summit-plain1 which expands inland into a great basin4 ,not having a barometer with me1 ,i did not ascertain the height of these plains1 but they appeared considerably higher than those at ,coquimbo4 ,their width varies much1 sometimes being very broad1 and sometimes contracting into mere fringes of separate flat-topped projections1 and then quite disappearing3 at the one spot1 where the fourth terrace was visible1 the whole six terraces were cut off for a short space by one single bold escarpment4 ,near ,ballenar 7thirty-seven miles from the mouth of the river71 the valley between the summit-edges of the highest escarpments is several miles in width1 and the five terraces on both sides are broadly developed3 the highest cannot be less than six hundred feet above the bed of the river1 which itself must1 ,i conceive1 be some hundred feet above the sea4 ,no4 #ab ,north and south section across the valley of ,guasco1 and of a plain north of it4 ,illustration3 ,north and south section across the valley of ,guasco4 ,on the northern side of the valley the summit-plain of gravel 7,a7 has two escarpments1 one facing the valley1 and the other a great basin-like plain 7,b71 which stretches for several leagues northward4 ,this narrow plain 7,a7 with the double escarpment1 evidently once formed a spit or promontory of gravel1 projecting into and dividing two great bays1 and subsequently was worn on both sides into steep cliffs4 ,whether the several escarpments in this valley were formed during the same stationary periods with those of ,coquimbo1 ,i will not pretend to conjecture2 but if so the intervening and subsequent elevatory movements must have been here much more energetic1 for these plains certainly stand at a much higher level than do those of ,coquimbo4 ,copiapo4--,from ,guasco to ,copiapo1 ,i followed the road near the foot of the ,cordillera1 and therefore saw no upraised remains4 ,at the mouth1 however1 of the valley of ,copiapo there is a plain1 estimated by ,meyen#bb between fifty and seventy feet in height1 of which the upper part consists chiefly of gravel1 abounding with recent shells1 chiefly of the ,concholepas1 ,venus ,dombeyi1 and ,calyptra trochiformis4 ,a little inland1 on a plain estimated by myself at nearly three hundred feet1 the upper stratum was formed of broken shells and sand cemented by white calcareous matter1 and abounding with embedded recent shells1 of which the ,mulinia ,byronensis and ,pecten purpuratus were the most numerous4 ,the lower plain stretches for some miles southward1 and for an unknown distance northward1 but not far up the valley2 its seaward face1 according to ,meyen1 is worn into caves above the level of the present beach4 ,the valley of ,copiapo is much less steeply inclined and less direct in its course than any other valley which ,i saw in ,chile2 and its bottom does not generally consist of gravel3 there are no step-formed terraces in it1 except at one spot near the mouth of the great lateral valley of the ,despoblado where there are only two1 one above the other3 lower down the valley1 in one place ,i observed that the solid rock had been cut into the shape of a beach1 and was smoothed over with shingle4 #bb 8,reise um die ,erde10 ,th4 ,i1 s4 #cgb1 et seq4 ,northward of ,copiapo1 in lat4 #bf ,s41 the old voyager ,wafer#bc found immense numbers of sea-shells some miles from the coast4 ,at ,cobija 7lat4 #bb #cd7 ,m4 d',orbigny observed beds of gravel and broken shells1 containing ten species of recent shells2 he also found1 on projecting points of porphyry1 at a height of #cjj feet1 shells of ,concholepas1 ,chiton1 ,calyptra1 ,fissurella1 and ,patella1 still attached to the spots on which they had lived4 ,m4 d',orbigny argues from this fact1 that the elevation must have been great and sudden3#bd to me it appears far more probable that the movement was gradual1 with small starts as during the earthquakes of #ahbb and #ahce1 by which whole beds of shells attached to the rocks were lifted above the subsequent reach of the waves4 ,m4 d',orbigny also found rolled pebbles extending up the mountain to a height of at least six hundred feet4 ,at ,iquique 7lat4 #bj #ab ,s471 in a great accumulation of sand1 at a height estimated between one hundred and fifty and two hundred feet1 ,i observed many large sea-shells which ,i thought could not have been blown up by the wind to that height4 ,mr4 ,j4 ,h4 ,blake has lately#be described these shells3 he states that 8inland toward the mountains they form a compact uniform bed1 scarcely a trace of the original shells being discernible2 but as we approach the shore1 the forms become gradually more distinct till we meet with the living shells on the coast40 ,this interesting observation1 showing by the gradual decay of the shells how slowly and gradually the coast must have been uplifted1 we shall presently see fully confirmed at ,lima4 ,at ,arica 7lat4 #ah #bh71 ,m4 d',orbigny#bf found a great range of sand-dunes1 fourteen leagues in length1 stretching towards ,tacna1 including recent shells and bones of ,cetacea1 and reaching up to a height of #cjj feet above the sea4 ,lieutenant ,freyer has given some more precise facts3 he states#bg that the ,morro of ,arica is about four hundred feet high2 it is worn into obscure terraces1 on the bare rock of which he found ,balini and ,millepor adhering4 ,at the height of between twenty and thirty feet the shells and corals were in a quite fresh state1 but at fifty feet they were much abraded2 there were1 however1 traces of organic remains at greater heights4 ,on the road from ,tacna to ,arequipa1 between ,loquimbo and ,moquegua1 ,mr4 ,m4 ,hamilton#bh found numerous recent sea shells in sand1 at a considerable distance from the sea4 #bc ,burnett's 8,collection of ,voyages10 vol4 iv1 p4 #aic4 #bd 8,voyage1 ,part ,golog410 p4 #id4 ,m4 d',orbigny 7p4 #ih71 in summing up1 says3 8,s'il est certain 7as he believes7 que tous les terrains en pente1 compris entre la mer et les montagnes sont l'ancien rivage de la mer1 on doit supposer1 pour l'ensemble1 un exhaussement que ce ne serait pas moindre de deux cent mtres2 il faudrait supposer encore que ce soulvement n'a point t graduel2 4 4 4 mais qu'il rsulterait d'une seule et mme cause fortuite10 etc4 ,now1 on this view1 when the sea was forming the beach at the foot of the mountains1 many shells of ,concholepas1 ,chiton1 ,calyptra1 ,fissurella1 and ,patella 7which are known to live close to the beach71 were attached to rocks at a depth of #cjj feet1 and at a depth of #fjj feet several of these same shells were accumulating in great numbers in horizontal beds4 ,from what ,i have myself seen in dredging1 ,i believe this to be improbable in the highest degree1 if not impossible2 and ,i think everyone who has read ,professor ,e4 ,forbes's excellent researches on the subject1 will without hesitation agree in this conclusion4 #be ,silliman's ,amer4 ,journ4 of ,science1 vol4 xliv1 p4 #b4 #bf 8,voyage10 etc41 p4 #aja4 #bg ,in a letter to ,mr4 ,lyell1 8,geolog4 ,proc410 vol4 ii1 p4 #agi4 #bh ,edin4 ,new ,phil4 ,journ41 vol4 xxx1 p4 #aee4 ,lima4 ,northward of ,arica1 ,i know nothing of the coast for about a space of five degrees of latitude2 but near ,callao1 the port of ,lima1 there is abundant and very curious evidence of the elevation of the land4 ,the island of ,san ,lorenzo is upwards of one thousand feet high2 the basset edges of the strata composing the lower part are worn into three obscure1 narrow1 sloping steps or ledges1 which can be seen only when standing on them3 they probably resemble those described by ,lieutenant ,freyer at ,arica4 ,the surface of the lower ledge1 which extends from a low cliff overhanging the sea to the foot of the next upper escarpment1 is covered by an enormous accumulation of recent shells4#bi ,the bed is level1 and in some parts more than two feet in thickness2 ,i traced it over a space of one mile in length1 and heard of it in other places3 the uppermost part is eighty-five feet by the barometer above high-water mark4 ,the shells are packed together1 but not stratified3 they are mingled with earth and stones1 and are generally covered by a few inches of detritus2 they rest on a mass of nearly angular fragments of the underlying sandstone1 sometimes cemented together by common salt4 ,i collected eighteen species of shells of all ages and sizes4 ,several of the univalves had evidently long lain dead at the bottom of the sea1 for their insides were incrusted with ,balani and ,serpul4 ,all1 according to ,mr4 ,g4,b4 ,sowerby1 are recent species3 they consist of3-- ,mytilus ,magellanicus3 same as that found at ,valparaiso1 and there stated to be probably distinct from the true ,m4 ,magellanicus of the east coast4 ,venus costellata1 ,sowerby 8,zoological ,proceedings40 ,pecten purpuratus1 ,lam4 ,chama1 probably echinulata1 ,brod4 ,calyptra ,byronensis1 ,gray4 ,calyptra radians 7,trochus1 ,lam47 ,fissurella affinis1 ,gray4 ,fissurella biradiata1 ,trembly4 ,purpura chocolatta1 ,duclos4 ,purpura ,peruviana1 ,gray4 ,purpura labiata1 ,gray4 ,purpura buxea 7,murex1 ,brod474 ,concholepas ,peruviana4 ,nassa1 related to reticulata4 ,triton rudis1 ,brod4 ,trochus1 not yet described1 but well-known and very common4 and #ah4 ,balanus1 two species1 both common on the coast4 #bi ,m4 ,chevalier1 in the 8,voyage of the ,bonite10 observed these shells2 but his specimens were lost4--8,l',institut10 #ahch1 p4 #aea4 ,these upraised shells appear to be nearly in the same proportional numbers--with the exception of the ,crepidul being more numerous--with those on the existing beach4 ,the state of preservation of the different species differed much2 but most of them were much corroded1 brittle1 and bleached3 the upper and lower surfaces of the ,concholepas had generally quite scaled off3 some of the ,trochi and ,fissurell still partially retain their colours4 ,it is remarkable that these shells1 taken all together1 have fully as ancient an appearance1 although the extremely arid climate appears highly favourable for their preservation1 as those from #a1cjj feet at ,valparaiso1 and certainly a more ancient appearance than those from five to six hundred feet from ,valparaiso and ,concepcion2 at which places ,i have seen grass and other vegetables actually growing out of the shells4 ,many of the univalves here at ,san ,lorenzo were filled with1 and united together by1 pure salt1 probably left by the evaporation of the sea-spray1 as the land slowly emerged4#cj ,on the highest parts of the ledge1 small fragments of the shells were mingled with1 and evidently in process of reduction into1 a yellowish-white1 soft1 calcareous powder1 tasting strongly of salt1 and in some places as fine as prepared medicinal chalk4 #cj ,the underlying sandstone contains true layers of salt2 so that the salt may possibly have come from the beds in the higher parts of the island2 but ,i think more probably from the sea-spray4 ,it is generally asserted that rain never falls on the coast of ,peru2 but this is not quite accurate2 for1 on several days1 during our visit1 the so-called ,peruvian dew fell in sufficient quantity to make the streets muddy1 and it would certainly have washed so deliquescent a substance as salt into the soil4 ,i state this because ,m4 d',orbigny1 in discussing an analogous subject1 supposes that ,i had forgotten that it never rains on this whole line of coast4 ,see ,ulloa's 8,voyage0 7vol4 ii1 ,eng4 ,trans41 p4 #fg7 for an account of the muddy streets of ,lima1 and on the continuance of the mists during the whole winter4 ,rain1 also1 falls at rare intervals even in the driest districts1 as1 for instance1 during forty days1 in #agbf1 at ,chocope 7#g #df72 this rain entirely ruined 78,ulloa10 etc41 p4 #ah7 the mud houses of the inhabitants4 ,fossil-remains of human art4--,in the midst of these shells on ,san ,lorenzo1 ,i found light corallines1 the horny ovule-cases of ,mollusca1 roots of seaweed1#ca bones of birds1 the heads of ,indian corn and other vegetable matter1 a piece of woven rushes1 and another of nearly decayed cotton string4 ,i extracted these remains by digging a hole1 on a level spot2 and they had all indisputably been embedded with the shells4 ,i compared the plaited rush1 the cotton string1 and ,indian corn1 at the house of an antiquary1 with similar objects1 taken from the ,huacas or burial-grounds of the ancient ,peruvians1 and they were undistinguishable2 it should be observed that the ,peruvians used string only of cotton4 ,the small quantity of sand or gravel with the shells1 the absence of large stones1 the width and thickness of the bed1 and the time requisite for a ledge to be cut into the sandstone1 all show that these remains were not thrown high up by an earthquake-wave3 on the other hand1 these facts1 together with the number of dead shells1 and of floating objects1 both marine and terrestrial1 both natural and human1 render it almost certain that they were accumulated on a true beach1 since upraised eighty-five feet1 and upraised this much since ,indian man inhabited ,peru4 ,the elevation may have been1 either by several small sudden starts1 or quite gradual2 in this latter case the unrolled shells having been thrown up during gales beyond the reach of the waves which afterwards broke on the slowly emerging land4 ,i have made these remarks1 chiefly because ,i was at first surprised at the complete difference in nature1 between this broad1 smooth1 upraised bed of shells1 and the present shingle-beach at the foot of the low sandstone-cliffs2 but a beach formed1 when the sea is cutting into the land1 as is shown now to be the case by the low bare sandstone-cliffs1 ought not to be compared with a beach accumulated on a gently inclined rocky surface1 at a period when the sea 7probably owing to the elevatory movement in process7 was not able to eat into the land4 ,with respect to the mass of nearly angular1 salt-cemented fragments of sandstone1 which lie under the shells1 and which are so unlike the materials of an ordinary sea-beach2 ,i think it probable after having seen the remarkable effects#cb of the earthquake of #ahce1 in absolutely shattering as if by gunpowder the surface of the primary rocks near ,concepcion1 that a smooth bare surface of stone was left by the sea covered by the shelly mass1 and that afterwards when upraised1 it was superficially shattered by the severe shocks so often experienced here4 #ca ,mr4 ,smith of ,jordan ,hill found pieces of seaweed in an upraised pleistocene deposit in ,scotland4 ,see his admirable ,paper in the ,edin4 ,new ,phil4 ,journal1 vol4 xxv1 p4 #chd4 #cb ,i have described this in my 8,journal of ,researches10 p4 #cjc1 #bnd edit4 ,the very low land surrounding the town of ,callao1 is to the south joined by an obscure escarpment to a higher plain 7south of ,bella ,vista71 which stretches along the coast for a length of about eight miles4 ,this plain appears to the eye quite level2 but the sea-cliffs show that its height varies 7as far as ,i could estimate7 from seventy to one hundred and twenty feet4 ,it is composed of thin1 sometimes waving1 beds of clay1 often of bright red and yellow colours1 of layers of impure sand1 and in one part with a great stratified mass of granitic pebbles4 ,these beds are capped by a remarkable mass1 varying from two to six feet in thickness1 of reddish loam or mud1 containing many scattered and broken fragments of recent marine shells1 sometimes though rarely single large round pebble1 more frequently short irregular layers of fine gravel1 and very many pieces of red coarse earthenware1 which from their curvatures must once have formed parts of large vessels4 ,the earthenware is of ,indian manufacture2 and ,i found exactly similar pieces accidentally included within the bricks1 of which the neighbouring ancient ,peruvian burial-mounds are built4 ,these fragments abounded in such numbers in certain spots1 that it appeared as if waggon-loads of earthenware had been smashed to pieces4 ,the broken sea-shells and pottery are strewed both on the surface1 and throughout the whole thickness of this upper loamy mass4 ,i found them wherever ,i examined the cliffs1 for a space of between two and three miles1 and for half a mile inland2 and there can be little doubt that this same bed extends with a smooth surface several miles further over the entire plain4 ,besides the little included irregular layers of small pebbles1 there are occasionally very obscure traces of stratification4 ,at one of the highest parts of the cliff1 estimated #abj feet above the sea1 where a little ravine came down1 there were two sections1 at right angles to each other1 of the floor of a shed or building4 ,in both sections or faces1 two rows1 one over the other1 of large round stones could be distinctly seen2 they were packed close together on an artificial layer of sand two inches thick1 which had been placed on the natural clay-beds2 the round stones were covered by three feet in thickness of the loam with broken sea-shells and pottery4 ,hence1 before this widely spread-out bed of loam was deposited1 it is certain that the plain was inhabited2 and it is probable1 from the broken vessels being so much more abundant in certain spots than in others1 and from the underlying clay being fitted for their manufacture1 that the kilns stood here4 ,the smoothness and wide extent of the plain1 the bulk of matter deposited1 and the obscure traces of stratification seem to indicate that the loam was deposited under water2 on the other hand1 the presence of sea-shells1 their broken state1 the pebbles of various sizes1 and the artificial floor of round stones1 almost prove that it must have originated in a rush of water from the sea over the land4 ,the height of the plain1 namely1 #abj feet1 renders it improbable that an earthquake-wave1 vast as some have here been1 could have broken over the surface at its present level2 but when the land stood eighty-five feet lower1 at the period when the shells were thrown up on the ledge at ,s4 ,lorenzo1 and when as we know man inhabited this district1 such an event might well have occurred2 and if we may further suppose1 that the plain was at that time converted into a temporary lake1 as actually occurred1 during the earthquakes of #agac and #agdf1 in the case of the low land round ,callao owing to its being encircled by a high shingle-beach1 all the appearances above described will be perfectly explained4 ,i must add1 that at a lower level near the point where the present low land round ,callao joins the higher plain1 there are appearances of two distinct deposits both apparently formed by debacles3 in the upper one1 a horse's tooth and a dog's jaw were embedded2 so that both must have been formed after the settlement of the ,spaniards3 according to ,acosta1 the earthquake-wave of #aehf rose eighty-four feet4 ,the inhabitants of ,callao do not believe1 as far as ,i could ascertain1 that any change in level is now in progress4 ,the great fragments of brickwork1 which it is asserted can be seen at the bottom of the sea1 and which have been adduced as a proof of a late subsidence1 are1 as ,i am informed by ,mr4 ,gill1 a resident engineer1 loose fragments2 this is probable1 for ,i found on the beach1 and not near the remains of any building1 masses of brickwork1 three and four feet square1 which had been washed into their present places1 and smoothed over with shingle during the earthquake of #agdf4 ,the spit of land1 on which the ruins of ,old ,callao stand1 is so extremely low and narrow1 that it is improbable in the highest degree that a town should have been founded on it in its present state2 and ,i have lately heard#cc that ,m4 ,tschudi has come to the conclusion1 from a comparison of old with modern charts1 that the coast both south and north of ,callao has subsided4 ,i have shown that the island of ,san ,lorenzo has been upraised eighty-five feet since the ,peruvians inhabited this country2 and whatever may have been the amount of recent subsidence1 by so much more must the elevation have exceeded the eighty-five feet4 ,in several places#cd in this neighbourhood1 marks of sea-action have been observed3 ,ulloa gives a detailed account of such appearances at a point five leagues northward of ,callao3 ,mr4 ,cruikshank found near ,lima successive lines of sea-cliffs1 with rounded blocks at their bases1 at a height of #gjj feet above the present level of the sea4 #cc ,i am indebted for this fact to ,dr4 ,e4 ,dieffenbach4 ,i may add that there is a tradition1 that the islands of ,san ,lorenzo and ,fronton were once joined1 and that the channel between ,san ,lorenzo and the mainland1 now above two miles in width1 was so narrow that cattle used to swim over4 #cd 8,observaciones sobre el ,clima del ,lima0 par ,dr4 ,h4 ,unane1 p4 #d4--,ulloa's 8,voyage10 vol4 ii1 ,eng4 ,trans41 p4 #ig4--;,for ,mr4 ,cruikshank's observations1 see ,mr4 ,lyell's 8,principles of ,geology0 7#ast edition7 vol4 iii1 p4 #acj4 ,on the decay of upraised sea-shells4--,i have stated that many of the shells on the lower inclined ledge or terrace of ,san ,lorenzo are corroded in a peculiar manner1 and that they have a much more ancient appearance than the same species at considerably greater heights on the coast of ,chile4 ,i have1 also1 stated that these shells in the upper part of the ledge1 at the height of eighty-five feet above the sea1 are falling1 and in some parts are quite changed into a fine1 soft1 saline1 calcareous powder4 ,the finest part of this powder has been analysed for me1 at the request of ,sir ,h4 ,de la ,beche1 by the kindness of ,mr4 ,trenham ,reeks of the ,museum of ,economic ,geology2 it consists of carbonate of lime in abundance1 of sulphate and muriate of lime1 and of muriate and sulphate of soda4 ,the carbonate of lime is obviously derived from the shells2 and common salt is so abundant in parts of the bed1 that1 as before remarked1 the univalves are often filled with it4 ,the sulphate of lime may have been derived1 as has probably the common salt1 from the evaporation of the sea-spray1 during the emergence of the land2 for sulphate of lime is now copiously deposited from the spray on the shores of ,ascension4#ce ,the other saline bodies may perhaps have been partially thus derived1 but chiefly1 as ,i conclude from the following facts1 through a different means4 #ce ,see 8,volcanic ,islands10 etc41 by the ,author4 ,on most parts of the second ledge or old sea-beach1 at a height of #agj feet1 there is a layer of white powder of variable thickness1 as much in some parts as two inches1 lying on the angular1 salt-cemented fragments of sandstone and under about four inches of earth1 which powder1 from its close resemblance in nature to the upper and most decayed parts of the shelly mass1 ,i can hardly doubt originally existed as a bed of shells1 now much collapsed and quite disintegrated4 ,i could not discover with the microscope a trace of organic structure in it2 but its chemical constituents1 according to ,mr4 ,reeks1 are the same as in the powder extracted from amongst the decaying shells on the lower ledge1 with the marked exception that the carbonate of lime is present in only very small quantity4 ,on the third and highest ledge1 ,i observed some of this powder in a similar position1 and likewise occasionally in small patches at considerably greater heights near the summit of the island4 ,at ,iquique1 where the whole face of the country is covered by a highly saliferous alluvium1 and where the climate is extremely dry1 we have seen that1 according to ,mr4 ,blake1 the shells which are perfect near the beach become1 in ascending1 gradually less and less perfect1 until scarcely a trace of their original structure can be discovered4 ,it is known that carbonate of lime and common salt left in a mass together1#cf and slightly moistened1 partially decompose each other3 now we have at ,san ,lorenzo and at ,iquique1 in the shells and salt packed together1 and occasionally moistened by the so-called ,peruvian dew1 the proper elements for this action4 ,we can thus understand the peculiar corroded appearance of the shells on ,san ,lorenzo1 and the great decrease of quantity in the carbonate of lime in the powder on the upper ledge4 ,there is1 however1 a great difficulty on this view1 for the resultant salts should be carbonate of soda and muriate of lime2 the latter is present1 but not the carbonate of soda4 ,hence ,i am led to the perhaps unauthorised conjecture 7which ,i shall hereafter have to refer to7 that the carbonate of soda1 by some unexplained means1 becomes converted into a sulphate4 ,if the above remarks be just1 we are led to the very unexpected conclusion1 that a dry climate1 by leaving the salt from the sea-spray undissolved1 is much less favourable to the preservation of upraised shells than a humid climate4 ,however this may be1 it is interesting to know the manner in which masses of shells1 gradually upraised above the sea-level1 decay and finally disappear4 #cf ,i am informed by ,dr4 ,kane1 through ,mr4 ,reeks1 that a manufactory was established on this principle in ,france1 but failed from the small quantity of carbonate of soda produced4 ,sprengel 7,gardeners' ,chron41 #ahde1 p4 #aeg7 states1 that salt and carbonate of lime are liable to mutual decomposition in the soil4 ,sir ,h4 ,de la ,beche informs me1 that calcareous rocks washed by the spray of the sea1 are often corroded in a peculiar manner2 see also on this latter subject ,gardeners' ,chron41 p4 #fge1 #ahdd4 ,summary on the recent elevation of the west coast of ,south ,america4--,we have seen that upraised marine remains occur at intervals1 and in some parts almost continuously1 from lat4 #de #ce to #ab ,s41 along the shores of the ,pacific4 ,this is a distance1 in a north and south line1 of #b1jge geographical miles4 ,from ,byron's observations1 the elevation has no doubt extended sixty miles further south2 and from the similarity in the form of the country near ,lima1 it has probably extended many leagues further north4#cg ,along this great line of coast1 besides the organic remains1 there are in very many parts1 marks of erosion1 caves1 ancient beaches1 sand-dunes1 and successive terraces of gravel1 all above the present level of the sea4 ,from the steepness of the land on this side of the continent1 shells have rarely been found at greater distances inland than from two to three leagues2 but the marks of sea-action are evident farther from the coast2 for instance1 in the valley of ,guasco1 at a distance of between thirty and forty miles4 ,judging from the upraised shells alone1 the elevation in ,chiloe has been #cej feet1 at ,concepcion certainly #fbe feet2 and by estimation #a1jjj feet2 at ,valparaiso #a1cjj feet2 at ,coquimbo #beb feet2 northward of this place1 sea-shells have not1 ,i believe1 been found above #cjj feet2 and at ,lima they were falling into decay 7hastened probably by the salt7 at #he feet4 ,not only has this amount of elevation taken place within the period of existing ,mollusca and ,cirripedes2 but their proportional numbers in the neighbouring sea have in most cases remained the same4 ,near ,lima1 however1 a small change in this respect between the living and the upraised was observed3 at ,coquimbo this was more evident1 all the shells being existing species1 but with those embedded in the uppermost calcareous plain not approximating so closely in proportional numbers1 as do those that lie loose on its surface at the height of #beb feet1 and still less closely than those which are strewed on the lower plains1 which latter are identical in proportional numbers with those now cast up on the beach4 ,from this circumstance1 and from not finding1 upon careful examination1 near ,coquimbo any shells at a greater height than #beb feet1 ,i believe that the recent elevation there has been much less than at ,valparaiso1 where it has been #a1cjj feet1 and ,i may add1 than at ,concepcion4 ,this considerable inequality in the amount of elevation at ,coquimbo and ,valparaiso1 places only #bjj miles apart1 is not improbable1 considering1 first1 the difference in the force and number of the shocks now yearly affecting different parts of this coast2 and1 secondly1 the fact of single areas1 such as that of the province of ,concepcion1 having been uplifted very unequally during the same earthquake4 ,it would1 in most cases1 be very hazardous to infer an inequality of elevation1 from shells being found on the surface or in superficial beds at different heights2 for we do not know on what their rate of decay depends2 and at ,coquimbo one instance out of many has been given1 of a promontory1 which1 from the occurrence of one very small collection of lime-cemented shells1 has indisputably been elevated #bdb feet1 and yet on which1 not even a fragment of shell could be found on careful examination between this height and the beach1 although many sites appeared very favourable for the preservation of organic remains3 the absence1 also1 of shells on the gravel-terraces a short distance up the valley of ,coquimbo1 though abundant on the corresponding terraces at its mouth1 should be borne in mind4 #cg ,i may take this opportunity of stating that in a ,,ms4 in the ,geological ,society by ,mr4 ,weaver1 it is stated that beds of oysters and other recent shells are found thirty feet above the level of the sea1 in many parts of ,tampico1 in the ,gulf of ,mexico4 ,there are other epochs1 besides that of the existence of recent ,mollusca1 by which to judge of the changes of level on this coast4 ,at ,lima1 as we have just seen1 the elevation has been at least eighty-five feet1 within the ,indo-human period2 and since the arrival of the ,spaniards in #aecj1 there has apparently been a sinking of the surface4 ,at ,valparaiso1 in the course of #bbj years1 the rise must have been less than nineteen feet2 but it has been as much as from ten to eleven feet in the seventeen years subsequently to #ahag1 and of this rise only a part can be attributed to the earthquake of #ahbb1 the remainder having been insensible and apparently still1 in #ahcd1 in progress4 ,at ,chiloe the elevation has been gradual1 and about four feet during four years4 ,at ,coquimbo1 also1 it has been gradual1 and in the course of #aej years has amounted to several feet4 ,the sudden small upheavals1 accompanied by earthquakes1 as in #ahbb at ,valparaiso1 in #ahce at ,concepcion1 and in #ahcg in the ,chonos ,archipelago1 are familiar to most geologists1 but the gradual rising of the coast of ,chile has been hardly noticed2 it is1 however1 very important1 as connecting together these two orders of events4 ,the rise of ,lima1 having been eighty-five feet within the period of man1 is the more surprising if we refer to the eastern coast of the continent1 for at ,port ,s4 ,julian1 in ,patagonia1 there is good evidence 7as we shall hereafter see7 that when the land stood ninety feet lower1 the ,macrauchenia1 a mammiferous beast1 was alive2 and at ,bahia ,blanca1 when it stood only a few feet lower than it now does1 many gigantic quadrupeds ranged over the adjoining country4 ,but the coast of ,patagonia is some way distant from the ,cordillera1 and the movement at ,bahia ,blanca is perhaps noways connected with this great range1 but rather with the tertiary volcanic rocks of ,banda ,oriental1 and therefore the elevation at these places may have been infinitely slower than on the coast of ,peru4 ,all such speculations1 however1 must be vague1 for as we know with certainty that the elevation of the whole coast of ,patagonia has been interrupted by many and long pauses1 who will pretend to say that1 in such cases1 many and long periods of subsidence may not also have been intercalated8 ,in many parts of the coast of ,chile and ,peru there are marks of the action of the sea at successive heights on the land1 showing that the elevation has been interrupted by periods of comparative rest in the upward movement1 and of denudation in the action of the sea4 ,these are plainest at ,chiloe1 where1 in a height of about five hundred feet1 there are three escarpments1--at ,coquimbo1 where in a height of #cfd feet1 there are five1--at ,guasco1 where there are six1 of which five may perhaps correspond with those at ,coquimbo1 but if so1 the subsequent and intervening elevatory movements have been here much more energetic1--at ,lima1 where1 in a height of about #bej feet there are three terraces1 and others1 as it is asserted1 at considerably greater heights4 ,the almost entire absence of ancient marks of sea-action at defined levels along considerable spaces of coast1 as near ,valparaiso and ,concepcion1 is highly instructive1 for as it is improbable that the elevation at these places alone should have been continuous1 we must attribute the absence of such marks to the nature and form of the coast-rocks4 ,seeing over how many hundred miles of the coast of ,patagonia1 and on how many places on the shores of the ,pacific1 the elevatory process has been interrupted by periods of comparative rest1 we may conclude1 conjointly with the evidence drawn from other quarters of the world1 that the elevation of the land is generally an intermittent action4 ,from the quantity of matter removed in the formation of the escarpments1 especially of those of ,patagonia1 it appears that the periods of rest in the movement1 and of denudation of the land1 have generally been very long4 ,in ,patagonia1 we have seen that the elevation has been equable1 and the periods of denudation synchronous over very wide spaces of coast2 on the shores of the ,pacific1 owing to the terraces chiefly occurring in the valleys1 we have not equal means of judging on this point2 and the very different heights of the upraised shells at ,coquimbo1 ,valparaiso1 and ,concepcion seem directly opposed to such a conclusion4 ,whether on this side of the continent the elevation1 between the periods of comparative rest when the escarpments were formed1 has been by small sudden starts1 such as those accompanying recent earthquakes1 or1 as is most probable1 by such starts conjointly with a gradual upward movement1 or by great and sudden upheavals1 ,i have no direct evidence4 ,but as on the eastern coast1 ,i was led to think1 from the analogy of the last hundred feet of elevation in ,la ,plata1 and from the nearly equal size of the pebbles over the entire width of the terraces1 and from the upraised shells being all littoral species1 that the elevation had been gradual2 so do ,i on this western coast1 from the analogy of the movements now in progress1 and from the vast numbers of shells now living exclusively on or close to the beach1 which are strewed over the whole surface of the land up to very considerable heights1 conclude1 that the movement here also has been slow and gradual1 aided probably by small occasional starts4 ,we know at least that at ,coquimbo1 where five escarpments occur in a height of #cfd feet1 the successive elevations1 if they have been sudden1 cannot have been very great4 ,it has1 ,i think1 been shown that the occasional preservation of shells1 unrolled and unbroken1 is not improbable even during a quite gradual rising of the land2 and their preservation1 if the movement has been aided by small starts1 is quite conformable with what actually takes place during recent earthquakes4 ,judging from the present action of the sea1 along the shores of the ,pacific1 on the deposits of its own accumulation1 the present time seems in most places to be one of comparative rest in the elevatory movement1 and of denudation of the land4 ,undoubtedly this is the case along the whole great length of ,patagonia4 ,at ,chiloe1 however1 we have seen that a narrow sloping fringe1 covered with vegetation1 separates the present sea-beach from a line of low cliffs1 which the waves lately reached2 here1 then1 the land is gaining in breadth and height1 and the present period is not one of rest in the elevation and of contingent denudation2 but if the rising be not prolonged at a quick rate1 there is every probability that the sea will soon regain its former horizontal limits4 ,i observed similar low sloping fringes on several parts of the coast1 both northward of ,valparaiso and near ,coquimbo2 but at this latter place1 from the change in form which the coast has undergone since the old escarpments were worn1 it may be doubted whether the sea1 acting for any length of time at its present level1 would eat into the land2 for it now rather tends to throw up great masses of sand4 ,it is from facts such as these that ,i have generally used the term comparative rest1 as applied to the elevation of the land2 the rest or cessation in the movement being comparative both with what has preceded it and followed it1 and with the sea's power of corrosion at each spot and at each level4 ,near ,lima1 the cliff-formed shores of ,san ,lorenzo1 and on the mainland south of ,callao1 show that the sea is gaining on the land2 and as we have here some evidence that its surface has lately subsided or is still sinking1 the periods of comparative rest in the elevation and of contingent denudation1 may probably in many cases include periods of subsidence4 ,it is only1 as was shown in detail when discussing the terraces of ,coquimbo1 when the sea with difficulty and after a long lapse of time has either corroded a narrow ledge into solid rock1 or has heaped up on a steep surface a narrow mound of detritus1 that we can confidently assert that the land at that level and at that period long remained absolutely stationary4 ,in the case of terraces formed of gravel or sand1 although the elevation may have been strictly horizontal1 it may well happen that no one level beach-line may be traceable1 and that neither the terraces themselves nor the summit nor basal edges of their escarpments may be horizontal4 ,finally1 comparing the extent of the elevated area1 as deduced from the upraised recent organic remains1 on the two sides of the continent1 we have seen that on the ,atlantic1 shells have been found at intervals from ,eastern ,tierra del ,fuego for #a1ahj miles northward1 and on the ,pacific for a space of #b1jge miles4 ,for a length of #gge miles1 they occur in the same latitudes on both sides of the continent4 ,without taking this circumstance into consideration1 it is probable from the reasons assigned in the last chapter1 that the entire breadth of the continent in ,central ,patagonia has been uplifted in mass2 but from other reasons there given1 it would be hazardous to extend this conclusion to ,la ,plata4 ,from the continent being narrow in the southern-most parts of ,patagonia1 and from the shells found at the ,inner ,narrows of the ,strait of ,magellan1 and likewise far up the valley of the ,santa ,cruz1 it is probable that the southern part of the western coast1 which was not visited by me1 has been elevated within the period of recent ,mollusca3 if so1 the shores of the ,pacific have been continuously1 recently1 and in a geological sense synchronously upraised1 from ,lima for a length of #b1dhj nautical miles southward1--a distance equal to that from the ,red ,sea to the ,north ,cape of ,scandinavia6 ,chapter ,,iii ,,on ,,the ,,plains ,,and ,,valleys ,,of ,,chile3--,,saliferous ,,superficial ,,deposits4 ,basin-like plains of ,chile2 their drainage1 their marine origin4--,marks of sea-action on the eastern flanks of the ,cordillera4--,sloping terrace-like fringes of stratified shingle within the valleys of the ,cordillera2 their marine origin4--,boulders in the valley of ,cachapual4--,horizontal elevation of the ,cordillera4--,formation of valleys4--,boulders moved by earthquake-waves4--,saline superficial deposits4--,bed of nitrate of soda at ,iquique4--,saline incrustations4--,salt-lakes of ,la ,plata and ,patagonia2 purity of the salt2 its origin4 ,the space between the ,cordillera and the coast of ,chile is on a rude average from eighty to above one hundred miles in width2 it is formed1 either of an almost continuous mass of mountains1 or more commonly of several nearly parallel ranges1 separated by plains2 in the more southern parts of this province the mountains are quite subordinate to the plains2 in the northern part the mountains predominate4 ,the basin-like plains at the foot of the ,cordillera are in several respects remarkable2 that on which the capital of ,chile stands is fifteen miles in width1 in an east and west line1 and of much greater length in a north and south line2 it stands #a1gej feet above the sea2 its surface appears smooth1 but really falls and rises in wide gentle undulations1 the hollows corresponding with the main valleys of the ,cordillera3 the striking manner in which it abruptly comes up to the foot of this great range has been remarked by every author#a since the time of ,molina4 ,near the ,cordillera it is composed of a stratified mass of pebbles of all sizes1 occasionally including rounded boulders3 near its western boundary1 it consists of reddish sandy clay1 containing some pebbles and numerous fragments of pumice1 and sometimes passes into pure sand or into volcanic ashes4 ,at ,podaguel1 on this western side of the plain1 beds of sand are capped by a calcareous tuff1 the uppermost layers being generally hard and substalagmitic1 and the lower ones white and friable1 both together precisely resembling the beds at ,coquimbo1 which contain recent marine shells4 ,abrupt1 but rounded1 hummocks of rock rise out of this plain3 those of ,sta4 ,lucia and ,s4 ,cristoval are formed of greenstone-porphyry almost entirely denuded of its original covering of porphyritic claystone breccia2 on their summits1 many fragments of rock 7some of them kinds not found in situ7 are coated and united together by a white1 friable1 calcareous tuff1 like that found at ,podaguel4 ,when this matter was deposited on the summit of ,s4 ,cristoval1 the water must have stood #idf feet#b above the surface of the surrounding plain4 #a ,this plain is partially separated into two basins by a range of hills2 the southern half1 according to ,meyen 78,reise um ,erde10 ,th4 i1 s4 #bgd71 falls in height1 by an abrupt step1 of between fifteen and twenty feet4 #b ,or #b1fij feet above the sea1 as measured barometrically by ,mr4 ,eck4 ,this tuff appears to the eye nearly pure2 but when placed in acid it leaves a considerable residue of sand and broken crystals1 apparently of feldspar4 ,dr4 ,meyen 78,reise10 ,th4 i1 s4 #bfi7 says he found a similar substance on the neighbouring hill of ,dominico 7and ,i found it also on the ,cerro ,blanco71 and he attributes it to the weathering of the stone4 ,in some places which ,i examined1 its bulk put this view of its origin quite out of the question2 and ,i should much doubt whether the decomposition of a porphyry would1 in any case1 leave a crust chiefly composed of carbonate of lime4 ,the white crust1 which is commonly seen on weathered feldspathic rocks1 does not appear to contain any free carbonate of lime4 ,to the south this basin-like plain contracts1 and rising scarcely perceptibly with a smooth surface1 passes through a remarkable level gap in the mountains1 forming a true land-strait1 and called the ,angostura4 ,it then immediately expands into a second basin-formed plain3 this again to the south contracts into another land-strait1 and expands into a third basin1 which1 however1 falls suddenly in level about forty feet4 ,this third basin1 to the south1 likewise contracts into a strait1 and then again opens into the great plain of ,san ,fernando1 stretching so far south that the snowy peaks of the distant ,cordillera are seen rising above its horizon as above the sea4 ,these plains1 near the ,cordillera1 are generally formed of a thick stratified mass of shingle2#c in other parts1 of a red sandy clay1 often with an admixture of pumiceous matter4 ,although these basins are connected together like a necklace1 in a north and south line1 by smooth land-straits1 the streams which drain them do not all flow north and south1 but mostly westward1 through breaches worn in the bounding mountains2 and in the case of the second basin1 or that of ,rancagua1 there are two distinct breaches4 ,each basin1 moreover1 is not drained singly2 thus1 to give the most striking instance1 but not the only one1 in proceeding southward over the plain of ,rancagua1 we first find the water flowing northward to and through the northern land-strait2 then1 without crossing any marked ridge or watershed1 we see it flowing south-westward towards the northern one of the two breaches in the western mountainous boundary2 and lastly1 again without any ridge1 it flows towards the southern breach in these same mountains4 ,hence the surface of this one basin-like plain1 appearing to the eye so level1 has been modelled with great nicety1 so that the drainage1 without any conspicuous watersheds1 is directed towards three openings in the encircling mountains4#d ,the streams flowing from the southern basin-like plains1 after passing through the breaches to the west1 unite and form the river ,rapel1 which enters the ,pacific near ,navidad4 ,i followed the southernmost branch of this river1 and found that the basin or plain of ,san ,fernando is continuously and smoothly united with those plains1 which were described in the ,second ,chapter1 as being worn near the coast into successive cave-eaten escarpments1 and still nearer to the coast1 as being strewed with upraised recent marine remains4 ,i might have given descriptions of numerous other plains of the same general form1 some at the foot of the ,cordillera1 some near the coast1 and some halfway between these points4 ,i will allude only to one other1 namely1 the plain of ,uspallata1 lying on the eastern or opposite side of the ,cordillera1 between that great range and the parallel lower range of ,uspallata4 ,according to ,miers1 its surface is #f1jjj feet above the level of the sea3 it is from ten to fifteen miles in width1 and is said to extend with an unbroken surface for #ahj miles northwards3 it is drained by two rivers passing through breaches in the mountains to the east4 ,on the banks of the ,river ,mendoza it is seen to be composed of a great accumulation of stratified shingle1 estimated at #djj feet in thickness4 ,in general appearance1 and in numerous points of structure1 this plain closely resembles those of ,chile4 #c ,the plain of ,san ,fernando has1 according to ,,mm4 ,meyen and ,gay 8,reise10 etc41 ,th4 i1 ss4 #bie and #bih1 near the ,cordillera1 an upper step-formed plain of clay1 on the surface of which they found numerous blocks of rocks1 from two to three feet long1 either lying single or piled in heaps1 but all arranged in nearly straight lines4 #d ,it appears from ,captain ,herbert's account of the ,diluvium of the ,himalaya1 78,gleanings of ,science10 ,calcutta1 vol4 ii1 p4 #afd71 that precisely similar remarks apply to the drainage of the plains or valleys between those great mountains4 ,the origin and manner of formation of the thick beds of gravel1 sandy clay1 volcanic detritus1 and calcareous tuff1 composing these basin-like plains1 is very important2 because1 as we shall presently show1 they send arms or fringes far up the main valleys of the ,cordillera4 ,many of the inhabitants believe that these plains were once occupied by lakes1 suddenly drained2 but ,i conceive that the number of the separate breaches at nearly the same level in the mountains surrounding them quite precludes this idea4 ,had not such distinguished naturalists as ,,mm4 ,meyen and ,gay stated their belief that these deposits were left by great debacles rushing down from the ,cordillera1 ,i should not have noticed a view1 which appears to me from many reasons improbable in the highest degree--namely1 from the vast accumulation of well-rounded pebbles--their frequent stratification with layers of sand--the overlying beds of calcareous tuff--this same substance coating and uniting the fragments of rock on the hummocks in the plain of ,santiago--and lastly even from the worn1 rounded1 and much denuded state of these hummocks1 and of the headlands which project from the surrounding mountains4 ,on the other hand1 these several circumstances1 as well as the continuous union of the basins at the foot of the ,cordillera1 with the great plain of the ,rio ,rapel which still retains the marks of sea-action at various levels1 and their general similarity in form and composition with the many plains near the coast1 which are either similarly marked or are strewed with upraised marine remains1 fully convince me that the mountains bounding these basin-plains were breached1 their islet-like projecting rocks worn1 and the loose stratified detritus forming their now level surfaces deposited1 by the sea1 as the land slowly emerged4 ,it is hardly possible to state too strongly the perfect resemblance in outline between these basin-like1 long1 and narrow plains of ,chile 7especially when in the early morning the mists hanging low represented water71 and the creeks and fiords now intersecting the southern and western shores of the continent4 ,we can on this view of the sea1 when the land stood lower1 having long and tranquilly occupied the spaces between the mountain-ranges1 understand how the boundaries of the separate basins were breached in more than one place2 for we see that this is the general character of the inland bays and channels of ,tierra del ,fuego2 we there1 also1 see in the sawing action of the tides1 which flow with great force in the cross channels1 a power sufficient to keep the breaches open as the land emerged4 ,we can further see that the waves would naturally leave the smooth bottom of each great bay or channel1 as it became slowly converted into land1 gently inclined to as many points as there were mouths1 through which the sea finally retreated1 thus forming so many watersheds1 without any marked ridges1 on a nearly level surface4 ,the absence of marine remains in these high inland plains cannot be properly adduced as an objection to their marine origin3 for we may conclude1 from shells not being found in the great shingle beds of ,patagonia1 though copiously strewed on their surfaces1 and from many other analogous facts1 that such deposits are eminently unfavourable for the embedment of such remains2 and with respect to shells not being found strewed on the surface of these basin-like plains1 it was shown in the last chapter that remains thus exposed in time decay and disappear4 ,no4 #ac ,section of the plain at the eastern foot of the ,chilian ,cordillera4 ,illustration3 ,section of the plain at the eastern foot of the ,chilian ,cordillera4 ,i observed some appearances on the plains at the eastern and opposite foot of the ,cordillera which are worth notice1 as showing that the sea there long acted at nearly the same level as on the basin-plains of ,chile4 ,the mountains on this eastern side are exceedingly abrupt2 they rise out of a smooth1 talus-like1 very gentle1 slope1 from five to ten miles in width 7as represented in ,figure #ac71 entirely composed of perfectly rounded pebbles1 often white-washed with an aluminous substance like decomposed feldspar4 ,this sloping plain or talus blends into a perfectly flat space a few miles in width1 composed of reddish impure clay1 with small calcareous concretions as in the ,pampean deposit1--of fine white sand with small pebbles in layers1--and of the above-mentioned white aluminous earth1 all interstratified together4 ,this flat space runs as far as ,mendoza1 thirty miles northward1 and stands probably at about the same height1 namely1 #b1gjj feet 7,pentland and ,miers7 above the sea4 ,to the east it is bounded by an escarpment1 eighty feet in height1 running for many miles north and south1 and composed of perfectly round pebbles1 and loose1 white-washed1 or embedded in the aluminous earth3 behind this escarpment there is a second and similar one of gravel4 ,northward of ,mendoza1 these escarpments become broken and quite obliterated2 and it does not appear that they ever enclosed a lake-like area3 ,i conclude1 therefore1 that they were formed by the sea1 when it reached the foot of the ,cordillera1 like the similar escarpments occurring at so many points on the coasts of ,chile and ,patagonia4 ,the talus-like plain slopes up with a smooth surface into the great dry valleys of the ,cordillera4 ,on each hand of the ,portillo valley1 the mountains are formed of red granite1 mica-slate1 and basalt1 which all have suffered a truly astonishing amount of denudation2 the gravel in the valley1 as well as on the talus-like plain in front of it1 is composed of these rocks2 but at the mouth of the valley1 in the middle 7height probably about three thousand five hundred feet above the sea71 a few small isolated hillocks of several varieties of porphyry project1 round which1 on all sides1 smooth and often white-washed pebbles of these same porphyries1 to the exclusion of all others1 extend to a circumscribed distance4 ,now1 it is difficult to conceive any other agency1 except the quiet and long-continued action of the sea on these hillocks1 which could have rounded and whitewashed the fragments of porphyry1 and caused them to radiate from such small and quite insignificant centres1 in the midst of that vast stream of stones which has descended from the main ,cordillera4 ,sloping terraces of gravel in the valleys of the ,cordillera4--,all the main valleys on both flanks of the ,chilean ,cordillera have formerly had1 or still have1 their bottoms filled up to a considerable thickness by a mass of rudely stratified shingle4 ,in ,central ,chile the greater part of this mass has been removed by the torrents2 cliff-bounded fringes1 more or less continuous1 being left at corresponding heights on both sides of the valleys4 ,these fringes1 or as they may be called terraces1 have a smooth surface1 and as the valleys rise1 they gently rise with them3 hence they are easily irrigated1 and afford great facilities for the construction of the roads4 ,from their uniformity1 they give a remarkable character to the scenery of these grand1 wild1 broken valleys4 ,in width1 the fringes vary much1 sometimes being only broad enough for the roads1 and sometimes expanding into narrow plains4 ,their surfaces1 besides gently rising up the valley1 are slightly inclined towards its centre in such a manner as to show that the whole bottom must once have been filled up with a smooth and slightly concave mass1 as still are the dry unfurrowed valleys of ,northern ,chile4 ,where two valleys unite into one1 these terraces are particularly well exhibited1 as is represented in ,figure #ad4 ,the thickness of the gravel forming these fringes1 on a rude average1 may be said to vary from thirty to sixty or eighty feet2 but near the mouths of the valleys it was in several places from two to three hundred feet4 ,the amount of matter removed by the torrents has been immense2 yet in the lower parts of the valleys the terraces have seldom been entirely worn away on either side1 nor has the solid underlying rock been reached3 higher up the valleys1 the terraces have frequently been removed on one or the other side1 and sometimes on both sides2 but in this latter case they reappear after a short interval on the line1 which they would have held had they been unbroken4 ,where the solid rock has been reached1 it has been cut into deep and narrow gorges4 ,still higher up the valleys1 the terraces gradually become more and more broken1 narrower1 and less thick1 until1 at a height of from seven to nine thousand feet1 they become lost1 and blended with the piles of fallen detritus4 ,illustration3 ,ground plan of a bifurcating valley in the ,cordillera4 ,i carefully examined in many places the state of the gravel1 and almost everywhere found the pebbles equally and perfectly rounded1 occasionally with great blocks of rock1 and generally distinctly stratified1 often with parting seams of sand4 ,the pebbles were sometimes coated with a white aluminous1 and less frequently with a calcareous1 crust4 ,at great heights up the valleys the pebbles become less rounded2 and as the terraces become obliterated1 the whole mass passes into the nature of ordinary detritus4 ,i was repeatedly struck with the great difference between this detritus high up the valleys1 and the gravel of the terraces low down1 namely1 in the greater number of the quite angular fragments in the detritus1--in the unequal degree to which the other fragments have been rounded1--in the quantity of associated earth1--in the absence of stratification1--and in the irregularity of the upper surfaces4 ,this difference was likewise well shown at points low down the valleys1 where precipitous ravines1 cutting through mountains of highly coloured rock1 have thrown down wide1 fan-shaped accumulations of detritus on the terraces3 in such cases1 the line of separation between the detritus and the terrace could be pointed out to within an inch or two2 the detritus consisting entirely of angular and only partially rounded fragments of the adjoining coloured rocks2 the stratified shingle 7as ,i ascertained by close inspection1 especially in one case1 in the valley of the ,river ,mendoza7 containing only a small proportion of these fragments1 and those few well rounded4 ,i particularly attended to the appearance of the terraces where the valleys made abrupt and considerable bends1 but ,i could perceive no difference in their structure3 they followed the bends with their usual nearly equable inclination4 ,i observed1 also1 in several valleys1 that wherever large blocks of any rock became numerous1 either on the surface of the terrace or embedded in it1 this rock soon appeared higher up in situ3 thus ,i have noticed blocks of porphyry1 of andesitic syenite1 of porphyry and of syenite1 alternately becoming numerous1 and in each case succeeded by mountains thus constituted4 ,there is1 however1 one remarkable exception to this rule2 for along the valley of the ,cachapual1 ,m4 ,gay found numerous large blocks of white granite1 which does not occur in the neighbourhood4 ,i observed these blocks1 as well as others of andesitic syenite 7not occurring here in situ71 near the baths of ,cauquenes at a height of between two and three hundred feet above the river1 and therefore quite above the terrace or fringe which borders that river2 some miles up the valleys there were other blocks at about the same height4 ,i also noticed1 at a less height1 just above the terrace1 blocks of porphyries 7apparently not found in the immediately impending mountains71 arranged in rude lines1 as on a sea-beach4 ,all these blocks were rounded1 and though large1 not gigantic1 like the true erratic boulders of ,patagonia and ,fuegia4 ,m4 ,gay#e states that the granite does not occur in situ within a distance of twenty leagues2 ,i suspect1 for several reasons1 that it will ultimately be found at a much less distance1 though certainly not in the immediate neighbourhood4 ,the boulders found by ,,mm4 ,meyen and ,gay on the upper plain of ,san ,fernando 7mentioned in a previous note7 probably belong to this same class of phenomena4 #e 8,annales des ,science ,nat40 7,i4 sries1 tome #bh74 ,m4 ,gay1 as ,i was informed1 penetrated the ,cordillera by the great oblique valley of ,los ,cupressos1 and not by the most direct line4 ,these fringes of stratified gravel occur along all the great valleys of the ,cordillera1 as well as along their main branches2 they are strikingly developed in the valleys of the ,maypu1 ,mendoza1 ,aconcagua1 ,cachapual1 and according to ,meyen1#f in the ,tinguirica4 ,x,in the valleys1 however1 of ,northern ,chile1 and in some on the eastern flank of the ,cordillera1 as in the ,portillo ,valley1 where streams have never flowed1 or are quite insignificant in volume1 the presence of a mass of stratified gravel can be inferred only from the smooth slightly concave form of the bottom4 ,one naturally seeks for some explanation of so general and striking a phenomenon2 that the matter forming the fringes along the valleys1 or still filling up their entire beds1 has not fallen from the adjoining mountains like common detritus1 is evident from the complete contrast in every respect between the gravel and the piles of detritus1 whether seen high up the valleys on their sides1 or low down in front of the more precipitous ravines2 that the matter has not been deposited by debacles1 even if we could believe in debacles having rushed down every valley1 and all their branches1 eastward and westward from the central pinnacles of the ,cordillera1 we must admit from the following reasons1--from the distinct stratification of the mass1--its smooth upper surface1--the well-rounded and sometimes encrusted state of the pebbles1 so different from the loose debris on the mountains1--and especially from the terraces preserving their uniform inclination round the most abrupt bends4 ,to suppose that as the land now stands1 the rivers deposited the shingle along the course of every valley1 and all their main branches1 appears to me preposterous1 seeing that these same rivers not only are now removing and have removed much of this deposit1 but are everywhere tending to cut deep and narrow gorges in the hard underlying rocks4 #f 8,reise10 etc41 ,th4 ,i1 s4 #cjb4 ,i have stated that these fringes of gravel1 the origin of which are inexplicable on the notion of debacles or of ordinary alluvial action1 are directly continuous with the similarly-composed basin-like plains at the foot of the ,cordillera1 which1 from the several reasons before assigned1 ,i cannot doubt were modelled by the agency of the sea4 ,now if we suppose that the sea formerly occupied the valleys of the ,chilean ,cordillera1 in precisely the same manner as it now does in the more southern parts of the continent1 where deep winding creeks penetrate into the very heart of1 and in the case of ,obstruction ,sound quite through1 this great range2 and if we suppose that the mountains were upraised in the same slow manner as the eastern and western coasts have been upraised within the recent period1 then the origin and formation of these sloping1 terrace-like fringes of gravel can be simply explained4 ,for every part of the bottom of each valley will1 on this view1 have long stood at the head of a sea creek1 into which the then existing torrents will have delivered fragments of rocks1 where1 by the action of the tides1 they will have been rolled1 sometimes encrusted1 rudely stratified1 and the whole surface levelled by the blending together of the successive beach lines4#g ,as the land rose1 the torrents in every valley will have tended to have removed the matter which just before had been arrested on1 or near1 the beach-lines2 the torrents1 also1 having continued to gain in force by the continued elevation increasing their total descent from their sources to the sea4 ,this slow rising of the ,cordillera1 which explains so well the otherwise inexplicable origin and structure of the terraces1 judging from all known analogies1 will probably have been interrupted by many periods of rest2 but we ought not to expect to find any evidence of these periods in the structure of the gravel-terraces3 for1 as the waves at the heads of deep creeks have little erosive power1 so the only effect of the sea having long remained at the same level will be that the upper parts of the creeks will have become filled up at such periods to the level of the water with gravel and sand2 and that afterwards the rivers will have thrown down on the filled-up parts a talus of similar matter1 of which the inclination 7as at the head of a partially filled-up lake7 will have been determined by the supply of detritus1 and the force of the stream4#h ,hence1 after the final conversion of the creeks into valleys1 almost the only difference in the terraces at those points at which the sea stood long1 will be a somewhat more gentle inclination1 with river-worn instead of sea-worn detritus on the surface4 #g ,sloping terraces of precisely similar structure have been described by me 78,philosoph4 ,transactions10 #ahci1 p4 #eh7 in the valleys of ,lochaber in ,scotland1 where1 at higher levels1 the parallel roads of ,glen ,roy show the marks of the long and quiet residence of the sea4 ,i have no doubt that these sloping terraces would have been present in the valleys of most of the ,european ranges1 had not every trace of them1 and all wrecks of sea-action1 been swept away by the glaciers which have since occupied them4 ,i have shown that this is the case with the mountains 7,london and ,edin4 ,phil4 ,journal1 vol4 xxi1 p4 #ahg7 of ,north ,wales4 #h ,i have attempted to explain this process in a more detailed manner1 in a letter to ,mr4 ,maclaren1 published in the ,edinburgh ,new ,phil4 ,journal1 vol4 xxxv1 p4 #bhh4 ,i know of only one difficulty on the foregoing view1 namely1 the far-transported blocks of rock high on the sides of the valley of the ,cachapual3 ,i will not attempt any explanation of this phenomenon1 but ,i may state my belief that a mountain-ridge near the ,baths of ,cauquenes has been upraised long subsequently to all the other ranges in the neighbourhood1 and that when this was effected the whole face of the country must have been greatly altered4 ,in the course of ages1 moreover1 in this and other valleys1 events may have occurred like1 but even on a grander scale than1 that described by ,molina1#i when a slip during the earthquake of #agfb banked up for ten days the great ,river ,lontue1 which then bursting its barrier 8inundated the whole country10 and doubtless transported many great fragments of rock4 ,finally1 notwithstanding this one case of difficulty1 ,i cannot entertain any doubt1 that these terrace-like fringes1 which are continuously united with the basin-shaped plains at the foot of the ,cordillera1 have been formed by the arrestment of river-borne detritus at successive levels1 in the same manner as we see now taking place at the heads of all those many1 deep1 winding fiords intersecting the southern coasts4 ,to my mind1 this has been one of the most important conclusions to which my observations on the geology of ,south ,america have led me2 for we thus learn that one of the grandest and most symmetrical mountain-chains in the world1 with its several parallel lines1#aj has been together uplifted in mass between seven and nine thousand feet1 in the same gradual manner as have the eastern and western coasts within the recent period4 #i 8,compendio de la ,hist410 etc41 tome i1 p4 #cj4 ,m4 ,brongniart1 in his report on ,m4 ,gay's labours 78,annales des ,sciences0 #ahcc71 considers that the boulders in the ,cachapual belong to the same class with the erratic boulders of ,europe4 ,as the blocks which ,i saw are not gigantic1 and especially as they are not angular1 and as they have not been transported fairly across low spaces or wide valleys1 ,i am unwilling to class them with those which1 both in the northern and southern hemisphere 78,geolog4 ,transac410 vol4 vi1 p4 #dae71 have been transported by ice4 ,it is to be hoped that when ,m4 ,gay's long-continued and admirable labours in ,chile are published1 more light will be thrown on this subject4 ,however1 the boulders may have been primarily transported2 the final position of those of porphyry1 which have been described as arranged at the foot of the mountain in rude lines1 ,i cannot doubt1 has been due to the action of waves on a beach4 ,the valley of the ,cachapual1 in the part where the boulders occur1 bursts through the high ridge of ,cauquenes1 which runs parallel to1 but at some distance from1 the ,cordillera4 ,this ridge has been subjected to excessive violence2 trachytic lava has burst from it1 and hot springs yet flow at its base4 ,seeing the enormous amount of denudation of solid rock in the upper and much broader parts of this valley where it enters the ,cordillera1 and seeing to what extent the ridge of ,cauquenes now protects the great range1 ,i could not help believing 7as alluded to in the text7 that this ridge with its trachytic eruptions had been thrown up at a much later period than the ,cordillera4 ,if this has been the case1 the boulders1 after having been transported to a low level by the torrents 7which exhibit in every valley proofs of their power of moving great fragments71 may have been raised up to their present height1 with the land on which they rested4 #aj ,i do not wish to affirm that all the lines have been uplifted quite equally2 slight differences in the elevation would leave no perceptible effect on the terraces4 ,it may1 however1 be inferred1 perhaps with one exception1 that since the period when the sea occupied these valleys1 the several ranges have not been dislocated by great and abrupt faults or upheavals2 for if such had occurred1 the terraces of gravel at these points would not have been continuous4 ,the one exception is at the lower end of a plain in the ,valle del ,yeso 7a branch of the ,maypu71 where1 at a great height1 the terraces and valley appear to have been broken through by a line of upheaval1 of which the evidence is plain in the adjoining mountains2 this dislocation1 perhaps1 occurred after the elevation of this part of the valley above the level of the sea4 ,the valley here is almost blocked up by a pile about one thousand feet in thickness1 formed1 as far as ,i could judge1 from three sides1 entirely1 or at least in chief part1 of gravel and detritus4 ,on the south side1 the river has cut quite through this mass2 on the northern side1 and on the very summit1 deep ravines1 parallel to the line of the valley1 are worn1 as if the drainage from the valley above had passed by these two lines before following its present course4 ,formation of ,valleys4 ,the bulk of solid rock which has been removed in the lower parts of the valleys of the ,cordillera has been enormous4 ,it is only by reflecting on such cases as that of the gravel beds of ,patagonia1 covering so many thousand square leagues of surface1 and which1 if heaped into a ridge1 would form a mountain-range almost equal to the ,cordillera1 that the amount of denudation becomes credible4 ,the valleys within this range often follow anticlinal but rarely synclinal lines2 that is1 the strata on the two sides more often dip from the line of valley than towards it4 ,on the flanks of the range1 the valleys most frequently run neither along anticlinal nor synclinal axes1 but along lines of flexure or faults3 that is1 the strata on both sides dip in the same direction1 but with different1 though often only slightly different1 inclinations4 ,as most of the nearly parallel ridges which together form the ,cordillera run approximately north and south1 the east and west valleys cross them in zig-zag lines1 bursting through the points where the strata have been least inclined4 ,no doubt the greater part of the denudation was affected at the periods when tidal-creeks occupied the valleys1 and when the outer flanks of the mountains were exposed to the full force of an open ocean4 ,i have already alluded to the power of the tidal action in the channels connecting great bays2 and ,i may here mention that one of the surveying vessels in a channel of this kind1 though under sail1 was whirled round and round by the force of the current4 ,we shall hereafter see1 that of the two main ridges forming the ,chilean ,cordillera1 the eastern and loftiest one owes the greater part of its angular upheaval to a period subsequent to the elevation of the western ridge2 and it is likewise probable that many of the other parallel ridges have been angularly upheaved at different periods2 consequently many parts of the surfaces of these mountains must formerly have been exposed to the full force of the waves1 which1 if the ,cordillera were now sunk into the sea1 would be protected by parallel chains of islands4 ,the torrents in the valleys certainly have great power in wearing the rocks2 as could be told by the dull rattling sound of the many fragments night and day hurrying downwards2 and as was attested by the vast size of certain fragments1 which ,i was assured had been carried onwards during floods2 yet we have seen in the lower parts of the valleys1 that the torrents have seldom removed all the sea-checked shingle forming the terraces1 and have had time since the last elevation in mass only to cut in the underlying rocks1 gorges1 deep and narrow1 but quite insignificant in dimensions compared with the entire width and depth of the valleys4 ,along the shores of the ,pacific1 ,i never ceased during my many and long excursions to feel astonished at seeing every valley1 ravine1 and even little inequality of surface1 both in the hard granitic and soft tertiary districts1 retaining the exact outline1 which they had when the sea left their surfaces coated with organic remains4 ,when these remains shall have decayed1 there will be scarcely any difference in appearance between this line of coast-land and most other countries1 which we are accustomed to believe have assumed their present features chiefly through the agency of the weather and fresh-water streams4 ,in the old granitic districts1 no doubt it would be rash to attribute all the modifications of outline exclusively to the sea-action2 for who can say how often this lately submerged coast may not previously have existed as land1 worn by running streams and washed by rain8 ,this source of doubt1 however1 does not apply to the districts superficially formed of the modern tertiary deposits4 ,the valleys worn by the sea1 through the softer formations1 both on the ,atlantic and ,pacific sides of the continent1 are generally broad1 winding1 and flat-bottomed3 the only district of this nature now penetrated by arms of the sea1 is the island of ,chiloe4 ,finally1 the conclusion at which ,i have arrived1 with respect to the relative powers of rain and sea water on the land1 is1 that the latter is far the most efficient agent1 and that its chief tendency is to widen the valleys2 whilst torrents and rivers tend to deepen them1 and to remove the wreck of the sea's destroying action4 ,as the waves have more power1 the more open and exposed the space may be1 so will they always tend to widen more and more the mouths of valleys compared with their upper parts3 hence1 doubtless1 it is1 that most valleys expand at their mouths1--that part1 at which the rivers flowing in them1 generally have the least wearing power4 ,when reflecting on the action of the sea on the land at former levels1 the effect of the great waves1 which generally accompany earthquakes1 must not be overlooked3 few years pass without a severe earthquake occurring on some part of the west coast of ,south ,america2 and the waves thus caused have great power4 ,at ,concepcion1 after the shock of #ahce1 ,i saw large slabs of sandstone1 one of which was six feet long1 three in breadth1 and two in thickness1 thrown high up on the beach2 and from the nature of the marine animals still adhering to it1 it must have been torn up from a considerable depth4 ,on the other hand1 at ,callao1 the recoil-wave of the earthquake of #agdf carried great masses of brickwork1 between three and four feet square1 some way out seaward4 ,during the course of ages1 the effect thus produced at each successive level1 cannot have been small2 and in some of the tertiary deposits on this line of coast1 ,i observed great boulders of granite and other neighbouring rocks1 embedded in fine sedimentary layers1 the transportal of which1 except by the means of earthquake-waves1 always appeared to me inexplicable4 ,superficial ,saline ,deposits4 ,this subject may be here conveniently treated of3 ,i will begin with the most interesting case1 namely1 the superficial saline beds near ,iquique in ,peru4 ,the porphyritic mountains on the coast rise abruptly to a height of between one thousand nine hundred and three thousand feet3 between their summits and an inland plain1 on which the celebrated deposit of nitrate of soda lies1 there is a high undulatory district1 covered by a remarkable superficial saliferous crust1 chiefly composed of common salt1 either in white1 hard1 opaque nodules1 or mingled with sand1 in this latter case forming a compact sandstone4 ,this saliferous superficial crust extends from the edge of the coast-escarpment1 over the whole face of the country2 but never attains1 as ,i am assured by ,mr4 ,bollaert 7long resident here7 any great thickness4 ,although a very slight shower falls only at intervals of many years1 yet small funnel-shaped cavities show that the salt has been in some parts dissolved4#aa ,in several places ,i saw large patches of sand1 quite moist1 owing to the quantity of muriate of lime 7as ascertained by ,mr4 ,t4 ,reeks7 contained in them4 ,from the compact salt-cemented sand being either red1 purplish1 or yellow1 according to the colour of the rocky strata on which it rested1 ,i imagined that this#ab substance had probably been derived through common alluvial action from the layers of salt which occur interstratified in the surrounding mountains3 but from the interesting details given by ,m4 d',orbigny1 and from finding on a fresh examination of this agglomerated sand1 that it is not irregularly cemented1 but consists of thin layers of sand of different tints of colour1 alternating with excessively fine parallel layers of salt1 ,i conclude that it is not of alluvial origin4 ,m4 d',orbigny#ac observed analogous saline beds extending from ,cobija for five degrees of latitude northward1 and at heights varying from six hundred to nine hundred feet3 from finding recent sea-shells strewed on these saliferous beds1 and under them1 great well-rounded blocks1 exactly like those on the existing beach1 he believes that the salt1 which is invariably superficial1 has been left by the evaporation of the sea-water4 ,this same conclusion must1 ,i now believe1 be extended to the superficial saliferous beds of ,iquique1 though they stand about three thousand feet above the level of the sea4 #aa ,it is singular how slowly1 according to the observations of ,m4 ,cordier on the salt-mountain of ,cardona in ,spain 78,ann4 des ,mines1 ,transl4 of ,geolog4 ,mem40 by ,de la ,beche1 p4 #fj71 salt is dissolved1 where the amount of rain is supposed to be as much as #cad of an inch in the year4 ,it is calculated that only five feet in thickness is dissolved in the course of a century4 #ab 8,journal of ,researches10 p4 #ddd1 first edit4 #ac 8,voyage10 etc41 p4 #ajb4 ,m4 d',orbigny found this deposit intersected1 in many places1 by deep ravines1 in which there was no salt4 ,streams must once1 though historically unknown1 have flowed in them2 and ,m4 d',orbigny argues from the presence of undissolved salt over the whole surrounding country1 that the streams must have arisen from rain or snow having fallen1 not in the adjoining country1 but on the now arid ,cordillera4 ,i may remark1 that from having observed ruins of ,indian buildings in absolutely sterile parts of the ,chilian ,cordillera 78,journal10 #bnd edit41 p4 #ceg71 ,i am led to believe that the climate1 at a time when ,indian man inhabited this part of the continent1 was in some slight degree more humid than it is at present4 ,associated with the salt in the superficial beds1 there are numerous1 thin1 horizontal layers of impure1 dirty-white1 friable1 gypseous and calcareous tuffs4 ,the gypseous beds are very remarkable1 from abounding with1 so as sometimes to be almost composed of1 irregular concretions1 from the size of an egg to that of a man's head1 of very hard1 compact1 heavy gypsum1 in the form of anhydrite4 ,this gypsum contains some foreign particles of stone2 it is stained1 judging from its action with borax1 with iron1 and it exhales a strong aluminous odour4 ,the surfaces of the concretions are marked by sharp1 radiating1 or bifurcating ridges1 as if they had been 7but not really7 corroded3 internally they are penetrated by branching veins 7like those of calcareous spar in the septaria of the ,london clay7 of pure white anhydrite4 ,these veins might naturally have been thought to have been formed by subsequent infiltration1 had not each little embedded fragment of rock been likewise edged in a very remarkable manner by a narrow border of the same white anhydrite3 this shows that the veins must have been formed by a process of segregation1 and not of infiltration4 ,some of the little included and cracked fragments of foreign rock are penetrated by the anhydrite1 and portions have evidently been thus mechanically displaced3 at ,st4 ,helena1 ,i observed that calcareous matter1 deposited by rain water1 also had the power to separate small fragments of rock from the larger masses4#ad ,i believe the superficial gypseous deposit is widely extended3 ,i received specimens of it from ,pisagua1 forty miles north of ,iquique1 and likewise from ,arica1 where it coats a layer of pure salt4 ,m4 d',orbigny#ae found at ,cobija a bed of clay1 lying above a mass of upraised recent shells1 which was saturated with sulphate of soda1 and included thin layers of fibrous gypsum4 ,these widely extended1 superficial1 beds of salt and gypsum1 appear to me an interesting geological phenomenon1 which could be presented only under a very dry climate4 #ad 8,volcanic ,islands10 etc41 p4 #hg4 #ae 8,voyage ,golog410 etc41 p4 #ie4 ,the plain or basin1 on the borders of which the famous bed of nitrate of soda lies1 is situated at the distance of about thirty miles from the sea1 being separated from it by the saliferous district just described4 ,it stands at a height of #c1cjj feet2 its surface is level1 and some leagues in width2 it extends forty miles northward1 and has a total length 7as ,i was informed by ,mr4 ,belford ,wilson1 the ,consul-,general at ,lima7 of #dbj miles4 ,in a well near the works1 thirty-six yards in depth1 sand1 earth1 and a little gravel were found3 in another well1 near ,almonte1 fifty yards deep1 the whole consisted1 according to ,mr4 ,blake1#af of clay1 including a layer of sand two feet thick1 which rested on fine gravel1 and this on coarse gravel1 with large rounded fragments of rock4 ,in many parts of this now utterly desert plain1 rushes and large prostrate trees in a hardened state1 apparently ,mimosas1 are found buried1 at a depth from three to six feet2 according to ,mr4 ,blake1 they have all fallen to the south-west4 ,the bed of nitrate of soda is said to extend for forty to fifty leagues along the western margin of the plain1 but is not found in its central parts3 it is from two to three feet in thickness1 and is so hard that it is generally blasted with gunpowder2 it slopes gently upwards from the edge of the plain to between ten and thirty feet above its level4 ,it rests on sand in which1 it is said1 vegetable remains and broken shells have been found2 shells have also been found1 according to ,mr4 ,blake1 both on and in the nitrate of soda4 ,it is covered by a superficial mass of sand1 containing nodules of common salt1 and1 as ,i was assured by a miner1 much soft gypseous matter1 precisely like that in the superficial crust already described3 certainly this crust1 with its characteristic concretions of anhydrite1 comes close down to the edge of the plain4 #af ,see an admirable paper 8,geolog4 and ,miscell4 ,notices of ,tarapaca10 in ,silliman's ,american ,journal1 vol4 xliv1 p4 #a4 ,the nitrate of soda varies in purity in different parts1 and often contains nodules of common salt4 ,according to ,mr4 ,blake1 the proportion of nitrate of soda varies from #bj to #ge per cent4 ,an analysis by ,mr4 ,a4 ,hayes1 of an average specimen1 gave3-- ,nitrate of ,soda #fdih ,sulphate of ,soda #cjj ,chloride of ,soda #bhfi ,iodic ,salts #jfc ,shells and ,marl #bfj ------ #ii4ij ,the 8mother-water0 at some of the refineries is very rich in iodic salts1 and is supposed#ag to contain much muriate of lime4 ,in an unrefined specimen brought home by myself1 ,mr4 ,t4 ,reeks has ascertained that the muriate of lime is very abundant4 ,with respect to the origin of this saline mass1 from the manner in which the gently inclined1 compact bed follows for so many miles the sinuous margin of the plain1 there can be no doubt that it was deposited from a sheet of water3 from the fragments of embedded shells1 from the abundant iodic salts1 from the superficial saliferous crust occurring at a higher level and being probably of marine origin1 and from the plain resembling in form those of ,chile and that of ,uspallata1 there can be little doubt that this sheet of water was1 at least originally1 connected with the sea4#ah #ag ,literary ,gazette1 #ahda1 p4 #dge4 #ah 7,from an official document1 shown me by ,mr4 ,belford ,wilson1 it appears that the first export of nitrate of soda to ,europe was in ,july #ahcj1 on ,french account1 in a ,british ship3-- ,entire export in ,quintals #ahcj #ag1cjj #ahca #dj1hhe #ahcb #ea1djj #ahcc #ia1cce #ahcd #adi1ech ,the ,spanish quintal nearly equals #ajj ,english pounds4 ,thin1 superficial1 saline incrustations4--,these saline incrustations are common in many parts of ,america3 ,humboldt met with them on the tableland of ,mexico1 and the ,jesuit ,falkner and other authors#ai state that they occur at intervals over the vast plains extending from the mouth of the ,plata to ,rioja and ,catamarca4 ,hence it is that during droughts1 most of the streams in the ,pampas are saline4 ,i nowhere met with these incrustations so abundantly as near ,bahia ,blanca3 square miles of the mud-flats1 which near that place are raised only a few feet above the sea1 just enough to protect them from being overflowed1 appear1 after dry weather1 whiter than the ground after the thickest hoar-frost4 ,after rain the salts disappear1 and every puddle of water becomes highly saline2 as the surface dries1 the capillary action draws the moisture up pieces of broken earth1 dead sticks1 and tufts of grass1 where the salt effloresces4 ,the incrustation1 where thickest1 does not exceed a quarter of an inch4 ,m4 ,parchappe#bj has analysed it2 and finds that the specimens collected at the extreme head of the low plain1 near the ,river ,manuello1 consist of #ic per cent of sulphate of soda1 and #g of common salt2 whilst the specimens taken close to the coast contain only #fc per cent of the sulphate1 and #cg of the muriate of soda4 ,this remarkable fact1 together with our knowledge that the whole of this low muddy plain has been covered by the sea within the recent period1 must lead to the suspicion that the common salt1 by some unknown process1 becomes in time changed into the sulphate4 ,friable1 calcareous matter is here abundant1 and the case of the apparent double decomposition of the shells and salt on ,san ,lorenzo1 should not be forgotten4 #ai ,azara 78,travels10 vol4 i1 p4 #ee7 considers that the ,parana is the eastern boundary of the saliferous region2 but ,i heard of 8salitrales0 in the ,province of ,entre ,rios4 #bj ,m4 d',orbigny's 8,voyage10 etc41 ,part4 ,hist41 tome i1 p4 #ffd4 ,the saline incrustations1 near ,bahia ,blanca1 are not confined to1 though most abundant on1 the low muddy flats2 for ,i noticed some on a calcareous plain between thirty and forty feet above the sea1 and even a little occurs in still higher valleys4 ,low alluvial tracts in the valleys of the ,rivers ,negro and ,colorado are also encrusted1 and in the latter valley such spaces appeared to be occasionally overflowed by the river4 ,i observed saline incrustations in some of the valleys of ,southern ,patagonia4 ,at ,port ,desire a low1 flat1 muddy valley was thickly incrusted by salts1 which on analysis by ,mr4 ,t4 ,reeks1 are found to consist of a mixture of sulphate and muriate of soda1 with carbonate of lime and earthy matter4 ,on the western side of the continent1 the southern coasts are much too humid for this phenomenon2 but in ,northern ,chile ,i again met with similar incrustations4 ,on the hardened mud1 in parts of the broad1 flat-bottomed valley of ,copiapo1 the saline matter encrusts the ground to the thickness of some inches3 specimens1 sent by ,mr4 ,bingley to ,apothecaries' ,hall for analysis1 were said to consist of carbonate and sulphate of soda4 ,much sulphate of soda is found in the desert of ,atacama4 ,in all parts of ,south ,america1 the saline incrustations occur most frequently on low damp surfaces of mud1 where the climate is rather dry2 and these low surfaces have1 in almost every case1 been upraised above the level of the sea1 within the recent period4 ,salt-lakes of ,patagonia and ,la ,plata4--,salinas1 or natural salt-lakes1 occur in various formations on the eastern side of the continent1--in the argillaceo-calcareous deposit of the ,pampas1 in the sandstone of the ,rio ,negro1 where they are very numerous1 in the pumiceous and other beds of the ,patagonian tertiary formation1 and in small primary districts in the midst of this latter formation4 ,port ,s4 ,julian is the most southerly point 7lat4 #di to #ej7 at which salinas are known to occur4#ba ,the depressions1 in which these salt-lakes lie1 are from a few feet to sixty metres1 as asserted by ,m4 d',orbigny1#bb below the surface of the surrounding plains2 and1 according to this same author1 near the ,rio ,negro they all trend1 either in the ,n4,e4 and ,s4,w4 or in ,e4 and ,w4 lines1 coincident with the general slope of the plain4 ,these depressions in the plain generally have one side lower than the others1 but there are no outlets for drainage4 ,under a less dry climate1 an outlet would soon have been formed1 and the salt washed away4 ,the salinas occur at different elevations above the sea2 they are often several leagues in diameter2 they are generally very shallow1 but there is a deep one in a quartz-rock formation near ,c4 ,blanco4 ,in the wet season1 the whole1 or a part1 of the salt is dissolved1 being redeposited during the succeeding dry season4 ,at this period the appearance of the snow-white expanse of salt crystallised in great cubes1 is very striking4 ,in a large salina1 northward of the ,rio ,negro1 the salt at the bottom1 during the whole year1 is between two and three feet in thickness4 #ba ,according to ,azara 78,travels10 vol4 i1 p4 #ef7 there are salt-lakes as far north as ,chaco 7lat4 #be71 on the banks of the ,vermejo4 ,the salt-lakes of ,siberia appear 7,pallas's 8,travels10 ,english ,trans41 vol4 i1 p4 #bhd7 to occur in very similar depressions to those of ,patagonia4 #bb 8,voyage ,golog410 p4 #fc4 ,the salt rests almost always on a thick bed of black muddy sand1 which is fetid1 probably from the decay of the burrowing worms inhabiting it4#bc ,in a salina1 situated about fifteen miles above the town of ,el ,carmen on the ,rio ,negro1 and three or four miles from the banks of that river1 ,i observed that this black mud rested on gravel with a calcareous matrix1 similar to that spread over the whole surrounding plains3 at ,port ,s4 ,julian the mud1 also1 rested on the gravel3 hence the depressions must have been formed anteriorly to1 or contemporaneously with1 the spreading out of the gravel4 ,i was informed that one small salina occurs in an alluvial plain within the valley of the ,rio ,negro1 and therefore its origin must be subsequent to the excavation of that valley4 ,when ,i visited the salina1 fifteen miles above the town1 the salt was beginning to crystallise1 and on the muddy bottom there were lying many crystals1 generally placed crossways of sulphate of soda 7as ascertained by ,mr4 ,reeks71 and embedded in the mud1 numerous crystals of sulphate of lime1 from one to three inches in length3 ,m4 d',orbigny#bd states that some of these crystals are acicular and more than even nine inches in length2 others are macled and of great purity3 those ,i found all contained some sand in their centres4 ,as the black and fetid sand overlies the gravel1 and that overlies the regular tertiary strata1 ,i think there can be no doubt that these remarkable crystals of sulphate of lime have been deposited from the waters of the lake4 ,the inhabitants call the crystals of selenite1 the padre del sal1 and those of the sulphate of soda1 the madre del sal2 they assured me that both are found under the same circumstances in several of the neighbouring salinas2 and that the sulphate of soda is annually dissolved1 and is always crystallised before the common salt on the muddy bottom4#be ,the association of gypsum and salt in this case1 as well as in the superficial deposits of ,iquique1 appears to me interesting1 considering how generally these substances are associated in the older stratified formations4 #bc ,professor ,ehrenberg examined some of this muddy sand1 but was unable to find in it any infusoria4 #bd 8,voyage ,golog410 p4 #fd4 #be ,this is what might have been expected2 for ,m4 ,ballard asserts 7,acad4 des ,sciences1 ,oct4 #g1 #ahdd1 that sulphate of soda is precipitated from solution more readily from water containing muriate of soda in excess1 than from pure water4 ,mr4 ,reeks has analysed for me some of the salt from the salina near the ,rio ,negro2 he finds it composed entirely of chloride of sodium1 with the exception of #jbf of sulphate of lime and of #jbb of earthy matter3 there are no traces of iodic salts4 ,some salt from the salina ,chiquitos1 in the ,pampean formation1 is equally pure4 ,it is a singular fact1 that the salt from these salinas does not serve so well for preserving meat1 as sea-salt from the ,cape de ,verde ,islands2 and a merchant at ,buenos ,ayres told me that he considered it as #ej per cent less valuable4 ,the purity of the ,patagonian salt1 or absence from it of those other saline bodies found in all sea-water1 is the only assignable cause for this inferiority2 a conclusion which is supported by the fact lately ascertained1#bf that those salts answer best for preserving cheese which contain most of the deliquescent chlorides4#bg #bf ,hort4 and ,agricult4 ,gazette1 #ahde1 p4 #ic4 #bg ,it would probably well answer for the merchants of ,buenos ,ayres 7considering the great consumption there of salt for preserving meat7 to import the deliquescent chlorides to mix with the salt from the salinas3 ,i may call attention to the fact1 that at ,iquique1 a large quantity of muriate of lime1 left in the mother-water during the refinement of the nitrate of soda1 is annually thrown away4 ,with respect to the origin of the salt in the salinas1 the foregoing analysis seems opposed to the view entertained by ,m4 d',orbigny and others1 and which seems so probable considering the recent elevation of this line of coast1 namely1 that it is due to the evaporation of sea-water and to the drainage from the surrounding strata impregnated with sea-salt4 ,i was informed 7,i know not whether accurately7 that on the northern side of the salina on the ,rio ,negro1 there is a small brine spring which flows at all times of the year3 if this be so1 the salt in this case at least1 probably is of subterranean origin4 ,it at first appears very singular that fresh water can often be procured in wells1#bh and is sometimes found in small lakes1 quite close to these salinas4 ,i am not aware that this fact bears particularly on the origin of the salt2 but perhaps it is rather opposed to the view of the salt having been washed out of the surrounding superficial strata1 but not to its having been the residue of sea-water1 left in depressions as the land was slowly elevated4 #bh ,sir ,w4 ,parish states 78,buenos ,ayres10 etc41 pp4 #abb and #agj7 that this is the case near the great salinas westward of the ,s4 ,ventana4 ,i have seen similar statements in an ancient ,,ms4 ,journal lately published by ,s4 ,angelis4 ,at ,iquique1 where the surface is so thickly encrusted with saline matter1 ,i tasted water only slightly brackish1 procured in a well thirty-six yards deep2 but here one feels less surprise at its presence1 as pure water might percolate under ground from the not very distant ,cordillera4 ,chapter ,,iv ,,on ,,the ,,formations ,,of ,,the ,,pampas4 ,mineralogical constitution4--,microscopical structure4--,buenos ,ayres1 shells embedded in tosca-rock4--,buenos ,ayres to the ,colorado4--,san ,ventana4--,bahia ,blanca2 ,m4 ,hermoso1 bones and infusoria of2 ,p4 ,alta1 shells1 bones1 and infusoria of2 co-existence of the recent shells and extinct mammifers4--,buenos ,ayres to ,santa ,f4--,skeletons of ,mastodon4--,infusoria4--,inferior marine tertiary strata1 their age4--,horse's tooth4 ,,banda ,,oriental4--,superficial ,pampean formation4--,inferior tertiary strata1 variation of1 connected with volcanic action2 ,macrauchenia ,patachonica at ,san ,julian in ,patagonia1 age of1 subsequent to living mollusca and to the erratic block period4 ,,summary4--,area of ,pampean formation4--,theories of origin4--,source of sediment4--,estuary origin4-- ,contemporaneous with existing mollusca4--,relations to underlying tertiary strata4--,ancient deposit of estuary origin4--,elevation and successive deposition of the ,pampean formation4--,number and state of the remains of mammifers2 their habitation1 food1 extinction1 and range4--,conclusion4--,localities in ,pampas at which mammiferous remains have been found4 ,the ,pampean formation is highly interesting from its vast extent1 its disputed origin1 and from the number of extinct gigantic mammifers embedded in it4 ,it has upon the whole a very uniform character3 consisting of a more or less dull reddish1 slightly indurated1 argillaceous earth or mud1 often1 but not always1 including in horizontal lines concretions of marl1 and frequently passing into a compact marly rock4 ,the mud1 wherever ,i examined it1 even close to the concretions1 did not contain any carbonate of lime4 ,the concretions are generally nodular1 sometimes rough externally1 sometimes stalactiformed2 they are of a compact structure1 but often penetrated 7as well as the mud7 by hair-like serpentine cavities1 and occasionally with irregular fissures in their centres1 lined with minute crystals of carbonate of lime2 they are of white1 brown1 or pale pinkish tints1 often marked by black dendritic manganese or iron2 they are either darker or lighter tinted than the surrounding mass2 they contain much carbonate of lime1 but exhale a strong aluminous odour1 and leave1 when dissolved in acids1 a large but varying residue1 of which the greater part consists of sand4 ,these concretions often unite into irregular strata2 and over very large tracts of country1 the entire mass consists of a hard1 but generally cavernous marly rock3 some of the varieties might be called calcareous tuffs4 ,dr4 ,carpenter has kindly examined under the microscope1 sliced and polished specimens of these concretions1 and of the solid marl-rock1 collected in various places between the ,colorado and ,santa ,fe ,bajada4 ,in the greater number1 ,dr4 ,carpenter finds that the whole substance presents a tolerably uniform amorphous character1 but with traces of incipient crystalline metamorphosis2 in other specimens he finds microscopically minute rounded concretions of an amorphous substance 7resembling in size those in oolitic rocks1 but not having a concentric structure71 united by a cement which is often crystalline4 ,in some1 ,dr4 ,carpenter can perceive distinct traces of shells1 corals1 ,polythalamia1 and rarely of spongoid bodies4 ,for the sake of comparison1 ,i sent ,dr4 ,carpenter specimens of the calcareous rock1 formed chiefly of fragments of recent shells1 from ,coquimbo in ,chile3 in one of these specimens1 ,dr4 ,carpenter finds1 besides the larger fragments1 microscopical particles of shells1 and a varying quantity of opaque amorphous matter2 in another specimen from the same bed1 he finds the whole composed of the amorphous matter1 with layers showing indications of an incipient crystalline metamorphosis3 hence these latter specimens1 both in external appearance and in microscopical structure1 closely resemble those of the ,pampas4 ,dr4 ,carpenter informs me that it is well known that chemical precipitation throws down carbonate of lime in the opaque amorphous state2 and he is inclined to believe that the long-continued attrition of a calcareous body in a state of crystalline or semi-crystalline aggregation 7as1 for instance1 in the ordinary shells of ,mollusca1 which1 when sliced1 are transparent7 may yield the same result4 ,from the intimate relations between all the ,coquimbo specimens1 ,i can hardly doubt that the amorphous carbonate of lime in them has resulted from the attrition and decay of the larger fragments of shell3 whether the amorphous matter in the marly rocks of the ,pampas has likewise thus originated1 it would be hazardous to conjecture4 ,for convenience' sake1 ,i will call the marly rock by the name given to it by the inhabitants1 namely1 ,tosca-rock2 and the reddish argillaceous earth1 ,pampean mud4 ,this latter substance1 ,i may mention1 has been examined for me by ,professor ,ehrenberg1 and the result of his examination will be given under the proper localities4 ,i will commence my descriptions at a central spot1 namely1 at ,buenos ,ayres1 and thence proceed first southward to the extreme limit of the deposit1 and afterwards northward4 ,the plain on which ,buenos ,ayres stands is from thirty to forty feet in height4 ,the ,pampean mud is here of a rather pale colour1 and includes small nearly white nodules1 and other irregular strata of an unusually arenaceous variety of tosca-rock4 ,in a well at the depth of seventy feet1 according to ,ignatio ,nunez1 much tosca-rock was met with1 and at several points1 at one hundred feet deep1 beds of sand have been found4 ,i have already given a list of the recent marine and estuary shells found in many parts on the surface near ,buenos ,ayres1 as far as three or four leagues from the ,plata4 ,specimens from near ,ensenada1 given me by ,sir ,w4 ,parish1 where the rock is quarried just beneath the surface of the plain1 consist of broken bivalves1 cemented by and converted into white crystalline carbonate of lime4 ,i have already alluded1 in the first chapter1 to a specimen 7also given me by ,sir ,w4 ,parish7 from the ,a4 del ,tristan1 in which shells1 resembling in every respect the ,azara labiata1 d',orbigny1 as far as their worn condition permits of comparison1 are embedded in a reddish1 softish1 somewhat arenaceous marly rock3 after careful comparison1 with the aid of a microscope and acids1 ,i can perceive no difference between the basis of this rock and the specimens collected by me in many parts of the ,pampas4 ,i have also stated1 on the authority of ,sir ,w4 ,parish1 that northward of ,buenos ,ayres1 on the highest parts of the plain1 about forty feet above the ,plata1 and two or three miles from it1 numerous shells of the ,azara labiata 7and ,i believe of ,venus sinuosa7 occur embedded in a stratified earthy mass1 including small marly concretions1 and said to be precisely like the great ,pampean deposit4 ,hence we may conclude that the mud of the ,pampas continued to be deposited to within the period of this existing estuary shell4 ,although this formation is of such immense extent1 ,i know of no other instance of the presence of shells in it4 ,buenos ,ayres to the ,rio ,colorado4--,with the exception of a few metamorphic ridges1 the country between these two points1 a distance of #djj geographical miles1 belongs to the ,pampean formation1 and in the southern part is generally formed of the harder and more calcareous varieties4 ,i will briefly describe my route3 about twenty-five miles ,s4,s4,w4 of the capital1 in a well forty yards in depth1 the upper part1 and1 as ,i was assured1 the entire thickness1 was formed of dark red ,pampean mud without concretions4 ,north of the ,river ,salado1 there are many lakes2 and on the banks of one 7near the ,guardia7 there was a little cliff similarly composed1 but including many nodular and stalactiform concretions3 ,i found here a large piece of tessellated armour1 like that of the ,glyptodon1 and many fragments of bones4 ,the cliffs on the ,salado consist of pale-coloured ,pampean mud1 including and passing into great masses of tosca-rock3 here a skeleton of the ,megatherium and the bones of other extinct quadrupeds 7see the list at the end of this chapter7 were found4 ,large quantities of crystallised gypsum 7of which specimens were given me7 occur in the cliffs of this river2 and likewise 7as ,i was assured by ,mr4 ,lumb7 in the ,pampean mud on the ,river ,chuelo1 seven leagues from ,buenos ,ayres3 ,i mention this because ,m4 d',orbigny lays some stress on the supposed absence of this mineral in the ,pampean formation4 ,southward of the ,salado the country is low and swampy1 with tosca-rock appearing at long intervals at the surface4 ,on the banks1 however1 of the ,tapalguen 7sixty miles south of the ,salado7 there is a large extent of tosca-rock1 some highly compact and even semi-crystalline1 overlying pale ,pampean mud with the usual concretions4 ,thirty miles further south1 the small quartz-ridge of ,tapalguen is fringed on its northern and southern flank1 by little1 narrow1 flat-topped hills of tosca-rock1 which stand higher than the surrounding plain4 ,between this ridge and the ,sierra of ,guitru-gueyu1 a distance of sixty miles1 the country is swampy1 with the tosca-rock appearing only in four or five spots3 this sierra1 precisely like that of ,tapalguen1 is bordered by horizontal1 often cliff-bounded1 little hills of tosca-rock1 higher than the surrounding plain4 ,here1 also1 a new appearance was presented in some extensive and level banks of alluvium or detritus of the neighbouring metamorphic rocks2 but ,i neglected to observe whether it was stratified or not4 ,between ,guitru-gueyu and the ,sierra ,ventana1 ,i crossed a dry plain of tosca-rock higher than the country hitherto passed over1 and with small pieces of denuded tableland of the same formation1 standing still higher4 ,the marly or calcareous beds not only come up nearly horizontally to the northern and southern foot of the great quartzose mountains of the ,sierra ,ventana1 but interfold between the parallel ranges4 ,the superficial beds 7for ,i nowhere obtained sections more than twenty feet deep7 retain1 even close to the mountains1 their usual character3 the uppermost layer1 however1 in one place included pebbles of quartz1 and rested on a mass of detritus of the same rock4 ,at the very foot of the mountains1 there were some few piles of quartz and tosca-rock detritus1 including land-shells2 but at the distance of only half a mile from these lofty1 jagged1 and battered mountains1 ,i could not1 to my great surprise1 find on the boundless surface of the calcareous plain even a single pebble4 ,quartz-pebbles1 however1 of considerable size have at some period been transported to a distance of between forty and fifty miles to the shores of ,bahia ,blanca4#a #a ,schmidtmeyer 78,travels in ,chile10 p4 #aej7 states that he first noticed on the ,pampas1 very small bits of red granite1 when fifty miles distant from the southern extremity of the mountains of ,cordova1 which project on the plain1 like a reef into the sea4 ,the highest peak of the ,st4 ,ventana is1 by ,captain ,fitzroy's measurement1 #c1cdj feet1 and the calcareous plain at its foot 7from observations taken by some ,spanish officers#b7 #hdj feet above the sea-level4 ,on the flanks of the mountains1 at a height of three hundred or four hundred feet above the plain1 there were a few small patches of conglomerate and breccia1 firmly cemented by ferruginous matter to the abrupt and battered face of the quartz--traces being thus exhibited of ancient sea-action4 ,the high plain round this range sinks quite insensibly to the eye on all sides1 except to the north1 where its surface is broken into low cliffs4 ,round the ,sierras ,tapalguen1 ,guitru-gueyu1 and between the latter and the ,ventana we have seen 7and shall hereafter see round some hills in ,banda ,oriental71 that the tosca-rock forms low1 flat-topped1 cliff-bounded hills1 higher than the surrounding plains of similar composition4 ,from the horizontal stratification and from the appearance of the broken cliffs1 the greater height of the ,pampean formation round these primary hills ought not to be altogether or in chief part attributed to these several points having been uplifted more energetically than the surrounding country1 but to the argillaceo-calcareous mud having collected round them1 when they existed as islets or submarine rocks1 at a greater height1 than at the bottom of the adjoining open sea2--the cliffs having been subsequently worn during the elevation of the whole country in mass4 #b 8,la ,plata10 etc41 by ,sir ,w4 ,parish1 p4 #adf4 ,southward of the ,ventana1 the plain extends farther than the eye can range2 its surface is not very level1 having slight depressions with no drainage exits2 it is generally covered by a few feet in thickness of sandy earth2 and in some places1 according to ,m4 ,parchappe1#c beds of clay two yards thick4 ,on the banks of the ,sauce1 four leagues ,s4,e4 of the ,ventana1 there is an imperfect section about two hundred feet in height1 displaying in the upper part tosca-rock and in the lower part red ,pampean mud4 ,at the settlement of ,bahia ,blanca1 the uppermost plain is composed of very compact1 stratified tosca-rock1 containing rounded grains of quartz distinguishable by the naked eye3 the lower plain1 on which the fortress stands1 is described by ,m4 ,parchappe#d as composed of solid tosca-rock2 but the sections which ,i examined appeared more like a redeposited mass of this rock1 with small pebbles and fragments of quartz4 ,i shall immediately return to the important sections on the shores of ,bahia ,blanca4 ,twenty miles southward of this place1 there is a remarkable ridge extending ,w4 by ,n4 and ,e4 by ,s41 formed of small1 separate1 flat-topped1 steep-sided hills1 rising between one hundred and two hundred feet above the ,pampean plain at its southern base1 which plain is a little lower than that to the north4 ,the uppermost stratum in this ridge consists of pale1 highly calcareous1 compact tosca-rock1 resting 7as seen in one place7 on reddish ,pampean mud1 and this again on a paler kind3 at the foot of the ridge1 there is a well in reddish clay or mud4 ,i have seen no other instance of a chain of hills belonging to the ,pampean formation2 and as the strata show no signs of disturbance1 and as the direction of the ridge is the same with that common to all the metamorphic lines in this whole area1 ,i suspect that the ,pampean sediment has in this instance been accumulated on and over a ridge of hard rocks1 instead of1 as in the case of the above-mentioned ,sierras1 round their submarine flanks4 ,south of this little chain of tosca-rock1 a plain of ,pampean mud declines towards the banks of the ,colorado3 in the middle a well has been dug in red ,pampean mud1 covered by two feet of white1 softish1 highly calcareous tosca-rock1 over which lies sand with small pebbles three feet in thickness--the first appearance of that vast shingle formation described in the ,first ,chapter4 ,in the first section after crossing the ,colorado1 an old tertiary formation1 namely1 the ,rio ,negro sandstone 7to be described in the next chapter71 is met with3 but from the accounts given me by the ,gauchos1 ,i believe that at the mouth of the ,colorado the ,pampean formation extends a little further southwards4 #c ,m4 d',orbigny1 8,voyage10 ,part4 ,golog41 pp4 #dg1 #dh4 #d ,ibid4 ,bahia ,blanca4--,to return to the shores of this bay4 ,at ,monte ,hermoso there is a good section1 about one hundred feet in height1 of four distinct strata1 appearing to the eye horizontal1 but thickening a little towards the ,n4,w4 ,the uppermost bed1 about twenty feet in thickness1 consists of obliquely laminated1 soft sandstone1 including many pebbles of quartz1 and falling at the surface into loose sand4 ,the second bed1 only six inches thick1 is a hard1 dark-coloured sandstone4 ,the third bed is pale-coloured ,pampean mud2 and the fourth is of the same nature1 but darker coloured1 including in its lower part horizontal layers and lines of concretions of not very compact pinkish tosca-rock4 ,the bottom of the sea1 ,i may remark1 to a distance of several miles from the shore1 and to a depth of between sixty and one hundred feet1 was found by the anchors to be composed of tosca-rock and reddish ,pampean mud4 ,professor ,ehrenberg has examined for me specimens of the two lower beds1 and finds in them three ,polygastrica and six ,phytolitharia4#e ,of these1 only one 7,spongolithis ,fustis87 is a marine form2 five of them are identical with microscopical structures of brackish-water origin1 hereafter to be mentioned1 which form a central point in the ,pampean formation4 ,in these two beds1 especially in the lower one1 bones of extinct mammifers1 some embedded in their proper relative positions and others single1 are very numerous in a small extent of the cliffs4 ,these remains consist of1 first1 the head of ,ctenomys antiquus1 allied to the living ,ctenomys ,braziliensis2 secondly1 a fragment of the remains of a rodent2 thirdly1 molar teeth and other bones of a large rodent1 closely allied to1 but distinct from1 the existing species of ,hydrochoerus1 and therefore probably an inhabitant of fresh water2 fourth and fifthly1 portions of vertebr1 limbs1 ribs1 and other bones of two rodents2 sixthly1 bones of the extremities of some great megatheroid quadruped4#f ,the number of the remains of rodents gives to this collection a peculiar character1 compared with those found in any other locality4 ,all these bones are compact and heavy2 many of them are stained red1 with their surfaces polished2 some of the smaller ones are as black as jet4 #e ,the following list is given in the 8,monatsberichten der knig4 ,akad4 zu ,berlin10 ,april #ahde3-- ,,polygastrica4 ,fragilaria rhabdosoma4 ,gallionella distans4 ,pinnularia8 ,,phytolitharia4 ,lithodontium ,bursa4 ,lithodontium furcatum4 ,lithostylidium exesum4 ,lithostylidium rude4 ,lithostylidium ,serra4 ,spongolithis ,fustis8 #f ,see 8,fossil ,mammalia0 7p4 #aji7 by ,professor ,owen1 in the 8,zoology of the ,voyage of the ,beagle20 and ,catalogue 7p4 #cf7 of ,fossil ,remains in ,museum of ,royal ,college of ,surgeons4 ,monte ,hermoso is between fifty and sixty miles distant in a ,s4,e4 line from the ,ventana1 with the intermediate country gently rising towards it1 and all consisting of the ,pampean formation4 ,what relation1 then1 do these beds1 at the level of the sea and under it1 bear to those on the flanks of the ,ventana1 at the height of #hdj feet1 and on the flanks of the other neighbouring sierras1 which1 from the reasons already assigned1 do not appear to owe their greater height to unequal elevation8 ,when the tosca-rock was accumulating round the ,ventana1 and when1 with the exception of a few small rugged primary islands1 the whole wide surrounding plains must have been under water1 were the strata at ,monte ,hermoso depositing at the bottom of a great open sea1 between eight hundred and one thousand feet in depth8 ,i much doubt this2 for if so1 the almost perfect carcasses of the several small rodents1 the remains of which are so very numerous in so limited a space1 must have been drifted to this spot from the distance of many hundred miles4 ,it appears to me far more probable1 that during the ,pampean period this whole area had commenced slowly rising 7and in the cliffs1 at several different heights we have proofs of the land having been exposed to sea-action at several levels71 and that tracts of land had thus been formed of ,pampean sediment round the ,ventana and the other primary ranges1 on which the several rodents and other quadrupeds lived1 and that a stream 7in which perhaps the extinct aquatic ,hydrochoerus lived7 drifted their bodies into the adjoining sea1 into which the ,pampean mud continued to be poured from the north4 ,as the land continued to rise1 it appears that this source of sediment was cut off2 and in its place sand and pebbles were borne down by stronger currents1 and conformably deposited over the ,pampean strata4 ,punta ,alta is situated about thirty miles higher up on the northern side of this same bay3 it consists of a small plain1 between twenty and thirty feet in height1 cut off on the shore by a line of low cliffs about a mile in length1 represented in figure ,no4 #ae with its vertical scale necessarily exaggerated4 ,the lower bed 7,a7 is more extensive than the upper ones2 it consists of stratified gravel or conglomerate1 cemented by calcareo-arenaceous matter1 and is divided by curvilinear layers of pinkish marl1 of which some are precisely like tosca-rock1 and some more sandy4 ,the beds are curvilinear1 owing to the action of currents1 and dip in different directions2 they include an extraordinary number of bones of gigantic mammifers and many shells4 ,the pebbles are of considerable size1 and are of hard sandstone1 and of quartz1 like that of the ,ventana3 there are also a few well-rounded masses of tosca-rock4 ,no4 #ae ,section of beds with recent shells and extinct mammifers1 at ,punta ,alta in ,bahia ,blanca4 ,illustration3 ,section of beds at ,punta ,alta in ,bahia ,blanca4 ,the second bed 7,b7 is about fifteen feet in thickness1 but towards both extremities of the cliff 7not included in the diagram7 it either thins out and dies away1 or passes insensibly into an overlying bed of gravel4 ,it consists of red1 tough clayey mud1 with minute linear cavities2 it is marked with faint horizontal shades of colour2 it includes a few pebbles1 and rarely a minute particle of shell3 in one spot1 the dermal armour and a few bones of a ,dasypoid quadruped were embedded in it3 it fills up furrows in the underlying gravel4 ,with the exception of the few pebbles and particles of shells1 this bed resembles the true ,pampean mud2 but it still more closely resembles the clayey flats 7mentioned in the ,first ,chapter7 separating the successively rising parallel ranges of sand-dunes4 ,the bed 7,c7 is of stratified gravel1 like the lowest one2 it fills up furrows in the underlying red mud1 and is sometimes interstratified with it1 and sometimes insensibly passes into it2 as the red mud thins out1 this upper gravel thickens4 ,shells are more numerous in it than in the lower gravel2 but the bones1 though some are still present1 are less numerous4 ,in one part1 however1 where this gravel and the red mud passed into each other1 ,i found several bones and a tolerably perfect head of the ,megatherium4 ,some of the large ,volutas1 though embedded in the gravel-bed 7,c71 were filled with the red mud1 including great numbers of the little recent ,paludestrina australis4 ,these three lower beds are covered by an unconformable mantle 7,d7 of stratified sandy earth1 including many pebbles of quartz1 pumice and phonolite1 land and sea-shells4 ,m4 d',orbigny has been so obliging as to name for me the twenty species of ,mollusca embedded in the two gravel beds3 they consist of3-- ,volutella angulata1 d',orbigny1 8,voyage0 ,mollusq4 and ,pal4 ,voluta ,braziliana1 ,sol4 ,olicancilleria ,braziliensis1 d',orbigny4 ,olicancilleria auricularia1 d',orbigny4 ,olivina puelchana1 d',orbigny4 ,buccinanops cochlidium1 d',orbigny4 ,buccinanops globulosum1 d',orbigny4 ,colombella sertulariarum1 d',orbigny4 ,trochus ,patagonicus1 and var4 of ditto1 d',orbigny4 ,paludestrina ,australis1 d',orbigny4 ,fissurella ,patagonica1 d',orbigny4 ,crepidula muricata1 ,lam4 ,venus purpurata1 ,lam4 ,venus rostrata1 ,phillippi4 ,mytilus ,darwinianus1 d',orbigny4 ,nucula semiornata1 d',orbigny4 ,cardita ,patagonica1 d',orbigny4 ,corbula ,patagonica 7871 d',orbigny4 ,pecten tethuelchus1 d',orbigny4 ,ostrea puelchana1 d',orbigny4 ,a living species of ,balanus4 and #bc4 ,an ,astr and encrusting ,flustra1 apparently identical with species now living in the bay4 ,all these shells now live on this coast1 and most of them in this same bay4 ,i was also struck with the fact1 that the proportional numbers of the different kinds appeared to be the same with those now cast up on the beach3 in both cases specimens of ,voluta1 ,crepidula1 ,venus1 and ,trochus are the most abundant4 ,four or five of the species are the same with the upraised shells on the ,pampas near ,buenos ,ayres4 ,all the specimens have a very ancient and bleached appearance1#g and do not emit1 when heated1 an animal odour3 some of them are changed throughout into a white1 soft1 fibrous substance2 others have the space between the external walls1 either hollow1 or filled up with crystalline carbonate of lime4 #g ,a ,bulinus1 mentioned in the ,introduction to the 8,fossil ,mammalia0 in the 8,zoology of the ,voyage of the ,beagle0 has so much fresher an appearance1 than the marine species1 that ,i suspect it must have fallen amongst the others1 and been collected by mistake4 ,the remains of the extinct mammiferous animals1 from the two gravel beds have been described by ,professor ,owen in the 8,zoology of the ,voyage of the ,beagle30 they consist of1 #ast1 one nearly perfect head and three fragments of heads of the ,megatherium ,cuvierii2 #bnd1 a lower jaw of ,megalonyx ,jeffersonii2 #crd1 lower jaw of ,mylodon ,darwinii2 #dth1 fragments of a head of some gigantic ,edental quadruped2 #eth1 an almost entire skeleton of the great ,scelidotherium leptocephalum1 with most of the bones1 including the head1 vertebr1 ribs1 some of the extremities to the claw-bone1 and even1 as remarked by ,professor ,owen1 the knee-cap1 all nearly in their proper relative positions2 #fth1 fragments of the jaw and a separate tooth of a ,toxodon1 belonging either to ,t4 ,platensis1 or to a second species lately discovered near ,buenos ,ayres2 #gth1 a tooth of ,equus curvidens2 #hth1 tooth of a ,pachyderm1 closely allied to ,palotherium1 of which parts of the head have been lately sent from ,buenos ,ayres to the ,british ,museum2 in all probability this pachyderm is identical with the ,macrauchenia ,patagonica from ,port ,s4 ,julian1 hereafter to be referred to4 ,lastly1 and #ithly1 in a cliff of the red clayey bed 7,b71 there was a double piece1 about three feet long and two wide1 of the bony armour of a large ,dasypoid quadruped1 with the two sides pressed nearly close together3 as the cliff is now rapidly washing away1 this fossil probably was lately much more perfect2 from between its doubled-up sides1 ,i extracted the middle and ungual phalanges1 united together1 of one of the feet1 and likewise a separate phalanx3 hence one or more of the limbs must have been attached to the dermal case1 when it was embedded4 ,besides these several remains in a distinguishable condition1 there were very many single bones3 the greater number were embedded in a space #bjj yards square4 ,the preponderance of the ,edental quadrupeds is remarkable2 as is1 in contrast with the beds of ,monte ,hermoso1 the absence of ,rodents4 ,most of the bones are now in a soft and friable condition1 and1 like the shells1 do not emit when burnt an animal odour4 ,the decayed state of the bones may be partly owing to their late exposure to the air and tidal-waves4 ,barnacles1 ,serpul1 and corallines are attached to many of the bones1 but ,i neglected to observe#h whether these might not have grown on them since being exposed to the present tidal action2 but ,i believe that some of the barnacles must have grown on the ,scelidotherium1 soon after being deposited1 and before being wholly covered up by the gravel4 ,besides the remains in the condition here described1 ,i found one single fragment of bone very much rolled1 and as black as jet1 so as perfectly to resemble some of the remains from ,monte ,hermoso4 #h ,after having packed up my specimens at ,bahia ,blanca1 this point occurred to me1 and ,i noted it2 but forgot it on my return1 until the remains had been cleaned and oiled3 my attention has been lately called to the subject by some remarks by ,m4 d',orbigny4 ,very many of the bones had been broken1 abraded1 and rolled1 before being embedded4 ,others1 even some of those included in the coarsest parts of the the now hard conglomerate1 still retain all their minutest prominences perfectly preserved2 so that ,i conclude that they probably were protected by skin1 flesh1 or ligaments1 whilst being covered up4 ,in the case of the ,scelidotherium1 it is quite certain that the whole skeleton was held together by its ligaments1 when deposited in the gravel in which ,i found it4 ,some cervical vertebr and a humerus of corresponding size lay so close together1 as did some ribs and the bones of a leg1 that ,i thought that they must originally have belonged to two skeletons1 and not have been washed in single2 but as remains were here very numerous1 ,i will not lay much stress on these two cases4 ,we have just seen that the armour of the ,dasypoid quadruped was certainly embedded together with some of the bones of the feet4 ,professor ,ehrenberg#i has examined for me specimens of the finer matter from in contact with these mammiferous remains3 he finds in them two ,polygastrica1 decidedly marine forms2 and six ,phytolitharia1 of which one is probably marine1 and the others either of fresh-water or terrestrial origin4 ,only one of these eight microscopical bodies is common to the nine from ,monte ,hermoso3 but five of them are in common with those from the ,pampean mud on the banks of the ,parana4 ,the presence of any fresh-water infusoria1 considering the aridity of the surrounding country1 is here remarkable3 the most probable explanation appears to be1 that these microscopical organisms were washed out of the adjoining great ,pampean formation during its denudation1 and afterwards redeposited4 #i 8,monatsberichten der ,akad4 zu ,berlin10 ,april #ahde4 ,the list consists of3-- ,,polygastrica4 ,gallionella sulcata4 ,stauroptera aspera8 fragm4 ,,phytolitharia4 ,lithasteriscus tuberculatus4 ,lithostylidium ,clepsammidium4 ,lithostylidium quadratum4 ,lithostylidium rude4 ,lithostylidium unidentatum4 ,spongolithis acicularis4 ,we will now see what conclusions may be drawn from the facts above detailed4 ,it is certain that the gravel-beds and intermediate red mud were deposited within the period1 when existing species of ,mollusca held to each other nearly the same relative proportions as they do on the present coast4 ,these beds1 from the number of littoral species1 must have been accumulated in shallow water2 but not1 judging from the stratification of the gravel and the layers of marl1 on a beach4 ,from the manner in which the red clay fills up furrows in the underlying gravel1 and is in some parts itself furrowed by the overlying gravel1 whilst in other parts it either insensibly passes into1 or alternates with1 this upper gravel1 we may infer several local changes in the currents1 perhaps caused by slight changes1 up or down1 in the level of the land4 ,by the elevation of these beds1 to which period the alluvial mantle with pumice-pebbles1 land and sea-shells belongs1 the plain of ,punta ,alta1 from twenty to thirty feet in height1 was formed4 ,in this neighbourhood there are other and higher sea-formed plains and lines of cliffs in the ,pampean formation worn by the denuding action of the waves at different levels4 ,hence we can easily understand the presence of rounded masses of tosca-rock in this lowest plain2 and likewise1 as the cliffs at ,monte ,hermoso with their mammiferous remains stand at a higher level1 the presence of the one much-rolled fragment of bone which was as black as jet3 possibly some few of the other much-rolled bones may have been similarly derived1 though ,i saw only the one fragment1 in the same condition with those from ,monte ,hermoso4 ,m4 d',orbigny has suggested#aj that all these mammiferous remains may have been washed out of the ,pampean formation1 and afterwards redeposited together with the recent shells4 ,undoubtedly it is a marvellous fact that these numerous gigantic quadrupeds1 belonging1 with the exception of the ,equus curvidens1 to seven extinct genera1 and one1 namely1 the ,toxodon1 not falling into any existing family1 should have co-existed with ,mollusca1 all of which are still living species2 but analogous facts have been observed in ,north ,america and in ,europe4 ,in the first place1 it should not be overlooked1 that most of the co-embedded shells have a more ancient and altered appearance than the bones4 ,in the second place1 is it probable that numerous bones not hardened by silex or any other mineral1 could have retained their delicate prominences and surfaces perfect if they had been washed out of one deposit1 and re-embedded in another3--this later deposit being formed of large1 hard pebbles1 arranged by the action of currents or breakers in shallow water into variously curved and inclined layers8 ,the bones which are now in so perfect a state of preservation1 must1 ,i conceive1 have been fresh and sound when embedded1 and probably were protected by skin1 flesh1 or ligaments4 ,the skeleton of the ,scelidotherium indisputably was deposited entire3 shall we say that when held together by its matrix it was washed out of an old gravel-bed 7totally unlike in character to the ,pampean formation71 and re-embedded in another gravel-bed1 composed 7,i speak after careful comparison7 of exactly the same kind of pebbles1 in the same kind of cement8 ,i will lay no stress on the two cases of several ribs and bones of the extremities having apparently been embedded in their proper relative position3 but will any one be so bold as to affirm that it is possible1 that a piece of the thin tessellated armour of a ,dasypoid quadruped1 at least three feet long and two in width1 and now so tender that ,i was unable with the utmost care to extract a fragment more than two or three inches square1 could have been washed out of one bed1 and re-embedded in another1 together with some of the small bones of the feet1 without having been dashed into atoms8 ,we must then wholly reject ,m4 d',orbigny's supposition1 and admit as certain1 that the ,scelidotherium and the large ,dasypoid quadruped1 and as highly probable1 that the ,toxodon1 ,megatherium1 etc41 some of the bones of which are perfectly preserved1 were embedded for the first time1 and in a fresh condition1 in the strata in which they were found entombed4 ,these gigantic quadrupeds1 therefore1 though belonging to extinct genera and families1 coexisted with the twenty above-enumerated ,mollusca1 the barnacle and two corals1 still living on this coast4 ,from the rolled fragment of black bone1 and from the plain of ,punta ,alta being lower than that of ,monte ,hermoso1 ,i conclude that the coarse sub-littoral deposits of ,punta ,alta1 are of subsequent origin to the ,pampean mud of ,monte ,hermoso2 and the beds at this latter place1 as we have seen1 are probably of subsequent origin to the high tosca-plain round the ,sierra ,ventana3 we shall1 however1 return1 at the end of this chapter1 to the consideration of these several stages in the great ,pampean formation4 #aj 8,voyage10 ,part4 ,golog41 p4 #di4 ,buenos ,ayres to ,st4 ,f ,bajada1 in ,entre ,rios4--,for some distance northward of ,buenos ,ayres1 the escarpment of the ,pampean formation does not approach very near to the ,plata1 and it is concealed by vegetation3 but in sections on the banks of the ,rios ,luxan1 ,areco1 and ,arrecifes1 ,i observed both pale and dark reddish ,pampean mud1 with small1 whitish concretions of tosca2 at all these places mammiferous remains have been found4 ,in the cliffs on the ,parana1 at ,san ,nicolas1 the ,pampean mud contains but little tosca2 here ,m4 d',orbigny found the remains of two rodents 7,ctenomys ,bonariensis and ,kerodon antiquus7 and the jaw of a ,canis3 when on the river ,i could clearly distinguish in this fine line of cliffs1 8horizontal lines of variation both in tint and compactness40#aa ,the plain northward of this point is very level1 but with some depressions and lakes2 ,i estimated its height at from forty to sixty feet above the ,parana4 ,at the ,a4 ,medio the bright red ,pampean mud contains scarcely any tosca-rock2 whilst at a short distance the stream of the ,pabon1 forms a cascade1 about twenty feet in height1 over a cavernous mass of two varieties of tosca-rock2 of which one is very compact and semi-crystalline1 with seams of crystallised carbonate of lime3 similar compact varieties are met with on the ,salidillo and ,seco4 ,the absolute identity 7,i speak after a comparison of my specimens7 between some of these varieties1 and those from ,tapalguen1 and from the ridge south of ,bahia ,blanca1 a distance of #djj miles of latitude1 is very striking4 #aa ,i quote these words from my note-book1 as written down on the spot1 on account of the general absence of stratification in the ,pampean formation having been insisted on by ,m4 d',orbigny as a proof of the diluvial origin of this great deposit4 ,at ,rosario there is but little tosca-rock3 near this place ,i first noticed at the edge of the river traces of an underlying formation1 which1 twenty-five miles higher up in the estancia of ,gorodona1 consists of a pale yellowish clay1 abounding with concretionary cylinders of a ferruginous sandstone4 ,this bed1 which is probably the equivalent of the older tertiary marine strata1 immediately to be described in ,entre ,rios1 only just rises above the level of the ,parana when low4 ,the rest of the cliff at ,gorodona1 is formed of red ,pampean mud1 with1 in the lower part1 many concretions of tosca1 some stalacti-formed1 and with only a few in the upper part3 at the height of six feet above the river1 two gigantic skeletons of the ,mastodon ,andium were here embedded2 their bones were scattered a few feet apart1 but many of them still held their proper relative positions3 they were much decayed and as soft as cheese1 so that even one of the great molar teeth fell into pieces in my hand4 ,we here see that the ,pampean deposit contains mammiferous remains close to its base4 ,on the banks of the ,carcarana1 a few miles distant1 the lowest bed visible was pale ,pampean mud1 with masses of tosca-rock1 in one of which ,i found a much decayed tooth of the ,mastodon3 above this bed1 there was a thin layer almost composed of small concretions of white tosca1 out of which ,i extracted a well preserved1 but slightly broken tooth of ,toxodon ,platensis3 above this there was an unusual bed of very soft impure sandstone4 ,in this neighbourhood ,i noticed many single embedded bones1 and ,i heard of others having been found in so perfect a state that they were long used as gate-posts3 the ,jesuit ,falkner found here the dermal armour of some gigantic ,edental quadruped4 ,in some of the red mud scraped from a tooth of one of the ,mastodons at ,gorodona1 ,professor ,ehrenberg finds seven ,polygastrica and thirteen ,phytolitharia1#ab all of them1 ,i believe1 with two exceptions1 already known species4 ,of these twenty1 the preponderating number are of fresh-water origin2 only two species of ,coscinodiscus and a ,spongolithis show the direct influence of the sea2 therefore ,professor ,ehrenberg arrives at the important conclusion that the deposit must have been of brackish-water origin4 ,of the thirteen ,phytolitharia1 nine are met with in the two deposits in ,bahia ,blanca1 where there is evidence from two other species of ,polygastrica that the beds were accumulated in brackish water4 ,the traces of coral1 sponges1 and ,polythalamia1 found by ,dr4 ,carpenter in the tosca-rock 7of which ,i must observe the greater number of specimens were from the upper beds in the southern parts of the formation71 apparently show a more purely marine origin4 #ab 8,monatsberichten der knig4 ,akad4 zu ,berlin10 ,april #ahde4 ,the list consists of3-- ,,polygastrica4 ,campylodiscus clypeus4 ,coscinodiscus subtilis4 ,coscinodiscus al4 sp4 ,eunotia4 ,gallionella granulata4 ,himantidium gracile4 ,pinnularia borealis4 ,at ,st4 ,f ,bajada1 in ,entre ,rios1 the cliffs1 estimated at between sixty and seventy feet in height1 expose an interesting section3 the lower half consists of tertiary strata with marine shells1 and the upper half of the ,pampean formation4 ,the lowest bed is an obliquely laminated1 blackish1 indurated mud1 with distinct traces of vegetable remains4#ac ,above this there is a thick bed of yellowish sandy clay1 with much crystallised gypsum and many shells of ,ostre1 ,pectens1 and ,arc3 above this there generally comes an arenaceous crystalline limestone1 but there is sometimes interposed a bed1 about twelve feet thick1 of dark green1 soapy clay1 weathering into small angular fragments4 ,the limestone1 where purest1 is white1 highly crystalline1 and full of cavities3 it includes small pebbles of quartz1 broken shells1 teeth of sharks1 and sometimes1 as ,i was informed1 large bones3 it often contains so much sand as to pass into a calcareous sandstone1 and in such parts the great ,ostrea ,patagonica#ad chiefly abounds4 ,in the upper part1 the limestone alternates with layers of fine white sand4 ,the shells included in these beds have been named for me by ,m4 d',orbigny3 they consist of3-- ,ostrea ,patagonica1 d',orbigny1 8,voyage10 ,part4 ,pal4 ,ostrea ,alvarezii1 d',orbigny1 8,voyage10 ,part4 ,pal4 ,pecten ,paranensis1 d',orbigny1 8,voyage10 ,part4 ,pal4 ,pecten ,darwinianus1 d',orbigny1 8,voyage10 ,part4 ,pal4 ,venus ,munsterii1 d',orbigny1 8,voyage10 ,pal4 ,arca ,bonplandiana1 d',orbigny1 8,voyage10 ,pal4 ,cardium ,platense1 d',orbigny1 8,voyage10 ,pal4 ,tellina1 probably nov4 spec41 but too imperfect for description4 ,,phytolitharia4 ,lithasteriscus tuberculatus4 ,lithodontium bursa4 ,lithodontium furcatum4 ,lithodontium rostratum4 ,lithostylidium ,amphiodon4 ,lithostylidium ,clepsammidium4 ,lithostylidium ,hamus4 ,lithostylidium polyedrum4 ,lithostylidium quadratum4 ,lithostylidium rude4 ,lithostylidium ,serra4 ,lithostylidium unidentatum4 ,spongolithis ,fustis4 #ac ,m4 d',orbigny has given 78,voyage10 ,part4 ,golog41 p4 #cg7 a detailed description of this section1 but as he does not mention this lowest bed1 it may have been concealed when he was there by the river4 ,there is a considerable discrepancy between his description and mine1 which ,i can only account for by the beds themselves varying considerably in short distances4 #ad ,captain ,sulivan1 ,r4,n41 has given me a specimen of this shell1 which he found in the cliffs at ,point ,cerrito1 between twenty and thirty miles above the ,bajada4 ,these species are all extinct3 the six first were found by ,m4 d',orbigny and myself in the formations of the ,rio ,negro1 ,s4 ,josef1 and other parts of ,patagonia2 and therefore1 as first observed by ,m4 d',orbigny1 these beds certainly belong to the great ,patagonian formation1 which will be described in the ensuing chapter1 and which we shall see must be considered as a very ancient tertiary one4 ,north of the ,bajada1 ,m4 d',orbigny found1 in beds which he considers as lying beneath the strata here described1 remains of a ,toxodon1 which he has named as a distinct species from the ,t4 ,platensis of the ,pampean formation4 ,much silicified wood is found on the banks of the ,parana 7and likewise on the ,uruguay71 and ,i was informed that they come out of these lower beds2 four specimens collected by myself are dicotyledonous4 ,the upper half of the cliff1 to a thickness of about thirty feet1 consists of ,pampean mud1 of which the lower part is pale-coloured1 and the upper part of a brighter red1 with some irregular layers of an arenaceous variety of tosca1 and a few small concretions of the ordinary kind4 ,close above the marine limestone1 there is a thin stratum with a concretionary outline of white hard tosca-rock or marl1 which may be considered either as the uppermost bed of the inferior deposits1 or the lowest of the ,pampean formation2 at one time ,i considered this bed as marking a passage between the two formations3 but ,i have since become convinced that ,i was deceived on this point4 ,in the section on the ,parana1 ,i did not find any mammiferous remains2 but at two miles distance on the ,a4 ,tapas 7a tributary of the ,conchitas71 they were extremely numerous in a low cliff of red ,pampean mud with small concretions1 precisely like the upper bed on the ,parana4 ,most of the bones were solitary and much decayed2 but ,i saw the dermal armour of a gigantic ,edental quadruped1 forming a caldron-like hollow1 four or five feet in diameter1 out of which1 as ,i was informed1 the almost entire skeleton had been lately removed4 ,i found single teeth of the ,mastodon ,andium1 ,toxodon ,platensis1 and ,equus curvidens1 near to each other4 ,as this latter tooth approaches closely to that of the common horse1 ,i paid particular attention to its true embedment1 for ,i did not at that time know that there was a similar tooth hidden in the matrix with the other mammiferous remains from ,punta ,alta4 ,it is an interesting circumstance1 that ,professor ,owen finds that the teeth of this horse approach more closely in their peculiar curvature to a fossil specimen brought by ,mr4 ,lyell#ae from ,north ,america1 than to those of any other species of ,equus4 #ae ,lyell's 8,travels in ,north ,america10 vol4 i1 p4 #afd and 8,proc4 of ,geolog4 ,soc410 vol4 iv1 p4 #ci4 ,the underlying marine tertiary strata extend over a wide area3 ,i was assured that they can be traced in ravines in an east and west line across ,entre ,rios to the ,uruguay1 a distance of about #ace miles4 ,in a ,s4,e4 direction ,i heard of their existence at the head of the ,r4 ,nankay2 and at ,p4 ,gorda in ,banda ,oriental1 a distance of #agj miles1 ,i found the same limestone1 containing the same fossil shells1 lying at about the same level above the river as at ,st4 ,fe4 ,in a southerly direction1 these beds sink in height1 for at another ,p4 ,gorda in ,entre ,rios1 the limestone is seen at a much less height2 and there can be little doubt that the yellowish sandy clay1 on a level with the river1 between the ,carcarana and ,s4 ,nicholas1 belongs to this same formation2 as perhaps do the beds of sand at ,buenos ,ayres1 which lie at the bottom of the ,pampean formation1 about sixty feet beneath the surface of the ,plata4 ,the southerly declination of these beds may perhaps be due1 not to unequal elevation1 but to the original form of the bottom of the sea1 sloping from land situated to the north2 for that land existed at no great distance1 we have evidence in the vegetable remains in the lowest bed at ,st4 ,fe2 and in the silicified wood and in the bones of ,toxodon ,paranensis1 found 7according to ,m4 d',orbigny7 in still lower strata4 ,banda ,oriental4--,this province lies on the northern side of the ,plata1 and eastward of the ,uruguay3 it has a gentle undulatory surface1 with a basis of primary rocks2 and is in most parts covered up with an unstratified mass1 of no great thickness1 of reddish ,pampean mud4 ,in the eastern half1 near ,maldonado1 this deposit is more arenaceous than in the ,pampas1 it contains many though small concretions of marl or tosca-rock1 and others of highly ferruginous sandstone2 in one section1 only a few yards in depth1 it rested on stratified sand4 ,near ,monte ,video this deposit in some spots appears to be of greater thickness2 and the remains of the ,glyptodon and other extinct mammifers have been found in it4 ,in the long line of cliffs1 between fifty and sixty feet in height1 called the ,barrancas de ,s4 ,gregorio1 which extend westward of the ,rio ,s4 ,lucia1 the lower half is formed of coarse sand of quartz and feldspar without mica1 like that now cast up on the beach near ,maldonado2 and the upper half of ,pampean mud1 varying in colour and containing honeycombed veins of soft calcareous matter and small concretions of tosca-rock arranged in lines1 and likewise a few pebbles of quartz4 ,this deposit fills up hollows and furrows in the underlying sand2 appearing as if water charged with mud had invaded a sandy beach4 ,these cliffs extend far westward1 and at a distance of sixty miles1 near ,colonia del ,sacramiento1 ,i found the ,pampean deposit resting in some places on this sand1 and in others on the primary rocks3 between the sand and the reddish mud1 there appeared to be interposed1 but the section was not a very good one1 a thin bed of shells of an existing ,mytilus1 still partially retaining their colour4 ,the ,pampean formation in ,banda ,oriental might readily be mistaken for an alluvial deposit3 compared with that of the ,pampas1 it is often more sandy1 and contains small fragments of quartz2 the concretions are much smaller1 and there are no extensive masses of tosca-rock4 ,in the extreme western parts of this province1 between the ,uruguay and a line drawn from ,colonia to the ,r4 ,perdido 7a tributary of the ,r4 ,negro71 the formations are far more complicated4 ,besides primary rocks1 we meet with extensive tracts and many flat-topped1 horizontally stratified1 cliff-bounded1 isolated hills of tertiary strata1 varying extraordinarily in mineralogical nature1 some identical with the old marine beds of ,st4 ,f ,bajada1 and some with those of the much more recent ,pampean formation4 ,there are1 also1 extensive low tracts of country covered with a deposit containing mammiferous remains1 precisely like that just described in the more eastern parts of the province4 ,although from the smooth and unbroken state of the country1 ,i never obtained a section of this latter deposit close to the foot of the higher tertiary hills1 yet ,i have not the least doubt that it is of quite subsequent origin2 having been deposited after the sea had worn the tertiary strata into the cliff-bounded hills4 ,this later formation1 which is certainly the equivalent of that of the ,pampas1 is well seen in the valleys in the estancia of ,berquelo1 near ,mercedes2 it here consists of reddish earth1 full of rounded grains of quartz1 and with some small concretions of tosca-rock arranged in horizontal lines1 so as perfectly to resemble1 except in containing a little calcareous matter1 the formation in the eastern parts of ,banda ,oriental1 in ,entre ,rios1 and at other places3 in this estancia the skeleton of a great ,edental quadruped was found4 ,in the valley of the ,sarandis1 at the distance of only a few miles1 this deposit has a somewhat different character1 being whiter1 softer1 finer-grained1 and full of little cavities1 and consequently of little specific gravity2 nor does it contain any concretions or calcareous matter3 ,i here procured a head1 which when first discovered must have been quite perfect1 of the ,toxodon ,platensis1 another of a ,mylodon1#af perhaps ,m4 ,darwinii1 and a large piece of dermal armour1 differing from that of the ,glyptodon clavipes4 ,these bones are remarkable from their extraordinarily fresh appearance2 when held over a lamp of spirits of wine1 they give out a strong odour and burn with a small flame2 ,mr4 ,t4 ,reeks has been so kind as to analyse some of the fragments1 and he finds that they contain about #g per cent of animal matter1 and #h per cent of water4#ag #af ,this head was at first considered by ,professor ,owen 7in the 8,zoology of the ,beagle's ,voyage07 as belonging to a distinct genus1 namely1 ,glossotherium4 #ag ,liebig 78,chemistry of ,agriculture10 p4 #aid7 states that fresh dry bones contain from #cb to #cc per cent of dry gelatine4 ,see also ,dr4 ,daubeny1 in ,edin4 ,new ,phil4 ,journ41 vol4 xxxvii1 p4 #bic4 ,the older tertiary strata1 forming the higher isolated hills and extensive tracts of country1 vary1 as ,i have said1 extraordinarily in composition3 within the distance of a few miles1 ,i sometimes passed over crystalline limestone with agate1 calcareous tuffs1 and marly rocks1 all passing into each other1--red and pale mud with concretions of tosca-rock1 quite like the ,pampean formation1--calcareous conglomerates and sandstones1--bright red sandstones passing either into red conglomerate1 or into white sandstone1--hard siliceous sandstones1 jaspery and chalcedonic rocks1 and numerous other subordinate varieties4 ,i was unable to mark out the relations of all these strata1 and will describe only a few distinct sections3--in the cliffs between ,p4 ,gorda on the ,uruguay and the ,a4 de ,vivoras1 the upper bed is crystalline cellular limestone often passing into calcareous sandstone1 with impressions of some of the same shells as at ,st4 ,f ,bajada2 at ,p4 ,gorda1#ah this limestone is interstratified with and rests on1 white sand1 which covers a bed about thirty feet thick of pale-coloured clay1 with many shells of the great ,ostrea ,patagonica3 beneath this1 in the vertical cliff1 nearly on a level with the river1 there is a bed of red mud absolutely like the ,pampean deposit1 with numerous often large concretions of perfectly characterised white1 compact tosca-rock4 ,at the mouth of the ,vivoras1 the river flows over a pale cavernous tosca-rock1 quite like that in the ,pampas1 and this appeared to underlie the crystalline limestone2 but the section was not unequivocal like that at ,p4 ,gorda4 ,these beds now form only a narrow and much denuded strip of land2 but they must once have extended much further2 for on the next stream1 south of the ,s4 ,juan1 ,captain ,sulivan1 ,r4,n41 found a little cliff1 only just above the surface of the river1 with numerous shells of the ,venus ,munsterii1 ,d',orbigny1--one of the species occurring at ,st4 ,f1 and of which there are casts at ,p4 ,gorda3 the line of cliffs of the subsequently deposited true ,pampean mud1 extend from ,colonia to within half a mile of this spot1 and no doubt once covered up this denuded marine stratum4 ,again at ,colonia1 a ,frenchman found1 in digging the foundations of a house1 a great mass of the ,ostrea ,patagonica 7of which ,i saw many fragments71 packed together just beneath the surface1 and directly superimposed on the gneiss4 ,these sections are important3 ,m4 d',orbigny is unwilling to believe that beds of the same nature with the ,pampean formation ever underlie the ancient marine tertiary strata2 and ,i was as much surprised at it as he could have been2 but the vertical cliff at ,p4 ,gorda allowed of no mistake1 and ,i must be permitted to affirm1 that after having examined the country from the ,colorado to ,st4 ,f ,bajada1 ,i could not be deceived in the mineralogical character of the ,pampean deposit4 #ah ,in my 8,journal0 7p4 #aga1 #ast edit471 ,i have hastily and inaccurately stated that the ,pampean mud1 which is found over the eastern part of ,b4 ,oriental1 lies over the limestone at ,p4 ,gorda2 ,i should have said that there was reason to infer that it was a subsequent or superior deposit4 ,moreover1 in a precipitous part of the ravine of ,las ,bocas1 a red sandstone is distinctly seen to overlie a thick bed of pale mud1 also quite like the ,pampean formation1 abounding with concretions of true tosca-rock4 ,this sandstone extends over many miles of country3 it is as red as the brightest volcanic scori2 it sometimes passes into a coarse red conglomerate composed of the underlying primary rocks2 and often passes into a soft white sandstone with red streaks4 ,at the ,calera de los ,huerfanos1 only a quarter of a mile south of where ,i first met with the red sandstone1 the crystalline white limestone is quarried3 as this bed is the uppermost1 and as it often passes into calcareous sandstone1 interstratified with pure sand2 and as the red sandstone likewise passes into soft white sandstone1 and is also the uppermost bed1 ,i believe that these two beds1 though so different1 are equivalents4 ,a few leagues southward of these two places1 on each side of the low primary range of ,s4 ,juan1 there are some flat-topped1 cliff-bounded1 separate little hills1 very similar to those fringing the primary ranges in the great plain south of ,buenos ,ayres3 they are composed--#ast1 of calcareous tuff with many particles of quartz1 sometimes passing into a coarse conglomerate2 #bnd1 of a stone undistinguishable on the closest inspection from the compacter varieties of tosca-rock2 and #crd1 of semi-crystalline limestone1 including nodules of agate3 these three varieties pass insensibly into each other1 and as they form the uppermost stratum in this district1 ,i believe that they1 also1 are the equivalents of the pure crystalline limestone1 and of the red and white sandstones and conglomerates4 ,between these points and ,mercedes on the ,rio ,negro1 there are scarcely any good sections1 the road passing over limestone1 tosca-rock1 calcareous and bright red sandstones1 and near the source of the ,san ,salvador over a wide extent of jaspery rocks1 with much milky agate1 like that in the limestone near ,san ,juan4 ,in the estancia of ,berquelo1 the separate1 flat-topped1 cliff-bounded hills are rather higher than in the other parts of the country2 they range in a ,n4,e4 and ,s4,w4 direction2 their uppermost beds consist of the same bright red sandstone1 passing sometimes into a conglomerate1 and in the lower part into soft white sandstone1 and even into loose sand3 beneath this sandstone1 ,i saw in two places layers of calcareous and marly rocks1 and in one place red ,pampean-like earth2 at the base of these sections1 there was a hard1 stratified1 white sandstone1 with chalcedonic layers4 ,near ,mercedes1 beds of the same nature and apparently of the same age1 are associated with compact1 white1 crystalline limestone1 including much botryoidal agate1 and singular masses1 like porcelain1 but really composed of a calcareo-siliceous paste4 ,in sinking wells in this district the chalcedonic strata seem to be the lowest4 ,beds1 such as there described1 occur over the whole of this neighbourhood2 but twenty miles further up the ,r4 ,negro1 in the cliffs of ,perika1 which are about fifty feet in height1 the upper bed is a prettily variegated chalcedony1 mingled with a pure white tallowy limestone2 beneath this there is a conglomerate of quartz and granite2 beneath this many sandstones1 some highly calcareous2 and the whole lower two-thirds of the cliff consists of earthy calcareous beds of various degrees of purity1 with one layer of reddish ,pampean-like mud4 ,when examining the agates1 the chalcedonic and jaspery rocks1 some of the limestones1 and even the bright red sandstones1 ,i was forcibly struck with their resemblance to deposits formed in the neighbourhood of volcanic action4 ,i now find that ,m4 ,isabelle1 in his 8,voyage a ,buenos ,ayres10 has described closely similar beds on ,itaquy and ,ibicuy 7which enter the ,uruguay some way north of the ,r4 ,negro7 and these beds include fragments of red decomposed true scori hardened by zeolite1 and of black retinite3 we have then here good evidence of volcanic action during our tertiary period4 ,still further north1 near ,s4 ,anna1#ai where the ,parana makes a remarkable bend1 ,m4 ,bonpland found some singular amygdaloidal rocks1 which perhaps may belong to this same epoch4 ,i may remark that1 judging from the size and well-rounded condition of the blocks of rock in the above-described conglomerates1 masses of primary formation probably existed at this tertiary period above water3 there is1 also1 according to ,m4 ,isabelle1 much conglomerate further north1 at ,salto4 #ai ,m4 d',orbigny1 8,voyage10 ,part4 ,golog41 p4 #bi4 ,from whatever source and through whatever means the great ,pampean formation originated1 we here have1 ,i must repeat1 unequivocal evidence of a similar action at a period before that of the deposition of the marine tertiary strata with extinct shells1 at ,santa ,f and ,p4 ,gorda4 ,during also the deposition of these strata1 we have in the intercalated layers of red ,pampean-like mud and tosca-rock1 and in the passage near ,s4 ,juan of the semi-crystalline limestones with agate into tosca undistinguishable from that of the ,pampas1 evidence of the same action1 though continued only at intervals and in a feeble manner4 ,we have further seen that in this district1 at a period not only subsequent to the deposition of the tertiary strata1 but to their upheavement and most extensive denudation1 true ,pampean mud with its usual characters and including mammiferous remains1 was deposited round and between the hills or islets formed of these tertiary strata1 and over the whole eastern and low primary districts of ,banda ,oriental4 ,no4 #af ,section of the lowest plain at ,port ,s4 ,julian4 ,illustration3 ,section of the lowest plain at ,port ,s4 ,julian4 ,earthy mass1 with extinct mammiferous remains1 over the porphyritic gravel at ,s4 ,julian1 lat4 #di #ad ,s41 in ,patagonia4--,this case1 though not coming strictly under the ,pampean formation1 may be conveniently given here4 ,on the south side of the harbour1 there is a nearly level plain 7mentioned in the ,first ,chapter7 about seven miles long1 and three or four miles wide1 estimated at ninety feet in height1 and bordered by perpendicular cliffs1 of which a section is represented above4 ,the lower old tertiary strata 7to be described in the next chapter7 are covered by the usual gravel bed2 and this by an irregular earthy1 sometimes sandy mass1 seldom more than two or three feet in thickness1 except where it fills up furrows or gullies worn not only through the underlying gravel1 but even through the upper tertiary beds4 ,this earthy mass is of a pale reddish colour1 like the less pure varieties of ,pampean mud in ,banda ,oriental2 it includes small calcareous concretions1 like those of tosca-rock but more arenaceous1 and other concretions of a greenish1 indurated argillaceous substance3 a few pebbles1 also1 from the underlying gravel-bed are also included in it1 and these being occasionally arranged in horizontal lines1 show that the mass is of sub-aqueous origin4 ,on the surface and embedded in the superficial parts1 there are numerous shells1 partially retaining their colours1 of three or four of the now commonest littoral species4 ,near the bottom of one deep furrow 7represented in figure ,no4 #af71 filled up with this earthy deposit1 ,i found a large part of the skeleton of the ,macrauchenia ,patachonica--a gigantic and most extraordinary pachyderm1 allied1 according to ,professor ,owen1 to the ,palotherium1 but with affinities to the ,ruminants1 especially to the ,american division of the ,camelid4 ,several of the vertebr in a chain1 and nearly all the bones of one of the limbs1 even to the smallest bones of the foot1 were embedded in their proper relative positions3 hence the skeleton was certainly united by its flesh or ligaments1 when enveloped in the mud4 ,this earthy mass1 with its concretions and mammiferous remains1 filling up furrows in the underlying gravel1 certainly presents a very striking resemblance to some of the sections 7for instance1 at ,p4 ,alta in ,b4 ,blanca1 or at the ,barrancas de ,s4 ,gregorio7 in the ,pampean formation2 but ,i must believe that this resemblance is only accidental4 ,i suspect that the mud which at the present day is accumulating in deep and narrow gullies at the head of the harbour1 would1 after elevation1 present a very similar appearance4 ,the southernmost part of the true ,pampean formation1 namely1 on the ,colorado1 lies #efj miles of latitude north of this point4#bj #bj ,in the succeeding chapter ,i shall have to refer to a great deposit of extinct mammiferous remains1 lately discovered by ,captain ,sulivan1 ,r4,n41 at a point still further south1 namely1 at the ,r4 ,gallegos2 their age must at present remain doubtful4 ,with respect to the age of the ,macrauchenia1 the shells on the surface prove that the mass in which the skeleton was enveloped has been elevated above the sea within the recent period3 ,i did not see any of the shells embedded at a sufficient depth to assure me 7though it be highly probable7 that the whole thickness of the mass was contemporaneous with these individual specimens4 ,that the ,macrauchenia lived subsequently to the spreading out of the gravel on this plain is certain2 and that this gravel1 at the height of ninety feet1 was spread out long after the existence of recent shells1 is scarcely less certain4 ,for1 it was shown in the ,first ,chapter1 that this line of coast has been upheaved with remarkable equability1 and that over a vast space both north and south of ,s4 ,julian1 recent species of shells are strewed on 7or embedded in7 the surface of the #bej feet plain1 and of the #cej feet plain up to a height of #djj feet4 ,these wide step-formed plains have been formed by the denuding action of the coast-waves on the old tertiary strata2 and therefore1 when the surface of the #cej feet plain1 with the shells on it1 first rose above the level of the sea1 the #bej feet plain did not exist1 and its formation1 as well as the spreading out of the gravel on its summit1 must have taken place subsequently4 ,so also the denudation and the gravel-covering of the #ij feet plain must have taken place subsequently to the elevation of the #bej feet plain1 on which recent shells are also strewed4 ,hence there cannot be any doubt that the ,macrauchenia1 which certainly was entombed in a fresh state1 and which must have been alive after the spreading out of the gravel on the #ij feet plain1 existed1 not only subsequently to the upraised shells on the surface of the #bej feet plain1 but also to those on the #cej to #djj feet plain3 these shells1 eight in number 7namely1 three species of ,mytilus1 two of ,patella1 one ,fusus1 ,voluta1 and ,balanus71 are undoubtedly recent species1 and are the commonest kinds now living on this coast4 ,at ,punta ,alta in ,b4 ,blanca1 ,i remarked how marvellous it was1 that the ,toxodon1 a mammifer so unlike to all known genera1 should have co-existed with twenty-three still living marine animals2 and now we find that the ,macrauchenia1 a quadruped only a little less anomalous than the ,toxodon1 also co-existed with eight other still existing ,mollusca3 it should1 moreover1 be borne in mind1 that a tooth of a pachydermatous animal was found with the other remains at ,punta ,alta1 which ,professor ,owen thinks almost certainly belonged to the ,macrauchenia4 ,mr4 ,lyell#ba has arrived at a highly important conclusion with respect to the age of the ,north ,american extinct mammifers 7many of which are closely allied to1 and even identical with1 those of the ,pampean formation71 namely1 that they lived subsequently to the period when erratic boulders were transported by the agency of floating ice in temperate latitudes4 ,now in the valley of the ,santa ,cruz1 only fifty miles of latitude south of the spot where the ,macrauchenia was entombed1 vast numbers of gigantic1 angular boulders1 which must have been transported from the ,cordillera on icebergs1 lie strewed on the plain1 at the height of #a1djj feet above the level of the sea4 ,in ascending to this level1 several step-formed plains must be crossed1 all of which have necessarily required long time for their formation2 hence the lowest or ninety feet plain1 with its superficial bed containing the remains of the ,macrauchenia1 must have been formed very long subsequently to the period when the #a1djj feet plain was beneath the sea1 and boulders were dropped on it from floating masses of ice4#bb ,mr4 ,lyell's conclusion1 therefore1 is thus far confirmed in the southern hemisphere2 and it is the more important1 as one is naturally tempted to admit so simple an explanation1 that it was the ice-period that caused the extinction of the numerous great mammifers which so lately swarmed over the two ,americas4 #ba 8,geological ,proceedings10 vol4 iv1 p4 #cf4 #bb ,it must not be inferred from these remarks1 that the ice-action ceased in ,south ,america at this comparatively ancient period2 for in ,tierra del ,fuego boulders were probably transported contemporaneously with1 if not subsequently to1 the formation of the ninety feet plain at ,s4 ,julian1 and at other parts of the coast of ,patagonia4 ,summary and concluding remarks on the ,pampean formation4--,one of its most striking features is its great extent2 ,i passed continuously over it from the ,colorado to ,st4 ,fe ,bajada1 a distance of #ejj geographical miles2 and ,m4 d',orbigny traced it for #bej miles further north4 ,in the latitude of the ,plata1 ,i examined this formation at intervals over an east and west line of #cjj miles from ,maldonado to the ,r4 ,carcarana2 and ,m4 d',orbigny believes it extends #ajj miles further inland3 from ,mr4 ,caldcleugh's travels1 however1 ,i should have thought that it had extended1 south of the ,cordovese range1 to near ,mendoza1 and ,i may add that ,i heard of great bones having been found high up the ,r4 ,quinto4 ,hence the area of the ,pampean formation1 as remarked by ,m4 d',orbigny1 is probably at least equal to that of ,france1 and perhaps twice or thrice as great4 ,in a basin1 surrounded by gravel-cliff 7at a height of nearly three thousand feet71 south of ,mendoza1 there is1 as described in the ,third ,chapter1 a deposit very like the ,pampean1 interstratified with other matter2 and again at ,s4 ,julian's1 in ,patagonia1 #efj miles south of the ,colorado1 a small irregular bed of a nearly similar nature contains1 as we have just seen1 mammiferous remains4 ,in the provinces of ,moxos and ,chiquitos 7#a1jjj miles northward of the ,pampas71 and in ,bolivia1 at a height of #d1jjj metres1 ,m4 d',orbigny has described similar deposits1 which he believes to have been formed by the same agency contemporaneously with the ,pampean formation4 ,considering the immense distances between these several points1 and their different heights1 it appears to me infinitely more probable1 that this similarity has resulted not from contemporaneousness of origin1 but from the similarity of the rocky framework of the continent3 it is known that in ,brazil an immense area consists of gneissic rocks1 and we shall hereafter see1 over how great a length the plutonic rocks of the ,cordillera1 the overlying purple porphyries1 and the trachytic ejections1 are almost identical in nature4 ,three theories on the origin of the ,pampean formation have been propounded3--,first1 that of a great debacle by ,m4 d',orbigny2 this seems founded chiefly on the absence of stratification1 and on the number of embedded remains of terrestrial quadrupeds4 ,although the ,pampean formation 7like so many argillaceous deposits7 is not divided into distinct and separate strata1 yet we have seen that in one good section it was striped with horizontal zones of colour1 and that in several specified places the upper and lower parts differed1 not only considerably in colour1 but greatly in constitution4 ,in the southern part of the ,pampas the upper mass 7to a certain extent stratified7 generally consists of hard tosca-rock1 and the lower part of red ,pampean mud1 often itself divided into two or more masses1 varying in colour and in the quantity of included calcareous matter4 ,in ,western ,banda ,oriental1 beds of a similar nature1 but of a greater age1 conformably underlie and are intercalated with the regularly stratified tertiary formation4 ,as a general rule1 the marly concretions are arranged in horizontal lines1 sometimes united into irregular strata3 surely1 if the mud had been tumultuously deposited in mass1 the included calcareous matter would have segregated itself irregularly1 and not into nodules arranged in horizontal lines1 one above the other and often far apart3 this arrangement appears to me to prove that mud1 differing slightly in composition1 was successively and quietly deposited4 ,on the theory of a debacle1 a prodigious amount of mud1 without a single pebble1 is supposed to have been borne over the wide surface of the ,pampas1 when under water3 on the other hand1 over the whole of ,patagonia1 the same or another debacle is supposed to have borne nothing but gravel1--the gravel and the fine mud in the neighbourhood of the ,rios ,negro and ,colorado having been borne to an equal distance from the ,cordillera1 or imagined line of disturbance3 assuredly directly opposite effects ought not to be attributed to the same agency4 ,where1 again1 could a mass of fine sediment1 charged with calcareous matter in a fit state for chemical segregation1 and in quantity sufficient to cover an area at least #gej miles long1 and #djj miles broad1 to a depth of from twenty to thirty feet to a hundred feet1 have been accumulated1 ready to be transported by the supposed debacle8 ,to my mind it is little short of demonstration1 that a great lapse of time was necessary for the production and deposition of the enormous amount of mudlike matter forming the ,pampas2 nor should ,i have noticed the theory of a debacle1 had it not been adduced by a naturalist so eminent as ,m4 d',orbigny4 ,a second theory1 first suggested1 ,i believe1 by ,sir ,w4 ,parish1 is that the ,pampean formation was thrown down on low and marshy plains by the rivers of this country before they assumed their present courses4 ,the appearance and composition of the deposit1 the manner in which it slopes up and round the primary ranges1 the nature of the underlying marine beds1 the estuary and sea-shells on the surface1 the overlying sandstone beds at ,m4 ,hermoso1 are all quite opposed to this view4 ,nor do ,i believe that there is a single instance of a skeleton of one of the extinct mammifers having been found in an upright position1 as if it had been mired4 ,the third theory1 of the truth of which ,i cannot entertain the smallest doubt1 is that the ,pampean formation was slowly accumulated at the mouth of the former estuary of the ,plata and in the sea adjoining it4 ,i have come to this conclusion from the reasons assigned against the two foregoing theories1 and from simple geographical considerations4 ,from the numerous shells of the ,azara labiata lying loose on the surface of the plains1 and near ,buenos ,ayres embedded in the tosca-rock1 we know that this formation not only was formerly covered by1 but that the uppermost parts were deposited in1 the brackish water of the ancient ,la ,plata4 ,southward and seaward of ,buenos ,ayres1 the plains were upheaved from under water inhabited by true marine shells4 ,we further know from ,professor ,ehrenberg's examination of the twenty microscopical organisms in the mud round the tooth of the ,mastodon high up the course of the ,parana1 that the bottom-most part of this formation was of brackish-water origin4 ,a similar conclusion must be extended to the beds of like composition1 at the level of the sea and under it1 at ,m4 ,hermoso in ,bahia ,blanca4 ,dr4 ,carpenter finds that the harder varieties of tosca-rock1 collected chiefly to the south1 contain marine spongoid bodies1 minute fragments of shells1 corals1 and ,polythalamia2 these perhaps may have been drifted inwards by the tides1 from the more open parts of the sea4 ,the absence of shells1 throughout this deposit1 with the exception of the uppermost layers near ,buenos ,ayres1 is a remarkable fact3 can it be explained by the brackish condition of the water1 or by the deep mud at the bottom8 ,i have stated that both the reddish mud and the concretions of tosca-rock are often penetrated by minute1 linear cavities1 such as frequently may be observed in fresh-water calcareous deposits3--were they produced by the burrowing of small worms8 ,only on this view of the ,pampean formation having been of estuary origin1 can the extraordinary numbers 7presently to be alluded to7 of the embedded mammiferous remains be explained4#bc #bc ,it is almost superfluous to give the numerous cases 7for instance1 in ,sumatra2 ,lyell's 8,principles10 vol4 iii1 p4 #cbe1 sixth edit471 of the carcasses of animals having been washed out to sea by swollen rivers2 but ,i may refer to a recent account by ,mr4 ,bettington 78,asiatic ,soc410 #ahde1 ,june #bast71 of oxen1 deer1 and bears being carried into the ,gulf of ,cambray2 see also the account in my 8,journal0 7#bnd edit41 p4 #acc71 of the numbers of animals drowned in the ,plata during the great1 often recurrent1 droughts4 ,with respect to the first origin of the reddish mud1 ,i will only remark1 that the enormous area of ,brazil consists in chief part of gneissic and other granitic rocks1 which have suffered decomposition1 and been converted into a red1 gritty1 argillaceous mass1 to a greater depth than in any other country which ,i have seen4 ,the mixture of rounded grains1 and even of small fragments and pebbles of quartz1 in the ,pampean mud of ,banda ,oriental1 is evidently due to the neighbouring and underlying primary rocks4 ,the estuary mud was drifted during the ,pampean period in a much more southerly course1 owing probably to the east and west primary ridges south of the ,plata not having been then elevated1 than the mud of the ,plata at present is2 for it was formerly deposited as far south as the ,colorado4 ,the quantity of calcareous matter in this formation1 especially in those large districts where the whole mass passes into tosca-rock1 is very great3 ,i have already remarked on the close resemblance in external and microscopical appearance1 between this tosca-rock and the strata at ,coquimbo1 which have certainly resulted from the decay and attrition of recent shells3#bd ,i dare not1 however1 extend this conclusion to the calcareous rocks of the ,pampas1 more especially as the underlying tertiary strata in western ,banda ,oriental show that at that period there was a copious emission of carbonate of lime1 in connection with volcanic action4 #bd ,i may add1 that there are nearly similar superficial calcareous beds at ,king ,george's ,sound in ,australia2 and these undoubtedly have been formed by the disintegration of marine remains 7see 8,volcanic ,islands10 etc41 p4 #add74 ,there is1 however1 something very remarkable in the frequency of superficial1 thin beds of earthy calcareous matter1 in districts where the surrounding rocks are not calcareous4 ,major ,charters1 in a ,paper read before the ,geographical ,society 7,april #acth1 #ahdj1 and abstracted in the ,athenum1 p4 #cag71 states that this is the case in parts of ,mexico1 and that he has observed similar appearances in many parts of ,south ,africa4 ,the circumstance of the uppermost stratum round the ragged ,sierra ,ventana1 consisting of calcareous or marly matter1 without any covering of alluvial matter1 strikes me as very singular1 in whatever manner we view the deposition and elevation of the ,pampean formation4 ,the ,pampean formation1 judging from its similar composition1 and from the apparent absolute specific identity of some of its mammiferous remains1 and from the generic resemblance of others1 belongs over its vast area--throughout ,banda ,oriental1 ,entre ,rios1 and the wide extent of the ,pampas as far south as the ,colorado1--to the same geological epoch4 ,the mammiferous remains occur at all depths from the top to the bottom of the deposit2 and ,i may add that nowhere in the ,pampas is there any appearance of much superficial denudation3 some bones which ,i found near the ,guardia del ,monte were embedded close to the surface2 and this appears to have been the case with many of those discovered in ,banda ,oriental3 on the ,matanzas1 twenty miles south of ,buenos ,ayres1 a ,glyptodon was embedded five feet beneath the surface2 numerous remains were found by ,s4 ,muniz1 near ,luxan1 at an average depth of eighteen feet2 in ,buenos ,ayres a skeleton was disinterred at sixty feet depth1 and on the ,parana ,i have described two skeletons of the ,mastodon only five or six feet above the very base of the deposit4 ,with respect to the age of this formation1 as judged of by the ordinary standard of the existence of ,mollusca1 the only evidence within the limits of the true ,pampas which is at all trustworthy1 is afforded by the still living ,azara labiata being embedded in tosca-rock near ,buenos ,ayres4 ,at ,punta ,alta1 however1 we have seen that several of the extinct mammifers1 most characteristic of the ,pampean formation1 co-existed with twenty species of ,mollusca1 a barnacle and two corals1 all still living on this same coast2--for when we remember that the shells have a more ancient appearance than the bones2 that many of the bones1 though embedded in a coarse conglomerate1 are perfectly preserved2 that almost all the parts of the skeleton of the ,scelidotherium1 even to the knee-cap1 were lying in their proper relative positions2 and that a large piece of the fragile dermal armour of a ,dasypoid quadruped1 connected with some of the bones of the foot1 had been entombed in a condition allowing the two sides to be doubled together1 it must assuredly be admitted that these mammiferous remains were embedded in a fresh state1 and therefore that the living animals co-existed with the co-embedded shells4 ,moreover1 the ,macrauchenia ,patachonica 7of which1 according to ,professor ,owen1 remains also occur in the ,pampas of ,buenos ,ayres1 and at ,punta ,alta7 has been shown by satisfactory evidence of another kind1 to have lived on the plains of ,patagonia long after the period when the adjoining sea was first tenanted by its present commonest molluscous animals4 ,we must1 therefore1 conclude that the ,pampean formation belongs1 in the ordinary geological sense of the word1 to the ,recent ,period4#be #be ,m4 d',orbigny believes 78,voyage10 ,part4 ,golog41 p4 #ha7 that this formation1 though 8trs voisine de la ntre1 est nanmoins de beaucoup antrieure notre cration40 ,at ,st4 ,f ,bajada1 the ,pampean estuary formation1 with its mammiferous remains1 conformably overlies the marine tertiary strata1 which 7as first shown by ,m4 d',orbigny7 are contemporaneous with those of ,patagonia1 and which1 as we shall hereafter see1 belong to a very ancient tertiary stage4 ,when examining the junction between these two formations1 ,i thought that the concretionary layer of marl marked a passage between the marine and estuary stages4 ,m4 d',orbigny disputes this view 7as given in my 8,journal071 and ,i admit that it is erroneous1 though in some degree excusable1 from their conformability and from both abounding with calcareous matter4 ,it would1 indeed1 have been a great anomaly if there had been a true passage between a deposit contemporaneous with existing species of mollusca1 and one in which all the mollusca appear to be extinct4 ,northward of ,santa ,fe1 ,m4 d',orbigny met with ferruginous sandstones1 marly rocks1 and other beds1 which he considers as a distinct and lower formation2 but the evidence that they are not parts of the same with an altered mineralogical character1 does not appear to me quite satisfactory4 ,in ,western ,banda ,oriental1 while the marine tertiary strata were accumulating1 there were volcanic eruptions1 much silex and lime were precipitated from solution1 coarse conglomerates were formed1 being derived probably from adjoining land1 and layers of red mud and marly rocks1 like those of the ,pampean formation1 were occasionally deposited4 ,the true ,pampean deposit1 with mammiferous remains1 instead of as at ,santa ,fe overlying conformably the tertiary strata1 is here seen at a lower level folding round and between the flat-topped1 cliff-bounded hills1 formed by a upheaval and denudation of these same tertiary strata4 ,the upheaval1 having occurred here earlier than at ,santa ,fe1 may be naturally accounted for by the contemporaneous volcanic action4 ,at the ,barrancas de ,s4 ,gregorio1 the ,pampean deposit1 as we have seen1 overlies and fills up furrows in coarse sand1 precisely like that now accumulating on the shores near the mouth of the ,plata4 ,i can hardly believe that this loose and coarse sand is contemporaneous with the old tertiary and often crystalline strata of the more western parts of the province2 and am induced to suspect that it is of subsequent origin4 ,if that section near ,colonia could be implicitly trusted1 in which1 at a height of only fifteen feet above the ,plata1 a bed of fresh-looking mussels1 of an existing littoral species1 appeared to lie between the sand and the ,pampean mud1 ,i should conclude that ,banda ,oriental must have stood1 when the coarse sand was accumulating1 at only a little below its present level1 and had then subsided1 allowing the estuary ,pampean mud to cover far and wide its surface up to a height of some hundred feet2 and that after this subsidence the province had been uplifted to its present level4 ,in ,western ,banda ,oriental1 we know1 from two unequivocal sections that there is a mass1 absolutely undistinguishable from the true ,pampean deposit1 beneath the old tertiary strata4 ,this inferior mass must be very much more ancient than the upper deposit with its mammiferous remains1 for it lies beneath the tertiary strata in which all the shells are extinct4 ,nevertheless1 the lower and upper masses1 as well as some intermediate layers1 are so similar in mineralogical character1 that ,i cannot doubt that they are all of estuary origin1 and have been derived from the same great source4 ,at first it appeared to me extremely improbable1 that mud of the same nature should have been deposited on nearly the same spot1 during an immense lapse of time1 namely1 from a period equivalent perhaps to the ,eocene of ,europe to that of the ,pampean formation4 ,but as1 at the very commencement of the ,pampean period1 if not at a still earlier period1 the ,sierra ,ventana formed a boundary to the south1--the ,cordillera or the plains in front of them to the west1--the whole province of ,corrientes probably to the north1 for1 according to ,m4 d',orbigny1 it is not covered by the ,pampean deposit1--and ,brazil1 as known by the remains in the caves1 to the north-east2 and as again1 during the older tertiary period1 land already existed in ,western ,banda ,oriental and near ,st4 ,f ,bajada1 as may be inferred from the vegetable debris1 from the quantities of silicified wood1 and from the remains of a ,toxodon found1 according to ,m4 d',orbigny1 in still lower strata1 we may conclude1 that at this ancient period a great expanse of water was surrounded by the same rocky framework which now bounds the plains of ,pampean formation4 ,this having been the case1 the circumstance of sediment of the same nature having been deposited in the same area during an immense lapse of time1 though highly remarkable1 does not appear incredible4 ,the elevation of the ,pampas1 at least of the southern parts1 has been slow and interrupted by several periods of rest1 as may be inferred from the plains1 cliffs1 and lines of sand-dunes 7with shells and pumice-pebbles7 standing at different heights4 ,i believe1 also1 that the ,pampean mud continued to be deposited1 after parts of this formation had already been elevated1 in the same manner as mud would continue to be deposited in the estuary of the ,plata1 if the mud-banks on its shores were now uplifted and changed into plains3 ,i believe in this from the improbability of so many skeletons and bones having been accumulated at one spot1 where ,m4 ,hermoso now stands1 at a depth of between eight hundred and one thousand feet1 and at a vast distance from any land except small rocky islets1--as must have been the case1 if the high tosca-plain round the ,ventana and adjoining ,sierras1 had not been already uplifted and converted into land1 supporting mammiferous animals4 ,at ,punta ,alta we have good evidence that the gravel-strata1 which certainly belong to the true ,pampean period1 were accumulated after the elevation in that neighbourhood of the main part of the ,pampean deposit1 whence the rounded masses of tosca-rock were derived1 and that rolled fragment of black bone in the same peculiar condition with the remains at ,monte ,hermoso4 ,the number of the mammiferous remains embedded in the ,pampas is1 as ,i have remarked1 wonderful3 it should be borne in mind that they have almost exclusively been found in the cliffs and steep banks of rivers1 and that1 until lately1 they excited no attention amongst the inhabitants3 ,i am firmly convinced that a deep trench could not be cut in any line across the ,pampas1 without intersecting the remains of some quadruped4 ,it is difficult to form an opinion in what part of the ,pampas they are most numerous2 in a limited spot they could not well have been more numerous than they were at ,p4 ,alta2 the number1 however1 lately found by ,senor ,f4 ,muniz1 near ,luxan1 in a central spot in the ,pampas1 is extraordinarily great3 at the end of this chapter ,i will give a list of all the localities at which ,i have heard of remains having been discovered4 ,very frequently the remains consist of almost perfect skeletons2 but there are1 also1 numerous single bones1 as for instance at ,st4 ,f4 ,their state of preservation varies much1 even when embedded near each other3 ,i saw none others so perfectly preserved as the heads of the ,toxodon and ,mylodon from the white soft earthy bed on the ,sarandis in ,banda ,oriental4 ,it is remarkable that in two limited sections ,i found no less than five teeth separately embedded1 and ,i heard of teeth having been similarly found in other parts3 may we suppose that the skeletons or heads were for a long time gently drifted by currents over the soft muddy bottom1 and that the teeth occasionally1 here and there1 dropped out8 ,it may be naturally asked1 where did these numerous animals live8 ,from the remarkable discoveries of ,,mm4 ,lund and ,clausen1 it appears that some of the species found in the ,pampas inhabited the highlands of ,brazil3 the ,mastodon ,andium is embedded at great heights in the ,cordillera from north of the equator#bf to at least as far south as ,tarija2 and as there is no higher land1 there can be little doubt that this ,mastodon must have lived on the plains and valleys of that great range4 ,these countries1 however1 appear too far distant for the habitation of the individuals entombed in the ,pampas3 we must probably look to nearer points1 for instance to the province of ,corrientes1 which1 as already remarked1 is said not to be covered by the ,pampean formation1 and may therefore1 at the period of its deposition1 have existed as dry land4 ,i have already given my reasons for believing that the animals embedded at ,m4 ,hermoso and at ,p4 ,alta in ,bahia ,blanca1 lived on adjoining land1 formed of parts of the already elevated ,pampean deposit4 ,with respect to the food of these many great extinct quadrupeds1 ,i will not repeat the facts given in my 8,journal0 7second edit41 p4 #he71 showing that there is no correlation between the luxuriance of the vegetation of a country and the size of its mammiferous inhabitants4 ,i do not doubt that large animals could now exist1 as far as the amount1 not kind1 of vegetation is concerned1 on the sterile plains of ,bahia ,blanca and of the ,r4 ,negro1 as well as on the equally1 if not more sterile plains of ,southern ,africa4 ,the climate1 however1 may perhaps have somewhat deteriorated since the mammifers embedded at ,bahia ,blanca lived there2 for we must not infer1 from the continued existence of the same shells on the present coasts1 that there has been no change in climate2 for several of these shells now range northward along the shores of ,brazil1 where the most luxuriant vegetation flourishes under a tropical temperature4 ,with respect to the extinction1 which at first fills the mind with astonishment1 of the many great and small mammifers of this period1 ,i may also refer to the work above cited 7second edit41 p4 #agc71 in which ,i have endeavoured to show1 that however unable we may be to explain the precise cause1 we ought not properly to feel more surprised at a species becoming extinct than at one being rare2 and yet we are accustomed to view the rarity of any particular species as an ordinary event1 not requiring any extraordinary agency4 #bf ,humboldt states that the ,mastodon has been discovered in ,new ,granada3 it has been found in ,quito4 ,when at ,lima1 ,i saw a tooth of a ,mastodon in the possession of ,don ,m4 ,rivero1 found at ,playa ,chica on the ,maranon1 near the ,guallaga4 ,every one has heard of the numerous remains of ,mastodon in ,bolivia4 ,the several mammifers embedded in the ,pampean formation1 which mostly belong to extinct genera1 and some even to extinct families or orders1 and which differ nearly1 if not quite1 as much as do the ,eocene mammifers of ,europe from living quadrupeds having existed contemporaneously with mollusca1 all still inhabiting the adjoining sea1 is certainly a most striking fact4 ,it is1 however1 far from being an isolated one2 for1 during the late tertiary deposits of ,britain1 an elephant1 rhinoceros1 and hippopotamus co-existed with many recent land and fresh-water shells2 and in ,north ,america1 we have the best evidence that a mastodon1 elephant1 megatherium1 megalonyx1 mylodon1 an extinct horse and ox1 likewise co-existed with numerous land1 fresh-water1 and marine recent shells4#bg ,the enumeration of these extinct ,north ,american animals naturally leads me to refer to the former closer relation of the mammiferous inhabitants of the two ,americas1 which ,i have discussed in my 8,journal10 and likewise to the vast extent of country over which some of them ranged3 thus the same species of the ,megatherium1 ,megalonyx1 ,equus 7as far as the state of their remains permits of identification71 extended from the ,southern ,united ,states of ,north ,america to ,bahia ,blanca1 in lat4 #ci ,s41 on the coast of ,patagonia4 ,the fact of these animals having inhabited tropical and temperate regions1 does not appear to me any great difficulty1 seeing that at the ,cape of ,good ,hope several quadrupeds1 such as the elephant and hippopotamus1 range from the equator to lat4 #ce south4 ,the case of the ,mastodon ,andium is one of more difficulty1 for it is found from lat4 #cf ,s41 over1 as ,i have reason to believe1 nearly the whole of ,brazil1 and up the ,cordillera to regions which1 according to ,m4 d',orbigny1 border on perpetual snow1 and which are almost destitute of vegetation3 undoubtedly the climate of the ,cordillera must have been different when the mastodon inhabited it2 but we should not forget the case of the ,siberian mammoth and rhinoceros1 as showing how severe a climate the larger pachydermata can endure2 nor overlook the fact of the guanaco ranging at the present day over the hot low deserts of ,peru1 the lofty pinnacles of the ,cordillera1 and the damp forest-clad land of ,southern ,tierra del ,fuego2 the puma1 also1 is found from the equator to the ,strait of ,magellan1 and ,i have seen its footsteps only a little below the limits of perpetual snow in the ,cordillera of ,chile4 #bg ,many original observations1 and a summary on this subject1 are given in ,mr4 ,lyell's paper in the 8,geolog4 ,proc410 vol4 iv1 p4 #c and in his 8,travels in ,north ,america10 vol4 i1 p4 #afd and vol4 ii1 p4 #fj4 ,for the ,european analogous cases see ,mr4 ,lyell's 8,principles of ,geology0 7#fth edit471 vol4 i1 p4 #cg4 ,at the period1 so recent in a geological sense1 when these extinct mammifers existed1 the two ,americas must have swarmed with quadrupeds1 many of them of gigantic size2 for1 besides those more particularly referred to in this chapter1 we must include in this same period those wonderfully numerous remains1 some few of them specifically1 and others generically related to those of the ,pampas1 discovered by ,,mm4 ,lund and ,clausen in the caves of ,brazil4 ,finally1 the facts here given show how cautious we ought to be in judging of the antiquity of a formation from even a great amount of difference between the extinct and living species in any one class of animals2--we ought even to be cautious in accepting the general proposition1 that change in organic forms and lapse of time are at all1 necessarily1 correlatives4 ,localities within the region of the ,pampas where great bones have been found4 ,the following list1 which includes every account which ,i have hitherto met with of the discovery of fossil mammiferous remains in the ,pampas1 may be hereafter useful to a geologist investigating this region1 and it tends to show their extraordinary abundance4 ,i heard of and saw many fossils1 the original position of which ,i could not ascertain2 and ,i received many statements too vague to be here inserted4 ,beginning to the south3--we have the two stations in ,bahia ,blanca1 described in this chapter1 where at ,p4 ,alta1 the ,megatherium1 ,megalonyx1 ,scelidotherium1 ,mylodon1 ,holophractus 7or an allied genus71 ,toxodon1 ,macrauchenia1 and an ,equus were collected2 and at ,m4 ,hermoso a ,ctenomys1 ,hydrochrus1 some other rodents and the bones of a great megatheroid quadruped4 ,close north-east of the ,s4 ,tapalguen1 we have the ,rios #8,huesos' 7i4e4 bones71 which probably takes its name from large fossil bones4 ,near ,villa ,nuevo1 and at ,las ,averias1 not far from the ,salado1 three nearly perfect skeletons1 one of the ,megatherium1 one of the ,glyptodon clavipes1 and one of some great ,dasypoid quadruped1 were found by the agent of ,sir ,w4 ,parish 7see his work 8,buenos ,ayres10 etc41 p4 #aga74 ,i have seen the tooth of a ,mastodon from the ,salado2 a little northward of this river1 on the borders of a lake near the ,g4 del ,monte1 ,i saw many bones1 and one large piece of dermal armour2 higher up the ,salado1 there is a place called ,monte 8,huesos40 ,on the ,matanzas1 about twenty miles south of ,buenos ,ayres1 the skeleton 7vide p4 #agh of 8,buenos ,ayres10 etc41 by ,sir ,w4 ,parish7 of a ,glyptodon was found about five feet beneath the surface2 here also 7see ,catalogue of ,royal ,college of ,surgeons7 remains of ,glyptodon clavipes1 ,g4 ornatus1 and ,g4 reticulatus were found4 ,signor ,angelis1 in a letter which ,i have seen1 refers to some great remains found in ,buenos ,ayres1 at a depth of twenty varas from the surface4 ,seven leagues north of this city the same author found the skeletons of ,mylodon robustus and ,glyptodon ornatus4 ,from this neighbourhood he has lately sent to the ,british ,museum the following fossils3--,remains of three or four individuals of ,megatherium2 of three species of ,glyptodon2 of three individuals of the ,mastodon ,andium2 of ,macrauchenia2 of a second species of ,toxodon1 different from ,t4 ,platensis2 and lastly1 of the ,machairodus1 a wonderful large carnivorous animal4 ,m4 d',orbigny has lately received from the ,recolate 78,voyage10 ,pal41 p4 #add71 near ,buenos ,ayres1 a tooth of ,toxodon ,platensis4 ,proceeding northward1 along the west bank of the ,parana1 we come to the ,rio ,luxan1 where two skeletons of the ,megatherium have been found2 and lately1 within eight leagues of the town of ,luxan1 ,dr4 ,f4 ,x4 ,muniz has collected 7,british ,packet1 ,buenos ,ayres1 ,september #be1 #ahda71 from an average depth of eighteen feet1 very numerous remains1 of no less than1 as he believes1 nine distinct species of mammifers4 ,at ,areco1 large bones have been found1 which are believed1 by the inhabitants1 to have been changed from small bones1 by the water of the river6 ,at ,arrecifes1 the ,glyptodon1 sent to the ,college of ,surgeons1 was found2 and ,i have seen two teeth of a ,mastodon from this quarter4 ,at ,s4 ,nicolas1 ,m4 d',orbigny found remains of a ,canis1 ,ctenomys1 and ,kerodon2 and ,m4 ,isabelle 78,voyage10 p4 #ccb7 refers to a gigantic ,armadillo found there4 ,at ,s4 ,carlos1 ,i heard of great bones4 ,a little below the mouth of the ,carcarana1 the two skeletons of ,mastodon were found2 on the banks of this river1 near ,s4 ,miguel1 ,i found teeth of the ,mastodon and ,toxodon2 and 8,falkner0 7p4 #ee7 describes the osseous armour of some great animal2 ,i heard of many other bones in this neighbourhood4 ,i have seen1 ,i may add1 in the possession of ,mr4 ,caldcleugh1 the tooth of a ,mastodon ,andium1 said to have been found in ,paraguay2 ,i may here also refer to a statement in this gentleman's travels 7vol4 i1 p4 #dh71 of a great skeleton having been found in the province of ,bolivia in ,brazil1 on the ,r4 de las ,contas4 ,the furthest point westward in the ,pampas1 at which ,i have heard of fossil bones1 was high up on the banks of ,r4 ,quinto4 ,in ,entre ,rios1 besides the remains of the ,mastodon1 ,toxodon1 ,equus1 and a great ,dasypoid quadruped near ,st4 ,fe ,bajada1 ,i received an account of bones having been found a little ,s4,e4 of ,p4 ,gorda 7on the ,parana71 and of an entire skeleton at ,matanzas1 on the ,arroyo del ,animal4 ,in ,banda ,oriental1 besides the remains of the ,toxodon1 ,mylodon1 and two skeletons of great animals with osseous armour 7distinct from that of the ,glyptodon71 found on the ,arroyos ,sarandis and ,berquelo1 ,m4 ,isabelle 78,voyage10 p4 #cbb7 says1 many bones have been found near the ,r4 ,negro1 and on the ,r4 ,arapey1 an affluent of the ,paraguay1 in lat4 #cj #dj south4 ,i heard of bones near the source of the ,a4 ,vivoras4 ,i saw the remains of a ,dasypoid quadruped from the ,arroyo ,seco1 close to ,m4 ,video2 and ,m4 d',orbigny refers 78,voyage10 ,golog41 p4 #bd71 to another found on the ,pedernal1 an affluent of the ,st4 ,lucia2 and ,signor ,angelis1 in a letter1 states that a third skeleton of this family has been found1 near ,canelones4 ,i saw a tooth of the ,mastodon from ,talas1 another affluent of the ,st4 ,lucia4 ,the most eastern point at which ,i heard of great bones having been found1 was at ,solis ,grande1 between ,m4 ,video and ,maldonado4 ,chapter ,v ,,on ,,the ,,older ,,tertiary ,,formations ,,of ,,patagonia ,,and ,,chile4 ,rio ,negro4--,s4 ,josef4--,port ,desire1 white pumiceous mudstone with infusoria4--,port ,s4 ,julian4--,santa ,cruz1 basaltic lava of4--,p4 ,gallegos4--,eastern ,tierra del ,fuego2 leaves of extinct beech-trees4--,summary on the ,patagonian tertiary formations4--,tertiary formations of the ,western ,coast4--,chonos and ,chiloe groups1 volcanic rocks of4--,concepcion4--,navidad4--,coquimbo4 --,summary4--,age of the tertiary formations4--,lines of elevation4--,silicified wood4--,comparative ranges of the extinct and living mollusca on the ,west ,coast of ,s4 ,america4--,climate of the tertiary period4--,on the causes of the absence of recent conchiferous deposits on the coast of ,s4 ,america4--,on the contemporaneous deposition and preservation of sedimentary formations4 ,rio ,negro4--,i can add little to the details given by ,m4 d',orbigny#a on the sandstone formation of this district4 ,the cliffs to the south of the river are about two hundred feet in height1 and are composed of sandstone of various tints and degrees of hardness4 ,one layer1 which thinned out at both ends1 consisted of earthy matter1 of a pale reddish colour1 with some gypsum1 and very like 7,i speak after comparison of the specimens brought home7 ,pampean mud3 above this was a layer of compact marly rock with dendritic manganese4 ,many blocks of a conglomerate of pumice-pebbles embedded in hard sandstone were strewed at the foot of the cliff1 and had evidently fallen from above4 ,a few miles ,n4,e4 of the town1 ,i found1 low down in the sandstone1 a bed1 a few inches in thickness1 of a white1 friable1 harsh-feeling sediment1 which adheres to the tongue1 is of easy fusibility1 and of little specific gravity2 examined under the microscope1 it is seen to be pumiceous tuff1 formed of broken transparent crystals4 ,in the cliffs south of the river1 there is1 also1 a thin layer of nearly similar nature1 but finer grained1 and not so white2 it might easily have been mistaken for a calcareous tuff1 but it contains no lime3 this substance precisely resembles a most widely extended and thick formation in ,southern ,patagonia1 hereafter to be described1 and which is remarkable for being partially formed of infusoria4 ,these beds1 conjointly with the conglomerate of pumice1 are interesting1 as showing the nature of the volcanic action in the ,cordillera during this old tertiary period4 #a 8,voyage10 ,part4 ,golog41 pp4 #eg-fe4 ,in a bed at the base of the southern cliffs1 ,m4 d',orbigny found two extinct fresh-water shells1 namely1 a ,unio and ,chilina4 ,this bed rested on one with bones of an extinct rodent1 namely1 the ,megamys ,patagoniensis2 and this again on another with extinct marine shells4 ,the species found by ,m4 d',orbigny in different parts of this formation consist of3-- ,ostrea ,patagonica1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f1 and whole coast of ,patagonia74 ,ostrea ,ferrarisi1 d',orbigny1 8,voyage1 ,pal40 ,ostrea ,alvarezii1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f1 and ,s4 ,josef74 ,pecten ,patagoniensis1 d',orbigny1 8,voyage1 ,pal40 ,venus ,munsterii1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f74 ,arca ,bonplandiana1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f74 ,according to ,m4 d',orbigny1 the sandstone extends westward along the coast as far as ,port ,s4 ,antonio1 and up the ,r4 ,negro far into the interior3 northward ,i traced it to the southern side of the ,rio ,colorado1 where it forms a low denuded plain4 ,this formation1 though contemporaneous with that of the rest of ,patagonia1 is quite different in mineralogical composition1 being connected with it only by the one thin white layer3 this difference may be reasonably attributed to the sediment brought down in ancient times by the ,rio ,negro2 by which agency1 also1 we can understand the presence of the fresh-water shells1 and of the bones of land animals4 ,judging from the identity of four of the above shells1 this formation is contemporaneous 7as remarked by ,m4 d',orbigny7 with that under the ,pampean deposit in ,entre ,rios and in ,banda ,oriental4 ,the gravel capping the sandstone plain1 with its calcareous cement and nodules of gypsum1 is probably1 from the reasons given in the ,first ,chapter1 contemporaneous with the uppermost beds of the ,pampean formation on the upper plain north of the ,colorado4 ,san ,josef4--,my examination here was very short3 the cliffs are about a hundred feet high2 the lower third consists of yellowish-brown1 soft1 slightly calcareous1 muddy sandstone1 parts of which when struck emit a fetid smell4 ,in this bed the great ,ostra ,patagonica1 often marked with dendritic manganese and small coral-lines1 were extraordinarily numerous4 ,i found here the following shells3-- ,ostrea ,patagonica1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f and whole coast of ,patagonia74 ,ostrea ,alvarezii1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f and ,r4 ,negro74 ,pecten ,paranensis1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f1 ,s4 ,julian1 and ,port ,desire74 ,pecten ,darwinianus1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f74 ,pecten actinodes1 ,g4 ,b4 ,sowerby4 ,terebratula ,patagonica1 ,g4 ,b4 ,sowerby 7also ,s4 ,julian74 ,casts of a ,turritella4 ,the four first of these species occur at ,st4 ,f in ,entre ,rios1 and the two first in the sandstone of the ,rio ,negro4 ,above this fossiliferous mass1 there is a stratum of very fine-grained1 pale brown mudstone1 including numerous lamin of selenite4 ,all the strata appear horizontal1 but when followed by the eye for a long distance1 they are seen to have a small easterly dip4 ,on the surface we have the porphyritic gravel1 and on it sand with recent shells4 ,nuevo ,gulf4--,from specimens and notes given me by ,lieutenant ,stokes1 it appears that the lower bed consists of soft muddy sandstone1 like that of ,s4 ,josef1 with many imperfect shells1 including the ,pecten ,paranensis1 d',orbigny1 casts of a ,turritella and ,scutella4 ,on this there are two strata of the pale brown mudstone1 also like that of ,s4 ,josef1 separated by a darker-coloured1 more argillaceous variety1 including the ,ostrea ,patagonica4 ,professor ,ehrenberg has examined this mudstone for me3 he finds in it three already known microscopic organisms1 enveloped in a fine-grained pumiceous tuff1 which ,i shall have immediately to describe in detail4 ,specimens brought to me from the uppermost bed1 north of the ,rio ,chupat1 consist of this same substance1 but of a whiter colour4 ,tertiary strata1 such as here described1 appear to extend along the whole coast between ,rio ,chupat and ,port ,desire1 except where interrupted by the underlying claystone porphyry1 and by some metamorphic rocks2 these hard rocks1 ,i may add1 are found at intervals over a space of about five degrees of latitude1 from ,point ,union to a point between ,port ,s4 ,julian and ,s4 ,cruz1 and will be described in the ensuing chapter4 ,many gigantic specimens of the ,ostrea ,patagonica were collected in the ,gulf of ,st4 ,george4 ,port ,desire4--,a good section of the lowest fossiliferous mass1 about forty feet in thickness1 resting on claystone porphyry1 is exhibited a few miles south of the harbour4 ,the shells sufficiently perfect to be recognised consist of3-- ,ostrea ,patagonica1 d',orbigny1 7also at ,st4 ,f1 and whole coast of ,patagonia74 ,pecten ,paranensis1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f1 ,s4 ,josef1 ,s4 ,julian74 ,pecten centralis1 ,g4 ,b4 ,sowerby 7also at ,s4 ,julian and ,s4 ,cruz74 ,cuculla alta1 ,g4 ,b4 ,sowerby 7also at ,s4 ,cruz74 ,nucula ornata1 ,g4 ,b4 ,sowerby4 ,turritella ,patagonica1 ,g4 ,b4 ,sowerby4 ,the fossiliferous strata1 when not denuded1 are conformably covered by a considerable thickness of the fine-grained pumiceous mudstone1 divided into two masses3 the lower half is very fine-grained1 slightly unctuous1 and so compact as to break with a semi-conchoidal fracture1 though yielding to the nail2 it includes lamin of selenite3 the upper half precisely resembles the one layer at the ,rio ,negro1 and with the exception of being whiter1 the upper beds at ,san ,josef and ,nuevo ,gulf4 ,in neither mass is there any trace to the naked eye of organic forms4 ,taking the entire deposit1 it is generally quite white1 or yellowish1 or feebly tinted with green2 it is either almost friable under the finger1 or as hard as chalk2 it is of easy fusibility1 of little specific gravity1 is not harsh to the touch1 adheres to the tongue1 and when breathed on exhales a strong aluminous odour2 it sometimes contains a very little calcareous matter1 and traces 7besides the included lamin7 of gypsum4 ,under the microscope1 according to ,professor ,ehrenberg1#b it consists of minute1 triturated1 cellular1 glassy fragments of pumice1 with some broken crystals4 ,in the minute glassy fragments1 ,professor ,ehrenberg recognises organic structures1 which have been affected by volcanic heat3 in the specimens from this place1 and from ,port ,s4 ,julian1 he finds sixteen ,polygastrica and twelve ,phytolitharia4 ,of these organisms1 seven are new forms1 the others being previously known3 all are of marine1 and chiefly of oceanic1 origin4 ,this deposit to the naked eye resembles the crust which often appears on weathered surfaces of feldspathic rocks2 it likewise resembles those beds of earthy feldspathic matter1 sometimes interstratified with porphyritic rocks1 as is the case in this very district with the underlying purple claystone porphyry4 ,from examining specimens under a common microscope1 and comparing them with other specimens undoubtedly of volcanic origin1 ,i had come to the same conclusion with ,professor ,ehrenberg1 namely1 that this great deposit1 in its first origin1 is of volcanic nature4 #b 8,monatsberichten de knig4 ,akad4 zu ,berlin10 vom ,april #ahde4 ,port ,s4 ,julian4--,on the south side of the harbour1 the following section 7figure ,no4 #ag7 gives the nature of the beds seen in the cliffs of the ninety feet plain4 ,beginning at the top3--first1 the earthy mass 7,,aa71 including the remains of the ,macrauchenia1 with recent shells on the surface2 second1 the porphyritic shingle 7,b71 which in its lower part is interstratified 7owing1 ,i believe1 to redisposition during denudation7 with the white pumiceous mudstone2 third1 this white mudstone1 about twenty feet in thickness1 and divided into two varieties 7,c and ,d71 both closely resembling the lower1 fine-grained1 more unctuous and compact kind at ,port ,desire2 and1 as at that place1 including much selenite2 fourth1 a fossiliferous mass1 divided into three main beds1 of which the uppermost is thin1 and consists of ferruginous sandstone1 with many shells of the great oyster and ,pecten ,paranensis2 the middle bed 7,e7 is a yellowish earthy sandstone abounding with ,scutell2 and the lowest bed 7,f7 is an indurated1 greenish1 sandy clay1 including large concretions of calcareous sandstone1 many shells of the great oyster1 and in parts almost made up of fragments of ,balanid4 ,out of these three beds1 ,i procured the following twelve species1 of which the two first were exceedingly numerous in individuals1 as were the ,terebratul and ,turritell in certain layers3-- ,ostrea ,patagonica1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f1 and whole coast of ,patagonia74 ,pecten ,paranensis1 d',orbigny1 8,voyage1 ,pal40 7,st4 ,f1 ,s4 ,josef1 ,port ,desire74 ,pecten centralis1 ,g4 ,b4 ,sowerby 7also at ,port ,desire and ,s4 ,cruz74 ,pecten geminatus1 ,g4 ,b4 ,sowerby4 ,terebratula ,patagonica1 ,g4 ,b4 ,sowerby 7also ,s4 ,josef74 ,struthiolaria ornata1 ,g4 ,b4 ,sowerby 7also ,s4 ,cruz74 ,fusus ,patagonicus1 ,g4 ,b4 ,sowerby4 ,fusus ,noachinus1 ,g4 ,b4 ,sowerby4 ,scalaria rugulosa1 ,g4 ,b4 ,sowerby4 ,turritella ambulacrum1 ,g4 ,b4 ,sowerby 7also ,s4 ,cruz74 ,pyrula1 cast of1 like ,p4 ventricosa of ,sowerby1 ,tank ,cat4 ,balanus varians1 ,g4 ,b4 ,sowerby4 ,scutella1 differing from the species from ,nuevo ,gulf4 ,no4 #ag ,section of the strata exhibited in the cliffs of the ninety feet plain at ,port ,s4 ,julian4 ,illustration3 ,section of the strata exhibited in the cliffs of the ninety feet plain at ,port ,s4 ,julian4 ,at the head of the inner harbour of ,port ,s4 ,julian1 the fossiliferous mass is not displayed1 and the sea-cliffs from the water's edge to a height of between one and two hundred feet are formed of the white pumiceous mudstone1 which here includes innumerable1 far-extended1 sometimes horizontal1 sometimes inclined or vertical lamin of transparent gypsum1 often about an inch in thickness4 ,further inland1 with the exception of the superficial gravel1 the whole thickness of the truncated hills1 which represent a formerly continuous plain #iej feet in height1 appears to be formed of this white mudstone3 here and there1 however1 at various heights1 thin earthy layers1 containing the great oyster1 ,pecten ,paranensis and ,turritella ambulacrum1 are interstratified2 thus showing that the whole mass belongs to the same epoch4 ,i nowhere found even a fragment of a shell actually in the white deposit1 and only a single cast of a ,turritella4 ,out of the eighteen microscopic organisms discovered by ,ehrenberg in the specimens from this place1 ten are common to the same deposit at ,port ,desire4 ,i may add that specimens of this white mudstone1 with the same identical characters were brought me from two points1--one twenty miles north of ,s4 ,julian1 where a wide gravel-capped plain1 #cej feet in height1 is thus composed2 and the other forty miles south of ,s4 ,julian1 where1 on the old charts1 the cliffs are marked as 8,chalk ,hills40 ,santa ,cruz4--,the gravel-capped cliffs at the mouth of the river are #cee feet in height3 the lower part1 to a thickness of fifty or sixty feet1 consists of a more or less hardened1 darkish1 muddy1 or argillaceous sandstone 7like the lowest bed of ,port ,desire71 containing very many shells1 some silicified and some converted into yellow calcareous spar4 ,the great oyster is here numerous in layers2 the ,trigonocelia and ,turritella are also very numerous3 it is remarkable that the ,pecten ,paranensis1 so common in all other parts of the coast1 is here absent3 the shells consist of3-- ,ostrea ,patagonica1 d',orbigny1 8,voyage1 ,pal40 7also at ,st4 ,f and whole coast of ,patagonia74 ,pecten centralis1 ,g4 ,b4 ,sowerby 7also ,p4 ,desire and ,s4 ,julian74 ,venus meridionalis of ,g4 ,b4 ,sowerby4 ,crassatella ,lyellii1 ,g4 ,b4 ,sowerby4 ,cardium puelchum1 ,g4 ,b4 ,sowerby4 ,cardita ,patagonica1 ,g4 ,b4 ,sowerby4 ,mactra rugata1 ,g4 ,b4 ,sowerby4 ,mactra ,darwinii1 ,g4 ,b4 ,sowerby4 ,cuculla alta1 ,g4 ,b4 ,sowerby 7also ,p4 ,desire74 ,trigonocelia insolita1 ,g4 ,b4 ,sowerby4 ,nucula 787 glabra1 ,g4 ,b4 ,sowerby4 ,crepidula gregaria1 ,g4 ,b4 ,sowerby4 ,voluta alta1 ,g4 ,b4 ,sowerby4 ,trochus collaris1 ,g4 ,b4 ,sowerby4 ,natica solida 7871 ,g4 ,b4 ,sowerby4 ,struthiolaria ornata1 ,g4 ,b4 ,sowerby 7also ,p4 ,desire74 ,turritella ambulacrum1 ,g4 ,b4 ,sowerby 7also ,p4 ,s4 ,julian74 ,imperfect fragments of the genera ,byssoarca1 ,artemis1 and ,fusus4 ,the upper part of the cliff is generally divided into three great strata1 differing slightly in composition1 but essentially resembling the pumiceous mudstone of the places farther north2 the deposit1 however1 here is more arenaceous1 of greater specific gravity1 and not so white3 it is interlaced with numerous thin veins1 partially or quite filled with transverse fibres of gypsum2 these fibres were too short to reach across the vein1 have their extremities curved or bent3 in the same veins with the gypsum1 and likewise in separate veins as well as in little nests1 there is much powdery sulphate of magnesia 7as ascertained by ,mr4 ,reeks7 in an uncompressed form3 ,i believe that this salt has not heretofore been found in veins4 ,of the three beds1 the central one is the most compact1 and more like ordinary sandstone3 it includes numerous flattened spherical concretions1 often united like a necklace1 composed of hard calcareous sandstone1 containing a few shells3 some of these concretions were four feet in diameter1 and in a horizontal line nine feet apart1 showing that the calcareous matter must have been drawn to the centres of attraction1 from a distance of four feet and a half on both sides4 ,in the upper and lower finer-grained strata1 there were other concretions of a grey colour1 containing calcareous matter1 and so fine-grained and compact1 as almost to resemble porcelain-rock3 ,i have seen exactly similar concretions in a volcanic tufaceous bed in ,chiloe4 ,although in this upper fine-grained strata1 organic remains were very rare1 yet ,i noticed a few of the great oyster2 and in one included soft ferruginous layer1 there were some specimens of the ,cuculla alta 7found at ,port ,desire in the lower fossiliferous mass7 and of the ,mactra rugata1 which latter shell has been partially converted into gypsum4 ,no4 #ah ,section of the plain at ,patagonia1 on the banks of the ,s4 ,cruz4 ,illustration3 ,section of the plain at ,patagonia1 on the banks of the ,s4 ,cruz4 ,in ascending the valley of the ,s4 ,cruz1 the upper strata of the coast-cliffs are prolonged1 with nearly the same characters1 for fifty miles3 at about this point1 they begin in the most gradual and scarcely perceptible manner1 to be banded with white lines2 and after ascending ten miles farther1 we meet with distinct thin layers of whitish1 greenish1 and yellowish fine-grained1 fusible sediments4 ,at eighty miles from the coast1#c in a cliff thus composed1 there were a few layers of ferruginous sandstone1 and of an argillaceous sandstone with concretions of marl like those in the ,pampas4 ,at one hundred miles from the coast1 that is at a central point between the ,atlantic and the ,cordillera1 we have the section in figure ,no4 #ah4 #c ,at this spot1 for a space of three-quarters of a mile along the north side of the river1 and for a width of half a mile1 there has been a great slip1 which has formed hills between sixty and seventy feet in height1 and has tilted the strata into highly inclined and even vertical positions4 ,the strata generally dipped at an angle of #de towards the cliff from which they had slided4 ,i have observed in slips1 both on a small and large scale1 that this inward dip is very general4 ,is it due to the hydrostatic pressure of water percolating with difficulty through the strata acting with greater force at the base of the mass than against the upper part8 ,the upper half of the sedimentary mass1 under the basaltic lava1 consists of innumerable zones of perfectly white bright green1 yellowish and brownish1 fine-grained1 sometimes incoherent1 sedimentary matter4 ,the white1 pumiceous1 trachytic tuff-like varieties are of rather greater specific gravity than the pumiceous mudstone on the coast to the north2 some of the layers1 especially the browner ones1 are coarser1 so that the broken crystals are distinguishable with a weak lens4 ,the layers vary in character in short distances4 ,with the exception of a few of the ,ostrea ,patagonica1 which appeared to have rolled down from the cliff above1 no organic remains were found4 ,the chief difference between these layers taken as a whole1 and the upper beds both at the mouth of the river and on the coast northward1 seems to lie in the occasional presence of more colouring matter1 and in the supply having been intermittent2 these characters1 as we have seen1 very gradually disappear in descending the valley1 and this fact may perhaps be accounted for by the currents of a more open sea having blended together the sediment from a distant and intermittent source4 ,the coloured layers in the foregoing section rest on a mass1 apparently of great thickness 7but much hidden by the talus71 of soft sandstone1 almost composed of minute pebbles1 from one-tenth to two-tenths of an inch in diameter1 of the rocks 7with the entire exception of the basaltic lava7 composing the great boulders on the surface of the plain1 and probably composing the neighbouring ,cordillera4 ,five miles higher up the valley1 and again thirty miles higher up#d 7that is twenty miles from the nearest range of the ,cordillera71 the lower plain included within the upper escarpments1 is formed1 as seen on the banks of the river1 of a nearly similar but finer-grained1 more earthy1 laminated sandstone1 alternating with argillaceous beds1 and containing numerous moderately sized pebbles of the same rocks1 and some shells of the great ,ostrea ,patagonica4 ,as most of these shells had been rolled before being here embedded1 their presence does not prove that the sandstone belongs to the great ,patagonian tertiary formation1 for they might have been redeposited in it1 when the valley existed as a sea-strait2 but as amongst the pebbles there were none of basalt1 although the cliffs on both sides of the valley are composed of this rock1 ,i believe that the sandstone does belong to this formation4 ,at the highest point to which we ascended1 twenty miles distant from the nearest slope of the ,cordillera1 ,i could see the horizontally zoned white beds1 stretching under the black basaltic lava1 close up to the mountains2 so that the valley of the ,s4 ,cruz gives a fair idea of the constitution of the whole width of ,patagonia4 #d ,i found at both places1 but not in situ1 quantities of coniferous and ordinary dicotyledonous silicified wood1 which was examined for me by ,mr4 ,r4 ,brown4 ,basaltic lava of the ,s4 ,cruz4--,this formation is first met with sixty-seven miles from the mouth of the river2 thence it extends uninterruptedly1 generally but not exclusively on the northern side of the valley1 close up to the ,cordillera4 ,the basalt is generally black and fine-grained1 but sometimes grey and laminated2 it contains some olivine1 and high up the valley much glassy feldspar1 where1 also1 it is often amygdaloidal2 it is never highly vesicular1 except on the sides of rents and on the upper and lower1 spherically laminated surfaces4 ,it is often columnar2 and in one place ,i saw magnificent columns1 each face twelve feet in width1 with their interstices filled up with calcareous tuff4 ,the streams rest conformably on the white sedimentary beds1 but ,i nowhere saw the actual junction2 nor did ,i anywhere see the white beds actually superimposed on the lava2 but some way up the valley at the foot of the uppermost escarpments1 they must be thus superimposed4 ,moreover1 at the lowest point down the valley1 where the streams thin out and terminate in irregular projections1 the spaces or intervals between these projections are filled up to the level of the now denuded and gravel-capped surfaces of the plains1 with the white-zoned sedimentary beds2 proving that this matter continued to be deposited after the streams had flowed4 ,hence we may conclude that the basalt is contemporaneous with the upper parts of the great tertiary formation4 ,the lava where first met with is #acj feet in thickness3 it there consists of two1 three1 or perhaps more streams1 divided from each other by vesicular spheroids like those on the surface4 ,from the streams having1 as it appears1 extended to different distances1 the terminal points are of unequal heights4 ,generally the surface of the basalt is smooth them in one part high up the valley1 it was so uneven and hummocky1 that until ,i afterwards saw the streams extending continuously on both sides of the valley up to a height of about three thousand feet close to the ,cordillera1 ,i thought that the craters of eruption were probably close at hand4 ,this hummocky surface ,i believe to have been caused by the crossing and heaping up of different streams4 ,in one place1 there were several rounded ridges about twenty feet in height1 some of them as broad as high1 and some broader1 which certainly had been formed whilst the lava was fluid1 for in transverse sections each ridge was seen to be concentrically laminated1 and to be composed of imperfect columns radiating from common centres1 like the spokes of wheels4 ,the basaltic mass where first met with is1 as ,i have said1 #acj feet in thickness1 and1 thirty-five miles higher up the valley1 it increases to #cbb feet4 ,in the first fourteen and a half miles of this distance1 the upper surface of the lava1 judging from three measurements taken above the level of the river 7of which the apparently very uniform inclination has been calculated from its total height at a point #ace miles from the mouth71 slopes towards the ,atlantic at an angle of only #j #g #bj3 this must be considered only as an approximate measurement1 but it cannot be far wrong4 ,taking the whole thirty-five miles1 the upper surface slopes at an angle of #j #aj #ec2 but this result is of no value in showing the inclination of any one stream1 for halfway between the two points of measurement1 the surface suddenly rises between one hundred and two hundred feet1 apparently caused by some of the uppermost streams having extended thus far and no farther4 ,from the measurement made at these two points1 thirty-five miles apart1 the mean inclination of the sedimentary beds1 over which the lava has flowed1 is now 7after elevation from under the sea7 only #j #g #eb3 for the sake of comparison1 it may be mentioned that the bottom of the present sea in a line from the mouth of the ,s4 ,cruz to the ,falkland ,islands1 from a depth of seventeen fathoms to a depth of eighty-five fathoms1 declines at an angle of #j #a #bb2 between the beach and the depth of seventeen fathoms1 the slope is greater4 ,from a point about half-way up the valley1 the basaltic mass rises more abruptly towards the foot of the ,cordillera1 namely1 from a height of #a1bjd feet1 to about #c1jjj feet above the sea4 ,this great deluge of lava is worthy1 in its dimensions1 of the great continent to which it belongs4 ,the aggregate streams have flowed from the ,cordillera to a distance 7unparalleled1 ,i believe1 in any case yet known7 of about one hundred geographical miles4 ,near their furthest extremity their total thickness is #acj feet1 which increase thirty-five miles farther inland1 as we have just seen1 to #cbb feet4 ,the least inclination given by ,m4 ,e4 de ,beaumont of the upper surface of a lava-stream1 namely #j #cj1 is that of the great subaerial eruption in #aghc from ,skaptar ,jukul in ,iceland2 and ,m4 ,e4 de ,beaumont shows#e that it must have flowed down a mean inclination of less than #j #bj4 ,but we now see that under the pressure of the sea1 successive streams have flowed over a smooth bottom with a mean inclination of not more than #j #g #eb2 and that the upper surface of the terminal portion 7over a space of fourteen and a half miles7 has an inclination of not more than #j #g #bj4 ,if the elevation of ,patagonia has been greater nearer the ,cordillera than near the ,atlantic 7as is probable71 then these angles are now all too large4 ,i must repeat1 that although the foregoing measurements1 which were all carefully taken with the barometer1 may not be absolutely correct1 they cannot be widely erroneous4 #e 8,mmoires pour servir10 etc41 pp4 #agh and #bag4 ,southward of the ,s4 ,cruz1 the cliffs of the #hdj feet plain extend to ,coy ,inlet1 and owing to the naked patches of the white sediment1 they are said on the charts to be 8like the coast of ,kent40 ,at ,coy ,inlet the high plain trends inland1 leaving flat-topped outliers4 ,at ,port ,gallegos 7lat4 #ea #ce1 and ninety miles south of ,s4 ,cruz71 ,i am informed by ,captain ,sulivan1 ,r4,n41 that there is a gravel-capped plain from two to three hundred feet in height1 formed of numerous strata1 some fine-grained and pale-coloured1 like the upper beds at the mouth of the ,s4 ,cruz1 others rather dark and coarser1 so as to resemble gritstones or tuffs2 these latter include rather large fragments of apparently decomposed volcanic rocks2 there are1 also1 included layers of gravel4 ,this formation is highly remarkable1 from abounding with mammiferous remains1 which have not as yet been examined by ,professor ,owen1 but which include some large1 but mostly small1 species of ,pachydermata1 ,edentata1 and ,rodentia4 ,from the appearance of the pale-coloured1 fine-grained beds1 ,i was inclined to believe that they corresponded with the upper beds of the ,s4 ,cruz2 but ,professor ,ehrenberg1 who has examined some of the specimens1 informs me that the included microscopical organisms are wholly different1 being fresh and brackish-water forms4 ,hence the two to three hundred feet plain at ,port ,gallegos is of unknown age1 but probably of subsequent origin to the great ,patagonian tertiary formation4 ,eastern ,tierra del ,fuego4--,judging from the height1 the general appearance1 and the white colour of the patches visible on the hill sides1 the uppermost plain1 both on the north and western side of the ,strait of ,magellan1 and along the eastern coast of ,tierra del ,fuego as far south as near ,port ,st4 ,polycarp1 probably belongs to the great ,patagonian tertiary formation4 ,these higher table-ranges are fringed by low1 irregular1 extensive plains1 belonging to the boulder formation1#f and composed of coarse unstratified masses1 sometimes associated 7as north of ,c4 ,virgin's7 with fine1 laminated1 muddy sandstones4 ,the cliffs in ,sebastian ,bay are #bjj feet in height1 and are composed of fine sandstones1 often in curvilinear layers1 including hard concretions of calcareous sandstone1 and layers of gravel4 ,in these beds there are fragments of wood1 legs of crabs1 barnacles encrusted with corallines still partially retaining their colour1 imperfect fragments of a ,pholas distinct from any known species1 and of a ,venus1 approaching very closely to1 but slightly different in form from1 the ,v4 lenticularis1 a species living on the coast of ,chile4 ,leaves of trees are numerous between the lamin of the muddy sandstone2 they belong1 as ,i am informed by ,dr4 ,j4 ,d4 ,hooker1#g to three species of deciduous beech1 different from the two species which compose the great proportion of trees in this forest-clad land4 ,from these facts it is difficult to conjecture1 whether we here see the basal part of the great ,patagonian formation1 or some later deposit4 #f ,described in the 8,geological ,transactions10 vol4 vi1 p4 #dae4 #g 8,botany of the ,antarctic ,voyage10 p4 #bab4 ,summary on the ,patagonian tertiary formation4--,four out of the seven fossil shells1 from ,st4 ,f in ,entre ,rios1 were found by ,m4 d',orbigny in the sandstone of the ,rio ,negro1 and by me at ,san ,josef4 ,three out of the six from ,san ,josef are identical with those from ,port ,desire and ,s4 ,julian1 which two places have together fifteen species1 out of which three are common to both4 ,santa ,cruz has seventeen species1 out of which five are common to ,port ,desire and ,s4 ,julian4 ,considering the difference in latitude between these several places1 and the small number of species altogether collected1 namely thirty-six1 ,i conceive the above proportional number of species in common1 is sufficient to show that the lower fossiliferous mass belongs nearly1 ,i do not say absolutely1 to the same epoch4 ,what this epoch may be1 compared with the ,european tertiary stages1 ,m4 d',orbigny will not pretend to determine4 ,the thirty-six species 7including those collected by myself and by ,m4 d',orbigny7 are all extinct1 or at least unknown2 but it should be borne in mind1 that the present coast consists of shingle1 and that no one1 ,i believe1 has dredged here for shells2 hence it is not improbable that some of the species may hereafter be found living4 ,some few of the species are closely related with existing ones2 this is especially the case1 according to ,m4 d',orbigny and ,mr4 ,sowerby1 with the ,fusus ,patagonicus2 and1 according to ,mr4 ,sowerby1 with the ,pyrula1 the ,venus meridionalis1 the ,crepidula gregaria1 and the ,turritella ambulacrum1 and ,t4 ,patagonica4 ,at least three of the genera1 namely1 ,cuculla1 ,crassatella1 and 7as determined by ,mr4 ,sowerby7 ,struthiolaria1 are not found in this quarter of the world2 and ,trigonocelia is extinct4 ,the evidence taken altogether indicates that this great tertiary formation is of considerable antiquity2 but when treating of the ,chilean beds1 ,i shall have to refer again to this subject4 ,the white pumiceous mudstone1 with its abundant gypsum1 belongs to the same general epoch with the underlying fossiliferous mass1 as may be inferred from the shells included in the intercalated layers at ,nuevo ,gulf1 ,s4 ,julian1 and ,s4 ,cruz4 ,out of the twenty-seven marine microscopic structures found by ,professor ,ehrenberg in the specimens from ,s4 ,julian and ,port ,desire1 ten are common to these two places3 the three found at ,nuevo ,gulf are distinct4 ,i have minutely described this deposit1 from its remarkable characters and its wide extension4 ,from ,coy ,inlet to ,port ,desire1 a distance of #bcj miles1 it is certainly continuous2 and ,i have reason to believe that it likewise extends to the ,rio ,chupat1 ,nuevo ,gulf1 and ,san ,josef1 a distance of #egj miles3 we have1 also1 seen that a single layer occurs at the ,rio ,negro4 ,at ,port ,s4 ,julian it is from eight to nine hundred feet in thickness2 and at ,s4 ,cruz it extends1 with a slightly altered character1 up to the ,cordillera4 ,from its microscopic structure1 and from its analogy with other formations in volcanic districts1 it must be considered as originally of volcanic origin3 it may have been formed by the long-continued attrition of vast quantities of pumice1 or judging from the manner in which the mass becomes1 in ascending the valley of ,s4 ,cruz1 divided into variously coloured layers1 from the long-continued eruption of clouds of fine ashes4 ,in either case1 we must conclude1 that the southern volcanic orifices of the ,cordillera1 now in a dormant state1 were at about this period over a wide space1 and for a great length of time1 in action4 ,we have evidence of this fact1 in the latitude of the ,rio ,negro1 in the sandstone-conglomerate with pumice1 and demonstrative proof of it1 at ,s4 ,cruz1 in the vast deluges of basaltic lava3 at this same tertiary period1 also1 there is distinct evidence of volcanic action in ,western ,banda ,oriental4 ,the ,patagonian tertiary formation extends continuously1 judging from fossils alone1 from ,s4 ,cruz to near the ,rio ,colorado1 a distance of above six hundred miles1 and reappears over a wide area in ,entre ,rios and ,banda ,oriental1 making a total distance of #a1ajj miles2 but this formation undoubtedly extends 7though no fossils were collected7 far south of the ,s4 ,cruz1 and1 according to ,m4 d',orbigny1 #abj miles north of ,st4 ,f4 ,at ,s4 ,cruz we have seen that it extends across the continent2 being on the coast about eight hundred feet in thickness 7and rather more at ,s4 ,julian71 and rising with the contemporaneous lava-streams to a height of about three thousand feet at the base of the ,cordillera4 ,it rests1 wherever any underlying formation can be seen1 on plutonic and metamorphic rocks4 ,including the newer ,pampean deposit1 and those strata in ,eastern ,tierra del ,fuego of doubtful age1 as well as the boulder formation1 we have a line of more than twenty-seven degrees of latitude1 equal to that from the ,straits of ,gibraltar to the south of ,iceland1 continuously composed of tertiary formations4 ,throughout this great space the land has been upraised1 without the strata having been in a single instance1 as far as my means of observation went1 unequally tilted or dislocated by a fault4 ,tertiary ,formations on the ,west ,coast4 ,chonos ,archipelago4--,the numerous islands of this group1 with the exception of ,lemus1 ,ypun1 consist of metamorphic schists2 these two islands are formed of softish grey and brown1 fusible1 often laminated sandstones1 containing a few pebbles1 fragments of black lignite1 and numerous mammillated concretions of hard calcareous sandstone4 ,out of these concretions at ,ypun 7lat4 #dj #cj ,s471 ,i extracted the four following extinct species of shells3-- ,turritella suturalis1 ,g4 ,b4 ,sowerby 7also ,navidad74 ,sigaretus subglobosus1 ,g4 ,b4 ,sowerby 7also ,navidad74 ,cythera 787 sulculosa 7871 ,g4 ,b4 ,sowerby 7also ,chiloe and ,huafo874 ,voluta1 fragments of4 ,in the northern parts of this group there are some cliffs of gravel and of the boulder formation4 ,in the southern part 7at ,p4 ,andres in ,tres ,montes71 there is a volcanic formation1 probably of tertiary origin4 ,the lavas attain a thickness of from two to three hundred feet2 they are extremely variable in colour and nature1 being compact1 or brecciated1 or cellular1 or amygdaloidal with zeolite1 agate and bole1 or porphyritic with glassy albitic feldspar4 ,there is also much imperfect rubbly pitchstone1 with the interstices charged with powdery carbonate of lime apparently of contemporaneous origin4 ,these lavas are conformably associated with strata of breccia and of brown tuff containing lignite4 ,the whole mass has been broken up and tilted at an angle of #de1 by a series of great volcanic dikes1 one of which was thirty yards in breadth4 ,this volcanic formation resembles one1 presently to be described1 in ,chiloe4 ,huafo4--,this island lies between the ,chonos and ,chiloe groups3 it is about eight hundred feet high1 and perhaps has a nucleus of metamorphic rocks4 ,the strata which ,i examined consisted of fine-grained muddy sandstones1 with fragments of lignite and concretions of calcareous sandstone4 ,i collected the following extinct shells1 of which the ,turritella was in great numbers3-- ,bulla cosmophila1 ,g4 ,b4 ,sowerby4 ,pleurotoma subqualis1 ,g4 ,b4 ,sowerby4 ,fusus cleryanus1 d',orbigny1 8,voyage ,pal40 7also at ,coquimbo74 ,triton leucostomoides1 ,g4 ,b4 ,sowerby4 ,turritella ,chilensis1 ,g4 ,b4 ,sowerby 7also ,mocha74 ,venus1 probably a distinct species1 but very imperfect4 ,cythera 787 sulculosa 7871 probably a distinct species1 but very imperfect4 ,dentalium majus1 ,g4 ,b4 ,sowerby4 ,chiloe4--,this fine island is about one hundred miles in length4 ,the entire southern part1 and the whole western coast1 consists of mica-schist1 which likewise is seen in the ravines of the interior4 ,the central mountains rise to a height of #c1jjj feet1 and are said to be partly formed of granite and greenstone3 there are two small volcanic districts4 ,the eastern coast1 and large parts of the northern extremity of the island are composed of gravel1 the boulder formation1 and underlying horizontal strata4 ,the latter are well displayed for twenty miles north and south of ,castro2 they vary in character from common sandstone to fine-grained1 laminated mudstones3 all the specimens which ,i examined are easily fusible1 and some of the beds might be called volcanic grit-stones4 ,these latter strata are perhaps related to a mass of columnar trachyte which occurs behind ,castro4 ,the sandstone occasionally includes pebbles1 and many fragments and layers of lignite2 of the latter1 some are apparently formed of wood and others of leaves3 one layer on the ,n4,w4 side of ,lemuy is nearly two feet in thickness4 ,there is also much silicified wood1 both common dicotyledonous and coniferous3 a section of one specimen in the direction of the medullary rays has1 as ,i am informed by ,mr4 ,r4 ,brown1 the discs in a double row placed alternately1 and not opposite as in the true ,araucaria4 ,i found marine remains only in one spot1 in some concretions of hard calcareous sandstone3 in several other districts ,i have observed that organic remains were exclusively confined to such concretions2 are we to account for this fact1 by the supposition that the shells lived only at these points1 or is it not more probable that their remains were preserved only where concretions were formed8 ,the shells here are in a bad state1 they consist of3-- ,tellinides 787 oblonga1 ,g4 ,b4 ,sowerby 7a solenella in ,m4 d',orbigny's opinion74 ,natica striolata1 ,g4 ,b4 ,sowerby4 ,natica 787 pumila1 ,g4 ,b4 ,sowerby4 ,cythera 787 sulculosa1 ,g4 ,b4 ,sowerby 7also ,ypun and ,huafo874 ,at the northern extremity of the island1 near ,s4 ,carlos1 there is a large volcanic formation1 between five and seven hundred feet in thickness4 ,the commonest lava is blackish-grey or brown1 either vesicular1 or amygdaloidal with calcareous spar and bole3 most even of the darkest varieties fuse into a pale-coloured glass4 ,the next commonest variety is a rubbly1 rarely well characterised pitchstone 7fusing into a white glass7 which passes in the most irregular manner into stony grey lavas4 ,this pitchstone1 as well as some purple claystone porphyry1 certainly flowed in the form of streams4 ,these various lavas often pass1 at a considerable depth from the surface1 in the most abrupt and singular manner into wacke4 ,great masses of the solid rock are brecciated1 and it was generally impossible to discover whether the recementing process had been an igneous or aqueous action4#h ,the beds are obscurely separated from each other2 they are sometimes parted by seams of tuff and layers of pebbles4 ,in one place they rested on1 and in another place were capped by1 tuffs and girt-stones1 apparently of submarine origin4 #h ,in a cliff of the hardest fragmentary mass1 ,i found several tortuous1 vertical veins1 varying in thickness from a few tenths of an inch to one inch and a half1 of a substance which ,i have not seen described4 ,it is glossy1 and of a brown colour2 it is thinly laminated1 with the lamin transparent and elastic2 it is a little harder than calcareous spar2 it is infusible under the blowpipe1 sometimes decrepitates1 gives out water1 curls up1 blackens1 and becomes magnetic4 ,borax easily dissolves a considerable quantity of it1 and gives a glass tinged with green4 ,i have no idea what its true nature is4 ,on first seeing it1 ,i mistook it for lignite6 ,the neighbouring peninsula of ,lacuy is almost wholly formed of tufaceous deposits1 connected probably in their origin with the volcanic hills just described4 ,the tuffs are pale-coloured1 alternating with laminated mudstones and sandstones 7all easily fusible71 and passing sometimes into fine-grained white beds strikingly resembling the great upper infusorial deposit of ,patagonia1 and sometimes into brecciolas with pieces of pumice in the last stage of decay2 these again pass into ordinary coarse breccias and conglomerates of hard rocks4 ,within very short distances1 some of the finer tuffs often passed into each other in a peculiar manner1 namely1 by irregular polygonal concretions of one variety increasing so much and so suddenly in size1 that the second variety1 instead of any longer forming the entire mass1 was left merely in thin veins between the concretions4 ,in a straight line of cliffs1 at ,point ,tenuy1 ,i examined the following remarkable section 7figure ,no4 #ai73-- ,no4 #ai ,illustration3 ,section at ,point ,tenuy ,on the left hand1 the lower part 7,,aa7 consists of regular1 alternating strata of brown tuffs and greenish laminated mudstone1 gently inclined to the right1 and conformably covered by a mass 7,b left7 of a white1 tufaceous and brecciolated deposit4 ,on the right hand1 the whole cliff 7,,bb right7 consists of the same white tufaceous matter1 which on this side presents scarcely a trace of stratification1 but to the left becomes very gradually and rather indistinctly divided into strata quite conformable with the underlying beds 7,,aa73 moreover1 a few hundred yards further to the left1 where the surface has been less denuded1 the tufaceous strata 7,b left7 are conformably covered by another set of strata1 like the underlying ones 7,,aa7 of this section4 ,in the middle of the diagram1 the beds 7,,aa7 are seen to be abruptly cut off1 and to abut against the tufaceous non-stratified mass2 but the line of junction has been accidentally not represented steep enough1 for ,i particularly noticed that before the beds had been tilted to the right1 this line must have been nearly vertical4 ,it appears that a current of water cut for itself a deep and steep submarine channel1 and at the same time or afterwards filled it up with the tufaceous and brecciolated matter1 and spread the same over the surrounding submarine beds2 the matter becoming stratified in these more distant and less troubled parts1 and being moreover subsequently covered up by other strata 7like ,,aa7 not shown in the diagram4 ,it is singular that three of the beds 7of ,,aa7 are prolonged in their proper direction1 as represented1 beyond the line of junction into the white tufaceous matter3 the prolonged portions of two of the beds are rounded2 in the third1 the terminal fragment has been pushed upwards3 how these beds could have been left thus prolonged1 ,i will not pretend to explain4 ,in another section on the opposite side of a promontory1 there was at the foot of this same line of junction1 that is at the bottom of the old submarine channel1 a pile of fragments of the strata 7,,aa71 with their interstices filled up with white tufaceous matter3 this is exactly what might have been anticipated under such circumstances4 ,no4 #bj ,ground plan showing the relation between veins and concretionary zones in a mass of tuff4 ,illustration3 ,ground plan showing the relation between veins and concretionary zones in a mass of tuff4 ,the various tufaceous and other beds at this northern end of ,chiloe probably belong to about the same age with those near ,castro1 and they contain1 as there1 many fragments of black lignite and of silicified and pyritous wood1 often embedded close together4 ,they also contain many and singular concretions3 some are of hard calcareous sandstone1 in which it would appear that broken volcanic crystals and scales of mica have been better preserved 7as in the case of the organic remains near ,castro7 than in the surrounding mass4 ,other concretions in the white brecciola are of a hard1 ferruginous1 yet fusible1 nature2 they are as round as cannon-balls1 and vary from two or three inches to two feet in diameter2 their insides generally consist either of fine1 scarcely coherent volcanic sand1#i or of an argillaceous tuff2 in this latter case1 the external crust was quite thin and hard4 ,some of these spherical balls were encircled in the line of their equators1 by a necklace-like row of smaller concretions4 ,again there were other concretions1 irregularly formed1 and composed of a hard1 compact1 ash-coloured stone1 with an almost porcelainous fracture1 adhesive to the tongue1 and without any calcareous matter4 ,these beds are1 also1 interlaced by many veins1 containing gypsum1 ferruginous matter1 calcareous spar1 and agate4 ,it was here seen with remarkable distinctness1 how intimately concretionary action and the production of fissures and veins are related together4 ,figure #bj is an accurate representation of a horizontal space of tuff1 about four feet long by two and a half in width3 the double lines represent the fissures partially filled with oxide of iron and agate3 the curvilinear lines show the course of the innumerable1 concentric1 concretionary zones of different shades of colour and of coarseness in the particles of tuff4 ,the symmetry and complexity of the arrangement gave the surface an elegant appearance4 ,it may be seen how obviously the fissures determine 7or have been determined by7 the shape1 sometimes of the whole concretion1 and sometimes only of its central parts4 ,the fissures also determine the curvatures of the long undulating zones of concretionary action4 ,from the varying composition of the veins and concretions1 the amount of chemical action which the mass has undergone is surprisingly great2 and it would likewise appear from the difference in size in the particles of the concretionary zones1 that the mass1 also1 has been subjected to internal mechanical movements4 #i ,the frequent tendency in iron to form hollow concretions or shell containing incoherent matter is singular2 ,d',aubuisson 78,trait de ,gogn40 tome i1 p4 #cah7 remarks on this circumstance4 ,in the peninsula of ,lacuy1 the strata over a width of four miles have been upheaved by three distinct1 and some other indistinct1 lines of elevation1 ranging within a point of north and south4 ,one line1 about two hundred feet in height1 is regularly anticlinal1 with the strata dipping away on both sides1 at an angle of #ae1 from a central 8valley of elevation10 about three hundred yards in width4 ,a second narrow steep ridge1 only sixty feet high1 is uniclinal1 the strata throughout dipping westward2 those on both flanks being inclined at an angle of from ten to fifteen degrees2 whilst those on the ridge dip in the same direction at an angle of between thirty and forty degrees4 ,this ridge1 traced northwards1 dies away2 and the beds at its terminal point1 instead of dipping westward1 are inclined #ab to the north4 ,this case interested me1 as being the first in which ,i found in ,south ,america1 formations perhaps of tertiary origin1 broken by lines of elevation4 ,valdivia3 ,island of ,mocha4--,the formations of ,chiloe seem to extend with nearly the same character to ,valdivia1 and for some leagues northward of it3 the underlying rocks are micaceous schists1 and are covered up with sandstone and other sedimentary beds1 including1 as ,i was assured1 in many places layers of lignite4 ,i did not land on ,mocha 7lat4 #ch #bj71 but ,mr4 ,stokes brought me specimens of the grey1 fine-grained1 slightly calcareous sandstone1 precisely like that of ,huafo1 containing lignite and numerous ,turritell4 ,the island is flat topped1 #a1bdj feet in height1 and appears like an outlier of the sedimentary beds on the mainland4 ,the few shells collected consist of3-- ,turritella ,chilensis1 ,g4 ,b4 ,sowerby 7also at ,huafo74 ,fusus1 very imperfect1 somewhat resembling ,f4 subreflexus of ,navidad1 but probably different4 ,venus1 fragments of4 ,concepcion4--,sailing northward from ,valdivia1 the coast-cliffs are seen1 first to assume near the ,r4 ,tolten1 and thence for #aej miles northward1 to be continued with the same mineralogical characters1 immediately to be described at ,concepcion4 ,i heard in many places of beds of lignite1 some of it fine and glossy1 and likewise of silicified wood2 near the ,tolten the cliffs are low1 but they soon rise in height2 and the horizontal strata are prolonged1 with a nearly level surface1 until coming to a more lofty tract between points ,rumena and ,lavapie4 ,here the beds have been broken up by at least eight or nine parallel lines of elevation1 ranging ,e4 or ,e4,n4,e4 and ,w4 or ,w4,s4,w4 ,these lines can be followed with the eye many miles into the interior2 they are all uniclinal1 the strata in each dipping to a point between ,s4 and ,s4,s4,e4 with an inclination in the central lines of about forty degrees1 and in the outer ones of under twenty degrees4 ,this band of symmetrically troubled country is about eight miles in width4 ,the island of ,quiriquina1 in the ,bay of ,concepcion1 is formed of various soft and often ferruginous sandstones1 with bands of pebbles1 and with the lower strata sometimes passing into a conglomerate resting on the underlying metamorphic schists4 ,these beds include subordinate layers of greenish impure clay1 soft micaceous and calcareous sandstones1 and reddish friable earthy matter with white specks like decomposed crystals of feldspar2 they include1 also1 hard concretions1 fragments of shells1 lignite1 and silicified wood4 ,in the upper part they pass into white1 soft sediments and brecciolas1 very like those described at ,chiloe2 as indeed is the whole formation4 ,at ,lirguen and other places on the eastern side of the bay1 there are good sections of the lower sandstones1 which are generally ferruginous1 but which vary in character1 and even pass into an argillaceous nature2 they contain hard concretions1 fragments of lignite1 silicified wood1 and pebbles 7of the same rocks with the pebbles in the sandstones of ,quiriquina71 and they alternate with numerous1 often very thin layers of imperfect coal1 generally of little specific gravity4 ,the main bed here is three feet thick2 and only the coal of this one bed has a glossy fracture4 ,another irregular1 curvilinear bed of brown1 compact lignite1 is remarkable for being included in a mass of coarse gravel4 ,these imperfect coals1 when placed in a heap1 ignite spontaneously4 ,the cliffs on this side of the bay1 as well as on the island of ,quiriquina1 are capped with red friable earth1 which1 as stated in the ,second ,chapter1 is of recent formation4 ,the stratification in this neighbourhood is generally horizontal2 but near ,lirguen the beds dip ,n4,w4 at an angle of #bc2 near ,concepcion they are also inclined3 at the northern end of ,quiriquina they have been tilted at an angle of #cj1 and at the southern end at angles varying from #ae to #dj3 these dislocations must have taken place under the sea4 ,a collection of shells1 from the island of ,quiriquina1 has been described by ,m4 d',orbigny3 they are all extinct1 and from their generic character1 ,m4 d',orbigny inferred that they were of tertiary origin3 they consist of3-- ,scalaria ,chilensis1 d',orbigny1 8,voyage1 ,part ,pal40 ,natica ,araucana1 d',orbigny1 8,voyage1 ,part ,pal40 ,natica australis1 d',orbigny1 8,voyage1 ,part ,pal40 ,fusus difficilis1 d',orbigny1 8,voyage1 ,part ,pal40 ,pyrula longirostra1 d',orbigny1 8,voyage1 ,part ,pal40 ,pleurotoma ,araucana1 d',orbigny1 8,voyage1 ,part ,pal40 ,cardium auca1 d',orbigny1 8,voyage1 ,part ,pal40 ,cardium acuticostatum1 d',orbigny1 8,voyage1 ,part ,pal40 ,venus auca1 d',orbigny1 8,voyage1 ,part ,pal40 ,mactra cecileana1 d',orbigny1 8,voyage1 ,part ,pal40 ,mactra ,araucana1 d',orbigny1 8,voyage1 ,part ,pal40 ,arca ,araucana1 d',orbigny1 8,voyage1 ,part ,pal40 ,nucula ,largillierti1 d',orbigny1 8,voyage1 ,part ,pal40 ,trigonia ,hanetiana1 d',orbigny1 8,voyage1 ,part ,pal40 ,during a second visit of the ,beagle to ,concepcion1 ,mr4 ,kent collected for me some silicified wood and shells out of the concretions in the sandstone from ,tome1 situated a short distance north of ,lirguen4 ,they consist of3-- ,natica australis1 d',orbigny1 8,voyage1 ,part ,pal40 ,mactra ,araucana1 d',orbigny1 8,voyage1 ,part ,pal40 ,trigonia ,hanetiana1 d',orbigny1 8,voyage1 ,part ,pal40 ,pecten1 fragments of1 probably two species1 but too imperfect for description4 ,baculites vagina1 ,e4 ,forbes4 ,nautilus d',orbignyanus1 ,e4 ,forbes4 ,besides these shells1 ,captain ,belcher#aj found here an ,ammonite1 nearly three feet in diameter1 and so heavy that he could not bring it away2 fragments are deposited at ,haslar ,hospital3 he also found the silicified vertebr of some very large animal4 ,from the identity in mineralogical nature of the rocks1 and from ,captain ,belcher's minute description of the coast between ,lirguen and ,tome1 the fossiliferous concretions at this latter place certainly belong to the same formation with the beds examined by myself at ,lirguen2 and these again are undoubtedly the same with the strata of ,quiriquina2 moreover2 the three first of the shells from ,tome1 though associated in the same concretions with the ,baculite1 are identical with the species from ,quiriquina4 ,hence all the sandstone and lignitiferous beds in this neighbourhood certainly belong to the same formation4 ,although the generic character of the ,quiriquina fossils naturally led ,m4 d',orbigny to conceive that they were of tertiary origin1 yet as we now find them associated with the ,baculites vagina and with an ,ammonite1 we must1 in the opinion of ,m4 d',orbigny1 and if we are guided by the analogy of the northern hemisphere1 rank them in the ,cretaceous system4 ,moreover1 the ,baculites vagina1 which is in a tolerable state of preservation1 appears to ,professor ,e4 ,forbes certainly to be identical with a species1 so named by him1 from ,pondicherry in ,india2 where it is associated with numerous decidedly cretaceous species1 which approach most nearly to ,lower ,greensand or ,neocomian forms3 this fact1 considering the vast distance between ,chile and ,india1 is truly surprising4 ,again1 the ,nautilus d',orbignyanus1 as far as its imperfect state allows of comparison1 resembles1 as ,i am informed by ,professor ,forbes1 both in its general form and in that of its chambers1 two species from the ,upper ,greensand4 ,it may be added that every one of the above-named genera from ,quiriquina1 which have an apparently tertiary character1 are found in the ,pondicherry strata4 ,there are1 however1 some difficulties on this view of the formations at ,concepcion being cretaceous1 which ,i shall afterwards allude to2 and ,i will here only state that the ,cardium auca is found also at ,coquimbo1 the beds at which place1 there can be no doubt1 are tertiary4 #aj 8,zoology of ,captain ,beechey's ,voyage10 p4 #afc4 ,navidad4#aa--,the ,concepcion formation extends some distance northward1 but how far ,i know not2 for the next point at which ,i landed was at ,navidad1 #afj miles north of ,concepcion1 and #fj miles south of ,valparaiso4 ,the cliffs here are about eight hundred feet in height3 they consist1 wherever ,i could examine them1 of fine-grained1 yellowish1 earthy sandstones1 with ferruginous veins1 and with concretions of hard calcareous sandstone4 ,in one part1 there were many pebbles of the common metamorphic porphyries of the ,cordillera3 and near the base of the cliff1 ,i observed a single rounded boulder of greenstone1 nearly a yard in diameter4 ,i traced this sandstone formation beneath the superficial covering of gravel1 for some distance inland3 the strata are slightly inclined from the sea towards the ,cordillera1 which apparently has been caused by their having been accumulated against or round outlying masses of granite1 of which some points project near the coast4 ,the sandstone contains fragments of wood1 either in the state of lignite or partially silicified1 sharks' teeth1 and shells in great abundance1 both high up and low down the sea-cliffs4 ,pectunculus and ,oliva were most numerous in individuals1 and next to them ,turritella and ,fusus4 ,i collected in a short time1 though suffering from illness1 the following thirty-one species1 all of which are extinct1 and several of the genera do not now range 7as we shall hereafter show7 nearly so far south3-- ,gastridium cepa1 ,g4 ,b4 ,sowerby4 ,monoceros1 fragments of1 considered by ,m4 d',orbigny as a new species4 ,voluta alta1 ,g4 ,b4 ,sowerby 7considered by ,m4 d',orbigny as distinct from the ,v4 alta of ,santa ,cruz74 ,voluta triplicata1 ,g4 ,b4 ,sowerby4 ,oliva dimidiata1 ,g4 ,b4 ,sowerby4 ,pleurotoma discors1 ,g4 ,b4 ,sowerby4 ,pleurotoma turbinelloides1 ,g4 ,b4 ,sowerby4 ,fusus subreflexus1 ,g4 ,b4 ,sowerby4 ,fusus pyruliformis1 ,g4 ,b4 ,sowerby4 ,fusus1 allied to ,f4 regularis 7considered by ,m4 d',orbigny as a distinct species74 ,turritella suturalis1 ,g4 ,b4 ,sowerby4 ,turritella ,patagonica1 ,g4 ,b4 ,sowerby 7fragments of74 ,trochus lvis1 ,g4 ,b4 ,sowerby4 ,trochus collaris1 ,g4 ,b4 ,sowerby 7considered by ,m4 d',orbigny as the young of the ,t4 lvis74 ,cassis monilifer1 ,g4 ,b4 ,sowerby4 ,pyrula distans1 ,g4 ,b4 ,sowerby4 ,triton verruculosus1 ,g4 ,b4 ,sowerby4 ,sigaretus subglobosus1 ,g4 ,b4 ,sowerby4 ,natica solida1 ,g4 ,b4 ,sowerby4 7,it is doubtful whether the ,natica solida of ,s4 ,cruz is the same species with this47 ,terebra undulifera1 ,g4 ,b4 ,sowerby4 ,terebra costellata1 ,g4 ,b4 ,sowerby4 ,bulla 7fragments of74 ,dentalium giganteum1 do4 ,dentalium sulcosum1 do4 ,corbis 787 lvigata1 do4 ,cardium multiradiatum1 do4 ,venus meridionalis1 do4 ,pectunculus dispar1 787 ,desh4 7considered by ,m4 d',orbigny as a distinct species74 and #cj4 ,cythera and ,mactra1 fragments of 7considered by ,m4 d',orbigny as new species74 ,pecten1 fragments of4 #aa ,i was guided to this locality by the ,report on ,m4 ,gay's 8,geological ,researches10 in the 8,annales des ,scienc4 ,nat40 7#ast series1 tome #bh4 ,coquimbo4--,for more than two hundred miles northward of ,navidad1 the coast consists of plutonic and metamorphic rocks1 with the exception of some quite insignificant superficial beds of recent origin4 ,at ,tonguay1 twenty-five miles south of ,coquimbo1 tertiary beds recommence4 ,i have already minutely described in the ,second ,chapter1 the step-formed plains of ,coquimbo1 and the upper calcareous beds 7from twenty to thirty feet in thickness7 containing shells of recent species1 but in different proportions from those on the beach4 ,there remains to be described only the underlying ancient tertiary beds1 represented in ,figure #ba by the letters ,f and ,e3-- ,no4 #ba ,section of the tertiary formation at ,coquimbo4 ,illustration3 ,section of the tertiary formation at ,coquimbo4 ,i obtained good sections of bed 7,f7 only in ,herradura ,bay3 it consists of soft whitish sandstone1 with ferruginous veins1 some pebbles of granite1 and concretionary layers of hard calcareous sandstone4 ,these concretions are remarkable from the great number of large silicified bones1 apparently of cetaceous animals1 which they contain2 and likewise of a shark's teeth1 closely resembling those of the ,carcharias megalodon4 ,shells of the following species1 of which the gigantic ,oyster and ,perna are the most conspicuous1 are numerously embedded in the concretions3-- ,bulla ambigua1 d',orbigny1 8,voyage10 ,pal4 ,monoceros ,blainvillii1 d',orbigny1 8,voyage10 ,pal4 ,cardium auca1 d',orbigny1 8,voyage10 ,pal4 ,panopa ,coquimbensis1 d',orbigny1 8,voyage10 ,pal4 ,perna ,gaudichaudi1 d',orbigny1 8,voyage10 ,pal4 ,artemis ponderosa2 ,mr4 ,sowerby can find no distinguishing character between this fossil and the recent ,a4 ponderosa2 it is certainly an ,artemis1 as shown by the pallial impression4 ,ostrea ,patagonica 7872 ,mr4 ,sowerby can point out no distinguishing character between this species and that so eminently characteristic of the great ,patagonian formation2 but he will not pretend to affirm that they are identical4 ,fragments of a ,venus and ,natica4 ,the cliffs on one side of ,herradura ,bay are capped by a mass of stratified shingle1 containing a little calcareous matter1 and ,i did not doubt that it belonged to the same recent formation with the gravel on the surrounding plains1 also cemented by calcareous matter1 until to my surprise1 ,i found in the midst of it1 a single thin layer almost entirely composed of the above gigantic oyster4 ,at a little distance inland1 ,i obtained several sections of the bed 7,e71 which1 though different in appearance from the lower bed 7,f71 belongs to the same formation3 it consists of a highly ferruginous sandy mass1 almost composed1 like the lowest bed at ,port ,s4 ,julian1 of fragments of ,balanid2 it includes some pebbles1 and layers of yellowish-brown mudstone4 ,the embedded shells consist of3-- ,monoceros ,blainvillii1 d',orbigny1 8,voyage0 ,pal4 ,monoceros ambiguus1 ,g4 ,b4 ,sowerby4 ,anomia alternans1 ,g4 ,b4 ,sowerby4 ,pecten rudis1 ,g4 ,b4 ,sowerby4 ,perna ,gaudichaudi1 d',orbigny1 8,voyage0 ,pal4 ,ostrea ,patagonica 7871 d',orbigny1 8,voyage0 ,pal4 ,ostrea1 small species1 in imperfect state2 it appeared to me like a small kind now living in1 but very rare in the bay4 ,mytilus ,chiloensis2 ,mr4 ,sowerby can find no distinguishing character between this fossil1 as far as its not very perfect condition allows of comparison1 and the recent species4 ,balanus ,coquimbensis1 ,g4 ,b4 ,sowerby4 ,balanus psittacus8 ,king4 ,this appears to ,mr4 ,sowerby and myself identical with a very large and common species now living on the coast4 ,the uppermost layers of this ferrugino-sandy mass are conformably covered by1 and impregnated to the depth of several inches with1 the calcareous matter of the bed 7,d7 called losa3 hence ,i at one time imagined that there was a gradual passage between them2 but as all the species are recent in the bed 7,d71 whilst the most characteristic shells of the uppermost layers of 7,e7 are the extinct ,perna1 ,pecten1 and ,monoceros1 ,i agree with ,m4 d',orbigny1 that this view is erroneous1 and that there is only a mineralogical passage between them1 and no gradual transition in the nature of their organic remains4 ,besides the fourteen species enumerated from these two lower beds1 ,m4 d',orbigny has described ten other species given to him from this locality2 namely3-- ,fusus ,cleryanus1 d',orbigny1 8,voyage0 ,pal4 ,fusus petitianus1 d',orbigny1 8,voyage0 ,pal4 ,venus hanetiana1 d',orbigny1 8,voyage0 ,pal4 ,venus incerta 787 d',orbigny1 8,voyage0 ,pal4 ,venus ,cleryana1 d',orbigny1 8,voyage0 ,pal4 ,venus petitiana1 d',orbigny1 8,voyage0 ,pal4 ,venus ,chilensis1 d',orbigny1 8,voyage0 ,pal4 ,solecurtus hanetianus1 d',orbigny1 8,voyage0 ,pal4 ,mactra auca1 d',orbigny1 8,voyage0 ,pal4 ,oliva serena1 d',orbigny1 8,voyage0 ,pal4 ,of these twenty-four shells1 all are extinct1 except1 according to ,mr4 ,sowerby1 the ,artemis ponderosa1 ,mytilus ,chiloensis1 and probably the great ,balanus4 ,coquimbo to ,copiapo4--,a few miles north of ,coquimbo1 ,i met with the ferruginous1 balaniferous mass 7,e7 with many silicified bones2 ,i was informed that these silicified bones occur also at ,tonguay1 south of ,coquimbo3 their number is certainly remarkable1 and they seem to take the place of the silicified wood1 so common on the coast-formations of ,southern ,chile4 ,in the valley of ,chaeral1 ,i again saw this same formation1 capped with the recent calcareous beds4 ,i here left the coast1 and did not see any more of the tertiary formations1 until descending to the sea at ,copiapo3 here in one place ,i found variously coloured layers of sand and soft sandstone1 with seams of gypsum1 and in another place1 a comminuted shelly mass1 with layers of rotten-stone and seams of gypsum1 including many of the extinct gigantic oyster3 beds with these oysters are said to occur at ,english ,harbour1 a few miles north of ,copiapo4 ,coast of ,peru4--,with the exception of deposits containing recent shells and of quite insignificant dimensions1 no tertiary formations have been observed on this coast1 for a space of twenty-two degrees of latitude north of ,copiapo1 until coming to ,payta1 where there is said to be a considerable calcareous deposit3 a few fossils have been described by ,m4 d',orbigny from this place1 namely3-- ,rostellaria ,gaudichaudi1 d',orbigny1 8,voyage0 ,pal4 ,pectunculus ,paytensis1 d',orbigny1 8,voyage0 ,pal4 ,venus petitiana1 d',orbigny1 8,voyage0 ,pal4 ,ostrea ,patagonica8 ,this great oyster 7of which specimens have been given me7 cannot be distinguished by ,mr4 ,sowerby from some of the varieties from ,patagonia2 though it would be hazardous to assert it is the same with that species1 or with that from ,coquimbo4 ,concluding ,remarks4--,the formations described in this chapter1 have1 in the case of ,chiloe and probably in that of ,concepcion and ,navidad1 apparently been accumulated in troughs formed by submarine ridges extending parallel to the ancient shores of the continent2 in the case of the islands of ,mocha and ,huafo it is highly probable1 and in that of ,ypun and ,lemus almost certain1 that they were accumulated round isolated rocky centres or nuclei1 in the same manner as mud and sand are now collecting round the outlying islets and reefs in the ,west ,indian ,archipelago4 ,hence1 ,i may remark1 it does not follow that the outlying tertiary masses of ,mocha and ,huafo were ever continuously united at the same level with the formations on the mainland1 though they may have been of contemporaneous origin1 and been subsequently upraised to the same height4 ,in the more northern parts of ,chile1 the tertiary strata seem to have been separately accumulated in bays1 now forming the mouths of valleys4 ,the relation between these several deposits on the shores of the ,pacific1 is not nearly so clear as in the case of the tertiary formations on the ,atlantic4 ,judging from the form and height of the land 7evidence which ,i feel sure is here much more trustworthy than it can ever be in such broken continents as that of ,europe71 from the identity of mineralogical composition1 from the presence of fragments of lignite and of silicified wood1 and from the intercalated layers of imperfect coal1 ,i must believe that the coast-formations from ,central ,chiloe to ,concepcion1 a distance of #djj miles1 are of the same age3 from nearly similar reasons1 ,i suspect that the beds of ,mocha1 ,huafo1 and ,ypun1 belong also to the same period4 ,the commonest shell in ,mocha and ,huafo is the same species of ,turritella2 and ,i believe the same ,cythera is found on the islands of ,huafo1 ,chiloe1 and ,ypun2 but with these trifling exceptions1 the few organic remains found at these places are distinct4 ,the numerous shells from ,navidad1 with the exception of two1 namely1 the ,sigaretus and ,turritella found at ,ypun1 are likewise distinct from those found in any other part of this coast4 ,coquimbo has ,cardium auca in common with ,concepcion1 and ,fusus ,cleryanus with ,huafo2 ,i may add1 that ,coquimbo has ,venus petitiana1 and a gigantic oyster 7said by ,m4 d',orbigny also to be found a little south of ,concepcion7 in common with ,payta1 though this latter place is situated twenty-two degrees northward of lat4 #bg1 to which point the ,coquimbo formation extends4 ,from these facts1 and from the generic resemblance of the fossils from the different localities1 ,i cannot avoid the suspicion that they all belong to nearly the same epoch1 which epoch1 as we shall immediately see1 must be a very ancient tertiary one4 ,but as the ,baculite1 especially considering its apparent identity with the ,cretaceous ,pondicherry species1 and the presence of an ,ammonite1 and the resemblance of the ,nautilus to two upper greensand species1 together afford very strong evidence that the formation of ,concepcion is a ,secondary one2 ,i will1 in my remarks on the fossils from the other localities1 put on one side those from ,concepcion and from ,eastern ,chiloe1 which1 whatever their age may be1 appear to me to belong to one group4 ,i must1 however1 again call attention to the fact that the ,cardium auca is found both at ,concepcion and in the undoubtedly tertiary strata of ,coquimbo3 nor should the possibility be overlooked1 that as ,trigonia1 though known in the northern hemisphere only as a ,secondary genus1 has living representatives in the ,australian seas1 so a ,baculite1 ,ammonite1 and ,trigonia may have survived in this remote part of the southern ocean to a somewhat later period than to the north of the equator4 ,before passing in review the fossils from the other localities1 there are two points1 with respect to the formations between ,concepcion and ,chiloe1 which deserve some notice4 ,first1 that though the strata are generally horizontal1 they have been upheaved in ,chiloe in a set of parallel anticlinal and uniclinal lines ranging north and south1--in the district near ,p4 ,rumena by eight or nine far-extended1 most symmetrical1 uniclinal lines ranging nearly east and west1--and in the neighbourhood of ,concepcion by less regular single lines1 directed both ,n4,e4 and ,s4,w41 and ,n4,w4 and ,s4,e4 ,this fact is of some interest1 as showing that within a period which cannot be considered as very ancient in relation to the history of the continent1 the strata between the ,cordillera and the ,pacific have been broken up in the same variously directed manner as have the old plutonic and metamorphic rocks in this same district4 ,the second point is1 that the sandstone between ,concepcion and ,southern ,chiloe is everywhere lignitiferous1 and includes much silicified wood2 whereas the formations in ,northern ,chile do not include beds of lignite or coal1 and in place of the fragments of silicified wood there are silicified bones4 ,now1 at the present day1 from ,cape ,horn to near ,concepcion1 the land is entirely concealed by forests1 which thin out at ,concepcion1 and in ,central and ,northern ,chile entirely disappear4 ,this coincidence in the distribution of the fossil wood and the living forests may be quite accidental2 but ,i incline to take a different view of it2 for1 as the difference in climate1 on which the presence of forests depends1 is here obviously in chief part due to the form of the land1 and as the ,cordillera undoubtedly existed when the lignitiferous beds were accumulating1 ,i conceive it is not improbable that the climate1 during the lignitiferous period1 varied on different parts of the coast in a somewhat similar manner as it now does4 ,looking to an earlier epoch1 when the strata of the ,cordillera were depositing1 there were islands which even in the latitude of ,northern ,chile1 where now all is irreclaimably desert1 supported large coniferous forests4 ,seventy-nine species of fossil shells1 in a tolerably recognisable condition1 from the coast of ,chile and ,peru1 are described in this volume1 and in the ,palontological part of ,m4 d',orbigny's 8,voyage03 if we put on one side the twenty species exclusively found at ,concepcion and ,chiloe1 fifty-nine species from ,navidad and the other specified localities remain4 ,of these fifty-nine species only an ,artemis1 a ,mytilus and ,balanus1 all from ,coquimbo1 are 7in the opinion of ,mr4 ,sowerby1 but not in that of ,m4 d',orbigny7 identical with living shells2 and it would certainly require a better series of specimens to render this conclusion certain4 ,only the ,turritella ,chilensis from ,huafo and ,mocha1 the ,t4 ,patagonica and ,venus meridionalis from ,navidad1 come very near to recent ,south ,american shells1 namely1 the two ,turritellas to ,t4 cingulata1 and the ,venus to ,v4 exalbida3 some few other species come rather less near2 and some few resemble forms in the older ,european tertiary deposits3 none of the species resemble secondary forms4 ,hence ,i conceive there can be no doubt that these formations are tertiary1--a point necessary to consider1 after the case of ,concepcion4 ,the fifty-nine species belong to thirty-two genera2 of these1 ,gastridium is extinct1 and three or four of the genera 7viz4 ,panopa1 ,rostellaria1 ,corbis 7871 and ,i believe ,solecurtus7 are not now found on the west coast of ,south ,america4 ,fifteen of the genera have on this coast living representatives in about the same latitudes with the fossil species2 but twelve genera now range very differently to what they formerly did4 ,the idea of the table on the following page1 in which the difference between the extension in latitude of the fossil and existing species is shown1 is taken from ,m4 d',orbigny's work2 but the range of the living shells is given on the authority of ,mr4 ,cuming1 whose long-continued researches on the conchology of ,south ,america are well-known4 ,when we consider that very few1 if any1 of the fifty-nine fossil shells are identical with1 or make any close approach to1 living species2 when we consider that some of the genera do not now exist on the west coast of ,south ,america1 and that no less than twelve genera out of the thirty-two formerly ranged very differently from the existing species of the same genera1 we must admit that these deposits are of considerable antiquity1 and that they probably verge on the commencement of the tertiary era4 ,may we not venture to believe1 that they are of nearly contemporaneous origin with the ,eocene formations of the northern hemisphere8 ,genera1 with living and tertiary species on the west coast of ,s4 ,america4#ab ,latitudes1 in which found fossil on the coasts of ,chile and ,peru4 ,southernmost latitude1 in which found living on the west coast of ,s4 ,america4 ,bulla #cj to #dc #cj #ab near ,lima4 ,cassis #cd #a #cg ,pyrula #cd 7and #cf #cj at ,concepcion7 #e ,payta ,fusus #cj to #dc #cj #bc ,mexillones2 reappears at the ,st4 of ,magellan ,pleurotoma #cd to #dc #cj #b #ah ,st4 ,elena ,terebra #cd #e ,payta ,sigaretus #cd to #dd #cj #ab ,lima ,anomia #cj #g #dh ,perna #cj #a #bc ,xixappa ,cardium #cj to #cd 7and #cf #cj at ,concepcion7 #e ,payta ,artemis #cj #e ,payta ,voluta #cd to #dd #cj ,mr4 ,cuming does not know of any species living on the west coast1 between the equator and lat4 #dc south2 from this latitude a species is found as far south as ,tierra del ,fuego4 #ab ,m4 d',orbigny states that the genus ,natica is not found on the coast of ,chile2 but ,mr4 ,cuming found it at ,valparaiso4 ,scalaria was found at ,valparaiso2 ,arca1 at ,iquique1 in lat4 #bj1 by ,mr4 ,cuming2 ,arca1 also1 was found by ,captain ,king1 at ,juan ,fernandez1 in lat4 #cc #cj4 ,comparing the fossil remains from the coast of ,chile 7leaving out1 as before1 ,concepcion and ,chiloe7 with those from ,patagonia1 we may conclude1 from their generic resemblance1 and from the small number of the species which from either coast approach closely to living forms1 that the formations of both belong to nearly the same epoch2 and this is the opinion of ,m4 ,d',orbigny4 ,had not a single fossil shell been common to the two coasts1 it could not have been argued that the formations belonged to different ages2 for ,messrs4 ,cuming and ,hinds have found1 on the comparison of nearly two thousand living species from the opposite sides of ,south ,america1 only one in common1 namely1 the ,purpura lapillus from both sides of the ,isthmus of ,panama3 even the shells collected by myself amongst the ,chonos ,islands and on the coast of ,patagonia1 are dissimilar1 and we must descend to the apex of the continent1 to ,tierra del ,fuego1 to find these two great conchological provinces united into one4 ,hence it is remarkable that four or five of the fossil shells from ,navidad1 namely1 ,voluta alta1 ,turritella ,patagonica1 ,trochus collaris1 ,venus meridionalis1 perhaps 7,natica solida71 and perhaps the large oyster from ,coquimbo1 are considered by ,mr4 ,sowerby as identical with species from ,santa ,cruz and ,p4 ,desire4 ,m4 d',orbigny1 however1 admits the perfect identity only of the ,trochus4 ,on the temperature of the ,tertiary period4--,as the number of the fossil species and genera from the western and eastern coasts is considerable1 it will be interesting to consider the probable nature of the climate under which they lived4 ,we will first take the case of ,navidad1 in lat4 #cd1 where thirty-one species were collected1 and which1 as we shall presently see1 must have inhabited shallow water1 and therefore will necessarily well exhibit the effects of temperature4 ,referring to the table given in the previous page1 we find that the existing species of the genera ,cassis1 ,pyrula1 ,pleurotoma1 ,terebra1 and ,sigaretus1 which are generally 7though by no means invariably7 characteristic of warmer latitudes1 do not at the present day range nearly so far south on this line of coast as the fossil species formerly did4 ,including ,coquimbo1 we have ,perna in the same predicament4 ,the first impression from this fact is1 that the climate must formerly have been warmer than it now is2 but we must be very cautious in admitting this1 for ,cardium1 ,bulla1 and ,fusus 7and1 if we include ,coquimbo1 ,anomia and ,artemis7 likewise formerly ranged farther south than they now do2 and as these genera are far from being characteristic of hot climates1 their former greater southern range may well have been owing to causes quite distinct from climate3 ,voluta1 again1 though generally so tropical a genus1 is at present confined on the west coast to colder or more southern latitudes than it was during the tertiary period4 ,the ,trochus collaris1 moreover1 and1 as we have just seen according to ,mr4 ,sowerby1 two or three other species1 formerly ranged from ,navidad as far south as ,santa ,cruz in latitude #ej degrees4 ,if1 instead of comparing the fossils of ,navidad1 as we have hitherto done1 with the shells now living on the west coast of ,south ,america1 we compare them with those found in other parts of the world1 under nearly similar latitudes2 for instance1 in the southern parts of the ,mediterranean or of ,australia1 there is no evidence that the sea off ,navidad was formerly hotter than what might have been expected from its latitude1 even if it was somewhat warmer than it now is when cooled by the great southern polar current4 ,several of the most tropical genera have no representative fossils at ,navidad2 and there are only single species of ,cassis1 ,pyrula1 and ,sigaretus1 two of ,pleurotoma and two of ,terebra1 but none of these species are of conspicuous size4 ,in ,patagonia1 there is even still less evidence in the character of the fossils1 of the climate having been formerly warmer4#ac ,as from the various reasons already assigned1 there can be little doubt that the formations of ,patagonia and at least of ,navidad and ,coquimbo in ,chile1 are the equivalents of an ancient stage in the tertiary formations of the northern hemisphere1 the conclusion that the climate of the southern seas at this period was not hotter than what might have been expected from the latitude of each place1 appears to me highly important2 for we must believe1 in accordance with the views of ,mr4 ,lyell1 that the causes which gave to the older tertiary productions of the quite temperate zones of ,europe a tropical character1 were of a local character and did not affect the entire globe4 ,on the other hand1 ,i have endeavoured to show1 in the 8,geological ,transactions10 that1 at a much later period1 ,europe and ,north and ,south ,america were nearly contemporaneously subjected to ice-action1 and consequently to a colder1 or at least more equable1 climate than that now characteristic of the same latitudes4 #ac ,it may be worth while to mention that the shells living at the present day on this eastern side of ,south ,america1 in lat4 #dj1 have perhaps a more tropical character than those in corresponding latitudes on the shores of ,europe3 for at ,bahia ,blanca and ,s4 ,blas1 there are two fine species of ,voluta and four of ,oliva4 ,on the absence of extensive modern conchiferous deposits in ,south ,america2 and on the contemporaneousness of the older ,tertiary deposits at distant points being due to contemporaneous movements of subsidence4--,knowing from the researches of ,professor ,e4 ,forbes1 that molluscous animals chiefly abound within a depth of #ajj fathoms and under1 and bearing in mind how many thousand miles of both coasts of ,south ,america have been upraised within the recent period by a slow1 long-continued1 intermittent movement1--seeing the diversity in nature of the shores and the number of shells now living on them1--seeing also that the sea off ,patagonia and off many parts of ,chile1 was during the tertiary period highly favourable to the accumulation of sediment1--the absence of extensive deposits including recent shells over these vast spaces of coast is highly remarkable4 ,the conchiferous calcareous beds at ,coquimbo1 and at a few isolated points northward1 offer the most marked exception to this statement2 for these beds are from twenty to thirty feet in thickness1 and they stretch for some miles along shore1 attaining1 however1 only a very trifling breadth4 ,at ,valdivia there is some sandstone with imperfect casts of shells1 which possibly may belong to the recent period3 parts of the boulder formation and the shingle-beds on the lower plains of ,patagonia probably belong to this same period1 but neither are fossiliferous3 it also so happens that the great ,pampean formation does not include1 with the exception of the ,azara1 any mollusca4 ,there cannot be the smallest doubt that the upraised shells along the shores of the ,atlantic and ,pacific1 whether lying on the bare surface1 or embedded in mould or in sand-hillocks1 will in the course of ages be destroyed by alluvial action3 this probably will be the case even with the calcareous beds of ,coquimbo1 so liable to dissolution by rain-water4 ,if we take into consideration the probability of oscillations of level and the consequent action of the tidal-waves at different heights1 their destruction will appear almost certain4 ,looking to an epoch as far distant in futurity as we now are from the past ,miocene period1 there seems to me scarcely a chance1 under existing conditions1 of the numerous shells now living in those zones of depths most fertile in life1 and found exclusively on the western and south-eastern coasts of ,s4 ,america1 being preserved to this imaginary distant epoch4 ,a whole conchological series will in time be swept away1 with no memorials of their existence preserved in the earth's crust4 ,can any light be thrown on this remarkable absence of recent conchiferous deposits on these coasts1 on which1 at an ancient tertiary epoch1 strata abounding with organic remains were extensively accumulated8 ,i think there can1 namely1 by considering the conditions necessary for the preservation of a formation to a distant age4 ,looking to the enormous amount of denudation which on all sides of us has been effected1--as evidenced by the lofty cliffs cutting off on so many coasts horizontal and once far-extended strata of no great antiquity 7as in the case of ,patagonia71--as evidenced by the level surface of the ground on both sides of great faults and dislocations1--by inland lines of escarpments1 by outliers1 and numberless other facts1 and by that argument of high generality advanced by ,mr4 ,lyell1 namely1 that every sedimentary formation1 whatever its thickness may be1 and over however many hundred square miles it may extend1 is the result and the measure of an equal amount of wear and tear of pre-existing formations2 considering these facts1 we must conclude that1 as an ordinary rule1 a formation to resist such vast destroying powers1 and to last to a distant epoch1 must be of wide extent1 and either in itself1 or together with superincumbent strata1 be of great thickness4 ,in this discussion1 we are considering only formations containing the remains of marine animals1 which1 as before mentioned1 live1 with some exceptions within 7most of them much within7 depths of #ajj fathoms4 ,how1 then1 can a thick and widely extended formation be accumulated1 which shall include such organic remains8 ,first1 let us take the case of the bed of the sea long remaining at a stationary level3 under these circumstances it is evident that conchiferous strata can accumulate only to the same thickness with the depth at which the shells can live2 on gently inclined coasts alone can they accumulate to any considerable width2 and from the want of superincumbent pressure1 it is probable that the sedimentary matter will seldom be much consolidated3 such formations have no very good chance1 when in the course of time they are upraised1 of long resisting the powers of denudation4 ,the chance will be less if the submarine surface1 instead of having remained stationary1 shall have gone on slowly rising during the deposition of the strata1 for in this case their total thickness must be less1 and each part1 before being consolidated or thickly covered up by superincumbent matter1 will have had successively to pass through the ordeal of the beach2 and on most coasts1 the waves on the beach tend to wear down and disperse every object exposed to their action4 ,now1 both on the south-eastern and western shores of ,s4 ,america1 we have had clear proofs that the land has been slowly rising1 and in the long lines of lofty cliffs1 we have seen that the tendency of the sea is almost everywhere to eat into the land4 ,considering these facts1 it ceases1 ,i think1 to be surprising1 that extensive recent conchiferous deposits are entirely absent on the southern and western shores of ,america4 ,let us take the one remaining case1 of the bed of the sea slowly subsiding during a length of time1 whilst sediment has gone on being deposited4 ,it is evident that strata might thus accumulate to any thickness1 each stratum being deposited in shallow water1 and consequently abounding with those shells which cannot live at great depths3 the pressure1 also1 ,i may observe1 of each fresh bed would aid in consolidating all the lower ones4 ,even on a rather steep coast1 though such must ever be unfavourable to widely extended deposits1 the formations would always tend to increase in breadth from the water encroaching on the land4 ,hence we may admit that periods of slow subsidence will commonly be most favourable to the accumulation of conchiferous deposits1 of sufficient thickness1 extension1 and hardness1 to resist the average powers of denudation4 ,we have seen that at an ancient tertiary epoch1 fossiliferous deposits were extensively deposited on the coasts of ,s4 ,america2 and it is a very interesting fact1 that there is evidence that these ancient tertiary beds were deposited during a period of subsidence4 ,thus1 at ,navidad1 the strata are about eight hundred feet in thickness1 and the fossil shells are abundant both at the level of the sea and some way up the cliffs2 having sent a list of these fossils to ,professor ,e4 ,forbes1 he thinks they must have lived in water between one and ten fathoms in depth3 hence the bottom of the sea on which these shells once lived must have subsided at least #gjj feet to allow of the superincumbent matter being deposited4 ,i must here remark1 that1 as all these and the following fossil shells are extinct species1 ,professor ,forbes necessarily judges of the depths at which they lived only from their generic character1 and from the analogical distribution of shells in the northern hemisphere2 but there is no just cause from this to doubt the general results4 ,at ,huafo the strata are about the same thickness1 namely1 #hjj feet1 and ,professor ,forbes thinks the fossils found there cannot have lived at a greater depth than fifty fathoms1 or #cjj feet4 ,these two points1 namely1 ,navidad and ,huafo1 are #egj miles apart1 but nearly halfway between them lies ,mocha1 an island #a1bjj feet in height1 apparently formed of tertiary strata up to its level summit1 and with many shells1 including the same ,turritella with that found at ,huafo1 embedded close to the level of the sea4 ,in ,patagonia1 shells are numerous at ,santa ,cruz1 at the foot of the #cej feet plain1 which has certainly been formed by the denudation of the #hdj feet plain1 and therefore was originally covered by strata that number of feet in thickness1 and these shells1 according to ,professor ,forbes1 probably lived at a depth of between seven and fifteen fathoms3 at ,port ,s4 ,julian1 sixty miles to the north1 shells are numerous at the foot of the ninety feet plain 7formed by the denudation of the #iej feet plain71 and likewise occasionally at the height of several hundred feet in the upper strata2 these shells must have lived in water somewhere between five and fifty fathoms in depth4 ,although in other parts of ,patagonia ,i have no direct evidence of shoal-water shells having been buried under a great thickness of superincumbent submarine strata1 yet it should be borne in mind that the lower fossiliferous strata with several of the same species of ,mollusca1 the upper tufaceous beds1 and the high summit-plain1 stretch for a considerable distance southward1 and for hundreds of miles northward2 seeing this uniformity of structure1 ,i conceive it may be fairly concluded that the subsidence by which the shells at ,santa ,cruz and ,s4 ,julian were carried down and covered up1 was not confined to these two points1 but was co-extensive with a considerable portion of the ,patagonian tertiary formation4 ,in a succeeding chapter it will be seen1 that we are led to a similar conclusion with respect to the secondary fossiliferous strata of the ,cordillera1 namely1 that they also were deposited during a long-continued and great period of subsidence4 ,from the foregoing reasoning1 and from the facts just given1 ,i think we must admit the probability of the following proposition3 namely1 that when the bed of the sea is either stationary or rising1 circumstances are far less favourable1 than when the level is sinking1 to the accumulation of conchiferous deposits of sufficient thickness and extension to resist1 when upheaved1 the average vast amount of denudation4 ,this result appears to me1 in several respects1 very interesting3 every one is at first inclined to believe that at innumerable points1 wherever there is a supply of sediment1 fossiliferous strata are now forming1 which at some future distant epoch will be upheaved and preserved2 but on the views above given1 we must conclude that this is far from being the case2 on the contrary1 we require 7#ast71 a long-continued supply of sediment2 7#bnd71 an extensive shallow area2 and 7#crd71 that this area shall slowly subside to a great depth1 so as to admit the accumulation of a widely extended thick mass of superincumbent strata4 ,in how few parts of the world1 probably1 do these conditions at the present day concur6 ,we can thus1 also1 understand the general want of that close sequence in fossiliferous formations which we might theoretically have anticipated2 for1 without we suppose a subsiding movement to go on at the same spot during an enormous period1 from one geological era to another1 and during the whole of this period sediment to accumulate at the proper rate1 so that the depth should not become too great for the continued existence of molluscous animals1 it is scarcely possible that there should be a perfect sequence at the same spot in the fossil shells of the two geological formations4#ad ,so far from a very long-continued subsidence being probable1 many facts lead to the belief that the earth's surface oscillates up and down2 and we have seen that during the elevatory movements there is but a small chance of durable fossiliferous deposits accumulating4 #ad ,professor ,h4 ,d4 ,rogers1 in his excellent address to the ,association of ,american ,geologists 7,silliman's ,journal1 vol4 xlvii1 p4 #bgg7 makes the following remark3 8,i question if we are at all aware how completely the whole history of all departed time lies indelibly recorded with the amplest minuteness of detail in the successive sediments of the globe1 how effectually1 in other words1 every period of time has written its own history1 carefully preserving every created form and every trace of action40 ,i think the correctness of such remarks is more than doubtful1 even if we except 7as ,i suppose he would7 all those numerous organic forms which contain no hard parts47 ,lastly1 these same considerations appear to throw some light on the fact that certain periods appear to have been favourable to the deposition1 or at least to the preservation1 of contemporaneous formations at very distant points4 ,we have seen that in ,s4 ,america an enormous area has been rising within the recent period2 and in other quarters of the globe immense spaces appear to have risen contemporaneously4 ,from my examination of the coral-reefs of the great oceans1 ,i have been led to conclude that the bed of the sea has gone on slowly sinking within the present era1 over truly vast areas3 this1 indeed1 is in itself probable1 from the simple fact of the rising areas having been so large4 ,in ,south ,america we have distinct evidence that at nearly the same tertiary period1 the bed of the sea off parts of the coast of ,chile and off ,patagonia was sinking1 though these regions are very remote from each other4 ,if1 then1 it holds good1 as a general rule1 that in the same quarter of the globe the earth's crust tends to sink and rise contemporaneously over vast spaces1 we can at once see1 that we have at distant points1 at the same period1 those very conditions which appear to be requisite for the accumulation of fossiliferous masses of sufficient extension1 thickness1 and hardness1 to resist denudation1 and consequently to last unto an epoch distant in futurity4#ae #ae ,professor ,forbes has some admirable remarks on this subject1 in his 8,report on the ,shells of the gean ,sea40 ,in a letter to ,mr4 ,maclaren 7,edinburgh ,new ,phil4 ,journal1 ,january #ahdc71 ,i partially entered into this discussion1 and endeavoured to show that it was highly improbable1 that upraised atolls or barrier-reefs1 though of great thickness1 should1 owing to their small extension or breadth1 be preserved to a distant future period4 ,chapter ,,vi ,,plutonic ,,and ,,metamorphic ,,rocks3--,,cleavage ,,and ,,foliation4 ,brazil1 ,bahia1 gneiss with disjointed metamorphosed dikes4--,strike of foliation4--,rio de ,janeiro1 gneiss-granite1 embedded fragment in1 decomposition of4--,la ,plata1 metamorphic and old volcanic rocks of4--,s4 ,ventana4--,claystone porphyry formation of ,patagonia2 singular metamorphic rocks2 pseudo-dikes4--,falkland ,islands1 ,palozoic fossils of4--,tierra del ,fuego1 clay-slate formation1 cretaceous fossils of2 cleavage and foliation2 form of land4--,chonos ,archipelago1 mica-schists1 foliation disturbed by granitic axis2 dikes4--,chiloe4--,concepcion1 dikes1 successive formation of4--,central and ,northern ,chile4--,concluding remarks on cleavage and foliation4--,their close analogy and similar origin4 --,stratification of metamorphic schists4--,foliation of intrusive rocks4--,relation of cleavage and foliation to the lines of tension during metamorphosis4 ,the metamorphic and plutonic formations of the several districts visited by the ,beagle will be here chiefly treated of1 but only such cases as appear to me new1 or of some special interest1 will be described in detail2 at the end of the chapter ,i will sum up all the facts on cleavage and foliation1--to which ,i particularly attended4 ,bahia1 ,brazil3 lat4 #ac south4--,the prevailing rock is gneiss1 often passing1 by the disappearance of the quartz and mica1 and by the feldspar losing its red colour1 into a brilliantly grey primitive greenstone4 ,not unfrequently quartz and hornblende are arranged in layers in almost amorphous feldspar4 ,there is some fine-grained syenitic granite1 orbicularly marked by ferruginous lines1 and weathering into vertical1 cylindrical holes1 almost touching each other4 ,in the gneiss1 concretions of granular feldspar and others of garnets with mica occur4 ,the gneiss is traversed by numerous dikes composed of black1 finely crystallised1 hornblendic rock1 containing a little glassy feldspar and sometimes mica1 and varying in thickness from mere threads to ten feet3 these threads1 which are often curvilinear1 could sometimes be traced running into the larger dikes4 ,one of these dikes was remarkable from having been in two or three places laterally disjointed1 with unbroken gneiss interposed between the broken ends1 and in one part with a portion of the gneiss driven1 apparently whilst in a softened state1 into its side or wall4 ,in several neighbouring places1 the gneiss included angular1 well-defined1 sometimes bent1 masses of hornblende rock1 quite like1 except in being more perfectly crystallised1 that forming the dikes1 and1 at least in one instance1 containing 7as determined by ,professor ,miller7 augite as well as hornblende4 ,in one or two cases these angular masses1 though now quite separate from each other by the solid gneiss1 had1 from their exact correspondence in size and shape1 evidently once been united2 hence ,i cannot doubt that most or all of the fragments have been derived from the breaking up of the dikes1 of which we see the first stage in the above-mentioned laterally disjointed one4 ,the gneiss close to the fragments generally contained many large crystals of hornblende1 which are entirely absent or rare in other parts3 its folia or lamin were gently bent round the fragments1 in the same manner as they sometimes are round concretions4 ,hence the gneiss has certainly been softened1 its composition modified1 and its folia arranged1 subsequently to the breaking up of the dikes1#a these latter also having been at the same time bent and softened4 #a ,professor ,hitchcock 78,geology of ,massachusetts10 vol4 ii1 p4 #fgc1 gives a closely similar case of a greenstone dike in syenite4 ,i must here take the opportunity of premising1 that by the term cleavage ,i imply those planes of division which render a rock1 appearing to the eye quite or nearly homogeneous1 fissile4 ,by the term foliation1 ,i refer to the layers or plates of different mineralogical nature of which most metamorphic schists are composed2 there are1 also1 often included in such masses1 alternating1 homogeneous1 fissile layers or folia1 and in this case the rock is both foliated and has a cleavage4 ,by stratification1 as applied to these formations1 ,i mean those alternate1 parallel1 large masses of different composition1 which are themselves frequently either foliated or fissile1--such as the alternating so-called strata of mica-slate1 gneiss1 glossy clay-slate1 and marble4 ,the folia of the gneiss within a few miles round ,bahia generally strike irregularly1 and are often curvilinear1 dipping in all directions at various angles3 but where best defined1 they extended most frequently in a ,n4,e4 by ,n4 7or ,east #ej ,n47 and ,s4,w4 by ,s4 line1 corresponding nearly with the coast-line northwards of the bay4 ,i may add that ,mr4 ,gardner#b found in several parts of the province of ,ceara1 which lies between four and five hundred miles north of ,bahia1 gneiss with the folia extending ,e4 #de ,n42 and in ,guyana according to ,sir ,r4 ,schomburgk1 the same rock strikes ,e4 #eg ,n4 ,again1 ,humboldt describes the gneiss-granite over an immense area in ,venezuela and even in ,colombia1 as striking ,e4 #ej ,n41 and dipping to the ,n4,w4 at an angle of fifty degrees4 ,hence all the observations hitherto made tend to show that the gneissic rocks over the whole of this part of the continent have their folia extending generally within almost a point of the compass of the same direction4#c #b 8,geological ,section of the ,brit4 ,assoc410 #ahdj4 ,for ,sir ,r4 ,schomburgk's observations see ,geograph4 ,journal1 #ahdb1 p4 #aij4 ,see also ,humboldt's discussion on ,loxodrism in the 8,personal ,narrative40 #c ,i landed at only one place north of ,bahia1 namely1 at ,pernambuco4 ,i found there only soft1 horizontally stratified matter1 formed from disintegrated granitic rocks1 and some yellowish impure limestone1 probably of a tertiary epoch4 ,i have described a most singular natural bar of hard sandstone1 which protects the harbour1 in the #aith vol4 7#ahda7 p4 #beh of the ,london and ,edin4 ,phil4 ,magazine4 ,,abrolhos ,,islets1 lat4 #ah ,s4 off the coast of ,brazil4--,although not strictly in place1 ,i do not know where ,i can more conveniently describe this little group of small islands4 ,the lowest bed is a sandstone with ferruginous veins2 it weathers into an extraordinary honeycombed mass2 above it there is a dark-coloured argillaceous shale2 above this a coarser sandstone--making a total thickness of about sixty feet2 and lastly1 above these sedimentary beds1 there is a fine conformable mass of greenstone1 in some parts having a columnar structure4 ,all the strata1 as well as the surface of the land1 dip at an angle of about #ab to ,n4 by ,w4 ,some of the islets are composed entirely of the sedimentary1 others of the trappean rocks1 generally1 however1 with the sandstone1 cropping out on the southern shores4 ,rio de ,janeiro4--,this whole district is almost exclusively formed of gneiss1 abounding with garnets1 and porphyritic with large crystals1 even three and four inches in length1 of orthoclase feldspar3 in these crystals mica and garnets are often enclosed4 ,at the western base of the ,corcovado1 there is some ferruginous carious quartz-rock2 and in the ,tijeuka range1 much fine-grained granite4 ,i observed boulders of greenstone in several places2 and on the islet of ,villegagnon1 and likewise on the coast some miles northward1 two large trappean dikes4 ,the porphyritic gneiss1 or gneiss-granite as it has been called by ,humboldt1 is only so far foliated that the constituent minerals are arranged with a certain degree of regularity1 and may be said to have a 8grain10 but they are not separated into distinct folia or lamin4 ,there are1 however1 several other varieties of gneiss regularly foliated1 and alternating with each other in so-called strata4 ,the stratification and foliation of the ordinary gneisses1 and the foliation or 8grain0 of the gneiss-granite1 are parallel to each other1 and generally strike within a point of ,n4,e4 and ,s4,w4 dipping at a high angle 7between #ej and #fj7 generally to ,s4,e43 so that here again we meet with the strike so prevalent over the more northern parts of this continent4 ,the mountains of gneiss-granite are to a remarkable degree abruptly conical1 which seems caused by the rock tending to exfoliate in thick1 conically concentric layers3 these peaks resemble in shape those of phonolite and other injected rocks on volcanic islands2 nor is the grain or foliation 7as we shall afterwards see7 any difficulty on the idea of the gneiss-granite having been an intrusive rather than a metamorphic formation4 ,the lines of mountains1 but not always each separate hill1 range nearly in the same direction with the foliation and so-called stratification1 but rather more easterly4 ,no4 #bb ,fragment of gneiss embedded in another variety of the same rock4 ,illustration3 ,fragment of gneiss embedded in another variety of the same rock4 ,on a bare gently inclined surface of the porphyritic gneiss in ,botofogo ,bay1 ,i observed the appearance here represented4 ,a fragment seven yards long and two in width1 with angular and distinctly defined edges1 composed of a peculiar variety of gneiss with dark layers of mica and garnets1 is surrounded on all sides by the ordinary gneiss-granite2 both having been dislocated by a granitic vein4 ,the folia in the fragment and in the surrounding rock strike in the same ,n4,n4,e4 and ,s4,s4,w4 line2 but in the fragment they are vertical1 whereas in the gneiss-granite they dip at a small angle1 as shown by the arrows1 to ,s4,s4,e4 ,this fragment1 considering its great size1 its solitary position1 and its foliated structure parallel to that of the surrounding rock1 is1 as far as ,i know1 a unique case3 and ,i will not attempt any explanation of its origin4 ,the numerous travellers#d in this country1 have all been greatly surprised at the depth to which the gneiss and other granitic rocks1 as well as the talcose slates of the interior1 have been decomposed4 ,near ,rio1 every mineral except the quartz has been completely softened1 in some places to a depth little less than one hundred feet4#e ,the minerals retain their positions in folia ranging in the usual direction2 and fractured quartz veins may be traced from the solid rock1 running for some distance into the softened1 mottled1 highly coloured1 argillaceous mass4 ,it is said that these decomposed rocks abound with gems of various kinds1 often in a fractured state1 owing1 as some have supposed1 to the collapse of geodes1 and that they contain gold and diamonds4 ,at ,rio1 it appeared to me that the gneiss had been softened before the excavation 7no doubt by the sea7 of the existing1 broad1 flat-bottomed valleys2 for the depth of decomposition did not appear at all conformable with the present undulations of the surface4 ,the porphyritic gneiss1 where now exposed to the air1 seems to withstand decomposition remarkably well2 and ,i could see no signs of any tendency to the production of argillaceous masses like those here described4 ,i was also struck with the fact1 that where a bare surface of this rock sloped into one of the quiet bays1 there were no marks of erosion at the level of the water1 and the parts both beneath and above it preserved a uniform curve4 ,at ,bahia1 the gneiss rocks are similarly decomposed1 with the upper parts insensibly losing their foliation1 and passing1 without any distinct line of separation1 into a bright red argillaceous earth1 including partially rounded fragments of quartz and granite4 ,from this circumstance1 and from the rocks appearing to have suffered decomposition before the excavation of the valleys1 ,i suspect that here1 as at ,rio1 the decomposition took place under the sea4 ,the subject appeared to me a curious one1 and would probably well repay careful examination by an able mineralogist4 #d ,spix and ,martius have collected in an ,appendix to their 8,travels10 the largest body of facts on this subject4 ,see also some remarks by ,m4 ,lund in his communications to the ,academy at ,copenhagen2 and others by ,m4 ,gaudichaud in ,freycinet's 8,voyage40 #e ,dr4 ,benza describes granitic rock 7,madras ,journal of ,lit41 etc41 ,oct4 #ahc8 p4 #bdf71 in the ,neelgherries1 decomposed to a depth of forty feet4 ,the ,northern ,provinces of ,la ,plata4--,according to some observations communicated to me by ,mr4 ,fox1 the coast from ,rio de ,janeiro to the mouth of the ,plata seems everywhere to be granitic1 with a few trappean dikes4 ,at ,port ,alegre1 near the boundary of ,brazil1 there are porphyries and diorites4#f ,at the mouth of the ,plata1 ,i examined the country for twenty-five miles west1 and for about seventy miles north of ,maldonado3 near this town1 there is some common gneiss1 and much1 in all parts of the country1 of a coarse-grained mixture of quartz and reddish feldspar1 often1 however1 assuming a little dark-green imperfect hornblende1 and then immediately becoming foliated4 ,the abrupt hillocks thus composed1 as well as the highly inclined folia of the common varieties of gneiss1 strike ,n4,n4,e4 or a little more easterly1 and ,s4,s4,w4 ,clay-slate is occasionally met with1 and near the ,l4 del ,potrero1 there is white marble1 rendered fissile from the presence of hornblende1 mica1 and asbestus2 the cleavage of these rocks and their stratification1 that is the alternating masses thus composed1 strike ,n4,n4,e4 and ,s4,s4,w4 like the foliated gneisses1 and have an almost vertical dip4 ,the ,sierra ,larga1 a low range five miles west of ,maldonado1 consists of quartzite1 often ferruginous1 having an arenaceous feel1 and divided into excessively thin1 almost vertical lamin or folia by microscopically minute scales1 apparently of mica1 and striking in the usual ,n4,n4,e4 and ,s4,s4,w4 direction4 ,the range itself is formed of one principal line with some subordinate ones2 and it extends with remarkable uniformity far northward 7it is said even to the confines of ,brazil71 in the same line with the vertically ribboned quartz rock of which it is composed4 ,the ,s4 de ,las ,animas is the highest range in the country2 ,i estimated it at #a1jjj feet2 it runs north and south1 and is formed of feldspathic porphyry2 near its base there is a ,n4,n4,w4 and ,s4,s4,e4 ridge of a conglomerate in a highly porphyritic basis4 #f ,m4 ,isabelle1 8,voyage ,buenos ,ayres10 p4 #dgi4 ,northward of ,maldonado1 and south of ,las ,minas1 there is an ,e4 and ,w4 hilly band of country1 some miles in width1 formed of siliceous clay-slate1 with some quartz1 rock1 and limestone1 having a tortuous irregular cleavage1 generally ranging east and west4 ,e4 and ,s4,e4 of ,las ,minas there is a confused district of imperfect gneiss and laminated quartz1 with the hills ranging in various directions1 but with each separate hill generally running in the same line with the folia of the rocks of which it is composed3 this confusion appears to have been caused by the intersection of the ,e4 and ,w4 and ,n4,n4,e4 and ,s4,s4,w4 strikes4 ,northward of ,las ,minas1 the more regular northerly ranges predominate3 from this place to near ,polanco1 we meet with the coarse-grained mixture of quartz and feldspar1 often with the imperfect hornblende1 and then becoming foliated in a ,n4 and ,s4 line--with imperfect clay-slate1 including lamin of red crystallised feldspar--with white or black marble1 sometimes containing asbestus and crystals of gypsum--with quartz-rock--with syenite--and lastly1 with much granite4 ,the marble and granite alternate repeatedly in apparently vertical masses3 some miles northward of the ,polanco1 a wide district is said to be entirely composed of marble4 ,it is remarkable1 how rare mica is in the whole range of country north and westward of ,maldonado4 ,throughout this district1 the cleavage of the clay-slate and marble--the foliation of the gneiss and the quartz--the stratification or alternating masses of these several rocks--and the range of the hills1 all coincide in direction2 and although the country is only hilly1 the planes of division are almost everywhere very highly inclined or vertical4 ,some ancient submarine volcanic rocks are worth mentioning1 from their rarity on this eastern side of the continent4 ,in the valley of the ,tapas 7fifty or sixty miles ,n4 of ,maldonado7 there is a tract three or four miles in length1 composed of various trappean rocks with glassy feldspar--of apparently metamorphosed grit-stones--of purplish amygdaloids with large kernels of carbonate of lime#g--;and much of a harshish rock with glassy feldspar intermediate in character between claystone porphyry and trachyte4 ,this latter rock was in one spot remarkable from being full of drusy cavities1 lined with quartz crystals1 and arranged in planes1 dipping at an angle of #ej to the east1 and striking parallel to the foliation of an adjoining hill composed of the common mixture of quartz1 feldspar1 and imperfect hornblende3 this fact perhaps indicates that these volcanic rocks have been metamorphosed1 and their constituent parts rearranged1 at the same time and according to the same laws1 with the granitic and metamorphic formations of this whole region4 ,in the valley of the ,marmaraya1 a few miles south of the ,tapas1 a band of trappean and amygdaloidal rock is interposed between a hill of granite and an extensive surrounding formation of red conglomerate1 which 7like that at the foot of the ,s4 ,animas7 has its basis porphyritic with crystals of feldspar1 and which hence has certainly suffered metamorphosis4 #g ,near the ,pan de ,azucar there is some greenish porphyry1 in one place amygdaloidal with agate4 ,monte ,video4--,the rocks here consist of several varieties of gneiss1 with the feldspar often yellowish1 granular and imperfectly crystallised1 alternating with1 and passing insensibly into1 beds1 from a few yards to nearly a mile in thickness1 of fine or coarse grained1 dark-green hornblendic slate2 this again often passing into chloritic schist4 ,these passages seem chiefly due to changes in the mica1 and its replacement by other minerals4 ,at ,rat ,island ,i examined a mass of chloritic schist1 only a few yards square1 irregularly surrounded on all sides by the gneiss1 and intricately penetrated by many curvilinear veins of quartz1 which gradually blend into the gneiss3 the cleavage of the chloritic schist and the foliation of the gneiss were exactly parallel4 ,eastward of the city there is much fine-grained1 dark-coloured gneiss1 almost assuming the character of hornblende-slate1 which alternates in thin lamin with lamin of quartz1 the whole mass being transversely intersected by numerous large veins of quartz3 ,i particularly observed that these veins were absolutely continuous with the alternating lamin of quartz4 ,in this case and at ,rat ,island1 the passage of the gneiss into imperfect hornblendic or into chloritic slate1 seemed to be connected with the segregation of the veins of quartz4#h #h ,mr4 ,greenough 7p4 #gh1 8,critical ,examination10 etc47 observes that quartz in mica-slate sometimes appears in beds and sometimes in veins4 ,von ,buch also in his 8,travels in ,norway0 7p4 #bcf71 remarks on alternating lamin of quartz and hornblende-slate replacing mica-schist4 ,the ,mount1 a hill believed to be #dej feet in height1 from which the place takes its name1 is much the highest land in this neighbourhood3 it consists of hornblendic slate1 which 7except on the eastern and disturbed base7 has an east and west nearly vertical cleavage2 the longer axis of the hill also ranges in this same line4 ,near the summit the hornblende-slate gradually becomes more and more coarsely crystallised1 and less plainly laminated1 until it passes into a heavy1 sonorous greenstone1 with a slaty conchoidal fracture2 the lamin on the north and south sides near the summit dip inwards1 as if this upper part had expanded or bulged outwards4 ,this greenstone must1 ,i conceive1 be considered as metamorphosed hornblende-slate4 ,the ,cerrito1 the next highest1 but much less elevated point1 is almost similarly composed4 ,in the more western parts of the province1 besides gneiss1 there is quartz-rock1 syenite1 and granite2 and at ,colla1 ,i heard of marble4 ,near ,m4 ,video1 the space which ,i more accurately examined was about fifteen miles in an east and west line1 and here ,i found the foliation of the gneiss and the cleavage of the slates generally well developed1 and extending parallel to the alternating strata composed of the gneiss1 hornblendic and chloritic schists4 ,these planes of division all range within one point of east and west1 frequently east by south and west by north2 their dip is generally almost vertical1 and scarcely anywhere under #de3 this fact1 considering how slightly undulatory the surface of the country is1 deserves attention4 ,westward of ,m4 ,video1 towards the ,uruguay1 wherever the gneiss is exposed1 the highly inclined folia are seen striking in the same direction2 ,i must except one spot where the strike was ,n4,w4 by ,w4 ,the little ,sierra de ,s4 ,juan1 formed of gneiss and laminated quartz1 must also be excepted1 for it ranges between ,n4 to ,n4,e4 and ,s4 to ,s4,w4 and seems to belong to the same system with the hills in the ,maldonado district4 ,finally1 we have seen that1 for many miles northward of ,maldonado and for twenty-five miles westward of it1 as far as the ,s4 de las ,animas1 the foliation1 cleavage1 so-called stratification and lines of hills1 all range ,n4,n4,e4 and ,s4,s4,w41 which is nearly coincident with the adjoining coast of the ,atlantic4 ,westward of the ,s4 de las ,animas1 as far as even the ,uruguay1 the foliation1 cleavage1 and stratification 7but not lines of hills1 for there are no defined ones7 all range about ,e4 by ,s4 and ,w4 by ,n41 which is nearly coincident with the direction of the northern shore of the ,plata2 in the confused country near ,las ,minas1 where these two great systems appear to intersect each other1 the cleavage1 foliation1 and stratification run in various directions1 but generally coincide with the line of each separate hill4 ,southern ,la ,plata4--,the first ridge1 south of the ,plata1 which projects through the ,pampean formation1 is the ,sierra ,tapalguen and ,vulcan1 situated #bjj miles southward of the district just described4 ,this ridge is only a few hundred feet in height1 and runs from ,c4 ,corrientes in a ,w4,n4,w4 line for at least #aej miles into the interior3 at ,tapalguen1 it is composed of unstratified granular quartz1 remarkable from forming tabular masses and small plains1 surrounded by precipitous cliffs3 other parts of the range are said to consist of granite3 and marble is found at the ,s4 ,tinta4 ,it appears from ,m4 ,parchappe's#i observations1 that at ,tandil there is a range of quartzose gneiss1 very like the rocks of the ,s4 ,larga near ,maldonado1 running in the same ,n4,n4,e4 and ,s4,s4,w4 direction2 so that the framework of the country here is very similar to that on the northern shore of the ,plata4 #i ,m4 d',orbigny's 8,voyage10 ,part4 ,golog41 p4 #df4 ,i have given a short account of the peculiar forms of the quartz hills of ,tapalguen1 so unusual in a metamorphic formation1 in my 8,journal of ,researches0 7#bnd edit471 p4 #aaf4 ,the ,sierra ,guitru-gueyu is situated sixty miles south of the ,s4 ,tapalguen3 it consists of numerous parallel1 sometimes blended together ridges1 about twenty-three miles in width1 and five hundred feet in height above the plain1 and extending in a ,n4,w4 and ,s4,e4 direction4 ,skirting round the extreme ,s4,e4 termination1 ,i ascended only a few points1 which were composed of a fine-grained gneiss1 almost composed of feldspar with a little mica1 and passing in the upper parts of the hills into a rather compact purplish clay-slate4 ,the cleavage was nearly vertical1 striking in a ,n4,w4 by ,w4 and ,s4,e4 by ,e4 line1 nearly1 though not quite1 coincident with the direction of the parallel ridges4 ,the ,sierra ,ventana lies close south of that of ,guitru-gueyu2 it is remarkable from attaining a height1 very unusual on this side of the continent1 of #c1cdj feet4 ,it consists up to its summit1 of quartz1 generally pure and white1 but sometimes reddish1 and divided into thick lamin or strata3 in one part there is a little glossy clay-slate with a tortuous cleavage4 ,the thick layers of quartz strike in a ,w4 #cj ,n4 line1 dipping southerly at an angle of #de and upwards4 ,the principal line of mountains1 with some quite subordinate parallel ridges1 range about ,w4 #de ,n43 but at their ,s4,e4 termination1 only ,w4 #be ,n4 ,this ,sierra is said to extend between twenty and thirty leagues into the interior4 ,patagonia4--,with the exception perhaps of the hill of ,s4 ,antonio 7#fjj feet high7 in the ,gulf of ,s4 ,matias1 which has never been visited by a geologist1 crystalline rocks are not met with on the coast of ,patagonia for a space of #chj miles south of the ,s4 ,ventana4 ,at this point 7lat4 #dc #ej71 at ,points ,union and ,tombo1 plutonic rocks are said to appear1 and are found1 at rather wide intervals1 beneath the ,patagonian tertiary formation for a space of about three hundred miles southward1 to near ,bird ,island1 in latitude #dh #ef4 ,judging from specimens kindly collected for me by ,mr4 ,stokes1 the prevailing rock at ,ports ,st4 ,elena1 ,camerones1 ,malaspina1 and as far south as the ,paps of ,pineda1 is a purplish-pink or brownish claystone porphyry1 sometimes laminated1 sometimes slightly vesicular1 with crystals of opaque feldspar and with a few grains of quartz2 hence these porphyries resemble those immediately to be described at ,port ,desire1 and likewise a series which ,i have seen from ,p4 ,alegre on the southern confines of ,brazil4 ,this porphyritic formation further resembles in a singularly close manner the lowest stratified formation of the ,cordillera of ,chile1 which1 as we shall hereafter see1 has a vast range1 and attains a great thickness4 ,at the bottom of the ,gulf of ,st4 ,george1 only tertiary deposits appear to be present4 ,at ,cape ,blanco1 there is quartz rock1 very like that of the ,falkland ,islands1 and some hard1 blue siliceous clay-slate4 ,at ,port ,desire there is an extensive formation of the claystone porphyry1 stretching at least twenty-five miles into the interior3 it has been denuded and deeply worn into gullies before being covered up by the tertiary deposits1 through which it here and there projects in hills2 those north of the bay being #ddj feet in height4 ,the strata have in several places been tilted at small angles1 generally either to ,n4,n4,w4 or ,s4,s4,e4 ,by gradual passages and alternations1 the porphyries change incessantly in nature4 ,i will describe only some of the principal mineralogical changes1 which are highly instructive1 and which ,i carefully examined4 ,the prevailing rock has a compact purplish base1 with crystals of earthy or opaque feldspar1 and often with grains of quartz4 ,there are other varieties1 with an almost truly trachytic base1 full of little angular vesicles and crystals of glassy feldspar2 and there are beds of black perfect pitchstone1 as well as of a concretionary imperfect variety4 ,on a casual inspection1 the whole series would be thought to be of the same plutonic or volcanic nature with the trachytic varieties and pitchstone2 but this is far from being the case1 as much of the porphyry is certainly of metamorphic origin4 ,besides the true porphyries1 there are many beds of earthy1 quite white or yellowish1 friable1 easily fusible matter1 resembling chalk1 which under the microscope is seen to consist of minute broken crystals1 and which1 as remarked in a former chapter1 singularly resembles the upper tufaceous beds of the ,patagonian tertiary formation4 ,this earthy substance often becomes coarser1 and contains minute rounded fragments of porphyries and rounded grains of quartz1 and in one case so many of the latter as to resemble a common sandstone4 ,these beds are sometimes marked with true lines of aqueous deposition1 separating particles of different degrees of coarseness2 in other cases there are parallel ferruginous lines not of true deposition1 as shown by the arrangement of the particles1 though singularly resembling them4 ,the more indurated varieties often include many small and some larger angular cavities1 which appear due to the removal of earthy matter3 some varieties contain mica4 ,all these earthy and generally white stones insensibly pass into more indurated sonorous varieties1 breaking with a conchoidal fracture1 yet of small specific gravity2 many of these latter varieties assume a pale purple tint1 being singularly banded and veined with different shades1 and often become plainly porphyritic with crystals of feldspar4 ,the formation of these crystals could be most clearly traced by minute angular and often partially hollow patches of earthy matter1 first assuming a fibrous structure1 then passing into opaque imperfectly shaped crystals1 and lastly1 into perfect glassy crystals4 ,when these crystals have appeared1 and when the basis has become compact1 the rock in many places could not be distinguished from a true claystone porphyry without a trace of mechanical structure4 ,in some parts1 these earthy or tufaceous beds pass into jaspery and into beautifully mottled and banded porcelain rocks1 which break into splinters1 translucent at their edges1 hard enough to scratch glass1 and fusible into white transparent beads3 grains of quartz included in the porcelainous varieties can be seen melting into the surrounding paste4 ,in other parts1 the earthy or tufaceous beds either insensibly pass into1 or alternate with1 breccias composed of large and small fragments of various purplish porphyries1 with the matrix generally porphyritic3 these breccias1 though their subaqueous origin is in many places shown both by the arrangement of their smaller particles and by an oblique or current lamination1 also pass into porphyries1 in which every trace of mechanical origin and stratification has been obliterated4 ,some highly porphyritic though coarse-grained masses1 evidently of sedimentary origin1 and divided into thin layers1 differing from each other chiefly in the number of embedded grains of quartz1 interested me much from the peculiar manner in which here and there some of the layers terminated in abrupt points1 quite unlike those produced by a layer of sediment naturally thinning out1 and apparently the result of a subsequent process of metamorphic aggregation4 ,in another common variety of a finer texture1 the aggregating process had gone further1 for the whole mass consisted of quite short1 parallel1 often slightly curved layers or patches1 of whitish or reddish finely granulo-crystalline feldspathic matter1 generally terminating at both ends in blunt points2 these layers or patches further tended to pass into wedge or almond-shaped little masses1 and these finally into true crystals of feldspar1 with their centres often slightly drusy4 ,the series was so perfect that ,i could not doubt that these large crystals1 which had their longer axes placed parallel to each other1 had primarily originated in the metamorphosis and aggregation of alternating layers of tuff2 and hence their parallel position must be attributed 7unexpected though the conclusion may be71 not to laws of chemical action1 but to the original planes of deposition4 ,i am tempted briefly to describe three other singular allied varieties of rock2 the first without examination would have passed for a stratified porphyritic breccia1 but all the included angular fragments consisted of a border of pinkish crystalline feldspathic matter1 surrounding a dark translucent siliceous centre1 in which grains of quartz not quite blended into the paste could be distinguished3 this uniformity in the nature of the fragments shows that they are not of mechanical1 but of concretionary origin1 having resulted perhaps from the self-breaking up and aggregation of layers of indurated tuff containing numerous grains of quartz1--into which1 indeed1 the whole mass in one part passed4 ,the second variety is a reddish non-porphyritic claystone1 quite full of spherical cavities1 about half an inch in diameter1 each lined with a collapsed crust formed of crystals of quartz4 ,the third variety also consists of a pale purple non-porphyritic claystone1 almost wholly formed of concretionary balls1 obscurely arranged in layers1 of a less compact and paler coloured claystone2 each ball being on one side partly hollow and lined with crystals of quartz4 ,pseudo-dikes4--,some miles up the harbour1 in a line of cliffs formed of slightly metamorphosed tufaceous and porphyritic claystone beds1 ,i observed three vertical dikes1 so closely resembling in general appearance ordinary volcanic dikes1 that ,i did not doubt1 until closely examining their composition1 that they had been injected from below4 ,the first is straight1 with parallel sides1 and about four feet wide2 it consists of whitish1 indurated tufaceous matter1 precisely like some of the beds intersected by it4 ,the second dike is more remarkable2 it is slightly tortuous1 about eighteen inches thick1 and can be traced for a considerable distance along the beach2 it is of a purplish-red or brown colour1 and is formed chiefly of rounded grains of quartz1 with broken crystals of earthy feldspar1 scales of black mica1 and minute fragments of claystone porphyry1 all firmly united together in a hard sparing base4 ,the structure of this dike shows obviously that it is of mechanical and sedimentary origin2 yet it thinned out upwards1 and did not cut through the uppermost strata in the cliffs4 ,this fact at first appears to indicate that the matter could not have been washed in from above2#aj but if we reflect on the suction which would result from a deep-seated fissure being formed1 we may admit that if the fissure were in any part open to the surface1 mud and water might well be drawn into it along its whole course4 ,the third dike consisted of a hard1 rough1 white rock1 almost composed of broken crystals of glassy feldspar1 with numerous scales of black mica1 cemented in a scanty base2 there was little in the appearance of this rock1 to preclude the idea of its having been a true injected feldspathic dike4 ,the matter composing these three pseudo-dikes1 especially the second one1 appears to have suffered1 like the surrounding strata1 a certain degree of metamorphic action2 and this has much aided the deceptive appearance4 ,at ,bahia1 in ,brazil1 we have seen that a true injected hornblendic dike1 not only has suffered metamorphosis1 but has been dislocated and even diffused in the surrounding gneiss1 under the form of separate crystals and of fragments4 #aj ,upfilled fissures are known to occur both in volcanic and in ordinary sedimentary formations4 ,at the ,galapagos ,archipelago 78,volcanic ,islands0 etc471 there are some striking examples of pseudo-dikes composed of hard tuff4 ,falkland ,islands4--,i have described these islands in a paper published in the third volume of the ,geological ,journal4 ,the mountain-ridges consist of quartz1 and the lower country of clay-slate and sandstone1 the latter containing ,palozoic fossils4 ,these fossils have been separately described by ,messrs4 ,morris and ,sharpe3 some of them resemble ,silurian1 and others ,devonian forms4 ,in the eastern part of the group the several parallel ridges of quartz extend in a west and east line2 but further westward the line becomes ,w4,n4,w4 and ,e4,s4,e41 and even still more northerly4 ,the cleavage-planes of the clay-slate are highly inclined1 generally at an angle of above #ej1 and often vertical2 they strike almost invariably in the same direction with the quartz ranges4 ,the outline of the indented shores of the two main islands1 and the relative positions of the smaller islets1 accord with the strike both of the main axes of elevation and of the cleavage of the clay-slate4 ,tierra del ,fuego4--,my notes on the geology of this country are copious1 but as they are unimportant1 and as fossils were found only in one district1 a brief sketch will be here sufficient4 ,the east coast from the ,s4 of ,magellan 7where the boulder formation is largely developed7 to ,st4 ,polycarp's ,bay is formed of horizontal tertiary strata1 bounded some way towards the interior by a broad mountainous band of clay-slate4 ,this great clay-slate formation extends from ,st4 ,le ,maire westward for #adj miles1 along both sides of the ,beagle ,channel to near its bifurcation4 ,south of this channel1 it forms all ,navarin ,island1 and the eastern half of ,hoste ,island and of ,hardy ,peninsula2 north of the ,beagle ,channel it extends in a north-west line on both sides of ,admiralty ,sound to ,brunswick ,peninsula in the ,st4 of ,magellan1 and ,i have reason to believe1 stretches far up the eastern side of the ,cordillera4 ,the western and broken side of ,tierra del ,fuego towards the ,pacific is formed of metamorphic schists1 granite and various trappean rocks3 the line of separation between the crystalline and clay-slate formations can generally be distinguished1 as remarked by ,captain ,king1#aa by the parallelism in the clay-slate districts of the shores and channels1 ranging in a line between ,w4 #bj to #dj ,n4 and ,e4 #bj to #dj ,s44 #aa ,geographical ,journal1 vol4 i1 p4 #aee4 ,the clay-slate is generally fissile1 sometimes siliceous or ferruginous1 with veins of quartz and calcareous spar2 it often assumes1 especially on the loftier mountains1 an altered feldspathic character1 passing into feldspathic porphyry3 occasionally it is associated with breccia and grauwacke4 ,at ,good ,success ,bay1 there is a little intercalated black crystalline limestone4 ,at ,port ,famine much of the clay-slate is calcareous1 and passes either into a mudstone or into grauwacke1 including odd-shaped concretions of dark argillaceous limestone4 ,here alone1 on the shore a few miles north of ,port ,famine1 and on the summit of ,mount ,tarn 7#b1fjj feet high71 ,i found organic remains2 they consist of3-- ,ancyloceras simplex1 d',orbigny1 8,pal ,franc10 ,mount ,tarn4 ,fusus 7in imperfect state71 d',orbigny1 8,pal ,franc10 ,mount ,tarn4 ,natica1 d',orbigny1 8,pal ,franc10 ,mount ,tarn4 ,pentacrimus1 d',orbigny1 8,pal ,franc10 ,mount ,tarn4 ,lucina excentrica1 ,g4 ,b4 ,sowerby1 ,port ,famine4 ,venus 7in imperfect state71 ,g4 ,b4 ,sowerby1 ,port ,famine4 ,turbinolia 7871 ,g4 ,b4 ,sowerby1 ,port ,famine4 ,hamites elatior1 ,g4 ,b4 ,sowerby1 ,port ,famine4 ,m4 d',orbigny states#ab that ,,mm4 ,hombron and ,grange found in this neighbourhood an ,ancyloceras1 perhaps ,a4 simplex1 an ,ammonite1 a ,plicatula and ,modiola4 ,m4 d',orbigny believes from the general character of these fossils1 and from the ,ancyloceras being identical 7as far as its imperfect condition allows of comparison7 with the ,a4 simplex of ,europe1 that the formation belongs to an early stage of the ,cretaceous system4 ,professor ,e4 ,forbes1 judging only from my specimens1 concurs in the probability of this conclusion4 ,the ,hamites elatior of the above list1 of which a description has been given by ,mr4 ,sowerby1 and which is remarkable from its large size1 has not been seen either by ,m4 d',orbigny or ,professor ,e4 ,forbes1 as1 since my return to ,england1 the specimens have been lost4 ,the great clay-slate formation of ,tierra del ,fuego being cretaceous1 is certainly a very interesting fact1--whether we consider the appearance of the country1 which1 without the evidence afforded by the fossils1 would form the analogy of most known districts1 probably have been considered as belonging to the ,palozoic series1--or whether we view it as showing that the age of this terminal portion of the great axis of ,south ,america1 is the same 7as will hereafter be seen7 with the ,cordillera of ,chile and ,peru4 #ab 8,voyage10 ,part ,golog41 p4 #bdb4 ,the clay-slate in many parts of ,tierra del ,fuego1 is broken by dikes#ac and by great masses of greenstone1 often highly hornblendic3 almost all the small islets within the clay-slate districts are thus composed4 ,the slate near the dikes generally becomes paler-coloured1 harder1 less fissile1 of a feldspathic nature1 and passes into a porphyry or greenstone3 in one case1 however1 it became more fissile1 of a red colour1 and contained minute scales of mica1 which were absent in the unaltered rock4 ,on the east side of ,ponsonby ,sound some dikes composed of a pale sonorous feldspathic rock1 porphyritic with a little feldspar1 were remarkable from their number1--there being within the space of a mile at least one hundred1--from their nearly equalling in bulk the intermediate slate1--and more especially from the excessive fineness 7like the finest inlaid carpentry7 and perfect parallelism of their junctions with the almost vertical lamin of clay-slate4 ,i was unable to persuade myself that these great parallel masses had been injected1 until ,i found one dike which abruptly thinned out to half its thickness1 and had one of its walls jagged1 with fragments of the slate embedded in it4 #ac ,in a greenstone-dike in the ,magdalen ,channel1 the feldspar cleaved with the angle of albite4 ,this dike was crossed1 as well as the surrounding slate1 by a large vein of quartz1 a circumstance of unusual occurrence4 ,in ,southern ,tierra del ,fuego1 the clay-slate towards its ,s4,w4 boundary1 becomes much altered and feldspathic4 ,thus on ,wollaston ,island slate and grauwacke can be distinctly traced passing into feldspathic rocks and greenstones1 including iron pyrites and epidote1 but still retaining traces of cleavage with the usual strike and dip4 ,one such metamorphosed mass was traversed by large vein-like masses of a beautiful mixture 7as ascertained by ,professor ,miller7 of green epidote1 garnets1 and white calcareous spar4 ,on the northern point of this same island1 there were various ancient submarine volcanic rocks1 consisting of amygdaloids with dark bole and agate1--of basalt with decomposed olivine--of compact lava with glassy feldspar1--and of a coarse conglomerate of red scori1 parts being amygdaloidal with carbonate of lime4 ,the southern part of ,wollaston ,island and the whole of ,hermite and ,horn ,islands1 seem formed of cones of greenstone2 the outlying islets of ,il ,defenso and ,d4 ,raminez are said#ad to consist of porphyritic lava4 ,in crossing ,hardy ,peninsula1 the slate still retaining traces of its usual cleavage1 passes into columnar feldspathic rocks1 which are succeeded by an irregular tract of trappean and basaltic rocks1 containing glassy feldspar and much iron pyrites3 there is1 also1 some harsh red claystone porphyry1 and an almost true trachyte1 with needles of hornblende1 and in one spot a curious slaty rock divided into quadrangular columns1 having a base almost like trachyte1 with drusy cavities lined by crystals1 too imperfect1 according to ,professor ,miller1 to be measured1 but resembling ,zeagonite4#ae ,in the midst of these singular rocks1 no doubt of ancient submarine volcanic origin1 a high hill of feldspathic clay-slate projected1 retaining its usual cleavage4 ,near this point1 there was a small hillock1 having the aspect of granite1 but formed of white albite1 brilliant crystals of hornblende 7both ascertained by the reflecting goniometer7 and mica2 but with no quartz4 ,no recent volcanic district has been observed in any part of ,tierra del ,fuego4 #ad ,determined by ,professor ,jameson4 ,weddell's 8,voyage10 p4 #afi4 #ae ,see ,mr4 ,brooke's ,paper in the ,london ,phil4 ,mag41 vol4 x4 ,this mineral occurs in an ancient volcanic rock near ,rome4 ,five miles west of the bifurcation of the ,beagle ,channel1 the slate-formation1 instead of becoming1 as in the more southern parts of ,tierra del ,fuego1 feldspathic1 and associated with trappean or old volcanic rocks1 passes by alternations into a great underlying mass of fine gneiss and glossy clay-slate1 which at no great distance is succeeded by a grand formation of mica-slate containing garnets4 ,the folia of these metamorphic schists strike parallel to the cleavage-planes of the clay-slate1 which have a very uniform direction over the whole of this part of the country3 the folia1 however1 are undulatory and tortuous1 whilst the cleavage-lamin of the slate are straight4 ,these schists compose the chief mountain-chain of ,southern ,tierra del ,fuego1 ranging along the north side of the northern arm of the ,beagle ,channel1 in a short ,w4,n4,w4 and ,e4,s4,e4 line1 with two points 7,mounts ,sarmiento and ,darwin7 rising to heights of #f1hjj and #f1ijj feet4 ,on the south-western side of this northern arm of the ,beagle ,channel1 the clay-slate is seen with its strata dipping from the great chain1 so that the metamorphic schists here form a ridge bordered on each side by clay-slate4 ,further north1 however1 to the west of this great range1 there is no clay-slate1 but only gneiss1 mica1 and hornblendic slates1 resting on great barren hills of true granite1 and forming a tract about sixty miles in width4 ,again1 westward of these rocks1 the outermost islands are of trappean formation1 which1 from information obtained during the voyages of the ,adventure and ,beagle1#af seem1 together with granite1 chiefly to prevail along the western coast as far north as the entrance of the ,st4 of ,magellan3 a little more inland1 on the eastern side of ,clarence ,island and ,s4 ,desolation1 granite1 greenstone1 mica-slate1 and gneiss appear to predominate4 ,i am tempted to believe1 that where the clay-slate has been metamorphosed at great depths beneath the surface1 gneiss1 mica-slate1 and other allied rocks have been formed1 but where the action has taken place nearer the surface1 feldspathic porphyries1 greenstones1 etc41 have resulted1 often accompanied by submarine volcanic eruptions4 #af ,see the ,paper by ,captain ,king in the ,geograph4 ,journal2 also a ,letter to ,dr4 ,fitton in 8,geolog4 ,proc410 vol4 i1 p4 #bi2 also some observations by ,captain ,fitzroy1 8,voyages10 vol4 i1 p4 #cge4 ,i am indebted also to ,mr4 ,lyell for a series of specimens collected by ,lieutenant ,graves4 ,only one other rock1 met with in both arms of the ,beagle ,channel1 deserves any notice1 namely a granulo-crystalline mixture of white albite1 black hornblende 7ascertained by measurement of the crystals1 and confirmed by ,professor ,miller71 and more or less of brown mica1 but without any quartz4 ,this rock occurs in large masses1 closely resembling in external form granite or syenite3 in the southern arm of the ,channel1 one such mass underlies the mica-slate1 on which clay-slate was superimposed3 this peculiar plutonic rock which1 as we have seen1 occurs also in ,hardy ,peninsula1 is interesting1 from its perfect similarity with that 7hereafter often to be referred to under the name of andesite7 forming the great injected axes of the ,cordillera of ,chile4 ,the stratification of the clay-slate is generally very obscure1 whereas the cleavage is remarkably well defined3 to begin with the extreme eastern parts of ,tierra del ,fuego2 the cleavage-planes near the ,st4 of ,le ,maire strike either ,w4 and ,e4 or ,w4,s4,w4 and ,e4,n4,e41 and are highly inclined2 the form of the land1 including ,staten ,island1 indicates that the axes of elevation have run in this same line1 though ,i was unable to distinguish the planes of stratification4 ,proceeding westward1 ,i accurately examined the cleavage of the clay-slate on the northern1 eastern1 and western sides 7thirty-five miles apart7 of ,navarin ,island1 and everywhere found the lamin ranging with extreme regularity1 ,w4,n4,w4 and ,e4,s4,e41 seldom varying more than one point of the compass from this direction4#ag ,both on the east and west coasts1 ,i crossed at right angles the cleavage-planes for a space of about eight miles1 and found them dipping at an angle of between #de and #ij1 generally to ,s4,s4,w41 sometimes to ,n4,n4,e41 and often quite vertically4 ,the ,s4,s4,w4 dip was occasionally succeeded abruptly by a ,n4,n4,e4 dip1 and this by a vertical cleavage1 or again by the ,s4,s4,w4 dip2 as in a lofty cliff on the eastern end of the island the lamin of slate were seen to be folded into very large steep curves1 ranging in the usual ,w4,n4,w4 line1 ,i suspect that the varying and opposite dips may possibly be accounted for by the cleavage-lamin1 though to the eye appearing straight1 being parts of large abrupt curves1 with their summits cut off and worn down4 #ag ,the clay-slate in this island was in many places crossed by parallel smooth joints4 ,out of five cases1 the angle of intersection between the strike of these joints and that of the cleavage-lamin was in two cases #de and in two others #gi4 ,in several places ,i was particularly struck with the fact1 that the fine lamin of the clay-slate1 where cutting straight through the bands of stratification1 and therefore indisputably true cleavage-planes1 differed slightly in their greyish and greenish tints of colour1 in compactness1 and in some of the lamin having a rather more jaspery appearance than others4 ,i have not seen this fact recorded1 and it appears to me important1 for it shows that the same cause which has produced the highly fissile structure1 has altered in a slight degree the mineralogical character of the rock in the same planes4 ,the bands of stratification1 just alluded to1 can be distinguished in many places1 especially in ,navarin ,island1 but only on the weathered surfaces of the slate2 they consist of slightly undulatory zones of different shades of colour and of thicknesses1 and resemble the marks 7more closely than anything else to which ,i can compare them7 left on the inside of a vessel by the draining away of some dirty slightly agitated liquid3 no difference in composition1 corresponding with these zones1 could be seen in freshly fractured surfaces4 ,in the more level parts of ,navarin ,island1 these bands of stratification were nearly horizontal2 but on the flanks of the mountains they were inclined from them1 but in no instance that ,i saw at a very high angle4 ,there can1 ,i think1 be no doubt that these zones1 which appear only on the weathered surfaces1 are the last vestiges of the original planes of stratification1 now almost obliterated by the highly fissile and altered structure which the mass has assumed4 ,the clay-slate cleaves in the same ,w4,n4,w4 and ,e4,s4,e4 direction1 as on ,navarin ,island1 on both sides of the ,beagle ,channel1 on the eastern side of ,hoste ,island1 on the ,n4,e4 side of ,hardy ,peninsula1 and on the northern point of ,wollaston ,island2 although in these two latter localities the cleavage has been much obscured by the metamorphosed and feldspathic condition of the slate4 ,within the area of these several islands1 including ,navarin ,island1 the direction of the stratification and of the mountain-chains is very obscure2 though the mountains in several places appeared to range in the same ,w4,n4,w4 line with the cleavage3 the outline of the coast1 however1 does not correspond with this line4 ,near the bifurcation of the ,beagle ,channel1 where the underlying metamorphic schists are first seen1 they are foliated 7with some irregularities71 in this same ,w4,n4,w4 line1 and parallel1 as before stated1 to the main mountain-axis of this part of the country4 ,westward of this main range1 the metamorphic schists are foliated1 though less plainly1 in the same direction1 which is likewise common to the zone of old erupted trappean rocks1 forming the outermost islets4 ,hence the area1 over which the cleavage of the slate and the foliation of the metamorphic schists extends with an average ,w4,n4,w4 and ,e4,s4,e4 strike1 is about forty miles in a north and south line1 and ninety miles in an east and west line4 ,further northward1 near ,port ,famine1 the stratification of the clay-slate and of the associated rocks1 is well defined1 and there alone the cleavage and strata-planes are parallel4 ,a little north of this port there is an anticlinal axis ranging ,n4,w4 7or a little more westerly7 and ,s4,e43 south of the port1 as far as ,admiralty ,sound and ,gabriel ,channel1 the outline of the land clearly indicates the existence of several lines of elevation in this same ,n4,w4 direction1 which1 ,i may add1 is so uniform in the western half of the ,st4 of ,magellan1 that1 as ,captain ,king#ah has remarked1 8a parallel ruler placed on the map upon the projecting points of the south shore1 and extended across the strait1 will also touch the headlands on the opposite coast40 ,it would appear1 from ,captain ,king's observations1 that over all this area the cleavage extends in the same line4 ,deep-water channels1 however1 in all parts of ,tierra del ,fuego have burst through the trammels both of stratification and cleavage2 most of them may have been formed during the elevation of the land by long-continued erosion1 but others1 for instance the ,beagle ,channel1 which stretches like a narrow canal for #abj miles obliquely through the mountains1 can hardly have thus originated4 #ah ,geograph4 ,journal1 vol4 i1 p4 #agj4 ,finally1 we have seen that in the extreme eastern point of ,tierra del ,fuego1 the cleavage and coast-lines extend ,w4 and ,e4 and even ,w4,s4,w4 and ,e4,n4,e43 over a large area westward1 the cleavage1 the main range of mountains1 and some subordinate ranges1 but not the outlines of the coast1 strike ,w4,n4,w41 and ,e4,s4,e43 in the central and western parts of the ,st4 of ,magellan1 the stratification1 the mountain-ranges1 the outlines of the coast1 and the cleavage all strike nearly ,n4,w4 and ,s4,e4 ,north of the strait1 the outline of the coast1 and the mountains on the mainland1 run nearly north and south4 ,hence we see1 at this southern point of the continent1 how gradually the ,cordillera bend1 from their north and south course of so many thousand miles in length1 into an ,e4 and even ,e4,n4,e4 direction4 ,west coast1 from the ,southern ,chonos ,islands to ,northern ,chile4--,the first place at which we landed north of the ,st4 of ,magellan was near ,cape ,tres ,montes1 in lat4 #dg ,s4 ,between this point and the ,northern ,chonos ,islands1 a distance of #bjj miles1 the ,beagle visited several points1 and specimens were collected for me from the intermediate spaces by ,lieutenant ,stokes4 ,the predominant rock is mica-slate1 with thick folia of quartz1 very frequently alternating with and passing into a chloritic1 or into a black1 glossy1 often striated1 slightly anthracitic schist1 which soils paper1 and becomes white under a great heat1 and then fuses4 ,thin layers of feldspar1 swelling at intervals into well crystallised kernels1 are sometimes included in these black schists2 and ,i observed one mass of the ordinary black variety insensibly lose its fissile structure1 and pass into a singular mixture of chlorite1 epidote1 feldspar1 and mica4 ,great veins of quartz are numerous in the mica-schists2 wherever these occur the folia are much convoluted4 ,in the southern part of the ,peninsula of ,tres ,montes1 a compact altered feldspathic rock with crystals of feldspar and grains of quartz is the commonest variety2 this rock#ai exhibits occasionally traces of an original brecciated structure1 and often presents 7like the altered state of ,tierra del ,fuego7 traces of cleavage-planes1 which strike in the same direction with the folia of mica-schist further northward4 ,at ,inchemo ,island1 a similar rock gradually becomes granulo-crystalline and acquires scales of mica2 and this variety at ,s4 ,estevan becomes highly laminated1 and though still exhibiting some rounded grains of quartz1 passes into the black1 glossy1 slightly anthracitic schist1 which1 as we have seen1 repeatedly alternates with and passes into the micaceous and chloritic schists4 ,hence all the rocks on this line of coast belong to one series1 and insensibly vary from an altered feldspathic clay-slate into largely foliated1 true mica-schist4 #ai ,the peculiar1 abruptly conical form of the hills in this neighbourhood1 would have led any one at first to have supposed that they had been formed of injected or intrusive rocks4 ,the cleavage of the homogeneous schists1 the foliation of those composed of more or less distinct minerals in layers1 and the planes of alternation of the different varieties or so-called stratification1 are all parallel1 and preserve over this #bjj miles of coast a remarkable degree of uniformity in direction4 ,at the northern end of the group1 at ,low's ,harbour1 the well-defined folia of mica-schist everywhere ranged within eight degrees 7or less than one point of the compass7 of ,n4 #ai ,w4 and ,s4 #ai ,e42 and even the point of dip varied very little1 being always directed to the west and generally at an angle of forty degrees2 ,i should mention that ,i had here good opportunities of observation1 for ,i followed the naked rock on the beach1 transversely to the strike1 for a distance of four miles and a half1 and all the way attended to the dip4 ,along the outer islands for #ajj miles south of ,low's ,harbour1 ,lieutenant ,stokes1 during his boat-survey1 kindly observed for me the strike of the foliation1 and he assures me that it was invariably northerly1 and the dip with one single exception to the west4 ,further south at ,vallenar ,bay1 the strike was almost universally ,n4 #be ,w4 and the dip1 generally at an angle of about #dj to ,w4 #be ,s41 but in some places almost vertical4 ,still farther south1 in the neighbourhood of the harbours of ,anna ,pink1 ,s4 ,estevan and ,s4 ,andres1 and 7judging from a distance7 along the southern part of ,tres ,montes1 the foliation and cleavage extended in a line between ,n4 #aa to #bb ,w4 and ,s4 #aa to #bb ,e42 and the planes dipped generally westerly1 but often easterly1 at angles varying from a gentle inclination to vertical4 ,at ,a4 ,pink's ,harbour1 where the schists generally dipped easterly1 wherever the angle became very high1 the strike changed from ,n4 #aa ,w4 to even as much as ,n4 #de ,w43 in an analogous manner at ,vallenar ,bay1 where the dip was westerly 7viz4 on an average directed to ,w4 #be ,s471 as soon as the angle became very high1 the planes struck in a line more than #be west of north4 ,the average result from all the observations on this #bjj miles of coast1 is a strike of ,n4 #ai ,w4 and ,s4 #ai ,e43 considering that in each specified place my examination extended over an area of several miles1 and that ,lieutenant ,stokes' observations apply to a length of #ajj miles1 ,i think this remarkable uniformity is pretty well established4 ,the prevalence1 throughout the northern half of this line of coast1 of a dip in one direction1 that is to the west1 instead of being sometimes west and sometimes east1 is1 judging from what ,i have elsewhere seen1 an unusual circumstance4 ,in ,brazil1 ,la ,plata1 the ,falkland ,islands1 and ,tierra del ,fuego1 there is generally an obvious relation between the axis of elevation1 the outline of the coast1 and the strike of the cleavage or foliation3 in the ,chonos ,archipelago1 however1 neither the minor details of the coast-line1 nor the chain of the ,cordillera1 nor the subordinate transverse mountain-axes1 accord with the strike of the foliation and cleavage3 the seaward face of the numerous islands composing this ,archipelago1 and apparently the line of the ,cordillera1 range ,n4 #aa ,e41 whereas1 as we have just seen1 the average strike of the foliation is ,n4 #ai ,w4 ,there is one interesting exception to the uniformity in the strike of the foliation4 ,at the northern point of ,tres ,montes 7lat4 #de #eb7 a bold chain of granite1 between two and three thousand feet in height1 runs from the coast far into the interior1#bj in an ,e4,s4,e4 line1 or more strictly ,e4 #bh ,s4 and ,w4 #bh ,n4 ,in a bay1 at the northern foot of this range1 there are a few islets of mica-slate1 with the folia in some parts horizontal1 but mostly inclined at an average angle of #bj to the north4 ,on the northern steep flank of the range1 there are a few patches 7some quite isolated1 and not larger than half-a-crown67 of the mica-schist1 foliated with the same northerly dip4 ,on the broad summit1 as far as the southern crest1 there is much mica-slate1 in some places even #djj feet in thickness1 with the folia all dipping north1 at angles varying from #e to #bj1 but sometimes mounting up to #cj4 ,the southern flank consists of bare granite4 ,the mica-slate is penetrated by small veins#ba of granite1 branching from the main body4 ,leaving out of view the prevalent strike of the folia in other parts of this ,archipelago1 it might have been expected that they would have dipped ,n4 #bh ,e41 that is directly from the ridge1 and1 considering its abruptness1 at a high inclination2 but the real dip1 as we have just seen1 both at the foot and on the northern flank1 and over the entire summit1 is at a small angle1 and directed nearly due north4 ,from these considerations it occurred to me1 that perhaps we here had the novel and curious case of already inclined lamin obliquely tilted at a subsequent period by the granitic axis4 ,mr4 ,hopkins1 so well known from his mathematical investigations1 has most kindly calculated the problem3 the proposition sent was1--,take a district composed of lamin1 dipping at an angle of #dj degrees to ,w4 #ai ,s41 and let an axis of elevation traverse it in an ,e4 #bh ,s4 line1 what will the position of the lamin be on the northern flank after a tilt1 we will first suppose1 of #de8 ,mr4 ,hopkins informs me1 that the angle of the dip will be #bh #ca1 and its direction to north #cj #cc west4#bb ,by varying the supposed angle of the tilt1 our previously inclined folia can be thrown into any angle between #bf1 which is the least possible angle1 and #ij2 but if a small inclination be thus given to them1 their point of dip will depart far from the north1 and therefore not accord with the actual position of the folia of mica-schist on our granitic range4 ,hence it appears very difficult1 without varying considerably the elements of the problem1 thus to explain the anomalous strike and dip of the foliated mica-schist1 especially in those parts1 namely1 at the base of the range1 where the folia are almost horizontal4 ,mr4 ,hopkins1 however1 adds1 that great irregularities and lateral thrusts might be expected in every great line of elevation1 and that these would account for considerable deviations from the calculated results3 considering that the granitic axis1 as shown by the veins1 has indisputably been injected after the perfect formation of the mica-slate1 and considering the uniformity of the strike of the folia throughout the rest of the ,archipelago1 ,i cannot but still think that their anomalous position at this one point is someway directly and mechanically related to the intrusion of this ,w4,n4,w4 and ,e4,s4,e4 mountain-chain of granite4 #bj ,in the distance1 other mountains could be seen apparently ranging ,n4,n4,e4 and ,s4,s4,w4 at right angles to this one4 ,i may add1 that not far from ,vallenar ,bay there is a fine range1 apparently of granite1 which has burst through the mica-slate in a ,n4,e4 by ,e4 and ,s4,w4 by ,s4 line4 #ba ,the granite within these veins1 as well as generally at the junction with the mica-slate1 is more quartzose than elsewhere4 ,the granite1 ,i may add1 is traversed by dikes running for a very great length in the line of the mountains2 they are composed of a somewhat laminated eurite1 containing crystals of feldspar1 hornblende1 and octagons of quartz4 #bb ,on the south side of the axis 7where1 however1 ,i did not see any mica-slate7 the dip of the folia would be at an angle of #gg #ee1 directed to west #ce #cc south4 ,hence the two points of dip on the opposite sides of the range1 instead of being as in ordinary cases directly opposed to each other at an angle of #ahj1 would here be only #hf #ej apart4 ,dikes are frequent in the metamorphic schists of the ,chonos ,islands1 and seem feebly to represent that great band of trappean and ancient volcanic rocks on the south-western coast of ,tierra del ,fuego4 ,at ,s4 ,andres ,i observed in the space of half-a-mile1 seven broad1 parallel dikes1 composed of three varieties of trap1 running in a ,n4,w4 and ,s4,e4 line1 parallel to the neighbouring mountain-ranges of altered clay-slate2 but they must be of long subsequent origin to these mountains2 for they intersected the volcanic formation described in the last chapter4 ,north of ,tres ,montes1 ,i noticed three dikes differing from each other in composition1 one of them having a euritic base including large octagons of quartz2 these dikes1 as well as several of porphyritic greenstone at ,vallenar ,bay1 extended ,n4,e4 and ,s4,w41 nearly at right angles to the foliation of the schists1 but in the line of their joints4 ,at ,low's ,harbour1 however1 a set of great parallel dikes1 one ninety yards and another sixty yards in width1 have been guided by the foliation of the mica-schist1 and hence are inclined westward at an angle of #de3 these dikes are formed of various porphyritic traps1 some of which are remarkable from containing numerous rounded grains of quartz4 ,a porphyritic trap of this latter kind1 passed in one of the dikes into a most curious hornstone1 perfectly white1 with a waxy fracture and pellucid edges1 fusible1 and containing many grains of quartz and specks of iron pyrites4 ,in the ninety-yard dike several large1 apparently now quite isolated1 fragments of mica-slate were embedded3 but as their foliation was exactly parallel to that of the surrounding solid rock1 no doubt these new separate fragments originally formed wedge-shaped depending portions of a continuous vault or crust1 once extending over the dike1 but since worn down and denuded4 ,chiloe1 ,valdivia1 ,concepcion4--,in ,chiloe1 a great formation of mica-schist strikingly resembles that of the ,chonos ,islands4 ,for a space of eleven miles on the ,s4,e4 coast1 the folia were very distinct1 though slightly convoluted1 and ranged within a point of ,n4,n4,w4 and ,s4,s4,e41 dipping either ,e4,n4,e4 or more commonly ,w4,s4,w41 at an average angle of #bb 7in one spot1 however1 at #fj71 and therefore decidedly at a lesser inclination than amongst the ,chonos ,islands4 ,on the west and north-western shores1 the foliation was often obscure1 though1 where best defined1 it ranged within a point of ,n4 by ,w4 and ,s4 by ,e41 dipping either easterly or westerly1 at varying and generally very small angles4 ,hence1 from the southern part of ,tres ,montes to the northern end of ,chiloe1 a distance of #cjj miles1 we have closely allied rocks with their folia striking on an average in the same direction1 namely between ,n4 #aa and #bb ,w4 ,again1 at ,valdivia1 we meet with the same mica-schist1 exhibiting nearly the same mineralogical passages as in the ,chonos ,archipelago1 often1 however1 becoming more ferruginous1 and containing so much feldspar as to pass into gneiss4 ,the folia were generally well defined2 but nowhere else in ,south ,america did ,i see them varying so much in direction3 this seemed chiefly caused by their forming parts1 as ,i could sometimes distinctly trace1 of large flat curves3 nevertheless1 both near the settlement and towards the interior1 a ,n4,w4 and ,s4,e4 strike seemed more frequent than any other direction2 the angle of the dip was generally small4 ,at ,concepcion1 a highly glossy clay-slate had its cleavage often slightly curvilinear1 and inclined1 seldom at a high angle1 towards various points of the compass3#bc but here1 as at ,valdivia1 a ,n4,w4 and ,s4,e4 strike seemed to be the most frequent one4 ,in certain spots large quartz veins were numerous1 and near them1 the cleavage1 as was the case with the foliation of the schists in the ,chonos ,archipelago1 became extremely tortuous4 #bc ,i observed in some parts that the tops of the lamin of the clay-slate 7b of the diagram7 under the superficial detritus and soil 7a7 were bent1 sometimes without being broken1 as represented in the accompanying diagram1 which is copied from one given by ,sir ,h4 ,de la ,beche 7p4 #db 8,geological ,manual07 of an exactly similar phenomenon in ,devonshire4 ,mr4 ,r4 ,a4 ,c4 ,austen1 also1 in his excellent paper on ,s4,e4 ,devon 78,geolog4 ,transact410 vol4 vi1 p4 #dcg71 has described this phenomenon2 he attributes it to the action of frosts1 but at the same time doubts whether the frosts of the present day penetrate to a sufficient depth4 ,as it is known that earthquakes particularly affect the surface of the ground1 it occurred to me that this appearance might perhaps be due1 at least at ,concepcion1 to their frequent occurrence2 the superficial layers of detritus being either jerked in one direction1 or1 where the surface was inclined1 pushed a little downwards during each strong vibration4 ,in ,north ,wales ,i have seen a somewhat analogous but less regular appearance1 though on a greater scale 7,london ,phil4 ,mag41 vol4 xxi1 p4 #ahd71 and produced by a quite different cause1 namely1 by the stranding of great icebergs2 this latter appearance has also been observed in ,n4 ,america4 ,illustration3 ,diagram described in note #bc4 ,at the northern end of ,quiriquina ,island1 in the ,bay of ,concepcion1 at least eight rudely parallel dikes1 which have been guided to a certain extent by the cleavage of the slate1 occur within the space of a quarter of a mile4 ,they vary much in composition1 resembling in many respects the dikes at ,low's ,harbour3 the greater number consist of feldspathic porphyries1 sometimes containing grains of quartz3 one1 however1 was black and brilliant1 like an augitic rock1 but really formed of feldspar2 others of a feldspathic nature were perfectly white1 with either an earthy or crystalline fracture1 and including grains and regular octagons of quartz2 these white varieties passed into ordinary greenstones4 ,although1 both here and at ,low's ,harbour1 the nature of the rock varied considerably in the same dike1 yet ,i cannot but think that at these two places and in other parts of the ,chonos group1 where the dikes1 though close to each other and running parallel1 are of different composition1 that they must have been formed at different periods4 ,in the case of ,quiriquina this is a rather interesting conclusion1 for these eight parallel dikes cut through the metamorphic schists in a ,n4,w4 and ,s4,e4 line1 and since their injection the overlying cretaceous or tertiary strata have been tilted 7whilst still under the sea7 from a ,n4,w4 by ,n4 and ,s4,e4 by ,s4 line2 and again1 during the great earthquake of ,february #ahce1 the ground in this neighbourhood was fissured in ,n4,w4 and ,s4,e4 lines2 and from the manner in which buildings were thrown down1 it was evident that the surface undulated in this same direction4#bd #bd 8,geolog4 ,trans410 vol4 vi1 pp4 #fjb and #fag4 8,journal of ,researches0 7#bnd edit471 p4 #cjg4 ,central and ,northern ,chile4--,northward of ,concepcion1 as far as ,copiapo1 the shores of the ,pacific consist1 with the exception of some small tertiary basins1 of gneiss1 mica-schist1 altered clay-slate1 granite1 greenstone and syenite3 hence the coast from ,tres ,montes to ,copiapo1 a distance of #a1bjj miles1 and ,i have reason to believe for a much greater space1 is almost similarly constituted4 ,near ,valparaiso the prevailing rock is gneiss1 generally including much hornblende3 concretionary balls formed of feldspar1 hornblende and mica1 from two or three feet in diameter1 are in very many places conformably enfolded by the foliated gneiss3 veins of quartz and feldspar1 including black schorl and well-crystallised epidote1 are numerous4 ,epidote likewise occurs in the gneiss in thin layers1 parallel to the foliation of the mass4 ,one large vein of a coarse granitic character was remarkable from in one part quite changing its character1 and insensibly passing into a blackish porphyry1 including acicular crystals of glassy feldspar and of hornblende3 ,i have never seen any other such case4#be #be ,humboldt 78,personal ,narrative10 vol4 iv1 p4 #fj7 has described with much surprise1 concretionary balls1 with concentric divisions1 composed of partially vitreous feldspar1 hornblende1 and garnets1 included within great veins of gneiss1 which cut across the mica-slate near ,venezuela4 ,i shall in the few following remarks on the rocks of ,chile allude exclusively to their foliation and cleavage4 ,in the gneiss round ,valparaiso the strike of the foliation is very variable1 but ,i think about ,n4 by ,w4 and ,s4 by ,e4 is the commonest direction2 this likewise holds good with the cleavage of the altered feldspathic clay-slates1 occasionally met with on the coast for ninety miles north of ,valparaiso4 ,some feldspathic slate1 alternating with strata of claystone porphyry in the ,bell of ,quillota and at ,jajuel1 and therefore1 perhaps1 belonging to a later period than the metamorphic schists on the coast1 cleaved in this same direction4 ,in the ,eastern ,cordillera1 in the ,portillo ,pass1 there is a grand mass of mica-slate1 foliated in a north and south line1 and with a high westerly dip3 in the ,uspallata range1 clay-slate and grauwacke have a highly inclined1 nearly north and south cleavage1 though in some parts the strike is irregular3 in the main or ,cumbre range1 the direction of the cleavage in the feldspathic clay-slate is ,n4,w4 and ,s4,e4 ,between ,coquimbo and ,guasco there are two considerable formations of mica-slate1 in one of which the rock passed sometimes into common clay-slate and sometimes into a glossy black variety1 very like that in the ,chonos ,archipelago4 ,the folia and cleavage of these rocks ranged between ,n4 and ,n4,w4 by ,n4 and ,s4 and ,s4,w4 by ,s44 ,near the ,port of ,guasco several varieties of altered clay-slate have a quite irregular cleavage4 ,between ,guasco and ,copiapo1 there are some siliceous and talcaceous slates cleaving in a north and south line1 with an easterly dip of between #fj and #gj3 high up1 also1 the main valley of ,copiapo1 there is mica-slate with a high easterly dip4 ,in the whole space between ,valparaiso and ,copiapo an easterly dip is much more common than an opposite or westerly one4 ,concluding ,remarks on ,cleavage and ,foliation4 ,in this southern part of the southern hemisphere1 we have seen that the cleavage-lamin range over wide areas with remarkable uniformity1 cutting straight through the planes of stratification1#bf but yet being parallel in strike to the main axes of elevation1 and generally to the outlines of the coast4 ,the dip1 however1 is as variable1 both in angle and in direction 7that is1 sometimes being inclined to the one side and sometimes to the directly opposite side71 as the strike is uniform4 ,in all these respects there is a close agreement with the facts given by ,professor ,sedgwick in his celebrated memoir in the 8,geological ,transactions10 and by ,sir ,r4 ,i4 ,murchison in his various excellent discussions on this subject4 ,the ,falkland ,islands1 and more especially ,tierra del ,fuego1 offer striking instances of the lines of cleavage1 the principle axes of elevation1 and the outlines of the coast1 gradually changing together their courses4 ,the direction which prevails throughout ,tierra del ,fuego and the ,falkland ,islands1 namely1 from west with some northing to east with some southing1 is also common to the several ridges in ,northern ,patagonia and in the western parts of ,banda ,oriental3 in this latter province1 in the ,sierra ,tapalguen1 and in the ,western ,falkland ,island1 the ,w4 by ,n41 or ,w4,n4,w4 and ,e4,s4,e41 ridges1 are crossed at right angles by others ranging ,n4,n4,e4 and ,s4,s4,w4 #bf ,in my paper on the ,falkland ,islands 7,geological ,journal1 vol4 iii1 p4 #bfg71 ,i have given a curious case on the authority of ,captain ,sulivan1 ,r4,n41 of much folded beds of clay-slate1 in some of which the cleavage is perpendicular to the horizon1 and in others it is perpendicular to each curvature or fold of the bed3 this appears a new case4 ,the fact of the cleavage-lamin in the clay-slate of ,tierra del ,fuego1 where seen cutting straight through the planes of stratification1 and where consequently there could be no doubt about their nature1 differing slightly in colour1 texture1 and hardness1 appears to me very interesting4 ,in a thick mass of laminated1 feldspathic and altered clay-slate1 interposed between two great strata of porphyritic conglomerate in ,central ,chile1 and where there could be but little doubt about the bedding1 ,i observed similar slight differences in composition1 and likewise some distinct thin layers of epidote1 parallel to the highly inclined cleavage of the mass4 ,again1 ,i incidentally noticed in ,north ,wales1#bg where glaciers had passed over the truncated edges of the highly inclined lamin of clay-slate1 that the surface1 though smooth1 was worn into small parallel undulations1 caused by the competent lamin being of slightly different degrees of hardness4 ,with reference to the slates of ,north ,wales1 ,professor ,sedgwick describes the planes of cleavage1 as 8coated over with chlorite and semi-crystalline matter1 which not only merely define the planes in question1 but strike in parallel flakes through the whole mass of the rock40#bh ,in some of those glossy and hard varieties of clay-slate1 which may often be seen passing into mica-schist1 it has appeared to me that the cleavage-planes were formed of excessively thin1 generally slighted convoluted1 folia1 composed of microscopically minute scales of mica4 ,from these several facts1 and more especially from the case of the clay-slate in ,tierra del ,fuego1 it must1 ,i think1 be concluded1 that the same power which has impressed on the slate its fissile structure or cleavage has tended to modify its mineralogical character in parallel planes4 #bg ,london ,phil4 ,mag41 vol4 xxi1 p4 #ahb4 #bh 8,geological ,trans410 vol4 iii1 p4 #dga4 ,let us now turn to the foliation of the metamorphic schists1 a subject which has been much less attended to4 ,as in the case of cleavage-lamin1 the folia preserve over very large areas a uniform strike3 thus ,humboldt#bi found for a distance of #cjj miles in ,venezuela1 and indeed over a much larger space1 gneiss1 granite1 mica1 and clay-slate1 striking very uniformly ,n4,e4 and ,s4,w41 and dipping at an angle of between #fj and #gj to ,n4,w42 it would even appear from the facts given in this chapter1 that the metamorphic rocks throughout the north-eastern part of ,south ,america are generally foliated within two points of ,n4,e4 and ,s4,w4 ,over the eastern parts of ,banda ,oriental1 the foliation strikes with a high inclination1 very uniformly ,n4,n4,e4 to ,s4,s4,w41 and over the western parts1 in a ,w4 by ,n4 and ,e4 by ,s4 line4 ,for a space of #cjj miles on the shores of the ,chonos and ,chiloe ,islands1 we have seen that the foliation seldom deviates more than a point of the compass from a ,n4 #ai ,w4 and ,s4 #ai ,e4 strike4 ,as in the case of cleavage1 the angle of the dip in foliated rocks is generally high but variable1 and alternates from one side of the line of strike to the other side1 sometimes being vertical3 in the ,northern ,chonos ,islands1 however1 the folia are inclined almost always to the west2 in nearly the same manner1 the cleavage-lamin in ,southern ,tierra del ,fuego certainly dip much more frequently to ,s4,s4,w4 than to the opposite point4 ,in ,eastern ,banda ,oriental1 in parts of ,brazil1 and in some other districts1 the foliation runs in the same direction with the mountain-ranges and adjoining coast-lines3 amongst the ,chonos ,islands1 however1 this coincidence fails1 and ,i have given my reasons for suspecting that one granitic axis has burst through and tilted the already inclined folia of mica-schist3 in the case of cleavage1#cj the coincidence between its strike and that of the main stratification seems sometimes to fail4 ,foliation and cleavage resemble each other in the planes winding round concretions1 and in becoming tortuous where veins of quartz abound4#ca ,on the flanks of the mountains both in ,tierra del ,fuego and in other countries1 ,i have observed that the cleavage-planes frequently dip at a high angle inwards2 and this was long ago observed by ,von ,buch to be the case in ,norway3 this fact is perhaps analogous to the folded1 fan-like or radiating structure in the metamorphic schists of the ,alps1#cb in which the folia in the central crests are vertical and on the two flanks inclined inwards4 ,where masses of fissile and foliated rocks alternate together1 the cleavage and foliation1 in all cases which ,i have seen1 are parallel4 ,where in one district the rocks are fissile1 and in another adjoining district they are foliated1 the planes of cleavage and foliation are likewise generally parallel3 this is the case with the feldspathic homogeneous slates in the southern part of the ,chonos group1 compared with the fine foliated mica-schists of the northern part2 so again the clay-slate of the whole eastern side of ,tierra del ,fuego cleaves in exactly the same line with the foliated gneiss and mica-slate of the western coast2 other analogous instances might have been adduced4#cc #bi 8,personal ,narrative10 vol4 vi1 p4 #ei et seq4 #cj ,cases are given by ,mr4 ,jukes in his 8,geology of ,newfoundland10 p4 #acj4 #ca ,i have seen in ,brazil and ,chile concretions thus enfolded by foliated gneiss2 and ,macculloch 78,highlands10 vol4 i1 p4 #fd7 has described a similar case4 ,for analogous cases in clay-slate1 see ,professor ,henslow's ,memoir in 8,cambridge ,phil4 ,trans410 vol4 i1 p4 #cgi1 and ,macculloch's 8,class4 of ,rocks10 p4 #cea4 ,with respect to both foliation and cleavage becoming tortuous where quartz-veins abound1 ,i have seen instances near ,monte ,video1 at ,concepcion1 and in the ,chonos ,islands4 ,see also ,mr4 ,greenough's 8,critical ,examination10 p4 #gh4 #cb ,studer in ,edin4 ,new ,phil4 ,journal1 vol4 xxiii1 p4 #add4 #cc ,i have given a case in ,australia4 ,see my 8,volcanic ,islands40 ,with respect to the origin of the folia of quartz1 mica1 feldspar1 and other minerals composing the metamorphic schists1 ,professor ,sedgwick1 ,mr4 ,lyell1 and most authors believe1 that the constituent parts of each layer were separately deposited as sediment1 and then metamorphosed4 ,this view1 in the majority of cases1 ,i believe to be quite untenable4 ,in those not uncommon instances1 where a mass of clay-slate1 in approaching granite1 gradually passes into gneiss1#cd we clearly see that folia of distinct minerals can originate through the metamorphosis of a homogeneous fissile rock4 ,the deposition1 it may be remarked1 of numberless alternations of pure quartz1 and of the elements of mica or feldspar does not appear a probable event4#ce ,in those districts in which the metamorphic schists are foliated in planes parallel to the cleavage of the rocks in an adjoining district1 are we to believe that the folia are due to sedimentary layers1 whilst the cleavage-lamin1 though parallel1 have no relation whatever to such planes of deposition8 ,on this view1 how can we reconcile the vastness of the areas over which the strike of the foliation is uniform1 with what we see in disturbed districts composed of true strata3 and especially1 how can we understand the high and even vertical dip throughout many wide districts1 which are not mountainous1 and throughout some1 as in ,western ,banda ,oriental1 which are not even hilly8 ,are we to admit that in the northern part of the ,chonos ,archipelago1 mica-slate was first accumulated in parallel horizontal folia to a thickness of about four geographical miles1 and then upturned at an angle of forty degrees2 whilst1 in the southern part of this same ,archipelago1 the cleavage-lamin of closely allied rocks1 which none would imagine had ever been horizontal1 dip at nearly the same angle1 to nearly the same point8 #cd ,i have described in 8,volcanic ,islands0 a good instance of such a passage at the ,cape of ,good ,hope4 #ce ,see some excellent remarks on this subject1 in ,d',aubuisson's 8,trait de ,gog410 tome i1 p4 #big4 ,also some remarks by ,mr4 ,dana in ,silliman's ,american ,journ41 vol4 xlv1 p4 #ajh4 ,seeing1 then1 that foliated schists indisputably are sometimes produced by the metamorphosis of homogeneous fissile rocks2 seeing that foliation and cleavage are so closely analogous in the several above-enumerated respects2 seeing that some fissile and almost homogeneous rocks show incipient mineralogical changes along the planes of their cleavage1 and that other rocks with a fissile structure alternate with1 and pass into varieties with a foliated structure1 ,i cannot doubt that in most cases foliation and cleavage are parts of the same process3 in cleavage there being only an incipient separation of the constituent minerals2 in foliation a much more complete separation and crystallisation4 ,the fact often referred to in this chapter1 of the foliation and the so-called strata in the metamorphic series1--that is1 the alternating masses of different varieties of gneiss1 mica-schist1 and hornblende-slate1 etc41--being parallel to each other1 at first appears quite opposed to the view1 that the folia have no relation to the planes of original deposition4 ,where the so-called beds are not very thick and of widely different mineralogical composition from each other1 ,i do not think that there is any difficulty in supposing that they have originated in an analogous manner with the separate folia4 ,we should bear in mind what thick strata1 in ordinary sedimentary masses1 have obviously been formed by a concretionary process4 ,in a pile of volcanic rocks on the ,island of ,ascension1 there are strata1 differing quite as much in appearance as the ordinary varieties of the metamorphic schists1 which undoubtedly have been produced1 not by successive flowings of lava1 but by internal molecular changes4 ,near ,monte ,video1 where the stratification1 as it would be called1 of the metamorphic series is1 in most parts1 particularly well developed1 being as usual1 parallel to the foliation1 we have seen that a mass of chloritic schist1 netted with quartz-veins1 is entangled in gneiss1 in such a manner as to show that it had certainly originated in some process of segregation3 again1 in another spot1 the gneiss tended to pass into hornblendic schist by alternating with layers of quartz2 but these layers of quartz almost certainly had never been separately deposited1 for they were absolutely continuous with the numerous intersecting veins of quartz4 ,i have never had an opportunity of tracing for any distance1 along the line both of strike and of dip1 the so-called beds in the metamorphic schists1 but ,i strongly suspect that they would not be found to extend with the same character1 very far in the line either of their dip or strike4 ,hence ,i am led to believe1 that most of the so-called beds are of the nature of complex folia1 and have not been separately deposited4 ,of course1 this view cannot be extended to thick masses included in the metamorphic series1 which are of totally different composition from the adjoining schists1 and which are far extended1 as is sometimes the case with quartz and marble2 these must generally be of the nature of true strata4#cf ,such strata1 however1 will almost always strike in the same direction with the folia1 owing to the axes of elevation being in most countries parallel to the strike of the foliation2 but they will generally dip at a different angle from that of the foliation2 and the angle of the foliation in itself almost always varies much3 hence1 in crossing a metamorphosed schistose district1 it would require especial attention to discriminate between true strata of deposition and complex foliated masses4 ,the mere presence of true strata in the midst of a set of metamorphic schists1 is no argument that the foliation is of sedimentary origin1 without it be further shown in each case1 that the folia not only strike1 but dip throughout in parallel planes with those of the true stratification4 #cf ,macculloch states 78,classification of ,rocks10 p4 #cfd7 states that primary limestones are often found in irregular masses or great nodules1 8which can scarcely be said to possess a stratified shape60 ,as in some cases it appears that where a fissile rock has been exposed to partial metamorphic action1 for instance from the irruption of granite1 the foliation has supervened on the already existing cleavage-planes2 so perhaps in some instances1 the foliation of a rock may have been determined by the original planes of deposition or of oblique current-lamin3 ,i have1 however1 myself1 never seen such a case1 and ,i must maintain that in most extensive metamorphic areas1 the foliation is the extreme result of that process1 of which cleavage is the first effect4 ,that foliation may arise without any previous structural arrangement in the mass1 we may infer from injected1 and therefore once liquified1 rocks1 both of volcanic and plutonic origin1 sometimes having a 8grain0 7as expressed by ,professor ,sedgwick71 and sometimes being composed of distinct folia or lamin of different compositions4 ,in my work on 8,volcanic ,islands10 ,i have given several instances of this structure in volcanic rocks1 and it is not uncommonly seen in plutonic masses--thus1 in the ,cordillera of ,chile1 there are gigantic mountain-like masses of red granite1 which have been injected whilst liquified1 and which1 nevertheless1 display in parts a decidedly laminar structure4#cg #cg ,as remarked in a former part of this chapter1 ,i suspect that the boldly conical mountains of gneiss-granite1 near ,rio de ,janeiro1 in which the constituent minerals are arranged in parallel planes1 are of intrusive origin4 ,we must not1 however1 forget the lesson of caution taught by the curious claystone porphyries of ,port ,desire1 in which we have seen that the breaking up and aggregation of a thinly stratified tufaceous mass1 has yielded a rock semi-porphyritic with crystals of feldspar1 arranged in the planes of original deposition4 ,finally1 we have seen that the planes of cleavage and of foliation1 that is1 of the incipient process and of the final result1 generally strike parallel to the principal axes of elevation1 and to the outline of the land3 the strike of the axes of elevation 7that is1 of the lines of fissures with the strata on their edges upturned71 according to the reasoning of ,mr4 ,hopkins1 is determined by the form of the area undergoing changes of level1 and the consequent direction of the lines of tension and fissure4 ,now1 in that remarkable pile of volcanic rocks at ,ascension1 which has several times been alluded to 7and in some other cases71 ,i have endeavoured to show1#ch that the lamination of the several varieties1 and their alternations1 have been caused by the moving mass1 just before its final consolidation1 having been subjected 7as in a glacier7 to planes of different tension2 this difference in the tension affecting the crystalline and concretionary processes4 ,one of the varieties of rock thus produced at ,ascension1 at first sight1 singularly resembles a fine-grained gneiss2 it consists of quite straight and parallel zones of excessive tenuity1 of more or less coloured crystallised feldspar1 of distinct crystals of quartz1 diopside1 and oxide of iron4 ,these considerations1 notwithstanding the experiments made by ,mr4 ,fox1 showing the influence of electrical currents in producing a structure like that of cleavage1 and notwithstanding the apparently inexplicable variation1 both in the inclination of the cleavage-lamin and in their dipping first to one side and then to the other side of the line of strike1 lead me to suspect that the planes of cleavage and foliation are intimately connected with the planes of different tension1 to which the area was long subjected1 after the main fissures or axes of upheavement had been formed1 but before the final consolidation of the mass and the total cessation of all molecular movement4 #ch ,in 8,volcanic ,islands40 ,illustration3 ,geological sections through the ,cordilleras4 ,for enlargements of the above plate use the following links3 left section center section right section ,chapter ,,vii ,,central ,,chile3--,,structure ,,of ,,the ,,cordillera4 ,central ,chile4--,basal formations of the ,cordillera4--,origin of the porphyritic clay-stone conglomerate4--,andesite4--,volcanic rocks4--,section of the ,cordillera by the ,peuquenes are ,portillo ,pass4--,great gypseous formation4--,peuquenes line2 thickness of strata1 fossils of4--,portillo line4--,conglomerate1 orthitic granite1 mica-schist1 volcanic rocks of4--,concluding remarks on the denudation and elevation of the ,portillo line4--,section by the ,cumbre1 or ,uspallata ,pass4--,porphyries4--,gypseous strata4--,section near the ,puente del ,inca2 fossils of4--,great subsidence4--,intrusive porphyries4--,plain of ,uspallata4--,section of the ,uspallata chain4--,structure and nature of the strata4--,silicified vertical trees4--,great subsidence4--,granitic rocks of axis4--,concluding remarks on the ,uspallata range2 origin subsequent to that of the main ,cordillera2 two periods of subsidence2 comparison with the ,portillo chain4 ,the district between the ,cordillera and the ,pacific1 on a rude average1 is from about eighty to one hundred miles in width4 ,it is crossed by many chains of mountains1 of which the principal ones1 in the latitude of ,valparaiso and southward of it1 range nearly north and south2 but in the more northern parts of the province1 they run in almost every possible direction4 ,near the ,pacific1 the mountain-ranges are generally formed of syenite or granite1 and or of an allied euritic porphyry2 in the low country1 besides these granitic rocks and greenstone1 and much gneiss1 there are1 especially northward of ,valparaiso1 some considerable districts of true clay-slate with quartz veins1 passing into a feldspathic and porphyritic slate2 there is also some grauwacke and quartzose and jaspery rocks1 the latter occasionally assuming the character of the basis of claystone porphyry3 trap-dikes are numerous4 ,nearer the ,cordillera the ranges 7such as those of ,s4 ,fernando1 the ,prado1#a and ,aconcagua7 are formed partly of granitic rocks1 and partly of purple porphyritic conglomerates1 claystone porphyry1 greenstone porphyry1 and other rocks1 such as we shall immediately see1 form the basal strata of the main ,cordillera4 ,in the more northern parts of ,chile1 this porphyritic series extends over large tracts of country far from the ,cordillera2 and even in ,central ,chile such occasionally occur in outlying positions4 #a ,meyen 8,reise um ,erde0 ,th4 ,i1 s4 #bce4 ,i will describe the ,campana of ,quillota1 which stands only fifteen miles from the ,pacific1 as an instance of one of these outlying masses4 ,this hill is conspicuous from rising to the height of #f1djj feet3 its summit shows a nucleus1 uncovered for a height of #hjj feet1 of fine greenstone1 including epidote and octahedral magnetic iron ore2 its flanks are formed of great strata of porphyritic claystone conglomerate associated with various true porphyries and amygdaloids1 alternating with thick masses of a highly feldspathic1 sometimes porphyritic1 pale-coloured slaty rock1 with its cleavage-lamin dipping inwards at a high angle4 ,at the base of the hill there are syenites1 a granular mixture of quartz and feldspar1 and harsh quartzose rocks1 all belonging to the basal metamorphic series4 ,i may observe that at the foot of several hills of this class1 where the porphyries are first seen 7as near ,s4 ,fernando1 the ,prado1 ,las ,vacas1 etc471 similar harsh quartzose rocks and granular mixtures of quartz and feldspar occur1 as if the more fusible constituent parts of the granitic series had been drawn off to form the overlying porphyries4 ,in ,central ,chile1 the flanks of the main ,cordillera1 into which ,i penetrated by four different valleys1 generally consist of distinctly stratified rocks4 ,the strata are inclined at angles varying from sometimes even under ten1 to twenty degrees1 very rarely exceeding forty degrees3 in some1 however1 of the quite small1 exterior1 spur-like ridges1 the inclination was not unfrequently greater4 ,the dip of the strata in the main outer lines was usually outwards or from the ,cordillera1 but in ,northern ,chile frequently inwards1--that is1 their basset-edges fronted the ,pacific4 ,dikes occur in extraordinary numbers4 ,in the great1 central1 loftiest ridges1 the strata1 as we shall presently see1 are almost always highly inclined and often vertical4 ,before giving a detailed account of my two sections across the ,cordillera1 it will1 ,i think1 be convenient to describe the basal strata as seen1 often to a thickness of four or five thousand feet1 on the flanks of the outer lines4 ,basal strata of the ,cordillera4--,the prevailing rock is a purplish or greenish1 porphyritic claystone conglomerate4 ,the embedded fragments vary in size from mere particles to blocks as much as six or eight inches 7rarely more7 in diameter2 in many places1 where the fragments were minute1 the signs of aqueous deposition were unequivocally distinct2 where they were large1 such evidence could rarely be detected4 ,the basis is generally porphyritic with perfect crystals of feldspar1 and resembles that of a true injected claystone porphyry3 often1 however1 it has a mechanical or sedimentary aspect1 and sometimes 7as at ,jajuel7 is jaspery4 ,the included fragments are either angular1 or partially or quite rounded2#b in some parts the rounded1 in others the angular fragments prevail1 and usually both kinds are mixed together3 hence the word breccia ought strictly to be appended to the term porphyritic conglomerate4 ,the fragments consist of many varieties of claystone porphyry1 usually of nearly the same colour with the surrounding basis1 namely1 purplish-reddish1 brownish1 mottled or bright green2 occasionally fragments of a laminated1 pale-coloured1 feldspathic rock1 like altered clay-slate are included2 as are sometimes grains of quartz1 but only in one instance in ,central ,chile 7namely1 at the mines of ,jajuel7 a few pebbles of quartz4 ,i nowhere observed mica in this formation1 and rarely hornblende2 where the latter mineral did occur1 ,i was generally in doubt whether the mass really belonged to this formation1 or was of intrusive origin4 ,calcareous spar occasionally occurs in small cavities2 and nests and layers of epidote are common4 ,in some few places in the finer-grained varieties 7for instance1 at ,quillota71 there were short1 interrupted layers of earthy feldspar1 which could be traced1 exactly as at ,port ,desire1 passing into large crystals of feldspar3 ,i doubt1 however1 whether in this instance the layers had ever been separately deposited as tufaceous sediment4 #b ,some of the rounded fragments in the porphyritic conglomerate near the ,baths of ,cauquenes1 were marked with radii and concentric zones of different shades of colour3 any one who did not know that pebbles1 for instance flint pebbles from the chalk1 are sometimes zoned concentrically with their worn and rounded surfaces1 might have been led to infer1 that these balls of porphyry were not true pebbles1 but had originated in concretionary action47 ,all the varieties of porphyritic conglomerates and breccias pass into each other1 and by innumerable gradations into porphyries no longer retaining the least trace of mechanical origin3 the transition appears to have been effected much more easily in the finer-grained1 than in the coarser-grained varieties4 ,in one instance1 near ,cauquenes1 ,i noticed that a porphyritic conglomerate assumed a spheroidal structure1 and tended to become columnar4 ,besides the porphyritic conglomerates and the perfectly characterised porphyries1 of metamorphic origin1 there are other porphyries1 which1 though differing not at all or only slightly in composition1 certainly have had a different origin3 these consist of pink or purple claystone porphyries1 sometimes including grains of quartz1--of greenstone porphyry1 and of other dusky rocks1 all generally porphyritic with fine1 large1 tabular1 opaque crystals1 often placed crosswise1 of feldspar cleaving like albite 7judging from several measurements71 and often amygdaloidal with silex1 agate1 carbonate of lime1 green and brown bole4#c ,these several porphyritic and amygdaloidal varieties never show any signs of passing into masses of sedimentary origin3 they occur both in great and small intrusive masses1 and likewise in strata alternating with those of the porphyritic conglomerate1 and with the planes of junction often quite distinct1 yet not seldom blended together4 ,in some of these intrusive masses1 the porphyries exhibit1 more or less plainly1 a brecciated structure1 like that often seen in volcanic masses4 ,these brecciated porphyries could generally be distinguished at once from the metamorphosed1 porphyritic breccia-conglomerates1 by all the fragments being angular and being formed of the same variety1 and by the absence of every trace of aqueous deposition4 ,one of the porphyries above specified1 namely1 the greenstone porphyry with large tabular crystals of albite1 is particularly abundant1 and in some parts of the ,cordillera 7as near ,st4 ,jago7 seemed more common even than the purplish porphyritic conglomerate4 ,numerous dikes likewise consist of this greenstone porphyry2 others are formed of various fine-grained trappean rocks2 but very few of claystone porphyry3 ,i saw no true basaltic dikes4 #c ,this bole is a very common mineral in the amygdaloidal rocks2 it is generally of a greenish-brown colour1 with a radiating structure2 externally it is black with an almost metallic lustre1 but often coated by a bright green film4 ,it is soft and can be scratched by a quill2 under the blowpipe swells greatly and becomes scaly1 then fuses easily into a black magnetic bead4 ,this substance is evidently similar to that which often occurs in submarine volcanic rocks4 ,an examination of some very curious specimens of a fine porphyry 7from ,jajuel7 leads me to suspect that some of these amygdaloidal balls1 instead of having been deposited in pre-existing air-vesicles1 are of concretionary origin2 for in these specimens1 some of the pea-shaped little masses 7often externally marked with minute pits7 are formed of a mixture of green earth with stony matter1 like the basis of the porphyry1 including minute imperfect crystals of feldspar2 and these pea-shaped little masses are themselves amygdaloidal with minute spheres of the green earth1 each enveloped by a film of white1 apparently feldspathic1 earthy matter3 so that the porphyry is doubly amygdaloidal4 ,it should not1 however1 be overlooked1 that all the strata here have undergone metamorphic action1 which may have caused crystals of feldspar to appear1 and other changes to be effected1 in the originally simple amygdaloidal balls4 ,mr4 ,j4 ,d4 ,dana1 in an excellent paper on ,trap-rocks 7,edin4 ,new ,phil4 ,journ41 vol4 xli1 p4 #aih71 has argued with great force1 that all amygdaloidal minerals have been deposited by aqueous infiltration4 ,i may take this opportunity of alluding to a curious case1 described in my work on 8,volcanic ,islands10 of an amygdaloid with many of its cells only half filled up with a mesotypic mineral4 ,m4 ,rose has described an amygdaloid1 brought by ,dr4 ,meyen 78,reise um ,erde10 ,th4 ,i1 s4 #caf7 from ,chile1 as consisting of crystallised quartz1 with crystals of stilbite within1 and lined externally by green earth4 ,in several places in the lower part of the series1 but not everywhere1 thick masses of a highly feldspathic1 often porphyritic1 slaty rock occur interstratified with the porphyritic conglomerate2 ,i believe in one or two cases blackish limestone has been found in a similar position4 ,the feldspathic rock is of a pale grey or greenish colour2 it is easily fusible2 where porphyritic1 the crystals of feldspar are generally small and vitreous3 it is distinctly laminated1 and sometimes includes parallel layers of epidote2#d the lamination appears to be distinct from stratification4 ,occasionally this rock is somewhat curious2 and at one spot1 namely1 at the ,c4 of ,quillota1 it had a brecciated structure4 ,near the mines of ,jajuel1 in a thick stratum of this feldspathic1 porphyritic slate1 there was a layer of hard1 blackish1 siliceous1 infusible1 compact clay-slate1 such as ,i saw nowhere else2 at the same place ,i was able to follow for a considerable distance the junction between the slate and the conformably underlying porphyritic conglomerate1 and they certainly passed gradually into each other4 ,wherever these slaty feldspathic rocks abound1 greenstone seems common2 at the ,c4 of ,quillota a bed of well-crystallised greenstone lay conformably in the midst of the feldspathic slate1 with the upper and lower junctions passing insensibly into it4 ,from this point1 and from the frequently porphyritic condition of the slate1 ,i should perhaps have considered this rock as an erupted one 7like certain laminated feldspathic lavas in the trachytic series71 had ,i not seen in ,tierra del ,fuego how readily true clay-slate becomes feldspathic and porphyritic1 and had ,i not seen at ,jajuel the included layer of black1 siliceous clay-slate1 which no one could have thought of igneous origin4 ,the gentle passage of the feldspathic slate1 at ,jajuel1 into the porphyritic conglomerate1 which is certainly of aqueous origin1 should also be taken in account4 #d ,this mineral is extremely common in all the formations of ,chile2 in the gneiss near ,valparaiso and in the granitic veins crossing it1 in the injected greenstone crowning the ,c4 of ,quillota1 in some granitic porphyries1 in the porphyritic conglomerate1 and in the feldspathic clay-slates4 ,the alternating strata of porphyries and porphyritic conglomerate1 and with the occasionally included beds of feldspathic slate1 together make a grand formation2 in several places within the ,cordillera1 ,i estimated its thickness at from six to seven thousand feet4 ,it extends for many hundred miles1 forming the western flank of the ,chilean ,cordillera2 and even at ,iquique in ,peru1 #hej miles north of the southernmost point examined by me in ,chile1 the coast-escarpment which rises to a height of between two and three thousand feet is thus composed4 ,in several parts of ,northern ,chile this formation extends much further towards the ,pacific1 over the granitic and metamorphic lower rocks1 than it does in ,central ,chile2 but the main ,cordillera may be considered as its central line1 and its breadth in an east and west direction is never great4 ,at first the origin of this thick1 massive1 long but narrow formation1 appeared to me very anomalous3 whence were derived1 and how were dispersed the innumerable fragments1 often of large size1 sometimes angular and sometimes rounded1 and almost invariably composed of porphyritic rocks8 ,seeing that the interstratified porphyries are never vesicular and often not even amygdaloidal1 we must conclude that the pile was formed in deep water2 how then came so many fragments to be well rounded and so many to remain angular1 sometimes the two kinds being equally mingled1 sometimes one and sometimes the other preponderating8 ,that the claystone1 greenstone1 and other porphyries and amygdaloids1 which lie conformably between the beds of conglomerate1 are ancient submarine lavas1 ,i think there can be no doubt2 and ,i believe we must look to the craters whence these streams were erupted1 as the source of the breccia-conglomerate2 after the great explosion1 we may fairly imagine that the water in the heated and scarcely quiescent crater would remain for a considerable time#e sufficiently agitated to triturate and round the loose fragments1 few or many in number1 would be shot forth at the next eruption1 associated with few or many angular fragments1 according to the strength of the explosion4 ,the porphyritic conglomerate being purple or reddish1 even when alternating with dusty-coloured or bright green porphyries and amygdaloids1 is probably an analogous circumstance to the scori of the blackish basalts being often bright red4 ,the ancient submarine orifices whence the porphyries and their fragments were ejected having been arranged in a band1 like most still active volcanoes1 accounts for the thickness1 the narrowness1 and linear extension of this formation4 #e ,this certainly seems to have taken place in some recent volcanic archipelagos1 as at the ,galapagos1 where numerous craters are exclusively formed of tuff and fragments of lava4 ,this whole great pile of rock has suffered much metamorphic action1 as is very obvious in the gradual formation and appearance of the crystals of albitic feldspar and of epidote--in the bending together of the fragments--in the appearance of a laminated structure in the feldspathic slate--and1 lastly1 in the disappearance of the planes of stratification1 which could sometimes be seen on the same mountain quite distinct in the upper part1 less and less plain on the flanks1 and quite obliterated at the base4 ,partly owing to this metamorphic action1 and partly to the close relationship in origin1 ,i have seen fragments of porphyries--taken from a metamorphosed conglomerate--from a neighbouring stream of lava--from the nucleus or centre 7as it appeared to me7 of the whole submarine volcano--and lastly from an intrusive mass of quite subsequent origin1 all of which were absolutely undistinguishable in external characters4 ,one other rock1 of plutonic origin1 and highly important in the history of the ,cordillera1 from having been injected in most of the great axes of elevation1 and from having apparently been instrumental in metamorphosing the superincumbent strata1 may be conveniently described in this preliminary discussion4 ,it has been called by some authors ,andesite3 it mainly consists of well-crystallised white albite#f 7as determined with the goniometer in numerous specimens both by ,professor ,miller and myself71 of less perfectly crystallised green hornblende1 often associated with much mica1 with chlorite and epidote1 and occasionally with a few grains of quartz3 in one instance in ,northern ,chile1 ,i found crystals of orthitic or potash feldspar1 mingled with those of albite4 ,where the mica and quartz are abundant1 the rock cannot be distinguished from granite2 and it may be called andesitic granite4 ,where these two minerals are quite absent1 and when1 as often then happens1 the crystals of albite are imperfect and blend together1 the rock may be called andesitic porphyry1 which bears nearly the same relation to andesitic granite that euritic porphyry does to common granite4 ,these andesitic rocks form mountain masses of a white colour1 which1 in their general outline and appearance--in their joints--in their occasionally including dark-coloured1 angular fragments1 apparently of some pre-existing rock--and in the great dikes branching from them into the superincumbent strata1 manifest a close and striking resemblance to masses of common granite and syenite3 ,i never1 however1 saw in these andesitic rocks1 those granitic veins of segregation which are so common in true granites4 ,we have seen that andesite occurs in three places in ,tierra del ,fuego2 in ,chile1 from ,s4 ,fernando to ,copiapo1 a distance of #dej miles1 ,i found it under most of the axes of elevation2 in a collection of specimens from the ,cordillera of ,lima in ,peru1 ,i immediately recognised it2 and ,erman#g states that it occurs in ,eastern ,kamtschatka4 ,from its wide range1 and from the important part it has played in the history of the ,cordillera1 ,i think this rock has well deserved its distinct name of ,andesite4 #f ,i here1 and elsewhere1 call by this name1 those feldspathic minerals which cleave like albite3 but it now appears 7,edin4 ,new ,phil4 ,journal41 vol4 xxiv1 p4 #aha7 that ,abich has analysed a mineral from the ,cordillera1 associated with hornblende and quartz 7probably the same rock with that here under discussion71 which cleaves like albite1 but which is a new and distinct kind1 called by him ,andesine4 ,it is allied to leucite1 with the greater proportion of its potash replaced by lime and soda4 ,this mineral seems scarcely distinguishable from albite1 except by analysis4 #g ,geograph4 ,journal1 vol4 ix1 p4 #eaj4 ,the few still active volcanoes in ,chile are confined to the central and loftiest ranges of the ,cordillera2 and volcanic matter1 such as appears to have been of subaerial eruption1 is everywhere rare4 ,according to ,meyen1#h there is a hill of pumice high up the valley of the ,maypu1 and likewise a trachytic formation at ,colina1 a village situated north of ,st4 ,jago4 ,close to this latter city1 there are two hills formed of a pale feldspathic porphyry1 remarkable from being doubly columnar1 great cylindrical columns being subdivided into smaller four- or five-sided ones2 and a third hillock 7,cerro ,blanco7 is formed of a fragmentary mass of rock1 which ,i believed to be of volcanic origin1 intermediate in character between the above feldspathic porphyry and common trachyte1 and containing needles of hornblende and granular oxide of iron4 ,near the ,baths of ,cauquenes1 between two short parallel lines of elevation1 where they are intersected by the valley1 there is a small1 though distinct volcanic district2 the rock is a dark grey 7andesitic7 trachyte1 which fuses into a greenish-grey bead1 and is formed of long crystals of fractured glassy albite 7judging from one measurement7 mingled with well-formed crystals1 often twin1 of augite4 ,the whole mass is vesicular1 but the surface is darker coloured and much more vesicular than any other part4 ,this trachyte forms a cliff-bounded1 horizontal1 narrow strip on the steep southern side of the valley1 at the height of four or five hundred feet above the river-bed2 judging from an apparently corresponding line of cliff on the northern side1 the valley must once have been filled up to this height by a field of lava4 ,on the summit of a lofty mountain some leagues higher up this same valley of the ,cachapual1 ,i found columnar pitchstone porphyritic with feldspar2 ,i do not suppose this rock to be of volcanic origin1 and only mention it here1 from its being intersected by masses and dikes of a vesicular rock1 approaching in character to trachyte2 in no other part of ,chile did ,i observe vesicular or amygdaloidal dikes1 though these are so common in ordinary volcanic districts4 #h 8,reise um ,erde10 ,th4 ,i1 ss4 #cch and #cfb4 ,passage of the ,andes by the ,portillo or ,pequenes ,pass4 ,although ,i crossed the ,cordillera only once by this pass1 and only once by that of the ,cumbre or ,uspallata 7presently to be described71 riding slowly and halting occasionally to ascend the mountains1 there are many circumstances favourable to obtaining a more faithful sketch of their structure than would at first be thought possible from so short an examination4 ,the mountains are steep and absolutely bare of vegetation2 the atmosphere is resplendently clear2 the stratification distinct2 and the rocks brightly and variously coloured3 some of the natural sections might be truly compared for distinctness to those coloured ones in geological works4 ,considering how little is known of the structure of this gigantic range1 to which ,i particularly attended1 most travellers having collected only specimens of the rocks1 ,i think my sketch-sections1 though necessarily imperfect1 possess some interest4 ,plate ,v sections 7between and #dda7 which ,i will now describe in detail1 is on a horizontal scale of a third of an inch to a nautical mile1 and on a vertical scale of one inch to a mile 7or #f1jjj feet74 ,the width of the range 7excluding a few outlying hillocks71 from the plain on which ,st4 ,jago the capital of ,chile stands1 to the ,pampas1 is sixty miles1 as far as ,i can judge from the maps1 which differ from each other and are all exceedingly imperfect4 ,the ,st4 ,jago plain at the mouth of the ,maypu1 ,i estimate from adjoining known points at #b1cjj feet1 and the ,pampas at #c1ejj feet1 both above the level of the sea4 ,the height of the ,pequenes line1 according to ,dr4 ,gillies1#i is #ac1baj feet2 and that of the ,portillo line 7both in the gaps where the road crosses them7 is #ad1cde feet2 the lowest part of the intermediate valley of ,tenuyan is #g1ecj feet--all above the level of the sea4 #i ,journal of ,nat4 and ,geograph4 ,science1 ,august #ahcj4 ,the ,cordillera here1 and indeed ,i believe throughout ,chile1 consist of several parallel1 anticlinal and uniclinal mountain-lines1 ranging north1 or north with a little westing1 and south4 ,some exterior and much lower ridges often vary considerably from this course1 projecting like oblique spurs from the main ranges3 in the district towards the ,pacific1 the mountains1 as before remarked1 extend in various directions1 even east and west4 ,in the main exterior lines1 the strata1 as also before remarked1 are seldom inclined at a high angle2 but in the central lofty ridges they are almost always highly inclined1 broken by many great faults1 and often vertical4 ,as far as ,i could judge1 few of the ranges are of great length3 and in the central parts of the ,cordillera1 ,i was frequently able to follow with my eye a ridge gradually becoming higher and higher1 as the stratification increased in inclination1 from one end where its height was trifling and its strata gently inclined to the other end where vertical strata formed snow-clad pinnacles4 ,even outside the main ,cordillera1 near the baths of ,cauquenes1 ,i observed one such case1 where a north and south ridge had its strata in the valley inclined at #cg1 and less than a mile south of it at #fg3 another parallel and similarly inclined ridge rose at the distance of about five miles1 into a lofty mountain with absolutely vertical strata4 ,within the ,cordillera1 the height of the ridges and the inclination of the strata often became doubled and trebled in much shorter distances than five miles2 this peculiar form of upheaval probably indicates that the stratified crust was thin1 and hence yielded to the underlying intrusive masses unequally1 at certain points on the lines of fissure4 ,the valleys1 by which the ,cordillera are drained1 follow the anticlinal or rarely synclinal troughs1 which deviate most from the usual north and south course2 or still more commonly those lines of faults or of unequal curvature 7that is1 lines with the strata on both hands dipping in the same direction1 but at a somewhat different angle7 which deviate most from a northerly course4 ,occasionally the torrents run for some distance in the north and south valleys1 and then recover their eastern or western course by bursting through the ranges at those points where the strata have been least inclined and the height consequently is less4 ,hence the valleys1 along which the roads run1 are generally zigzag2 and1 in drawing an east and west section1 it is necessary to contract greatly that which is actually seen on the road4 ,commencing at the western end of our section ,plate ,v where the ,r4 ,maypu debouches on the plain of ,st4 ,jago1 we immediately enter on the porphyritic conglomerate formation1 and in the midst of it find some hummocks ,a of granite and syenite1 which probably 7for ,i neglected to collect specimens7 belong to the andesitic class4 ,these are succeeded by some rugged hills ,b of dark-green1 crystalline1 feldspathic and in some parts slaty rocks1 which ,i believe belong to the altered clay-slate formation4 ,from this point1 great mountains of purplish and greenish1 generally thinly stratified1 highly porphyritic conglomerates1 including many strata of amygdaloidal and greenstone porphyries1 extend up the valley to the junction of the rivers ,yeso and ,volcan4 ,as the valley here runs in a very southerly course1 the width of the porphyritic conglomerate formation is quite conjectural2 and from the same cause1 ,i was unable to make out much about the stratification4 ,in most of the exterior mountains the dip was gentle and directed inwards2 and at only one spot ,i observed an inclination as high as #ej4 ,near the junction of the ,r4 ,colorado with the main stream1 there is a hill of whitish1 brecciated1 partially decomposed feldspathic porphyry1 having a volcanic aspect but not being really of that nature3 at ,tolla1 however1 in this valley1 ,dr4 ,meyen#aj met with a hill of pumice containing mica4 ,at the junction of the ,yeso and ,volcan ,d there is an extensive mass1 in white conical hillocks1 of andesite1 containing some mica1 and passing either into andesitic granite1 or into a spotted1 semi-granular mixture of albitic 787 feldspar and hornblende3 in the midst of this formation ,dr4 ,meyen found true trachyte4 ,the andesite is covered by strata of dark-coloured1 crystalline1 obscurely porphyritic rocks1 and above them by the ordinary porphyritic conglomerates1--the strata all dipping away at a small angle from the underlying mass4 ,the surrounding lofty mountains appear to be entirely composed of the porphyritic conglomerate1 and ,i estimated its thickness here at between six and seven thousand feet4 #aj 8,reise um ,erde10 ,th4 ,i1 ss4 #cch1 #cda47 ,beyond the junction of the ,yeso and ,volcan1 the porphyritic strata appear to dip towards the hillocks of andesite at an angle of #dj2 but at some distant points on the same ridge they are bent up and vertical4 ,following the valley of the ,yeso1 trending ,n4,e4 7and therefore still unfavourable for our transverse section71 the same porphyritic conglomerate formation is prolonged to near the ,cuestadel ,indio1 situated at the western end of the basin 7like a drained lake7 of ,yeso4 ,some way before arriving at this point1 distant lofty pinnacles capped by coloured strata belonging to the great gypseous formation could first be seen4 ,from the summit of the ,cuesta1 looking southward1 there is a magnificent sectional view of a mountain-mass1 at least #b1jjj feet in thickness ,e1 of fine andesite granite 7containing much black mica1 a little chlorite and quartz71 which sends great white dikes far into the superincumbent1 dark-coloured1 porphyritic conglomerates4 ,at the line of junction the two formations are wonderfully interlaced together3 in the lower part of the porphyritic conglomerate1 the stratification has been quite obliterated1 whilst in the upper part it is very distinct1 the beds composing the crests of the surrounding mountains being inclined at angles of between #gj and #hj degrees1 and some being even vertical4 ,on the northern side of the valley1 there is a great corresponding mass of andesitic granite1 which is encased by porphyritic conglomerate1 dipping both on the western and eastern sides1 at about #hj to west1 but on the eastern side with the tips of the strata bent in such a manner1 as to render it probable that the whole mass has been on that side thrown over and inverted4 ,in the valley basin of the ,yeso1 which ,i estimated at #g1jjj feet above the level of the sea1 we first reach at ,f the gypseous formation4 ,its thickness is very great4 ,it consists in most parts of snow-white1 hard1 compact gypsum1 which breaks with a saccharine fracture1 having translucent edges2 under the blowpipe gives out much vapour2 it frequently includes nests and exceedingly thin layers of crystallised1 blackish carbonate of lime4 ,large1 irregularly shaped concretions 7externally still exhibiting lines of aqueous deposition7 of blackish-grey1 but sometimes white1 coarsely and brilliantly crystallised1 hard anhydrite1 abound within the common gypsum4 ,hillocks1 formed of the hardest and purest varieties of the white gypsum1 stand up above the surrounding parts1 and have their surfaces cracked and marked1 just like newly baked bread4 ,there is much pale brown1 soft argillaceous gypsum2 and there were some intercalated green beds which ,i had not time to reach4 ,i saw only one fragment of selenite or transparent gypsum1 and that perhaps may have come from some subsequently formed vein4 ,from the mineralogical characters here given1 it is probable that these gypseous beds have undergone some metamorphic action4 ,the strata are much hidden by detritus1 but they appeared in most parts to be highly inclined2 and in an adjoining lofty pinnacle they could be distinctly seen bending up1 and becoming vertical1 conformably with the underlying porphyritic conglomerate4 ,in very many parts of the great mountain-face ,f1 composed of thin gypseous beds1 there were innumerable masses1 irregularly shaped and not like dikes1 yet with well-defined edges1 of an imperfectly granular1 pale greenish1 or yellowish-white rock1 essentially composed of feldspar1 with a little chlorite or hornblende1 epidote1 iron-pyrites1 and ferruginous powder3 ,i believe that these curious trappean masses have been injected from the not far distant mountain-mass ,e of andesite whilst still fluid1 and that owing to the softness of the gypseous strata they have not acquired the ordinary forms of dikes4 ,subsequently to the injection of these feldspathic rocks1 a great dislocation has taken place2 and the much shattered gypseous strata here overlie a hillock ,g1 composed of vertical strata of impure limestone and of black highly calcareous shale including threads of gypsum3 these rocks1 as we shall presently see1 belong to the upper parts of the gypseous series1 and hence must here have been thrown down by a vast fault4 ,proceeding up the valley-basin of the ,yeso1 and taking our section sometimes on one hand and sometimes on the other1 we come to a great hill of stratified porphyritic conglomerate ,h dipping at #de to the west2 and a few hundred yards farther on1 we have a bed between three or four hundred feet thick of gypsum ,i dipping eastward at a very high angle3 here then we have a fault and anticlinal axis4 ,on the opposite side of the valley1 a vertical mass of red conglomerate1 conformably underlying the gypsum1 appears gradually to lose its stratification and passes into a mountain of porphyry4 ,the gypsum ,i is covered by a bed ,k1 at least #a1jjj feet in thickness1 of a purplish-red1 compact1 heavy1 fine-grained sandstone or mudstone1 which fuses easily into a white enamel1 and is seen under a lens to contain triturated crystals4 ,this is succeeded by a bed ,l1 #a1jjj feet thick 7,i believe ,i understate the thickness7 of gypsum1 exactly like the beds before described2 and this again is capped by another great bed ,m of purplish-red sandstone4 ,all these strata dip eastward2 but the inclination becomes less and less1 as we leave the first and almost vertical bed ,i of gypsum4 ,leaving the basin-plain of ,yeso1 the road rapidly ascends1 passing by mountains composed of the gypseous and associated beds1 with their stratification greatly disturbed and therefore not easily intelligible3 hence this part of the section has been left uncoloured4 ,shortly before reaching the great ,pequenes ridge1 the lowest stratum visible ,n is a red sandstone or mudstone1 capped by a vast thickness of black1 compact1 calcareous1 shaly rock ,o1 which has been thrown into four lofty1 though small ridges3 looking northward1 the strata in these ridges are seen gradually to rise in inclination1 becoming in some distant pinnacles absolutely vertical4 ,the ridge of ,pequenes1 which divides the waters flowing into the ,pacific and ,atlantic ,oceans1 extends in a nearly ,n4,n4,w4 and ,s4,s4,e4 line2 its strata dip eastward at an angle of between #cj and #de1 but in the higher peaks bending up and becoming almost vertical4 ,where the road crosses this range1 the height is #ac1baj feet above the sea-level1 and ,i estimated the neighbouring pinnacles at from fourteen to fifteen thousand feet4 ,the lowest stratum visible in this ridge is a red stratified sandstone ,p2 on it are superimposed two great masses ,q and ,s of black1 hard1 compact1 even having a conchoidal fracture1 calcareous1 more or less laminated shale1 passing into limestone3 this rock contains organic remains1 presently to be enumerated4 ,the compacter varieties fuse easily in a white glass2 and this ,i may add is a very general character with all the sedimentary beds in the ,cordillera3 although this rock when broken is generally quite black1 it everywhere weathers into an ash-grey tint4 ,between these two great masses ,q and ,s1 a bed ,r of gypsum is interposed1 about three hundred feet in thickness1 and having the same characters as heretofore described4 ,i estimated the total thickness of these three beds ,q1 ,r1 ,s at nearly three thousand feet2 and to this must be added1 as will be immediately seen1 a great overlying mass of red sandstone4 ,in descending the eastern slope of this great central range1 the strata1 which in the upper part dip eastward at about an angle of #dj1 become more and more curved1 till they are nearly vertical2 and a little further onwards there is seen on the further side of a ravine1 a thick mass of strata of bright red sandstone ,t1 with their upper extremities slightly curved1 showing that they were once conformably prolonged over the beds ,s3 on the southern and opposite side of the road1 this red sandstone and the underlying black shaly rocks stand vertical1 and in actual juxtaposition4 ,continuing to descend1 we come to a synclinal valley filled with rubbish1 beyond which we have the red sandstone ,t#b corresponding with ,t1 and now dipping1 as is seen both north and south of the road1 at #de to the west2 and under it1 the beds ,s#b1 ,r#b1 ,q#b1 and ,i believe ,p#b in corresponding order and of similar composition1 with those on the western flank of the ,pequenes range1 but dipping westward4 ,close to the synclinal valley the dip of these strata is #de1 but at the eastern or farther end of the series it increases to #fj4 ,here the great gypseous formation abruptly terminates1 and is succeeded eastward by a pile of more modern strata4 ,considering how violently these central ranges have been dislocated1 and how very numerous dikes are in the exterior and lower parts of the ,cordillera1 it is remarkable that ,i did not here notice a single dike4 ,the prevailing rock in this neighbourhood is the black1 calcareous1 compact shale1 whilst in the valley-basin of the ,yeso the purplish red sandstone or mudstone predominates1--both being associated with gypseous strata of exactly the same nature4 ,it would be very difficult to ascertain the relative superposition of these several masses1 for we shall afterwards see in the ,cumbre ,pass that the gypseous and intercalated beds are lens-shaped1 and that they thin out1 even where very thick1 and disappear in short horizontal distances3 it is quite possible that the black shales and red sandstones may be contemporaneous1 but it is more probable that the former compose the uppermost parts of the series4 ,the fossils above alluded to in the black calcareous shales are few in number1 and are in an imperfect condition2 they consist1 as named for me by ,m4 d',orbigny1 of3-- ,ammonite1 indeterminable1 near to ,a4 recticostatus1 d',orbigny1 8,pal4 ,franc40 7,neocomian formation74 ,grypha1 near to ,g4 ,couloni 7,neocomian formations of ,france and ,neufchtel74 ,natica1 indeterminable4 ,cyprina rostrata1 d',orbigny1 8,pal4 ,franc40 7,neocomian formation74 ,rostellaria angulosa 7871 d',orbigny1 8,pal4 de l',amr4 ,mer40 ,terebratula 7874 ,some of the fragments of ,ammonites were as thick as a man's arm3 the ,grypha is much the most abundant shell4 ,these fossils ,m4 d',orbigny considers as belonging to the ,neocomian stage of the ,cretaceous system4 ,dr4 ,meyen1#aa who ascended the valley of the ,rio ,volcan1 a branch of the ,yeso1 found a nearly similar1 but apparently more calcareous formation1 with much gypsum1 and no doubt the equivalent of that here described3 the beds were vertical1 and were prolonged up to the limits of perpetual snow2 at the height of #i1jjj feet above the sea1 they abounded with fossils1 consisting1 according to ,von ,buch1#ab of3-- ,exogyra 7,grypha7 ,couloni1 absolutely identical with specimens from the ,jura and ,south of ,france4 ,trigonia costata1 identical with those found in the upper ,jurassic beds at ,hildesheim4 ,pecten striatus1 identical with those found in the upper ,jurassic beds at ,hildesheim4 ,cuculla1 corresponding in form to ,c4 longirostris1 so frequent in the upper ,jurassic beds of ,westphalia4 ,ammonites resembling ,a4 biplex4 #aa 8,reise um ,erde10 etc41 ,th4 ,i1 s4 #cee4 #ab 8,descript4 ,phys4 des ,iles ,canaries10 p4 #dga4 ,von ,buch concludes that this formation is intermediate between the limestone of the ,jura and the chalk1 and that it is analogous with the uppermost ,jurassic beds forming the plains of ,switzerland4 ,hence ,m4 ,d',orbigny and ,von ,buch1 under different terms1 compare these fossils to those from the same late stage in the secondary formations of ,europe4 ,some of the fossils which ,i collected were found a good way down the western slope of the main ridge1 and hence must originally have been covered up by a great thickness of the black shaly rock1 independently of the now denuded1 thick1 overlying masses of red sandstone4 ,i neglected at the time to estimate how many hundred or rather thousand feet thick the superincumbent strata must have been3 and ,i will not now attempt to do so4 ,this1 however1 would have been a highly interesting point1 as indicative of a great amount of subsidence1 of hich we shall hereafter find in other parts of the ,cordillera analogous evidence during this same period4 ,the altitude of the ,peuquenes ,range1 considering its not great antiquity1 is very remarkable2 many of the fossils were embedded at the height of #ac1baj feet1 and the same beds are prolonged up to at least from fourteen to fifteen thousand feet above the level of the sea4 ,the ,portillo or ,eastern ,chain4--,the valley of ,tenuyan1 separating the ,peuquenes and ,portillo lines1 is1 as estimated by ,dr4 ,gillies and myself1 about twenty miles in width2 the lowest part1 where the road crosses the river1 being #g1ejj feet above the sea-level4 ,the pass on the ,portillo line is #ad1cfe feet high 7#a1ajj feet higher than that on the ,peuquenes71 and the neighbouring pinnacles must1 ,i conceive1 rise to nearly #af1jjj feet above the sea4 ,the river draining the intermediate valley of ,tenuyan1 passes through the ,portillo line4 ,to return to our section3--shortly after leaving the lower beds ,p#b of the gypseous formation1 we come to grand masses of a coarse1 red conglomerate ,v1 totally unlike any strata hitherto seen in the ,cordillera4 ,this conglomerate is distinctly stratified1 some of the beds being well defined by the greater size of the pebbles3 the cement is calcareous and sometimes crystalline1 though the mass shows no signs of having been metamorphosed4 ,the included pebbles are either perfectly or only partially rounded3 they consist of purplish sandstones1 of various porphyries1 of brownish limestone1 of black calcareous1 compact shale precisely like that in situ in the ,peuquenes range1 and containing some of the same fossil shells2 also very many pebbles of quartz1 some of micaceous schist1 and numerous1 broken1 rounded crystals of a reddish orthitic or potash feldspar 7as determined by ,professor ,miller71 and these from their size must have been derived from a coarse-grained rock1 probably granite4 ,from this feldspar being orthitic1 and even from its external appearance1 ,i venture positively to affirm that it has not been derived from the rocks of the western ranges2 but1 on the other hand1 it may well have come1 together with the quartz and metamorphic schists1 from the eastern or ,portillo line1 for this line mainly consists of coarse orthitic granite4 ,the pebbles of the fossiliferous slate and of the purple sandstone1 certainly have been derived from the ,peuquenes or western ranges4 ,the road crosses the valley of ,tenuyan in a nearly east and west line1 and for several miles we have on both hands the conglomerate1 everywhere dipping west and forming separate great mountains4 ,the strata1 where first met with1 after leaving the gypseous formation1 are inclined westward at an angle of only #bj1 which further on increases to about #de4 ,the gypseous strata1 as we have seen1 are also inclined westward3 hence1 when looking from the eastern side of the valley towards the ,peuquenes range1 a most deceptive appearance is presented1 as if the newer beds of conglomerate dipped directly under the much older beds of the gypseous formation4 ,in the middle of the valley1 a bold mountain of unstratified lilac-coloured porphyry 7with crystals of hornblende7 projects2 and further on1 a little south of the road1 there is another mountain1 with its strata inclined at a small angle eastwards1 which in its general aspect and colour1 resembles the porphyritic conglomerate formation1 so rare on this side of the ,peuquenes line and so grandly developed throughout the western ranges4 ,the conglomerate is of great thickness3 ,i do not suppose that the strata forming the separate mountain-masses ,v1 ,v1 ,v have ever been prolonged over each other1 but that one mass has been broken up by several1 distinct1 parallel1 uniclinal lines of elevation4 ,judging therefore of the thickness of the conglomerate1 as seen in the separate mountain-masses1 ,i estimated it at least from one thousand five hundred to two thousand feet4 ,the lower beds rest conformably on some singularly coloured1 soft strata ,w1 which ,i could not reach to examine2 and these again rest conformably on a thick mass of micaceous1 thinly laminated1 siliceous sandstone ,x1 associated with a little black clay-slate4 ,these lower beds are traversed by several dikes of decomposing porphyry4 ,the laminated sandstone is directly superimposed on the vast masses of granite ,y1 ,y which mainly compose the ,portillo range4 ,the line of junction between this latter rock1 which is of a bright red colour1 and the whitish sandstone was beautifully distinct2 the sandstone being penetrated by numerous1 great1 tortuous dikes branching from the granite1 and having been converted into a granular quartz rock 7singularly like that of the ,falkland ,islands71 containing specks of an ochrey powder1 and black crystalline atoms1 apparently of imperfect mica4 ,the quartzose strata in one spot were folded into a regular dome4 ,the granite which composes the magnificent bare pinnacles and the steep western flank of the ,portillo chain1 is of a brick-red colour1 coarsely crystallised1 and composed of orthitic or potash feldspar1 quartz1 and imperfect mica in small quantity1 sometimes passing into chlorite4 ,these minerals occasionally assume a laminar or foliated arrangement4 ,the fact of the feldspar being orthitic in this range1 is very remarkable1 considering how rare1 or rather1 as ,i believe1 entirely absent1 this mineral is throughout the western ranges1 in which soda-feldspar1 or at least a variety cleaving like albite1 is so extremely abundant4 ,in one spot on the western flank1 and on the eastern flank near ,los ,manantiales and near the crest1 ,i noticed some great masses of a whitish granite1 parts of it fine- grained1 and parts containing large crystals of feldspar2 ,i neglected to collect specimens1 so ,i do not know whether this feldspar is also orthitic1 though ,i am inclined to think so from its general appearance4 ,i saw also some syenite and one mass which resembled andesite1 but of which ,i likewise neglected to collect specimens4 ,from the manner in which the whitish granites formed separate mountain-masses in the midst of the brick-red variety1 and from one such mass near the crest being traversed by numerous veins of flesh-coloured and greenish eurite 7into which ,i occasionally observed the brick-red granite insensibly passing71 ,i conclude that the white granites probably belong to an older formation1 almost overwhelmed and penetrated by the red granite4 ,on the crest ,i saw also1 at a short distance1 some coloured stratified beds1 apparently like those ,w at the western base1 but was prevented examining them by a snowstorm3 ,mr4 ,caldcleugh1#ac however1 collected here specimens of ribboned jasper1 magnesian limestone1 and other minerals4 ,a little way down the eastern slope a few fragments of quartz and mica-slate are met with2 but the great formation of this latter rock ,z1 which covers up much of the eastern flank and base of the ,portillo range1 cannot be conveniently examined until much lower down at a place called ,mal ,paso4 ,the mica-schist here consists of thick layers of quartz1 with intervening folia of finely-scaly mica1 often passing into a substance like black glossy clay-slate3 in one spot1 the layers of the quartz having disappeared1 the whole mass became converted into glossy clay-slate4 ,where the folia were best defined1 they were inclined at a high angle westward1 that is1 towards the range4 ,the line of junction between the dark mica-slate and the coarse red granite was most clearly distinguishable from a vast distance3 the granite sent many small veins into the mica-slate1 and included some angular fragments of it4 ,as the sandstone on the western base has been converted by the red granite into a granular quartz-rock1 so this great formation of mica-schist may possibly have been metamorphosed at the same time and by the same means2 but ,i think it more probable1 considering its more perfect metamorphic character and its well-pronounced foliation1 that it belongs to an anterior epoch1 connected with the white granites3 ,i am the more inclined to this view1 from having found at the foot of the range the mica-schist surrounding a hummock ,y#b1 exclusively composed of white granite4 ,near ,los ,arenales1 the mountains on all sides are composed of the mica-slate2 and looking backwards from this point up to the bare gigantic peaks above1 the view was eminently interesting4 ,the colours of the red granite and the black mica-slate are so distinct1 that with a bright light these rocks could be readily distinguished even from the ,pampas1 at a level of at least #i1jjj feet below4 ,the red granite1 from being divided by parallel joints1 has weathered into sharp pinnacles1 on some of which1 even on some of the loftiest1 little caps of mica-schist could be clearly seen3 here and there isolated patches of this rock adhered to the mountain-flanks1 and these often corresponded in height and position on the opposite sides of the immense valleys4 ,lower down the schist prevailed more and more1 with only a few quite small points of granite projecting through4 ,looking at the entire eastern face of the ,portillo range1 the red colour far exceeds in area the black2 yet it was scarcely possible to doubt that the granite had once been almost wholly encased by the mica-schist4 #ac 8,travels10 etc41 vol4 i1 p4 #cjh4 ,at ,los ,arenales1 low down on the eastern flank1 the mica-slate is traversed by several closely adjoining1 broad dikes1 parallel to each other and to the foliation of the schist4 ,the dikes are formed of three different varieties of rock1 of which a pale brown feldspathic porphyry with grains of quartz was much the most abundant4 ,these dikes with their granules of quartz1 as well as the mica-schist itself1 strikingly resemble the rocks of the ,chonos ,archipelago4 ,at a height of about twelve hundred feet above the dikes1 and perhaps connected with them1 there is a range of cliffs formed of successive lava-streams ,,aa1 between three and four hundred feet in thickness1 and in places finely columnar4 ,the lava consists of dark-greyish1 harsh rocks1 intermediate in character between trachyte and basalt1 containing glassy feldspar1 olivine1 and a little mica1 and sometimes amygdaloidal with zeolite3 the basis is either quite compact1 or crenulated with air-vesicles arranged in lamin4 ,the streams are separated from each other by beds of fragmentary brown scori1 firmly cemented together1 and including a few well-rounded pebbles of lava4 ,from their general appearance1 ,i suspect that these lava-streams flowed at an ancient period under the pressure of the sea1 when the ,atlantic covered the ,pampas and washed the eastern foot of the ,cordillera4#ad ,on the opposite and northern side of the valley there is another line of lava-cliffs at a corresponding height2 the valley between being of considerable breadth1 and as nearly as ,i could estimate #a1ejj feet in depth4 ,this field of lava is confined on both sides by the mountains of mica-schist1 and slopes down rapidly but irregularly to the edge of the ,pampas1 where1 having a thickness of about two hundred feet1 it terminates against a little range of claystone porphyry4 ,the valley in this lower part expands into a bay-like1 gentle slope1 bordered by the cliffs of lava1 which must certainly once have extended across this wide expanse4 ,the inclination of the streams from ,los ,arenales to the mouth of the valley is so great1 that at the time 7though ignorant of ,m4 ,elie de ,beaumont's researches on the extremely small slope over which lava can flow1 and yet retain a compact structure and considerable thickness7 ,i concluded that they must subsequently to their flowing have been upheaved and tilted from the mountains2 of this conclusion ,i can now entertain not the smallest doubt4 #ad ,this conclusion might1 perhaps1 even have been anticipated1 from the general rarity of volcanic action1 except near the sea or large bodies of water4 ,conformably with this rule1 at the present day1 there are no active volcanoes on this eastern side of the ,cordillera2 nor are severe earthquakes experienced here4 ,at the mouth of the valley1 within the cliffs of the above lava-field1 there are remnants1 in the form of separate small hillocks and of lines of low cliffs1 of a considerable deposit of compact white tuff 7quarried for filtering-stones71 composed of broken pumice1 volcanic crystals1 scales of mica1 and fragments of lava4 ,this mass has suffered much denudation2 and the hard mica-schist has been deeply worn1 since the period of its deposition2 and this period must have been subsequent to the denudation of the basaltic lava-streams1 as attested by their encircling cliffs standing at a higher level4 ,at the present day1 under the existing arid climate1 ages might roll past without a square yard of rock of any kind being denuded1 except perhaps in the rarely moistened drainage-channel of the valley4 ,must we then look back to that ancient period1 when the waves of the sea beat against the eastern foot of the ,cordillera1 for a power sufficient to denude extensively1 though superficially1 this tufaceous deposit1 soft although it be8 ,there remains only to mention some little water-worn hillocks ,,bb1 a few hundred feet in height1 and mere mole-hills compared with the gigantic mountains behind them1 which rise out of the sloping1 shingle-covered margin of the ,pampas4 ,the first little range is composed of a brecciated purple porphyritic claystone1 with obscurely marked strata dipping at #gj to the ,s4,w42 the other ranges consist of--a pale-coloured feldspathic porphyry1--a purple claystone porphyry with grains of quartz1--and a rock almost exclusively composed of brick-red crystals of feldspar4 ,these outermost small lines of elevation extend in a ,n4,w4 by ,w4 and ,s4,e4 by ,s4 direction4 ,concluding remarks on the ,portillo range4--,when on the ,pampas and looking southward1 and whilst travelling northward1 ,i could see for very many leagues the red granite and dark mica-schist forming the crest and eastern flank of the ,portillo line4 ,this great range1 according to ,dr4 ,gillies1 can be traced with little interruption for #adj miles southward to the ,r4 ,diamante1 where it unites with the western ranges3 northward1 according to this same author1 it terminates where the ,r4 ,mendoza debouches from the mountains2 but a little further north in the eastern part of the ,cumbre section1 there are1 as we shall hereafter see1 some mountain-masses of a brick-red porphyry1 the last injected amidst many other porphyries1 and having so close an analogy with the coarse red granite of the ,portillo line1 that ,i am tempted to believe that they belong to the same axis of injection2 if so1 the ,portillo line is at least #bjj miles in length4 ,its height1 even in the lowest gap in the road1 is #ad1cfe feet1 and some of the pinnacles apparently attain an elevation of about #af1jjj feet above the sea4 ,the geological history of this grand chain appears to me eminently interesting4 ,we may safely conclude1 that at a former period the valley of ,tenuyan existed as an arm of the sea1 about twenty-miles in width1 bordered on one hand by a ridge or chain of islets of the black calcareous shales and purple sandstones of the gypseous formation2 and on the other hand1 by a ridge or chain of islets composed of mica-slate1 white granite1 and perhaps to a partial extent of red granite4 ,these two chains1 whilst thus bordering the old sea-channel1 must have been exposed for a vast lapse of time to alluvial and littoral action1 during which the rocks were shattered1 the fragments rounded1 and the strata of conglomerate accumulated to a thickness of at least fifteen hundred or two thousand feet4 ,the red orthitic granite now forms1 as we have seen1 the main part of the ,portillo chain3 it is injected in dikes not only into the mica-schist and white granites1 but into the laminated sandstone1 which it has metamorphosed1 and which it has thrown off1 together with the conformably overlying coloured beds and stratified conglomerate1 at an angle of forty-five degrees4 ,to have thrown off so vast a pile of strata at this angle1 is a proof that the main part of the red granite 7whether or not portions1 as perhaps is probable1 previously existed7 was injected in a liquified state after the accumulation both of the laminated sandstone and of the conglomerate2 this conglomerate1 we know1 was accumulated1 not only after the deposition of the fossiliferous strata of the ,peuquenes line1 but after their elevation and long-continued denudation3 and these fossiliferous strata belong to the early part of the ,cretaceous system4 ,late1 therefore1 in a geological sense1 as must be the age of the main part of the red granite1 ,i can conceive nothing more impressive than the eastern view of this great range1 as forcing the mind to grapple with the idea of the thousands of thousands of years requisite for the denudation of the strata which originally encased it1--for that the fluidified granite was once encased1 its mineralogical composition and structure1 and the bold conical shape of the mountain-masses1 yield sufficient evidence4 ,of the encasing strata we see the last vestiges in the coloured beds on the crest1 in the little caps of mica-schist on some of the loftiest pinnacles1 and in the isolated patches of this same rock at corresponding heights on the now bare and steep flanks4 ,the lava-streams at the eastern foot of the ,portillo are interesting1 not so much from the great denudation which they have suffered at a comparatively late period as from the evidence they afford by their inclination taken conjointly with their thickness and compactness1 that after the great range had assumed its present general outline1 it continued to rise as an axis of elevation4 ,the plains extending from the base of the ,cordillera to the ,atlantic show that the continent has been upraised in mass to a height of #c1ejj feet1 and probably to a much greater height1 for the smooth shingle-covered margin of the ,pampas is prolonged in a gentle unbroken slope far up many of the great valleys4 ,nor let it be assumed that the ,peuquenes and ,portillo ranges have undergone only movements of elevation2 for we shall hereafter see1 that the bottom of the sea subsided several thousand feet during the deposition of strata1 occupying the same relative place in the ,cordillera1 with those of the ,peuquenes ridge2 moreover1 we shall see from the unequivocal evidence of buried upright trees1 that at a somewhat later period1 during the formation of the ,uspallata chain1 which corresponds geographically with that of the ,portillo1 there was another subsidence of many thousand feet3 here1 indeed1 in the valley of ,tenuyan1 the accumulation of the coarse stratified conglomerate to a thickness of fifteen hundred or two thousand feet1 offers strong presumptive evidence of subsidence2 for all existing analogies lead to the belief that large pebbles can be transported only in shallow water1 liable to be affected by currents and movements of undulation--and if so1 the shallow bed of the sea on which the pebbles were first deposited must necessarily have sunk to allow of the accumulation of the superincumbent strata4 ,what a history of changes of level1 and of wear and tear1 all since the age of the latter secondary formations of ,europe1 does the structure of this one great mountain-chain reveal6 ,passage of the ,andes by the ,cumbre or ,uspallata ,pass4 ,this ,pass crosses the ,andes about sixty miles north of that just described3 the section given in ,plate ,v1 ,section #a_/b1 see map page #ddj is on the same scale as before1 namely1 at one-third of an inch to a mile in distance1 and one inch to a mile 7or #f1jjj feet7 in height4 ,like the last section1 it is a mere sketch1 and cannot pretend to accuracy1 though made under favourable circumstances4 ,we will commence as before1 with the western half1 of which the main range bears the name of the ,cumbre 7that is the ,ridge71 and corresponds to the ,peuquenes line in the former section2 as does the ,uspallata range1 though on a much smaller scale1 to that of the ,portillo4 ,near the point where the river ,aconcagua debouches on the basin plain of the same name1 at a height of about two thousand three hundred feet above the sea1 we meet with the usual purple and greenish porphyritic claystone conglomerate4 ,beds of this nature1 alternating with numerous compact and amygdaloidal porphyries1 which have flowed as submarine lavas1 and associated with great mountain-masses of various1 injected1 non-stratified porphyries1 are prolonged the whole distance up to the ,cumbre or central ridge4 ,one of the commonest stratified porphyries is of a green colour1 highly amygdaloidal with the various minerals described in the preliminary discussion1 and including fine tabular crystals of albite4 ,the mountain-range north 7often with a little westing7 and south4 ,the stratification1 wherever ,i could clearly distinguish it1 was inclined westward or towards the ,pacific1 and1 except near the ,cumbre1 seldom at angles above #be4 ,only at one spot on this western side1 on a lofty pinnacle not far from the ,cumbre1 ,i saw strata apparently belonging to the gypseous formation1 and conformably capping a pile of stratified porphyries4 ,hence1 both in composition and in stratification1 the structure of the mountains on this western side of the divortium aquarum1 is far more simple than in the corresponding part of the ,peuquenes section4 ,in the porphyritic claystone conglomerate1 the mechanical structure and the planes of stratification have generally been much obscured and even quite obliterated towards the base of the series1 whilst in the upper parts1 near the summits of the mountains1 both are distinctly displayed4 ,in these upper portions the porphyries are generally lighter coloured4 ,in three places ,x1 ,y1 ,z masses of andesite are exposed3 at ,y1 this rock contained some quartz1 but the greater part consisted of andesitic porphyry1 with only a few well-developed crystals of albite1 and forming a great white mass1 having the external aspect of granite1 capped by much dark unstratified porphyry4 ,in many parts of the mountains1 there are dikes of a green colour1 and other white ones1 which latter probably spring from underlying masses of andesite4 ,the ,cumbre1 where the road crosses it1 is1 according to ,mr4 ,pentland1 #ab1ded feet above the sea2 and the neighbouring peaks1 composed of dark purple and whitish porphyries1 some obscurely stratified with a westerly dip1 and others without a trace of stratification1 must exceed #ac1jjj feet in height4 ,descending the eastern slope of the ,cumbre1 the structure becomes very complicated1 and generally differs on the two sides of the east and west line of road and section4 ,first we come to a great mass ,a of nearly vertical1 singularly contorted strata1 composed of highly compact red sandstones1 and of often calcareous conglomerates1 and penetrated by green1 yellow1 and reddish dikes2 but ,i shall presently have an opportunity of describing in some detail an analogous pile of strata4 ,these vertical beds are abruptly succeeded by others ,b1 of apparently nearly the same nature but more metamorphosed1 alternating with porphyries and limestones2 these dip for a short space westward1 but there has been here an extraordinary dislocation1 which1 on the north side of the road1 appears to have determined the excavation of the north and south valley of the ,r4 de las ,cuevas4 ,on this northern side of the road1 the strata ,b are prolonged till they come in close contact with a jagged lofty mountain ,d of dark-coloured1 unstratified1 intrusive porphyry1 where the beds have been more highly inclined and still more metamorphosed4 ,this mountain of porphyry seems to form a short axis of elevation1 for south of the road in its line there is a hill ,c of porphyritic conglomerate with absolutely vertical strata4 ,we now come to the gypseous formation3 ,i will first describe the structure of the several mountains1 and then give in one section a detailed account of the nature of the rocks4 ,on the north side of the road1 which here runs in an east and west valley1 the mountain of porphyry ,d is succeeded by a hill ,e formed of the upper gypseous strata tilted1 at an angle of between #gj and #hj to the west1 by a uniclinal axis of elevation which does not run parallel to the other neighbouring ranges1 and which is of short length2 for on the south side of the valley its prolongation is marked only by a small flexure in a pile of strata inclined by a quite separate axis4 ,a little further on the north and south valley of ,horcones enters at right angles our line of section2 its western side is bounded by a hill of gypseous strata ,f dipping westward at about #de1 and its eastern side by a mountain of similar strata ,g inclined westward at #gj1 and superimposed by an oblique fault on another mass of the same strata ,h1 also inclined westward1 but at an angle of about #cj3 the complicated relation of these three masses ,f1 ,g1 ,h is explained by the structure of a great mountain-range lying some way to the north1 in which a regular anticlinal axis 7represented in the section by dotted lines7 is seen1 with the strata on its eastern side again bending up and forming a distinct uniclinal axis1 of which the beds marked ,h form the lower part4 ,this great uniclinal line is intersected1 near the ,puente del ,inca1 by the valley along which the road runs1 and the strata composing it will be immediately described4 ,on the south side of the road1 in the space corresponding with the mountains ,e1 ,f1 and ,g1 the strata everywhere dip westward generally at an angle of #cj1 occasionally mounting up to #de1 but not in an unbroken line1 for there are several vertical faults1 forming separate uniclinal masses1 all dipping in the same direction1--a form of elevation common in the ,cordillera4 ,we thus see that within a narrow space1 the gypseous strata have been upheaved and crushed together by a great uniclinal1 anticlinal1 and one lesser uniclinal line ,e of elevation2 and that between these three lines and the ,cumbre1 in the sandstones1 conglomerates and porphyritic formation1 there have been at least two or three other great elevatory axes4 ,the uniclinal axis ,i intersected near the ,puente del ,inca#ae 7of which the strata at ,h form a part7 ranges ,n4 by ,w4 and ,s4 by ,e41 forming a chain of mountains1 apparently little inferior in height to the ,cumbre3 the strata1 as we have seen1 dip at an average angle of #cj to the west4 ,the flanks of the mountains are here quite bare and steep1 affording an excellent section2 so that ,i was able to inspect the strata to a thickness of about #d1jjj feet1 and could clearly distinguish their general nature for #a1jjj feet higher1 making a total thickness of #e1jjj feet1 to which must be added about #a1jjj feet of the inferior strata seen a little lower down the valley1 ,i will describe this one section in detail1 beginning at the bottom4 #ae ,at this place1 there are some hot and cold springs1 the warmest having a temperature1 according to ,lieutenant ,brand 78,travels10 p4 #bdj71 of #ia2 they emit much gas4 ,according to ,mr4 ,brande1 of the ,royal ,institution1 ten cubical inches contain forty-five grains of solid matter1 consisting chiefly of salt1 gypsum1 carbonate of lime1 and oxide of iron4 ,the water is charged with carbonic acid and sulphuretted hydrogen4 ,these springs deposit much tufa in the form of spherical balls4 ,they burst forth1 as do those of ,cauquenes1 and probably those of ,villa ,vicencio1 on a line of elevation4 #ast4 ,the lowest mass is the altered clay-slate described in the preliminary discussion1 and which in this line of section was here first met with4 ,lower down the valley1 at the ,r4 de las ,vacas1 ,i had a better opportunity of examining it2 it is there in some parts well characterised1 having a distinct1 nearly vertical1 tortuous cleavage1 ranging ,n4,w4 and ,s4,e41 and intersected by quartz veins3 in most parts1 however1 it is crystalline and feldspathic1 and passes into a true greenstone often including grains of quartz4 ,the clay-slate1 in its upper half1 is frequently brecciated1 the embedded angular fragments being of nearly the same nature with the paste4 #bnd4 ,several strata of purplish porphyritic conglomerate1 of no very great thickness1 rest conformably upon the feldspathic slate4 ,a thick bed of fine1 purple1 claystone porphyry1 obscurely brecciated 7but not of metamorphosed sedimentary origin71 and capped by porphyritic conglomerate1 was the lowest bed actually examined in this section at the ,puente del ,inca4 #crd4 ,a stratum1 eighty feet thick1 of hard and very compact impure whitish limestone1 weathering bright red1 with included layers brecciated and recemented4 ,obscure marks of shell are distinguishable in it4 #dth4 ,a red1 quartzose1 fine-grained conglomerate1 with grains of quartz1 and with patches of white earthy feldspar1 apparently due to some process of concretionary crystalline action2 this bed is more compact and metamorphosed than any of the overlying conglomerates4 #eth4 ,a whitish cherty limestone1 with nodules of bluish argillaceous limestone4 #fth4 ,a white conglomerate1 with many particles of quartz1 almost blending into the paste4 #gth4 ,highly siliceous1 fine-grained white sandstone4 #hth and #ith4 ,red and white beds not examined4 #ajth4 ,yellow1 fine-grained1 thinly stratified1 magnesian 7judging from its slow dissolution in acids7 limestone3 it includes some white quartz pebbles1 and little cavities1 lined with calcareous spar1 some retaining the form of shells4 #aath4 ,a bed between twenty and thirty feet thick1 quite conformable with the underlying ones1 composed of a hard basis1 tinged lilac-grey porphyritic with numerous crystals of whitish feldspar1 with black mica and little spots of soft ferruginous matter3 evidently a submarine lava4 #abth4 ,yellow magnesian limestone1 as before1 part-stained purple4 #acth4 ,a most singular rock2 basis purplish grey1 obscurely crystalline1 easily fusible into a dark green glass1 not hard1 thickly speckled with crystals more or less perfect of white carbonate of lime1 of red hydrous oxide of iron1 of a white and transparent mineral like analcime1 and of a green opaque mineral like soap-stone2 the basis is moreover amygdaloidal with many spherical balls of white crystallised carbonate of lime1 of which some are coated with the red oxide of iron4 ,i have no doubt1 from the examination of a superincumbent stratum 7#ai71 that this is a submarine lava2 though in ,northern ,chile1 some of the metamorphosed sedimentary beds are almost as crystalline1 and of as varied composition4 #adth4 ,red sandstone1 passing in the upper part into a coarse1 hard1 red conglomerate1 #cjj feet thick1 having a calcareous cement1 and including grains of quartz and broken crystals of feldspar2 basis infusible2 the pebbles consist of dull purplish porphyries1 with some of quartz1 from the size of a nut to a man's head4 ,this is the coarsest conglomerate in this part of the ,cordillera3 in the middle there was a white layer not examined4 #aeth4 ,grand thick bed1 of a very hard1 yellowish-white rock1 with a crystalline feldspathic base1 including large crystals of white feldspar1 many little cavities mostly full of soft ferruginous matter1 and numerous hexagonal plates of black mica4 ,the upper part of this great bed is slightly cellular2 the lower part compact3 the thickness varied a little in different parts4 ,manifestly a submarine lava2 and is allied to bed #aa4 #afth and #agth4 ,dull purplish1 calcareous1 fine-grained1 compact sandstones1 which pass into coarse white conglomerates with numerous particles of quartz4 #ahth4 ,several alternations of red conglomerate1 purplish sandstone1 and submarine lava1 like that singular rock forming bed #ac4 #aith4 ,a very heavy1 compact1 greenish-black stone1 with a fine-grained obviously crystalline basis1 containing a few specks of white calcareous spar1 many specks of the crystallised hydrous red oxide of iron1 and some specks of a green mineral2 there are veins and nests filled with epidote3 certainly a submarine lava4 #bjth4 ,many thin strata of compact1 fine-grained1 pale purple sandstone4 #bast4 ,gypsum in a nearly pure state1 about three hundred feet in thickness3 this bed1 in its concretions of anhydrite and layers of small blackish crystals of carbonate of lime1 exactly resembles the great gypseous beds in the ,peuquenes range4 #bbnd4 ,pale purple and reddish sandstone1 as in bed #bj3 about three hundred feet in thickness4 #bcrd4 ,a thick mass composed of layers1 often as thin as paper and convoluted1 of pure gypsum with others very impure1 of a purplish colour4 #bdth4 ,pure gypsum1 thick mass4 #beth4 ,red sandstones1 of great thickness4 #bfth4 ,pure gypsum1 of great thickness4 #bgth4 ,alternating layers of pure and impure gypsum1 of great thickness4 ,i was not able to ascend to these few last great strata1 which compose the neighbouring loftiest pinnacles4 ,the thickness1 from the lowest to the uppermost bed of gypsum1 cannot be less than #b1jjj feet3 the beds beneath ,i estimated at #c1jjj feet1 and this does not include either the lower parts of the porphyritic conglomerate1 or the altered clay-slate2 ,i conceive the total thickness must be about six thousand feet4 ,i distinctly observed that not only the gypsum1 but the alternating sandstones and conglomerates were lens-shaped1 and repeatedly thinned out and replaced each other3 thus in the distance of about a mile1 a bed #cjj feet thick of sandstone between two beds of gypsum1 thinned out to nothing and disappeared4 ,the lower part of this section differs remarkably1--in the much greater diversity of its mineralogical composition1--in the abundance of calcareous matter1--in the greater coarseness of some of the conglomerates1--and in the numerous particles and well-rounded pebbles1 sometimes of large size1 of quartz1-- from any other section hitherto described in ,chile4 ,from these peculiarities and from the lens-form of the strata1 it is probable that this great pile of strata was accumulated on a shallow and very uneven bottom1 near some pre-existing land formed of various porphyries and quartz-rock4 ,the formation of porphyritic claystone conglomerate does not in this section attain nearly its ordinary thickness2 this may be ,,partly attributed to the metamorphic action having been here much less energetic than usual1 though the lower beds have been affected to a certain degree4 ,if it had been as energetic as in most other parts of ,chile1 many of the beds of sandstone and conglomerate1 containing rounded masses of porphyry1 would doubtless have been converted into porphyritic conglomerate2 and these would have alternated with1 and even blended into1 crystalline and porphyritic strata without a trace of mechanical structure1--namely1 into those which1 in the present state of the section1 we see are unquestionably submarine lavas4 ,the beds of gypsum1 together with the red alternating sandstones and conglomerates1 present so perfect and curious a resemblance with those seen in our former section in the basin-valley of ,yeso1 that ,i cannot doubt the identity of the two formations3 ,i may add1 that a little westward of the ,p4 del ,inca1 a mass of gypsum passed into a fine-grained1 hard1 brown sandstone1 which contained some layers of black1 calcareous1 compact1 shaly rock1 precisely like that seen in such vast masses on the ,peuquenes range4 ,near the ,puente del ,inca1 numerous fragments of limestone1 containing some fossil remains1 were scattered on the ground3 these fragments so perfectly resemble the limestone of bed ,no4 #c1 in which ,i saw impressions of shells1 that ,i have no doubt they have fallen from it4 ,the yellow magnesian limestone of bed ,no4 #aj1 which also includes traces of shells1 has a different appearance4 ,these fossils 7as named by ,m4 d',orbigny7 consist of3-- ,grypha1 near to ,g4 ,couloni 7,neocomian formation74 ,arca1 perhaps ,a4 ,gabrielis1 d',orbigny1 8,pal4 ,franc40 7,neocomian formation74 ,mr4 ,pentland made a collection of shells from this same spot1 and ,von ,buch#af considers them as consisting of3-- ,trigonia1 resembling in form ,t4 costata4 ,pholadomya1 like one found by ,m4 ,dufresnoy near ,alencon4 ,isocardi excentrica1 ,voltz41 identical with that from the ,jura4 #af 8,description ,phys4 des ,iles ,can410 p4 #dgb4 ,two of these shells1 namely1 the ,grypha and ,trigonia1 appear to be identical with species collected by ,meyen and myself on the ,peuquenes range2 and in the opinion of ,von ,buch and ,m4 d',orbigny1 the two formations belong to the same age4 ,i must here add1 that ,professor ,e4 ,forbes1 who has examined my specimens from this place and from the ,peuquenes range1 has likewise a strong impression that they indicate the ,cretaceous period1 and probably an early epoch in it3 so that all the palontologists who have seen these fossils nearly coincide in opinion regarding their age4 ,the limestone1 however1 with these fossils here lies at the very base of the formation1 just above the porphyritic conglomerate1 and certainly several thousand feet lower in the series1 than the equivalent1 fossiliferous1 black1 shaly rocks high up on the ,peuquenes range4 ,it is well worthy of remark that these shells1 or at least those of which ,i saw impressions in the limestone 7bed ,no4 #c71 must have been covered up1 on the least computation1 by #d1jjj feet of strata3 now we know from ,professor ,e4 ,forbes's researches1 that the sea at greater depths than #fjj feet becomes exceedingly barren of organic beings1--a result quite in accordance with what little ,i have seen of deep-sea soundings4 ,hence1 after this limestone with its shells was deposited1 the bottom of the sea where the main line of the ,cordillera now stands1 must have subsided some thousand feet to allow of the deposition of the superincumbent submarine strata4 ,without supposing a movement of this kind1 it would1 moreover1 be impossible to understand the accumulation of the several lower strata of coarse1 well-rounded conglomerates1 which it is scarcely possible to believe were spread out in profoundly deep water1 and which1 especially those containing pebbles of quartz1 could hardly have been rounded in submarine craters and afterwards ejected from them1 as ,i believe to have been the case with much of the porphyritic conglomerate formation4 ,i may add that1 in ,professor ,forbes's opinion1 the above-enumerated species of mollusca probably did not live at a much greater depth than twenty fathoms1 that is only #abj feet4 ,to return to our section down the valley2 standing on the great ,n4 by ,w4 and ,s4 by ,e4 uniclinal axis of the ,puente del ,inca1 of which a section has just been given1 and looking north-east1 greater tabular masses of gypseous formation 7,,kk7 could be seen in the distance1 very slightly inclined towards the east4 ,lower down the valley1 the mountains are almost exclusively composed of porphyries1 many of them of intrusive origin and non-stratified1 others stratified1 but with the stratification seldom distinguishable except in the upper parts4 ,disregarding local disturbances1 the beds are either horizontal or inclined at a small angle eastwards3 hence1 when standing on the plain of ,uspallata and looking to the west or backwards1 the ,cordillera appear composed of huge1 square1 nearly horizontal1 tabular masses3 so wide a space1 with such lofty mountains so equably elevated1 is rarely met with within the ,cordillera4 ,in this line of section1 the interval between the ,puente del ,inca and the neighbourhood of the ,cumbre1 includes all the chief axes of dislocation4 ,the altered clay-slate formation1 already described1 is seen in several parts of the valley as far down as ,las ,vacas1 underlying the porphyritic conglomerate4 ,at the ,casa de ,pujios ,l1 there is a hummock of 7andesitic87 granite2 and the stratification of the surrounding mountains here changes from ,w4 by ,s4 to ,s4,w4 ,again1 near the ,r4 ,vacas there is a larger formation of 7andesitic87 granite ,m1 which sends a meshwork of veins into the superincumbent clay-slate1 and which locally throws off the strata1 on one side to ,n4,w4 and on the other to ,s4,e4 but not at a high angle3 at the junction1 the clay-slate is altered into fine-grained greenstone4 ,this granitic axis is intersected by a green dike1 which ,i mention1 because ,i do not remember having elsewhere seen dikes in this lowest and latest intrusive rock4 ,from the ,r4 ,vacas to the plain of ,uspallata1 the valley runs ,n4,e41 so that ,i have had to contract my section2 it runs exclusively through porphyritic rocks4 ,as far as the ,pass of ,jaula1 the claystone conglomerate formation1 in most parts highly porphyritic1 and crossed by numerous dikes of greenstone porphyry1 attains a great thickness3 there is also much intrusive porphyry4 ,from the ,jaula to the plain1 the stratification has been in most places obliterated1 except near the tops of some of the mountains2 and the metamorphic action has been extremely great4 ,in this space1 the number and bulk of the intrusive masses of differently coloured porphyries1 injected one into another and intersected by dikes1 is truly extraordinary4 ,i saw one mountain of whitish porphyry1 from which two huge dikes1 thinning out1 branched downwards into an adjoining blackish porphyry4 ,another hill of white porphyry1 which had burst through dark-coloured strata1 was itself injected by a purple1 brecciated1 and recemented porphyry1 both being crossed by a green dike1 and both having been upheaved and injected by a granitic dome4 ,one brick-red porphyry1 which above the ,jaula forms an isolated mass in the midst of the porphyritic conglomerate formation1 and lower down the valley a magnificent group of peaked mountains1 differs remarkably from all the other porphyries4 ,it consists of a red feldspathic base1 including some rather large crystals of red feldspar1 numerous large angular grains of quartz1 and little bits of a soft green mineral answering in most of its characters to soapstone4 ,the crystals of red feldspar resemble in external appearance those of orthite1 though1 from being partially decomposed1 ,i was unable to measure them2 and they certainly are quite unlike the variety1 so abundantly met with in almost all the other rocks of this line of section1 and which1 wherever ,i tried it1 cleaved like albite4 ,this brick-red porphyry appears to have burst through all the other porphyries1 and numerous red dikes traversing the neighbouring mountains have proceeded from it3 in some few places1 however1 it was intersected by white dikes4 ,from this posteriority of intrusive origin1--from the close general resemblance between this red porphyry and the red granite of the ,portillo line1 the only difference being that the feldspar here is less perfectly granular1 and that soapstone replaces the mica1 which is there imperfect and passes into chlorite1--and from the ,portillo line a little southward of this point appearing to blend 7according to ,dr4 ,gillies7 into the western ranges1--,i am strongly urged to believe 7as formerly remarked7 that the grand mountain-masses composed of this brick-red porphyry belong to the same axis of injection with the granite of the ,portillo line2 if so1 the injection of this porphyry probably took place1 as long subsequently to the several axes of elevation in the gypseous formation near the ,cumbre1 as the injection of the ,portillo granite has been shown to have been subsequent to the elevation of the gypseous strata composing the ,peuquenes range2 and this interval1 we have seen1 must have been a very long one4 ,the ,plain of ,uspallata has been briefly described in ,chapter #c2 it resembles the basin-plains of ,chile2 it is ten or fifteen miles wide1 and is said to extend for #ahj miles northward2 its surface is nearly six thousand feet above the sea2 it is composed1 to a thickness of some hundred feet of loosely aggregated1 stratified shingle1 which is prolonged with a gently sloping surface up the valleys in the mountains on both sides4 ,one section in this plain ,z is interesting1 from the unusual#ag circumstance of alternating layers of almost loose red and white sand with lines of pebbles 7from the size of a nut to that of an apple71 and beds of gravel1 being inclined at an angle of #de1 and in some spots even at a higher angle4 ,these beds are dislocated by small faults3 and are capped by a thick mass of horizontally stratified gravel1 evidently of subaqueous origin4 ,having been accustomed to observe the irregularities of beds accumulated under currents1 ,i feel sure that the inclination here has not been thus produced4 ,the pebbles consist chiefly of the brick-red porphyry just described and of white granite1 both probably derived from the ranges to the west1 and of altered clay-slate and of certain porphyries1 apparently belonging to the rocks of the ,uspallata chain4 ,this plain corresponds geographically with the valley of ,tenuyan between the ,portillo and ,peuquenes ranges2 but in that valley the shingle1 which likewise has been derived both from the eastern and western ranges1 has been cemented into a hard conglomerate1 and has been throughout tilted at a considerable inclination2 the gravel there apparently attains a much greater thickness1 and is probably of higher antiquity4 #ag ,i find that ,mr4 ,smith of ,jordan ,hill has described 7,edinburgh ,new ,phil4 ,journ41 vol4 xxv1 p4 #cib7 beds of sand and gravel1 near ,edinburgh1 tilted at an angle of #fj1 and dislocated by miniature faults4 ,the ,uspallata range4--,the road by the ,villa ,vicencio ,pass does not strike directly across the range1 but runs for some leagues northward along its western base3 and ,i must briefly describe the rocks here seen1 before continuing with the coloured east and west section4 ,at the mouth of the valley of ,canota1 and at several points northwards1 there is an extensive formation of a glossy and harsh1 and of a feldspathic clay-slate1 including strata of grauwacke1 and having a tortuous1 nearly vertical cleavage1 traversed by numerous metalliferous veins and others of quartz4 ,the clay-slate is in many parts capped by a thick mass of fragments of the same rock1 firmly recemented2 and both together have been injected and broken up by very numerous hillocks1 ranging north and south1 of lilac1 white1 dark and salmon-coloured porphyries3 one steep1 now denuded1 hillock of porphyry had its face as distinctly impressed with the angles of a fragmentary mass of the slate1 with some of the points still remaining embedded1 as sealing-wax could be by a seal4 ,at the mouth of this same valley of ,canota1 in a fine escarpment having the strata dipping from #ej to #fj to the ,n4,e41#ah the clay-slate formation is seen to be covered by--7#ast7 a purple1 claystone porphyry resting unconformably in some parts on the solid slate1 and in others on a thick fragmentary mass2 7#bnd71 a conformable stratum of compact blackish rock1 having a spheroidal structure1 full of minute acicular crystals of glassy feldspar1 with red spots of oxide of iron2 7#crd71 a great stratum of purplish-red claystone porphyry1 abounding with crystals of opaque feldspar1 and laminated with thin1 parallel1 often short1 layers1 and likewise with great irregular patches of white1 earthy1 semi-crystalline feldspar2 this rock 7which ,i noticed in other neighbouring places7 perfectly resembles a curious variety described at ,port ,desire1 and occasionally occurs in the great porphyritic conglomerate formation of ,chile2 7#dth71 a thin stratum of greenish white1 indurated tuff1 fusible and containing broken crystals and particles of porphyries2 7#eth71 a grand mass1 imperfectly columnar and divided into three parallel and closely joined strata1 of cream-coloured claystone porphyry2 7#fth71 a thick stratum of lilac-coloured porphyry1 which ,i could see was capped by another bed of the cream-coloured variety2 ,i was unable to examine the still higher parts of the escarpment4 ,these conformably stratified porphyries1 though none are either vesicular are amygdaloidal1 have evidently flowed as submarine lavas3 some of them are separated from each other by seams of indurated tuff1 which1 however1 are quite insignificant in thickness compared with the porphyries4 ,this whole pile resembles1 but not very closely1 some of the less brecciated parts of the great porphyritic conglomerate formation of ,chile2 but it does not probably belong to the same age1 as the porphyries here rest unconformably on the altered feldspathic clay-slate1 whereas the porphyritic conglomerate formation alternates with and rests conformably on it4 ,these porphyries1 moreover1 with the exception of the one blackish stratum1 and of the one indurated1 white tufaceous bed1 differ from the beds composing the ,uspallata range in the line of the ,villa ,vicencio ,pass4 #ah ,nearly opposite to this escarpment1 there is another corresponding one1 with the strata dipping not to the exactly opposite point1 or ,s4,w41 but to ,s4,s4,w43 consequently the two escarpments trend towards each other1 and some miles southward they become actually united3 this is a form of elevation which ,i have not elsewhere seen4 ,i will now give1 first1 a sketch of the structure of the range1 as represented in the section1 and will then describe its composition and interesting history4 ,at its western foot1 a hillock ,n is seen to rise out of the plain1 with its strata dipping at #gj to the west1 fronted by strata ,o inclined at #de to the east1 thus forming a little north and south anticlinal axis4 ,some other little hillocks of similar composition1 with their strata highly inclined1 range ,n4,e4 and ,s4,w41 obliquely to the main ,uspallata line4 ,the cause of these dislocations1 which1 though on a small scale1 have been violent and complicated1 is seen to lie in hummocks of lilac1 purple and red porphyries1 which have been injected in a liquified state through and into the underlying clay-slate formation4 ,several dykes were exposed here1 but in no other part1 that ,i saw of this range4 ,as the strata consist of black1 white1 greenish and brown-coloured rocks1 and as the intrusive porphyries are so brightly tinted1 a most extraordinary view was presented1 like a coloured geological drawing4 ,on the gently inclined main western slope ,,pp1 above the little anticlinal ridges just mentioned1 the strata dip at an average angle of #be to the west2 the inclination in some places being only #ai1 in some few others as much as #de4 ,the masses having these different inclinations1 are separated from each other by parallel vertical faults as represented at ,pa1 often giving rise to separate1 parallel1 uniclinal ridges4 ,the summit of the main range is broad and undulatory1 with the stratification undulatory and irregular3 in a few places granitic and porphyritic masses ,q protrude1 which1 from the small effect they have locally produced in deranging the strata1 probably form the upper points of a regular1 great underlying dome4 ,these denuded granitic points1 ,i estimated at about nine thousand feet in height above the sea4 ,on the eastern slope1 the strata in the upper part are regularly inclined at about #be to the east1 so that the summit of this chain1 neglecting small irregularities1 forms a broad anticlinal axis4 ,lower down1 however1 near ,los ,hornillos ,r1 there is a well-marked synclinal axis1 beyond which the strata are inclined at nearly the same angle1 namely from #bj to #cj1 inwards or westward4 ,owing to the amount of denudation which this chain has suffered1 the outline of the gently inclined eastern flank scarcely offers the slightest indication of this synclinal axis4 ,the stratified beds1 which we have hitherto followed across the range1 a little further down are seen to lie1 ,i believe unconformably1 on a broad mountainous band of clay-slate and grauwacke4 ,the strata and lamin of this latter formation1 on the extreme eastern flank1 are generally nearly vertical2 further inwards they become inclined from #de to #hj to the west3 near ,villa ,vicencio ,s there is apparently an anticlinal axis1 but the structure of this outer part of the clay-slate formation is so obscure1 that ,i have not marked the planes of stratification in the section4 ,on the margin of the ,pampas1 some low1 much dislocated spurs of this same formation1 project in a north-easterly line1 in the same oblique manner as do the ridges on the western foot1 and as is so frequently the case with those at the base of the main ,cordillera4 ,i will now describe the nature of the beds1 beginning at the base on the eastern side4 ,first1 for the clay-slate formation3 the slate is generally hard and bluish1 with the lamin coated by minute micaceous scales2 it alternates many times with a coarse-grained1 greenish grauwacke1 containing rounded fragments of quartz and bits of slate in a slightly calcareous basis4 ,the slate in the upper part generally becomes purplish1 and the cleavage so irregular that the whole consists of mere splinters4 ,transverse veins of quartz are numerous4 ,at the ,calera1 some leagues distant1 there is a dark crystalline limestone1 apparently included in this formation4 ,with the exception of the grauwacke being here more abundant1 and the clay-slate less altered1 this formation closely resembles that unconformably underlying the porphyries at the western foot of this same range2 and likewise that alternating with the porphyritic conglomerate in the main ,cordillera4 ,this formation is a considerable one1 and extends several leagues southward to near ,mendoza3 the mountains composed of it rise to a height of about two thousand feet above the edge of the ,pampas1 or about seven thousand feet above the sea4#ai #ai ,i infer this from the height of ,v4 ,vicencio1 which was ascertained by ,mr4 ,miers to be #e1cbh feet above the sea4 ,secondly3 the most usual bed on the clay-slate is a coarse1 white1 slightly calcareous conglomerate1 of no great thickness1 including broken crystals of feldspar1 grains of quartz1 and numerous pebbles of brecciated claystone porphyry1 but without any pebbles of the underlying clay-slate4 ,i nowhere saw the actual junction between this bed and the clay-slate1 though ,i spent a whole day in endeavouring to discover their relations4 ,in some places ,i distinctly saw the white conglomerate and overlying beds inclined at from #be to #cj to the west1 and at the bottom of the same mountain1 the clay-slate and grauwacke inclined to the same point1 but at an angle from #gj to #hj3 in one instance1 the clay-slate dipped not only at a different angle1 but to a different point from the overlying formation4 ,in these cases the two formations certainly appeared quite unconformable3 moreover1 ,i found in the clay-slate one great1 vertical1 dike-like fissure1 filled up with an indurated whitish tuff1 quite similar to some of the upper beds presently to be described2 and this shows that the clay-slate must have been consolidated and dislocated before their deposition4 ,on the other hand1 the stratification of the slate and grauwacke1#bj in some cases gradually and entirely disappeared in approaching the overlying white conglomerate2 in other cases the stratification of the two formations became strictly conformable2 and again in other cases1 there was some tolerably well characterised clay-slate lying above the conglomerate4 ,the most probable conclusion appears to be1 that after the clay-slate formation had been dislocated and tilted1 but whilst under the sea1 a fresh and more recent deposition of clay-slate took place1 on which the white conglomerate was conformably deposited1 with here and there a thin intercalated bed of clay-slate4 ,on this view the white conglomerates and the presently to be described tuffs and lavas are really unconformable to the main part of the clay-slate2 and this1 as we have seen1 certainly is the case with the clay-stone lavas in the valley of ,canota1 at the western and opposite base of the range4 #bj ,the coarse1 mechanical structure of many grauwackes has always appeared to me a difficulty2 for the texture of the associated clay-slate and the nature of the embedded organic remains where present1 indicate that the whole has been a deep-water deposit4 ,whence have the sometimes included angular fragments of clay-slate1 and the rounded masses of quartz and other rocks1 been derived8 ,many deep-water limestones1 it is well known1 have been brecciated1 and then firmly recemented4 ,thirdly3 on the white conglomerate1 strata several hundred feet in thickness are superimposed1 varying much in nature in short distances3 the commonest variety is a white1 much indurated tuff1 sometimes slightly calcareous1 with ferruginous spots and water-lines1 often passing into whitish or purplish compact1 fine-grained grit or sandstones2 other varieties become semi-porcellanic1 and tinted faint green or blue2 others pass into an indurated shale3 most of these varieties are easily fusible4 ,fourthly3 a bed1 about one hundred feet thick of a compact1 partially columnar1 pale-grey1 feldspathic lava1 stained with iron1 including very numerous crystals of opaque feldspar1 and with some crystallised and disseminated calcareous matter4 ,the tufaceous stratum on which this feldspathic lava rests is much hardened1 stained purple1 and has a spherico-concretionary structure2 it here contains a good many pebbles of claystone porphyry4 ,fifthly3 thin beds1 #djj feet in thickness1 varying much in nature1 consisting of white and ferruginous tuffs1 in some parts having a concretionary structure1 in others containing rounded grains and a few pebbles of quartz2 also passing into hard gritstones and into greenish mudstones3 there is1 also1 much of a bluish-grey and green semi-porcellanic stone4 ,sixthly3 a volcanic stratum1 #bej feet in thickness1 of so varying a nature that ,i do not believe a score of specimens would show all the varieties2 much is highly amygdaloidal1 much compact2 there are greenish1 blackish1 purplish1 and grey varieties1 rarely including crystals of green augite and minute acicular ones of feldspar1 but often crystals and amygdaloidal masses of white1 red1 and black carbonate of lime4 ,some of the blackish varieties of this rock have a conchoidal fracture and resemble basalt2 others have an irregular fracture4 ,some of the grey and purplish varieties are thickly speckled with green earth and with white crystalline carbonate of lime2 others are largely amygdaloidal with green earth and calcareous spar4 ,again1 other earthy varieties1 of greenish1 purplish and grey tints1 contain much iron1 and are almost half composed of amygdaloidal balls of dark brown bole1 of a whitish indurated feldspathic matter1 of bright green earth1 of agate1 and of black and white crystallised carbonate of lime4 ,all these varieties are easily fusible4 ,viewed from a distance1 the line of junction with the underlying semi-porcellanic strata was distinct2 but when examined closely1 it was impossible to point out within a foot where the lava ended and where the sedimentary mass began3 the rock at the time of junction was in most places hard1 of a bright green colour1 and abounded with irregular amygdaloidal masses of ferruginous and pure calcareous spar1 and of agate4 ,seventhly3 strata1 eighty feet in thickness1 of various indurated tuffs1 as before2 many of the varieties have a fine basis including rather coarse extraneous particles2 some of them are compact and semi-porcellanic1 and include vegetable impressions4 ,eighthly3 a bed1 about fifty feet thick1 of greenish-grey1 compact1 feldspathic lava1 with numerous small crystals of opaque feldspar1 black augite1 and oxide of iron4 ,the junction with the bed on which it rested1 was ill defined2 balls and masses of the feldspathic rock being enclosed in much altered tuff4 ,ninthly3 indurated tuffs1 as before4 ,tenthly3 a conformable layer1 less than two feet in thickness1 of pitchstone1 generally brecciated1 and traversed by veins of agate and of carbonate of lime3 parts are composed of apparently concretionary fragments of a more perfect variety1 arranged in horizontal lines in a less perfectly characterised variety4 ,i have much difficulty in believing that this thin layer of pitchstone flowed as lava4 ,eleventhly3 sedimentary and tufaceous beds as before1 passing into sandstone1 including some conglomerate3 the pebbles in the latter are of claystone porphyry1 well rounded1 and some as large as cricket-balls4 ,twelfthly3 a bed of compact1 sonorous1 feldspathic lava1 like that of bed ,no4 #h1 divided by numerous joints into large angular blocks4 ,thirteenthly3 sedimentary beds as before4 ,fourteenthly3 a thick bed of greenish or greyish black1 compact basalt 7fusing into a black enamel71 with small crystals1 occasionally distinguishable1 of feldspar and augite3 the junction with the underlying sedimentary bed1 differently from that in most of the foregoing streams1 here was quite distinct3--the lava and tufaceous matter preserving their perfect characters within two inches of each other4 ,this rock closely resembles certain parts of that varied and singular lava-stream ,no4 #f2 it likewise resembles1 as we shall immediately see1 many of the great upper beds on the western flank and on the summit of this range4 ,the pile of strata here described attains a great thickness2 and above the last-mentioned volcanic stratum1 there were several other great tufaceous beds alternating with submarine lavas1 which ,i had not time to examine2 but a corresponding series1 several thousand feet in thickness1 is well exhibited on the crest and western flank of the range4 ,most of the lava-streams on the western side are of a jet-black colour and basaltic nature2 they are either compact and fine-grained1 including minute crystals of augite and feldspar1 or they are coarse-grained and abound with rather large coppery-brown crystals of an augitic mineral4#ba ,another variety was of a dull-red colour1 having a claystone brecciated basis1 including specks of oxide of iron and of calcareous spar1 and amygdaloidal with green earth3 there were apparently several other varieties4 ,these submarine lavas often exhibit a spheroidal1 and sometimes an imperfect columnar structure3 their upper junctions are much more clearly defined than their lower junctions2 but the latter are not so much blended into the underlying sedimentary beds as is the case in the eastern flank4 ,on the crest and western flank of the range1 the streams1 viewed as a whole1 are mostly basaltic2 whilst those on the eastern side1 which stand lower in the series1 are1 as we have seen1 mostly feldspathic4 #ba ,very easily fusible into a jet-black bead1 attracted by the magnet3 the crystals are too much tarnished to be measured by the goniometer4 ,the sedimentary strata alternating with the lavas on the crest and western side1 are of an almost infinitely varying nature2 but a large proportion of them closely resemble those already described on the eastern flank3 there are white and brown1 indurated1 easily fusible tuffs1--some passing into pale blue and green semi-porcellanic rocks1--others into brownish and purplish sandstones and gritstones1 often including grains of quartz1--others into mudstone containing broken crystals and particles of rock1 and occasionally single large pebbles4 ,there was one stratum of a bright red1 coarse1 volcanic gritstone2 another of conglomerate2 another of a black1 indurated1 carbonaceous shale marked with imperfect vegetable impressions2 this latter bed1 which was thin1 rested on a submarine lava1 and followed all the considerable inequalities of its upper surface4 ,mr4 ,miers states that coal has been found in this range4 ,lastly1 there was a bed 7like ,no4 #aj on the eastern flank7 evidently of sedimentary origin1 and remarkable from closely approaching in character to an imperfect pitchstone1 and from including extremely thin layers of perfect pitchstone1 as well as nodules and irregular fragments 7but not resembling extraneous fragments7 of this same rock arranged in horizontal lines3 ,i conceive that this bed1 which is only a few feet in thickness1 must have assumed its present state through metamorphic and concretionary action4 ,most of these sedimentary strata are much indurated1 and no doubt have been partially metamorphosed3 many of them are extraordinarily heavy and compact2 others have agate and crystalline carbonate of lime disseminated throughout them4 ,some of the beds exhibit a singular concretionary arrangement1 with the curves determined by the lines of fissure4 ,there are many veins of agate and calcareous spar1 and innumerable ones of iron and other metals1 which have blackened and curiously affected the strata to considerable distances on both sides4 ,many of these tufaceous beds resemble1 with the exception of being more indurated1 the upper beds of the ,great ,patagonian tertiary formation1 especially those variously coloured layers high up the ,river ,santa ,cruz1 and in a remarkable degree the tufaceous formation at the northern end of ,chiloe4 ,i was so much struck with this resemblance1 that ,i particularly looked out for silicified wood1 and found it under the following extraordinary circumstances4 ,high up on this western flank1#bb at a height estimated at #g1jjj feet above the sea1 in a broken escarpment of thin strata1 composed of compact green gritstone passing into a fine mudstone1 and alternating with layers of coarser1 brownish1 very heavy mudstone1 including broken crystals and particles of rock almost blended together1 ,i counted the stumps of fifty-two trees4 ,they projected between two and five feet above the ground1 and stood at exactly right angles to the strata1 which were here inclined at an angle of about #be to the west4 ,eleven of these trees were silicified and well preserved2 ,mr4 ,r4 ,brown has been so kind as to examine the wood when sliced and polished2 he says it is coniferous1 partaking of the characters of the ,araucarian tribe1 with some curious points of affinity with the ,yew4 ,the bark round the trunks must have been circularly furrowed with irregular lines1 for the mudstone round them is thus plainly marked4 ,one cast consisted of dark argillaceous limestone2 and forty of them of coarsely crystallised carbonate of lime1 with cavities lined by quartz crystals3 these latter white calcareous columns do not retain any internal structure1 but their external form plainly shows their origin4 ,all the stumps have nearly the same diameter1 varying from one foot to eighteen inches2 some of them stand within a yard of each other2 they are grouped in a clump within a space of about sixty yards across1 with a few scattered round at the distance of #aej yards4 ,they all stand at about the same level4 ,the longest stump stood seven feet out of the ground3 the roots1 if they are still preserved1 are buried and concealed4 ,no one layer of the mudstone appeared much darker than the others1 as if it had formerly existed as soil1 nor could this be expected1 for the same agents which replaced with silex and lime the wood of the trees1 would naturally have removed all vegetable matter from the soil4 ,besides the fifty-two upright trees1 there were a few fragments1 like broken branches1 horizontally embedded4 ,the surrounding strata are crossed by veins of carbonate of lime1 agate1 and oxide of iron2 and a poor gold vein has been worked not far from the trees4 #bb ,for the information of any future traveller1 ,i will describe the spot in detail4 ,proceeding eastward from the ,agua del ,zorro1 and afterwards leaving on the north side of the road a rancho attached to some old goldmines1 you pass through a gully with low but steep rocks on each hand3 the road then bends1 and the ascent becomes steeper4 ,a few hundred yards farther on1 a stone's throw on the south side of the road1 the white calcareous stumps may be seen4 ,the spot is about half a mile east of the ,agua del ,zorro4 ,the green and brown mudstone beds including the trees1 are conformably covered by much indurated1 compact1 white or ferruginous tuffs1 which pass upwards into a fine-grained1 purplish sedimentary rock3 these strata1 which1 together1 are from four to five hundred feet in thickness1 rest on a thick bed of submarine lava1 and are conformably covered by another great mass of fine-grained basalt1#bc which ,i estimated at #a1jjj feet in thickness1 and which probably has been formed by more than one stream4 ,above this mass ,i could clearly distinguish five conformable alternations1 each several hundred feet in thickness1 of stratified sedimentary rocks and lavas1 such as have been previously described4 ,certainly the upright trees have been buried under several thousand feet in thickness of matter1 accumulated under the sea4 ,as the trees obviously must once have grown on dry land1 what an enormous amount of subsidence is thus indicated6 ,nevertheless1 had it not been for the trees there was no appearance which would have led any one even to have conjectured that these strata had subsided4 ,as the land1 moreover1 on which the trees grew1 is formed of subaqueous deposits1 of nearly if not quite equal thickness with the superincumbent strata1 and as these deposits are regularly stratified and fine-grained1 not like the matter thrown up on a sea-beach1 a previous upward movement1 aided no doubt by the great accumulation of lavas and sediment1 is also indicated4#bd #bc ,this rock is quite black1 and fuses into a black bead1 attracted strongly by the magnet2 it breaks with a conchoidal fracture2 the included crystals of augite are distinguishable by the naked eye1 but are not perfect enough to be measured3 there are many minute acicular crystals of glassy feldspar4 #bd ,at first ,i imagined1 that the strata with the trees might have been accumulated in a lake3 but this seems highly improbable2 for1 first1 a very deep lake was necessary to receive the matter below the trees1 then it must have been drained for their growth1 and afterwards re-formed and made profoundly deep1 so as to receive a subsequent accumulation of matter several thousand feet in thickness4 ,and all this must have taken place necessarily before the formation of the ,uspallata range1 and therefore on the margin of the wide level expanse of the ,pampas6 ,hence ,i conclude1 that it is infinitely more probable that the strata were accumulated under the sea3 the vast amount of denudation1 moreover1 which this range has suffered1 as shown by the wide valleys1 by the exposure of the very trees and by other appearances1 could have been effected1 ,i conceive1 only by the long-continued action of the sea2 and this shows that the range was either upheaved from under the sea1 or subsequently let down into it4 ,from the natural manner in which the stumps 7fifty-two in number7 are grouped in a clump1 and from their all standing vertically to the strata1 it is superfluous to speculate on the chance of the trees having been drifted from adjoining land1 and deposited upright3 ,i may1 however1 mention that the late ,dr4 ,malcolmson assured me1 that he once met in the ,indian ,ocean1 fifty miles from land1 several cocoa-nut trees floating upright1 owing to their roots being loaded with earth4 ,in nearly the middle of the range1 there are some hills ,q1 before alluded to1 formed of a kind of granite externally resembling andesite1 and consisting of a white1 imperfectly granular1 feldspathic basis1 including some perfect crystals apparently of albite 7but ,i was unable to measure them71 much black mica1 epidote in veins1 and very little or no quartz4 ,numerous small veins branch from this rock into the surrounding strata2 and it is a singular fact that these veins1 though composed of the same kind of feldspar and small scales of mica as in the solid rock1 abound with innumerable minute rounded grains of quartz3 in the veins or dikes also1 branching from the great granitic axis in the peninsula of ,tres ,montes1 ,i observed that quartz was more abundant in them than in the main rock3 ,i have heard of other analogous cases3 can we account for this fact1 by the long-continued vicinity of quartz#be when cooling1 and by its having been thus more easily sucked into fissures than the other constituent minerals of granite8 ,the strata encasing the flanks of these granitic or andesite masses1 and forming a thick cap on one of their summits1 appear originally to have been of the same tufaceous nature with the beds already described1 but they are now changed into porcellanic1 jaspery1 and crystalline rocks1 and into others of a white colour with a harsh texture1 and having a siliceous aspect1 though really of a feldspathic nature and fusible4 ,both the granitic intrusive masses and the encasing strata are penetrated by innumerable metallic veins1 mostly ferruginous and auriferous1 but some containing copper-pyrites and a few silver3 near the veins1 the rocks are blackened as if blasted by gunpowder4 ,the strata are only slightly dislocated close round these hills1 and hence1 perhaps1 it may be inferred that the granitic masses form only the projecting points of a broad continuous axis-dome1 which has given to the upper parts of this range its anticlinal structure4 #be ,see a paper by ,m4 ,elie de ,beaumont1 8,soc4 ,philomath410 ,may #ahci 78,l',institut410 #ahci1 p4 #afa47 ,concluding remarks on the ,uspallata range4--,i will not attempt to estimate the total thickness of the pile of strata forming this range1 but it must amount to many thousand feet4 ,the sedimentary and tufaceous beds have throughout a general similarity1 though with infinite variations4 ,the submarine lavas in the lower part of the series are mostly feldspathic1 whilst in the upper part1 on the summit and western flank1 they are mostly basaltic4 ,we are thus reminded of the relative position in most recent volcanic districts of the trachytic and basaltic lavas1--the latter from their greater weight having sunk to a lower level in the earth's crust1 and having consequently been erupted at a later period over the lighter and upper lavas of the trachytic series4#bf ,both the basaltic and feldspathic submarine streams are very compact2 none being vesicular1 and only a few amygdaloidal3 the effects which some of them1 especially those low in the series1 have produced on the tufaceous beds over which they have flowed is highly curious4 ,independently of this local metamorphic action1 all the strata undoubtedly display an indurated and altered character2 and all the rocks of this range--the lavas1 the alternating sediments1 the intrusive granite and porphyries1 and the underlying clay-slate--are intersected by metalliferous veins4 ,the lava-strata can often be seen extending for great distances1 conformably with the under and overlying beds2 and it was obvious that they thickened towards the west4 ,hence the points of eruption must have been situated westward of the present range1 in the direction of the main ,cordillera3 as1 however1 the flanks of the ,cordillera are entirely composed of various porphyries1 chiefly claystone and greenstone1 some intrusive1 and others belonging to the porphyritic conglomerate formation1 but all quite unlike these submarine lava-streams1 we must in all probability look to the plain of ,uspallata for the now deeply buried points of eruption4 #bf ,see on this subject1 8,volcanic ,islands10 etc41 by the ,author4 ,comparing our section of the ,uspallata range with that of the ,cumbre1 we see1 with the exception of the underlying clay-slate1 and perhaps of the intrusive rocks of the axes1 a striking dissimilarity in the strata composing them4 ,the great porphyritic conglomerate formation has not extended as far as this range2 nor have we here any of the gypseous strata1 the magnesian and other limestones1 the red sandstones1 the siliceous beds with pebbles of quartz1 and comparatively little of the conglomerates1 all of which form such vast masses over the basal series in the main ,cordillera4 ,on the other hand1 in the ,cordillera1 we do not find those endless varieties of indurated tuffs1 with their numerous veins and concretionary arrangement1 and those grit and mud stones1 and singular semi-porcellanic rocks1 so abundant in the ,uspallata range4 ,the submarine lavas1 also1 differ considerably2 the feldspathic streams of the ,cordillera contain much mica1 which is absent in those of the ,uspallata range3 in this latter range we have seen on how grand a scale1 basaltic lava has been poured forth1 of which there is not a trace in the ,cordillera4 ,this dissimilarity is the more striking1 considering that these two parallel chains are separated by a plain only between ten and fifteen miles in width2 and that the ,uspallata lavas1 as well as no doubt the alternating tufaceous beds1 have proceeded from the west1 from points apparently between the two ranges4 ,to imagine that these two piles of strata were contemporaneously deposited in two closely adjoining1 very deep1 submarine areas1 separated from each other by a lofty ridge1 where a plain now extends1 would be a gratuitous hypothesis4 ,and had they been contemporaneously deposited1 without any such dividing ridge1 surely some of the gypseous and other sedimentary matter forming such immensely thick masses in the ,cordillera1 would have extended this short distance eastwards2 and surely some of the ,uspallata tuffs and basalts also accumulated to so great a thickness1 would have extended a little westward4 ,hence ,i conclude1 that it is far from probable that these two series are not contemporaneous2 but that the strata of one of the chains were deposited1 and even the chain itself uplifted1 before the formation of the other3--which chain1 then1 is the oldest8 ,considering that in the ,uspallata range the lowest strata on the western flank lie unconformably on the clay-slate1 as probably is the case with those on the eastern flank1 whereas in the ,cordillera all the overlying strata lie conformably on this formation3--considering that in the ,uspallata range some of the beds1 both low down and high up in the series1 are marked with vegetable impressions1 showing the continued existence of neighbouring land2--considering the close general resemblance between the deposits of this range and those of tertiary origin in several parts of the continent2--and lastly1 even considering the lesser height and outlying position of the ,uspallata range1--,i conclude that the strata composing it are in all probability of subsequent origin1 and that they were accumulated at a period when a deep sea studded with submarine volcanoes washed the eastern base of the already partially elevated ,cordillera4 ,this conclusion is of much importance1 for we have seen that in the ,cordillera1 during the deposition of the ,neocomian strata1 the bed of the sea must have subsided many thousand feet3 we now learn that at a later period an adjoining area first received a great accumulation of strata1 and was upheaved into land on which coniferous trees grew1 and that this area then subsided several thousand feet to receive the superincumbent submarine strata1 afterwards being broken up1 denuded1 and elevated in mass to its present height4 ,i am strengthened in this conclusion of there having been two distinct1 great periods of subsidence1 by reflecting on the thick mass of coarse stratified conglomerate in the valley of ,tenuyan1 between the ,peuquenes and ,portillo lines2 for the accumulation of this mass seems to me1 as previously remarked1 almost necessarily to have required a prolonged subsidence2 and this subsidence1 from the pebbles in the conglomerate having been to a great extent derived from the gypseous or ,neocomian strata of the ,peuquenes line1 we know must have been quite distinct from1 and subsequent to1 that sinking movement which probably accompanied the deposition of the ,peuquenes strata1 and which certainly accompanied the deposition of the equivalent beds near the ,puente del ,inca1 in this line of section4 ,the ,uspallata chain corresponds in geographical position1 though on a small scale1 with the ,portillo line2 and its clay-slate formation is probably the equivalent of the mica-schist of the ,portillo1 there metamorphosed by the old white granites and syenites4 ,the coloured beds under the conglomerate in the valley of ,tenuyan1 of which traces are seen on the crest of the ,portillo1 and even the conglomerate itself1 may perhaps be synchronous with the tufaceous beds and submarine lavas of the ,uspallata range2 an open sea and volcanic action in the latter case1 and a confined channel between two bordering chains of islets in the former case1 having been sufficient to account for the mineralogical dissimilarity of the two series4 ,from this correspondence between the ,uspallata and ,portillo ranges1 perhaps in age and certainly in geographical position1 one is tempted to consider the one range as the prolongation of the other2 but their axes are formed of totally different intrusive rocks2 and we have traced the apparent continuation of the red granite of the ,portillo in the red porphyries diverging into the main ,cordillera4 ,whether the axis of the ,uspallata range was injected before1 or as perhaps is more probable1 after that of the ,portillo line1 ,i will not pretend to decide2 but it is well to remember that the highly inclined lava-streams on the eastern flank of the ,portillo line1 prove that its angular upheavement was not a single and sudden event2 and therefore that the anticlinal elevation of the ,uspallata range may have been contemporaneous with some of the later angular movements by which the gigantic ,portillo range gained its present height above the adjoining plain4 ,chapter ,,viii ,,northern ,,chile4--,,conclusion4 ,section from ,illapel to ,combarbala2 gypseous formation with silicified wood4--,panuncillo4--,coquimbo2 mines of ,arqueros2 section up valley2 fossils4--,guasco1 fossils of4--,copiapo1 section up valley2 ,las ,amolanas1 silicified wood4--,conglomerates1 nature of former land1 fossils1 thickness of strata1 great subsidence4--,valley of ,despoblado1 fossils1 tufaceous deposit1 complicated dislocations of4--,relations between ancient orifices of eruption and subsequent axes of injection4--,iquique1 ,peru1 fossils of1 salt-deposits4--,metalliferous veins4--,summary on the porphyritic conglomerate and gypseous formations4--,great subsidence with partial elevations during the cretaceo-oolitic period4--,on the elevation and structure of the ,cordillera4--,recapitulation on the tertiary series4--,relation between movements of subsidence and volcanic action4--,pampean formation4--,recent elevatory movements4 ,long-continued volcanic action in the ,cordillera4--,conclusion4 ,valparaiso to ,coquimbo4 ,i have already described the general nature of the rocks in the low country north of ,valparaiso1 consisting of granites1 syenites1 greenstones1 and altered feldspathic clay-slate4 ,near ,coquimbo there is much hornblendic rock and various dusky-coloured porphyries4 ,i will describe only one section in this district1 namely1 from near ,illapel in a ,n4,e4 line to the mines of ,los ,hornos1 and thence in a north by east direction to ,combarbala1 at the foot of the main ,cordillera4 ,near ,illapel1 after passing for some distance over granite1 andesite1 and andesitic porphyry1 we come to a greenish stratified feldspathic rock1 which ,i believe is altered clay-slate1 conformably capped by porphyries and porphyritic conglomerate of great thickness1 dipping at an average angle of #bj to ,n4,e4 by ,n4 ,the uppermost beds consist of conglomerates and sandstone only a little metamorphosed1 and conformably covered by a gypseous formation of very great thickness1 but much denuded4 ,this gypseous formation1 where first met with1 lies in a broad valley or basin1 a little southward of the mines of ,los ,hornos3 the lower half alone contains gypsum1 not in great masses as in the ,cordillera1 but in innumerable thin layers1 seldom more than an inch or two in thickness4 ,the gypsum is either opaque or transparent1 and is associated with carbonate of lime4 ,the layers alternate with numerous varying ones of a calcareous clay-shale 7with strong aluminous odour1 adhering to the tongue1 easily fusible into a pale green glass71 more or less indurated1 either earthy and cream-coloured1 or greenish and hard4 ,the more indurated varieties have a compact1 homogeneous1 almost crystalline fracture1 and contain granules of crystallised oxide of iron4 ,some of the varieties almost resemble honestones4 ,there is also a little black1 hardly fusible1 siliceo-calcareous clay-slate1 like some of the varieties alternating with gypsum on the ,peuquenes range4 ,the upper half of this gypseous formation is mainly formed of the same calcareous clay-shale rock1 but without any gypsum1 and varying extremely in nature3 it passes from a soft1 coarse1 earthy1 ferruginous state1 including particles of quartz1 into compact claystones with crystallised oxide of iron1--into porcellanic layers1 alternating with seams of calcareous matter1--and into green porcelain-jasper1 excessively hard1 but easily fusible4 ,strata of this nature alternate with much black and brown siliceo-calcareous slate1 remarkable from the wonderful number of huge embedded logs of silicified wood4 ,this wood1 according to ,mr4 ,r4 ,brown1 is 7judging from several specimens7 all coniferous4 ,some of the layers of the black siliceous slate contained irregular angular fragments of imperfect pitchstone1 which ,i believe1 as in the ,uspallata range1 has originated in a metamorphic process4 ,there was one bed of a marly tufaceous nature1 and of little specific gravity4 ,veins of agate and calcareous spar are numerous4 ,the whole of this gypseous formation1 especially the upper half1 has been injected1 metamorphosed1 and locally contorted by numerous hillocks of intrusive porphyries crowded together in an extraordinary manner4 ,these hillocks consist of purple claystone and of various other porphyries1 and of much white feldspathic greenstone passing into andesite2 this latter variety included in one case crystals of orthitic and albitic feldspar touching each other1 and others of hornblende1 chlorite1 and epidote4 ,the strata surrounding these intrusive hillocks at the mines of ,los ,hornos1 are intersected by many veins of copper-pyrites1 associated with much micaceous iron-ore1 and by some of gold3 in the neighbourhood of these veins the rocks are blackened and much altered4 ,the gypsum near the intrusive masses is always opaque4 ,one of these hillocks of porphyry was capped by some stratified porphyritic conglomerate1 which must have been brought up from below1 through the whole immense thickness of the overlying gypseous formation4 ,the lower beds of the gypseous formation resemble the corresponding and probably contemporaneous strata of the main ,cordillera2 whilst the upper beds in several respects resemble those of the ,uspallata chain1 and possibly may be contemporaneous with them2 for ,i have endeavoured to show that the ,uspallata beds were accumulated subsequently to the gypseous or ,neocomian formations of the ,cordillera4 ,this pile of strata dips at an angle of about #bj degrees to ,n4,e4 by ,n41 close up to the foot of the ,cuesta de ,los ,hornos1 a crooked range of mountains formed of intrusive rocks of the same nature with the above described hillocks4 ,only in one or two places1 on this south-eastern side of the range1 ,i noticed a narrow fringe of the upper gypseous strata brushed up and inclined south-eastward from it4 ,on its north-eastern flank1 and likewise on a few of the summits1 the stratified porphyritic conglomerate is inclined ,n4,e43 so that1 if we disregard the very narrow anticlinal fringe of gypseous strata at its ,s4,e4 foot1 this range forms a second uniclinal axis of elevation4 ,proceeding in a north-by-east direction to the village of ,combarbala1 we come to a third escarpment of the porphyritic conglomerate1 dipping eastwards1 and forming the outer range of the main ,cordillera4 ,the lower beds were here more jaspery than usual1 and they included some white cherty strata and red sandstones1 alternating with purple claystone porphyry4 ,higher up in the ,cordillera there appeared to be a line of andesitic rocks2 and beyond them1 a fourth escarpment of the porphyritic conglomerate1 again dipping eastwards or inwards4 ,the overlying gypseous strata1 if they ever existed here1 have been entirely removed4 ,copper mines of ,panuncillo4--,from ,combarbala to ,coquimbo1 ,i traversed the country in a zigzag direction1 crossing and recrossing the porphyritic conglomerate and finding in the granitic districts an unusual number of mountain-masses composed of various intrusive1 porphyritic rocks1 many of them andesitic4 ,one common variety was greenish-black1 with large crystals of blackish albite4 ,at ,panuncillo a short ,n4,n4,w4 and ,s4,s4,e4 ridge1 with a nucleus formed of greenstone and of a slate-coloured porphyry including crystals of glassy feldspar1 deserves notice1 from the very singular nature of the almost vertical strata composing it4 ,these consist chiefly of a finer and coarser granular mixture1 not very compact1 of white carbonate of lime1 of protoxide of iron and of yellowish garnets 7ascertained by ,professor ,miller71 each grain being an almost perfect crystal4 ,some of the varieties consist exclusively of granules of the calcareous spar2 and some contain grains of copper ore1 and1 ,i believe1 of quartz4 ,these strata alternate with a bluish1 compact1 fusible1 feldspathic rock4 ,much of the above granular mixture has1 also1 a pseudo-brecciated structure1 in which fragments are obscurely arranged in planes parallel to those of the stratification1 and are conspicuous on the weathered surfaces4 ,the fragments are angular or rounded1 small or large1 and consist of bluish or reddish compact feldspathic matter1 in which a few acicular crystals of feldspar can sometimes be seen4 ,the fragments often blend at their edges into the surrounding granular mass1 and seem due to a kind of concretionary action4 ,these singular rocks are traversed by many copper veins1 and appear to rest conformably on the granular mixture 7in parts as fine-grained as a sandstone7 of quartz1 mica1 hornblende1 and feldspar2 and this on fine-grained1 common gneiss2 and this on a laminated mass1 composed of pinkish orthitic feldspar1 including a few specks of hornblende2 and lastly1 this on granite1 which together with andesitic rocks1 form the surrounding district4 ,coquimbo3 ,mining district of ,arqueros4--,at ,coquimbo the porphyritic conglomerate formation approaches nearer to the ,pacific than in any other part of ,chile visited by me1 being separated from the coast by a tract only a few miles broad of the usual plutonic rocks1 with the addition of a porphyry having a red euritic base4 ,in proceeding to the mines of ,arqueros1 the strata of porphyritic conglomerate are at first nearly horizontal1 an unusual circumstance1 and afterwards they dip gently to ,s4,s4,e4 ,after having ascended to a considerable height1 we come to an undulatory district in which the famous silver mines are situated2 my examination was chiefly confined to those of ,s4 ,rosa4 ,most of the rocks in this district are stratified1 dipping in various directions1 and many of them are of so singular a nature1 that at the risk of being tedious ,i must briefly describe them4 ,the commonest variety is a dull-red1 compact1 finely brecciated stone1 containing much iron and innumerable white crystallised particles of carbonate of lime1 and minute extraneous fragments4 ,another variety is almost equally common near ,s4 ,rosa2 it has a bright green1 scanty basis1 including distinct crystals and patches of white carbonate of lime1 and grains of red1 semi-micaceous oxide of iron2 in parts the basis becomes dark green1 and assumes an obscure crystalline arrangement1 and occasionally in parts it becomes soft and slightly translucent like soapstone4 ,these red and green rocks are often quite distinct1 and often pass into each other2 the passage being sometimes affected by a fine brecciated structure1 particles of the red and green matter being mingled together4 ,some of the varieties appear gradually to become porphyritic with feldspar2 and all of them are easily fusible into pale or dark-coloured beads1 strongly attracted by the magnet4 ,i should perhaps have mistaken several of these stratified rocks for submarine lavas1 like some of those described at the ,puente del ,inca1 had ,i not examined1 a few leagues eastward of this point1 a fine series of analogous but less metamorphosed1 sedimentary beds belonging to the gypseous formation1 and probably derived from a volcanic source4 ,this formation is intersected by numerous metalliferous veins1 running1 though irregularly1 ,n4,w4 and ,s4,e41 and generally at right angles to the many dikes4 ,the veins consist of native silver1 of muriate of silver1 an amalgam of silver1 cobalt1 antimony1 and arsenic1#a generally embedded in sulphate of barytes4 ,i was assured by ,mr4 ,lambert1 that native copper without a trace of silver has been found in the same vein with native silver without a trace of copper4 ,at the mines of ,aristeas1 the silver veins are said to be unproductive as soon as they pass into the green strata1 whereas at ,s4 ,rosa1 only two or three miles distant1 the reverse happens2 and at the time of my visit1 the miners were working through a red stratum1 in the hope of the vein becoming productive in the underlying green sedimentary mass4 ,i have a specimen of one of these green rocks1 with the usual granules of white calcareous spar and red oxide of iron1 abounding with disseminated particles of glittering native and muriate of silver1 yet taken at the distance of one yard from any vein1--a circumstance1 as ,i was assured1 of very rare occurrence4 #a ,see the ,report on ,m4 ,domeyko's account of those mines1 in the 8,comptes ,rendus10 tome xiv1 p4 #efj4 ,section eastward1 up the ,valley of ,coquimbo4--,after passing for a few miles over the coast granitic series1 we come to the porphyritic conglomerate1 with its usual characters1 and with some of the beds distinctly displaying their mechanical origin4 ,the strata1 where first met with1 are1 as before stated1 only slightly inclined2 but near the ,hacienda of ,pluclaro1 we come to an anticlinal axis1 with the beds much dislocated and shifted by a great fault1 of which not a trace is externally seen in the outline of the hill4 ,i believe that this anticlinal axis can be traced northwards1 into the district of ,arqueros1 where a conspicuous hill called ,cerro ,blanco1 formed of a harsh1 cream-coloured euritic rock1 including a few crystals of reddish feldspar1 and associated with some purplish claystone porphyry1 seems to fall on a line of elevation4 ,in descending from the ,arqueros district1 ,i crossed on the northern border of the valley1 strata inclined eastward from the ,pluclaro axis3 on the porphyritic conglomerate there rested a mass1 some hundred feet thick1 of brown argillaceous limestone1 in parts crystalline1 and in parts almost composed of ,hippurites ,chilensis1 d',orbigny2 above this came a black calcareous shale1 and on it a red conglomerate4 ,in the brown limestone1 with the ,hippurites1 there was an impression of a ,pecten and a coral1 and great numbers of a large ,grypha1 very like1 and1 according to ,professor ,e4 ,forbes1 probably identical with ,g4 ,orientalis1 ,forbes ,,ms41--a cretaceous species 7probably upper greensand7 from ,verdachellum1 in ,southern ,india4 ,these fossils seem to occupy nearly the same position with those at the ,puente del ,inca1--namely1 at the top of the porphyritic conglomerate1 and at the base of the gypseous formation4 ,a little above the ,hacienda of ,pluclaro1 ,i made a detour on the northern side of the valley1 to examine the superincumbent gypseous strata1 which ,i estimated at #f1jjj feet in thickness4 ,the uppermost beds of the porphyritic conglomerate1 on which the gypseous strata conformably rest1 are variously coloured1 with one very singular and beautiful stratum composed of purple pebbles of various kinds of porphyry1 embedded in white calcareous spar1 including cavities lined with bright-green crystallised epidote4 ,the whole pile of strata belonging to both formations is inclined1 apparently from the above-mentioned axis of ,pluclaro1 at an angle of between #bj and #cj degrees to the east4 ,i will here give a section of the principal beds met with in crossing the entire thickness of the gypseous strata4 ,firstly3 above the porphyritic conglomerate formation1 there is a fine-grained1 red1 crystalline sandstone4 ,secondly3 a thick mass of smooth-grained1 calcareo-aluminous1 shaly rock1 often marked with dendritic manganese1 and having1 where most compact1 the external appearance of honestone4 ,it is easily fusible4 ,i shall for the future1 for convenience' sake1 call this variety pseudo-honestone4 ,some of the varieties are quite black when freshly broken1 but all weather into a yellowish-ash coloured1 soft1 earthy substance1 precisely as is the case with the compact shaly rocks of the ,peuquenes range4 ,this stratum is of the same general nature with many of the beds near ,los ,hornos in the ,illapel section4 ,in this second bed1 or in the underlying red sandstone 7for the surface was partially concealed by detritus71 there was a thick mass of gypsum1 having the same mineralogical characters with the great beds described in our sections across the ,cordillera4 ,thirdly3 a thick stratum of fine-grained1 red1 sedimentary matter1 easily fusible into a white glass1 like the basis of claystone porphyry2 but in parts jaspery1 in parts brecciated1 and including crystalline specks of carbonate of lime4 ,in some of the jaspery layers1 and in some of the black siliceous slaty bands1 there were irregular seams of imperfect pitchstone1 undoubtedly of metamorphic origin1 and other seams of brown1 crystalline limestone4 ,here1 also1 were masses1 externally resembling ill-preserved silicified wood4 ,fourthly and fifthly3 calcareous pseudo-honestone2 and a thick stratum concealed by detritus4 ,sixthly3 a thinly stratified mass of bright green1 compact1 smooth-grained1 calcareo-argillaceous stone1 easily fusible1 and emitting a strong aluminous odour3 the whole has a highly angulo-concretionary structure2 and it resembles1 to a certain extent1 some of the upper tufaceo-infusorial deposits of the ,patagonian tertiary formation4 ,it is in its nature allied to our pseudo-honestone1 and it includes well characterised layers of that variety2 and other layers of a pale green1 harder1 and brecciated variety2 and others of red sedimentary matter1 like that of bed ,three4 ,some pebbles of porphyries are embedded in the upper part4 ,seventhly3 red sedimentary matter or sandstone like that of bed ,one1 several hundred feet in thickness1 and including jaspery layers1 often having a finely brecciated structure4 ,eighthly3 white1 much indurated1 almost crystalline tuff1 several hundred feet in thickness1 including rounded grains of quartz and particles of green matter like that of bed ,six4 ,parts pass into a very pale green1 semi-porcellanic stone4 ,ninthly3 red or brown coarse conglomerate1 three or four hundred feet thick1 formed chiefly of pebbles of porphyries1 with volcanic particles1 in an arenaceous1 non-calcareous1 fusible basis3 the upper two feet are arenaceous without any pebbles4 ,tenthly3 the last and uppermost stratum here exhibited1 is a compact1 slate-coloured porphyry1 with numerous elongated crystals of glassy feldspar1 from one hundred and fifty to two hundred feet in thickness2 it lies strictly conformably on the underlying conglomerate1 and is undoubtedly a submarine lava4 ,this great pile of strata has been broken up in several places by intrusive hillocks of purple claystone porphyry1 and by dikes of porphyritic greenstone3 it is said that a few poor metalliferous veins have been discovered here4 ,from the fusible nature and general appearance of the finer-grained strata1 they probably owe their origin 7like the allied beds of the ,uspallata range1 and of the ,upper ,patagonian tertiary formations71 to gentle volcanic eruptions1 and to the abrasion of volcanic rocks4 ,comparing these beds with those in the mining district of ,arqueros1 we see at both places rocks easily fusible1 of the same peculiar bright green and red colours1 containing calcareous matter1 often having a finely brecciated structure1 often passing into each other1 and often alternating together3 hence ,i cannot doubt that the only difference between them1 lies in the ,arqueros beds having been more metamorphosed 7in conformity with their more dislocated and injected condition71 and consequently in the calcareous matter1 oxide of iron and green colouring matter1 having been segregated under a more crystalline form4 ,the strata are inclined1 as before stated1 from #bj to #cj eastward1 towards an irregular north and south chain of andesitic porphyry and of porphyritic greenstone1 where they are abruptly cut off4 ,in the valley of ,coquimbo1 near to the ,h4 of ,gualliguaca1 similar plutonic rocks are met with1 apparently a southern prolongation of the above chain2 and eastward of it we have an escarpment of the porphyritic conglomerate1 with the strata inclined at a small angle eastward1 which makes the third escarpment1 including that nearest the coast4 ,proceeding up the valley we come to another north and south line of granite1 andesite1 and blackish porphyry1 which seem to lie in an irregular trough of the porphyritic conglomerate4 ,again1 on the south side of the ,r4 ,claro1 there are some irregular granitic hills1 which have thrown off the strata of porphyritic conglomerate to the ,n4,w4 by ,w42 but the stratification here has been much disturbed4 ,i did not proceed any farther up the valley1 and this point is about two-thirds of the distance between the ,pacific and the main ,cordillera4 ,i will describe only one other section1 namely1 on the north side of the ,r4 ,claro1 which is interesting from containing fossils3 the strata are much dislocated by faults and dikes1 and are inclined to the north1 towards a mountain of andesite and porphyry1 into which they appear to become almost blended4 ,as the beds approach this mountain1 their inclination increases up to an angle of #gj1 and in the upper part1 the rocks become highly metamorphosed4 ,the lowest bed visible in this section1 is a purplish hard sandstone4 ,secondly1 a bed two or three hundred feet thick1 of a white siliceous sandstone1 with a calcareous cement1 containing seams of slaty sandstone1 and of hard yellowish-brown 7dolomitic87 limestone2 numerous1 well-rounded1 little pebbles of quartz are included in the sandstone4 ,thirdly1 a dark coloured limestone with some quartz pebbles1 from fifty to sixty feet in thickness1 containing numerous silicified shells1 presently to be enumerated4 ,fourthly1 very compact1 calcareous1 jaspery sandstone1 passing into 7fifthly7 a great bed1 several hundred feet thick1 of conglomerate1 composed of pebbles of white1 red1 and purple porphyries1 of sandstone and quartz1 cemented by calcareous matter4 ,i observed that some of the finer parts of this conglomerate were much indurated within a foot of a dike eight feet in width1 and were rendered of a paler colour with the calcareous matter segregated into white crystallised particles2 some parts were stained green from the colouring matter of the dike4 ,sixthly1 a thick mass1 obscurely stratified1 of a red sedimentary stone or sandstone1 full of crystalline calcareous matter1 imperfect crystals of oxide of iron1 and ,i believe of feldspar1 and therefore closely resembling some of the highly metamorphosed beds at ,arqueros3 this bed was capped by1 and appeared to pass in its upper part into1 rocks similarly coloured1 containing calcareous matter1 and abounding with minute crystals1 mostly elongated and glassy1 of reddish albite4 ,seventhly1 a conformable stratum of fine reddish porphyry with large crystals of 7albitic87 feldspar2 probably a submarine lava4 ,eighthly1 another conformable bed of green porphyry1 with specks of green earth and cream-coloured crystals of feldspar4 ,i believe that there are other superincumbent crystalline strata and submarine lavas1 but ,i had not time to examine them4 ,the upper beds in this section probably correspond with parts of the great gypseous formation2 and the lower beds of red sandstone conglomerate and fossiliferous limestone no doubt are the equivalents of the ,hippurite stratum1 seen in descending from ,arqueros to ,pluclaro1 which there lies conformably upon the porphyritic conglomerate formation4 ,the fossils found in the third bed1 consist of3-- ,pecten ,dufreynoyi1 d',orbigny1 8,voyage1 ,part ,pal40 ,this species1 which occurs here in vast numbers1 according to ,m4 ,d',orbigny1 resembles certain cretaceous forms4 ,ostrea hemispherica1 d',orbigny1 8,voyage0 etc4 ,also resembles1 according to the same author1 cretaceous forms4 ,terebratula nigma1 d',orbigny1 8,voyage0 etc4 7,pl4 ,,xxii1 ,figs4 #aj-ab47 ,is allied1 according to ,m4 d',orbigny1 to ,t4 concinna from the ,forest ,marble4 ,a series of this species1 collected in several localities hereafter to be referred to1 has been laid before ,professor ,forbes2 and he informs me that many of the specimens are almost undistinguishable from our oolitic ,t4 tetrdra1 and that the varieties amongst them are such as are found in that variable species4 ,generally speaking1 the ,american specimens of ,t4 nigma may be distinguished from the ,british ,t4 tetrdra1 by the surface having the ribs sharp and well-defined to the beak1 whilst in the ,british species they become obsolete and smoothed down2 but this difference is not constant4 ,professor ,forbes adds1 that1 possibly1 internal characters may exist1 which would distinguish the ,american species from its ,european allies4 ,spirifer linguiferoides1 ,e4 ,forbes4 ,professor ,forbes states that this species is very near to ,s4 linguifera of ,phillips 7a carboniferous limestone fossil71 but probably distinct4 ,m4 d',orbigny considers it as perhaps indicating the ,jurassic period4 ,ammonites1 imperfect impression of4 ,m4 ,domeyko has sent to ,france a collection of fossils1 which1 ,i presume1 from the description given1 must have come from the neighbourhood of ,arqueros2 they consist of3-- ,pecten ,dufreynoyi1 d',orbigny1 8,voyage0 ,part ,pal4 ,ostrea hemispherica1 d',orbigny1 8,voyage0 ,part ,pal4 ,turritella ,andii1 d',orbigny1 8,voyage0 ,part ,pal4 7,pleurotomaria ,humboldtii of ,von ,buch74 ,hippurites ,chilensis1 d',orbigny1 8,voyage0 ,part ,pal4 ,the specimens of this ,hippurite1 as well as those ,i collected in my descent from ,arqueros1 are very imperfect2 but in ,m4 d',orbigny's opinion they resemble1 as does the ,turritella ,andii1 cretaceous 7upper greensand7 forms4 ,nautilus ,domeykus1 d',orbigny1 8,voyage0 ,part ,pal4 ,terebratula nigma1 d',orbigny1 8,voyage0 ,part ,pal4 ,terebratula ignaciana1 d',orbigny1 8,voyage0 ,part ,pal4 ,this latter species was found by ,m4 ,domeyko in the same block of limestone with the ,t4 nigma4 ,according to ,m4 d',orbigny1 it comes near to ,t4 ornithocephala from the ,lias4 ,a series of this species collected at ,guasco1 has been examined by ,professor ,e4 ,forbes1 and he states that it is difficult to distinguish between some of the specimens and the ,t4 hastata from the mountain limestone2 and that it is equally difficult to draw a line between them and some ,marlstone ,terebratul4 ,without a knowledge of the internal structure1 it is impossible at present to decide on their identity with analogous ,european forms4 ,the remarks given on the several foregoing shells1 show that1 in ,m4 d',orbigny's opinion1 the ,pecten1 ,ostrea1 ,turritella1 and ,hippurite indicate the cretaceous period2 and the ,grypha appears to ,professor ,forbes to be identical with a species1 associated in ,southern ,india with unquestionably cretaceous forms4 ,on the other hand1 the two ,terebratul and the ,spirifer point1 in the opinion both of ,m4 d',orbigny and ,professor ,forbes1 to the oolitic series4 ,hence ,m4 d',orbigny1 not having himself examined this country1 has concluded that there are here two distinct formations2 but the ,spirifer and ,t4 nigma were certainly included in the same bed with the ,pecten and ,ostrea1 whence ,i extracted them2 and the geologist ,m4 ,domeyko sent home the two ,terebratul with the other-named shells1 from the same locality1 without specifying that they came from different beds4 ,again1 as we shall presently see1 in a collection of shells given me from ,guasco1 the same species1 and others presenting analogous differences1 are mingled together1 and are in the same condition2 and lastly1 in three places in the valley of ,copiapo1 ,i found some of these same species similarly grouped4 ,hence there cannot be any doubt1 highly curious though the fact be1 that these several fossils1 namely1 the ,hippurites1 ,grypha1 ,ostrea1 ,pecten1 ,turritella1 ,nautilus1 two ,terebratul1 and ,spirifer all belong to the same formation1 which would appear to form a passage between the oolitic and cretaceous systems of ,europe4 ,although aware how unusual the term must sound1 ,i shall1 for convenience' sake1 call this formation cretaceo-oolitic4 ,comparing the sections in this valley of ,coquimbo with those in the ,cordillera described in the last chapter1 and bearing in mind the character of the beds in the intermediate district of ,los ,hornos1 there is certainly a close general mineralogical resemblance between them1 both in the underlying porphyritic conglomerate1 and in the overlying gypseous formation4 ,considering this resemblance1 and that the fossils from the ,puente del ,inca at the base of the gypseous formation1 and throughout the greater part of its entire thickness on the ,peuquenes range1 indicate the ,neocomian period1--that is1 the dawn of the cretaceous system1 or1 as some have believed1 a passage between this latter and the oolitic series--,i conclude that probably the gypseous and associated beds in all the sections hitherto described1 belong to the same great formation1 which ,i have denominated--cretaceo-oolitic4 ,i may add1 before leaving ,coquimbo1 that ,m4 ,gay found in the neighbouring ,cordillera1 at the height of #ad1jjj feet above the sea1 a fossiliferous formation1 including a ,trigonia and ,pholadomya2#b--;both of which genera occur at the ,puente del ,inca4 #b ,d',orbigny1 8,voyage10 ,part ,golog41 p4 #bdb4 ,coquimbo to ,guasco4--,the rocks near the coast1 and some way inland1 do not differ from those described northwards of ,valparaiso3 we have much greenstone1 syenite1 feldspathic and jaspery slate1 and grauwackes having a basis like that of claystone2 there are some large tracts of granite1 in which the constituent minerals are sometimes arranged in folia1 thus composing an imperfect gneiss4 ,there are two large districts of mica-schists1 passing into glossy clay-slate1 and resembling the great formation in the ,chonos ,archipelago4 ,in the valley of ,guasco1 an escarpment of porphyritic conglomerate is first seen high up the valley1 about two leagues eastward of the town of ,ballenar4 ,i heard of a great gypseous formation in the ,cordillera2 and a collection of shells made there was given me4 ,these shells are all in the same condition1 and appear to have come from the same bed3 they consist of3-- ,turritella ,andii1 d',orbigny1 8,voyage10 ,part ,pal4 ,pecten ,dufreynoyi1 d',orbigny1 8,voyage10 ,part ,pal4 ,terebatula ignaciana1 d',orbigny1 8,voyage10 ,part ,pal4 ,the relations of these species have been given under the head of ,coquimbo4 ,terebratula nigma1 d',orbigny1 8,voyage10 ,part ,pal4 ,this shell ,m4 d',orbigny does not consider identical with his ,t4 nigma1 but near to ,t4 obsoleta4 ,professor ,forbes thinks that it is certainly a variety of ,t4 nigma3 we shall meet with this variety again at ,copiapo4 ,spirifer ,chilensis1 ,e4 ,forbes4 ,professor ,forbes remarks that this fossil resembles several carboniferous limestone ,spirifers2 and that it is also related to some liassic species1 as ,s4 ,wolcotii4 ,if these shells had been examined independently of the other collections1 they would probably have been considered1 from the characters of the two ,terebratul1 and from the ,spirifer1 as oolitic2 but considering that the first species1 and according to ,professor ,forbes1 the four first1 are identical with those from ,coquimbo1 the two formations no doubt are the same1 and may1 as ,i have said1 be provisionally called cretaceo-oolitic4 ,valley of ,copiapo4--,the journey from ,guasco to ,copiapo1 owing to the utterly desert nature of the country1 was necessarily so hurried1 that ,i do not consider my notes worth giving4 ,in the valley of ,copiapo some of the sections are very interesting4 ,from the sea to the town of ,copiapo1 a distance estimated at thirty miles1 the mountains are composed of greenstone1 granite1 andesite1 and blackish porphyry1 together with some dusky-green feldspathic rocks1 which ,i believe to be altered clay-slate3 these mountains are crossed by many brown-coloured dikes1 running north and south4 ,above the town1 the main valley runs in a south-east and even more southerly course towards the ,cordillera1 where it is divided into three great ravines1 by the northern one of which1 called ,jolquera1 ,i penetrated for a short distance4 ,the section1 ,fig4 #c in ,plate ,v1 gives an eye-sketch of the structure and composition of the mountains on both sides of this valley3 a straight east and west line from the town to the ,cordillera is perhaps not more than thirty miles1 but along the valley the distance is much greater4 ,wherever the valley trended very southerly1 ,i have endeavoured to contract the section into its true proportion4 ,this valley1 ,i may add1 rises much more gently than any other valley which ,i saw in ,chile4 ,to commence with our section1 for a short distance above the town we have hills of the granitic series1 together with some of that rock ,a1 which ,i suspect to be altered clay-slate1 but which ,professor ,g4 ,rose1 judging from specimens collected by ,meyen at ,p4 ,negro1 states is serpentine passing into greenstone4 ,we then come suddenly to the great gypseous formation ,b1 without having passed over1 differently from1 in all the sections hitherto described1 any of the porphyritic conglomerate4 ,the strata are at first either horizontal or gently inclined westward2 then highly inclined in various directions1 and contorted by underlying masses of intrusive rocks2 and lastly1 they have a regular eastward dip1 and form a tolerably well pronounced north and south line of hills4 ,this formation consists of thin strata1 with innumerable alternations1 of black1 calcareous slate-rock1 of calcareo-aluminous stones like those at ,coquimbo1 which ,i have called pseudo-honestones of green jaspery layers1 and of pale-purplish1 calcareous1 soft rotten-stone1 including seams and veins of gypsum4 ,these strata are conformably overlaid by a great thickness of thinly stratified1 compact limestone with included crystals of carbonate of lime4 ,at a place called ,tierra ,amarilla1 at the foot of a mountain thus composed there is a broad vein1 or perhaps stratum1 of a beautiful and curious crystallised mixture1 composed1 according to ,professor ,g4 ,rose1#c of sulphate of iron under two forms1 and of the sulphates of copper and alumina3 the section is so obscure that ,i could not make out whether this vein or stratum occurred in the gypseous formation1 or more probably in some underlying masses ,a1 which ,i believe are altered clay-slate4 #c ,meyen's 8,reise10 etc41 ,th4 ,i1 s4 #cid4 ,second axis of elevation4--,after the gypseous masses ,b1 we come to a line of hills of unstratified porphyry ,c1 which on their eastern side blend into strata of great thickness of porphyritic conglomerate1 dipping eastward4 ,this latter formation1 however1 here has not been nearly so much metamorphosed as in most parts of ,central ,chile2 it is composed of beds of true purple claystone porphyry1 repeatedly alternating with thick beds of purplish-red conglomerate with the well-rounded1 large pebbles of various porphyries1 not blended together4 ,third axis of elevation4--,near the ravine of ,los ,hornitos1 there is a well-marked line of elevation1 extending for many miles in a ,n4,n4,e4 and ,s4,s4,w4 direction1 with the strata dipping in most parts 7as in the second axis7 only in one direction1 namely1 eastward at an average angle of between #cj and #dj4 ,close to the mouth of the valley1 however1 there is1 as represented in the section1 a steep and high mountain ,d1 composed of various green and brown intrusive porphyries enveloped with strata1 apparently belonging to the upper parts of the porphyritic conglomerate1 and dipping both eastward and westward4 ,i will describe the section seen on the eastern side of this mountain ,d1 beginning at the base with the lowest bed visible in the porphyritic conglomerate1 and proceeding upwards through the gypseous formation4 ,bed #a consists of reddish and brownish porphyry varying in character1 and in many parts highly amygdaloidal with carbonate of lime1 and with bright green and brown bole4 ,its upper surface is throughout clearly defined1 but the lower surface is in most parts indistinct1 and towards the summit of the mountain ,d quite blended into the intrusive porphyries4 ,bed #b1 a pale lilac1 hard but not heavy stone1 slightly laminated1 including small extraneous fragments1 and imperfect as well as some perfect and glassy crystals of feldspar2 from one hundred and fifty to two hundred feet in thickness4 ,when examining it in situ1 ,i thought it was certainly a true porphyry1 but my specimens now lead me to suspect that it possibly may be a metamorphosed tuff4 ,from its colour it could be traced for a long distance1 overlying in one part1 quite conformably to the porphyry of bed #a1 and in another not distant part1 a very thick mass of conglomerate1 composed of pebbles of a porphyry chiefly like that of bed #a3 this fact shows how the nature of the bottom formerly varied in short horizontal distances4 ,bed #c1 white1 much indurated tuff1 containing minute pebbles1 broken crystals1 and scales of mica1 varies much in thickness4 ,this bed is remarkable from containing many globular and pear-shaped1 externally rusty balls1 from the size of an apple to a man's head1 of very tough1 slate-coloured porphyry1 with imperfect crystals of feldspar3 in shape these balls do not resemble pebbles1 and i believe that they are subaqueous volcanic bombs2 they differ from subaerial bombs only in not being vesicular4 ,bed #d2 a dull purplish-red1 hard conglomerate1 with crystallised particles and veins of carbonate of lime1 from three hundred to four hundred feet in thickness4 ,the pebbles are of claystone porphyries of many varieties2 they are tolerably well rounded1 and vary in size from a large apple to a man's head4 ,this bed includes three layers of coarse1 black1 calcareous1 somewhat slaty rock3 the upper part passes into a compact red sandstone4 ,in a formation so highly variable in mineralogical nature1 any division not founded on fossil remains1 must be extremely arbitrary3 nevertheless1 the beds below the last conglomerate may1 in accordance with all the sections hitherto described1 be considered as belonging to the porphyritic conglomerate1 and those above it to the gypseous formation1 marked ,e in the section4 ,the part of the valley in which the following beds are seen is near ,potrero ,seco4 ,bed #e1 compact1 fine-grained1 pale greenish-grey1 non-calcareous1 indurated mudstone1 easily fusible into a pale green and white glass4 ,bed #f1 purplish1 coarse-grained1 hard sandstone1 with broken crystals of feldspar and crystallised particles of carbonate of lime2 it possesses a slightly nodular structure4 ,bed #g1 blackish-grey1 much indurated1 calcareous mudstone1 with extraneous particles of unequal size2 the whole being in parts finely brecciated4 ,in this mass there is a stratum1 twenty feet in thickness1 of impure gypsum4 ,bed #h1 a greenish mudstone1 with several layers of gypsum4 ,bed #i1 a highly indurated1 easily fusible1 white tuff1 thickly mottled with ferruginous matter1 and including some white semi-porcellanic layers1 which are interlaced with ferruginous veins4 ,this stone closely resembles some of the commonest varieties in the ,uspallata chain4 ,bed #aj1 a thick bed of rather bright green1 indurated mudstone or tuff1 with a concretionary nodular structure so strongly developed that the whole mass consists of balls4 ,i will not attempt to estimate the thickness of the strata in the gypseous formation hitherto described1 but it must certainly be very many hundred feet4 ,bed #aa is at least #hjj feet in thickness3 it consists of thin layers of whitish1 greenish1 or more commonly brown1 fine-grained1 indurated tuffs1 which crumble into angular fragments3 some of the layers are semi-porcellanic1 many of them highly ferruginous1 and some are almost composed of carbonate of lime and iron with drusy cavities lined with quartzf-crystals4 ,bed #ab1 dull purplish or greenish or dark-grey1 very compact and much indurated mudstone3 estimated at #a1ejj feet in thickness3 in some parts this rock assumes the character of an imperfect coarse clay-slate2 but viewed under a lens1 the basis always has a mottled appearance1 with the edges of the minute component particles blending together4 ,parts are calcareous1 and there are numerous veins of highly crystalline carbonate of lime charged with iron4 ,the mass has a nodular structure1 and is divided by only a few planes of stratification3 there are1 however1 two layers1 each about eighteen inches thick1 of a dark brown1 finer-grained stone1 having a conchoidal1 semi-porcellanic fracture1 which can be followed with the eye for some miles across the country4 ,i believe this last great bed is covered by other nearly similar alternations2 but the section is here obscured by a tilt from the next porphyritic chain1 presently to be described4 ,i have given this section in detail1 as being illustrative of the general character of the mountains in this neighbourhood2 but it must not be supposed that any one stratum long preserves the same character4 ,at a distance of between only two and three miles the green mudstones and white indurated tuffs are to a great extent replaced by red sandstone and black calcareous shaly rocks1 alternating together4 ,the white indurated tuff1 bed #aa1 here contains little or no gypsum1 whereas on the northern and opposite side of the valley1 it is of much greater thickness and abounds with layers of gypsum1 some of them alternating with thin seams of crystalline carbonate of lime4 ,the uppermost1 dark-coloured1 hard mudstone1 bed #ab1 is in this neighbourhood the most constant stratum4 ,the whole series differs to a considerable extent1 especially in its upper part1 from that met with at ,,bb1 in the lower part of the valley2 nevertheless1 ,i do not doubt that they are equivalents4 ,fourth axis of elevation 7,valley of ,copiapo74--,this axis is formed of a chain of mountains ,f1 of which the central masses 7near ,la ,punta7 consist of andesite containing green hornblende and coppery mica1 and the outer masses of greenish and black porphyries1 together with some fine lilac-coloured claystone porphyry2 all these porphyries being injected and broken up by small hummocks of andesite4 ,the central great mass of this latter rock1 is covered on the eastern side by a black1 fine-grained1 highly micaceous slate1 which1 together with the succeeding mountains of porphyry1 are traversed by numerous white dikes1 branching from the andesite1 and some of them extending in straight lines1 to a distance of at least two miles4 ,the mountains of porphyry eastward of the micaceous schist soon1 but gradually1 assume 7as observed in so many other cases7 a stratified structure1 and can then be recognised as a part of the porphyritic conglomerate formation4 ,these strata ,g are inclined at a high angle to the ,s4,e41 and form a mass from fifteen hundred to two thousand feet in thickness4 ,the gypseous masses to the west already described1 dip directly towards this axis1 with the strata only in a few places 7one of which is represented in the section7 thrown from it3 hence this fourth axis is mainly uniclinal towards the ,s4,e41 and just like our third axis1 only locally anticlinal4 ,the above strata of porphyritic conglomerate ,g with their south-eastward dip1 come abruptly up against beds of the gypseous formation ,h1 which are gently1 but irregularly1 inclined westward3 so that there is here a synclinal axis and great fault4 ,further up the valley1 here running nearly north and south1 the gypseous formation is prolonged for some distance2 but the stratification is unintelligible1 the whole being broken up by faults1 dikes1 and metalliferous veins4 ,the strata consist chiefly of red calcareous sandstones1 with numerous veins in the place of layers1 of gypsum2 the sandstone is associated with some black calcareous slate-rock1 and with green pseudo-honestones1 passing into porcelain-jasper4 ,still further up the valley1 near ,las ,amolanas ,i1 the gypseous strata become more regular1 dipping at an angle of between #cj and #dj degrees to ,w4,s4,w41 and conformably overlying1 near the mouth of the ravine of ,jolquera1 strata ,k of porphyritic conglomerate4 ,the whole series has been tilted by a partially concealed axis ,l1 of granite1 andesite1 and a granitic mixture of white feldspar1 quartz1 and oxide of iron4 ,fifth axis of elevation 7,valley of ,copiapo1 near ,las ,amolanas74--,i will describe in some detail the beds ,i seen here1 which1 as just stated1 dip to ,w4,s4,w41 at an angle of from #cj to #dj4 ,i had not time to examine the underlying porphyritic conglomerate1 of which the lowest beds1 as seen at the mouth of the ,jolquera1 are highly compact1 with crystals of red oxide of iron2 and ,i am not prepared to say whether they are chiefly of volcanic or metamorphic origin4 ,on these beds there rests a coarse purplish conglomerate1 very little metamorphosed1 composed of pebbles of porphyry1 but remarkable from containing one pebble of granite2--of which fact no instance has occurred in the sections hitherto described4 ,above this conglomerate1 there is a black siliceous claystone1 and above it numerous alternations of dark-purplish and green porphyries1 which may be considered as the uppermost limit of the porphyritic conglomerate formation4 ,above these porphyries comes a coarse1 arenaceous conglomerate1 the lower half white and the upper half of a pink colour1 composed chiefly of pebbles of various porphyries1 but with some of red sandstone and jaspery rocks4 ,in some of the more arenaceous parts of the conglomerate1 there was an oblique or current lamination2 a circumstance which ,i did not elsewhere observe4 ,above this conglomerate1 there is a vast thickness of thinly stratified1 pale-yellowish1 siliceous sandstone1 passing into a granular quartz-rock1 used for grindstones 7hence the name of the place ,las ,amolanas71 and certainly belonging to the gypseous formation1 as does probably the immediately underlying conglomerate4 ,in this yellowish sandstone there are layers of white and pale-red siliceous conglomerate2 other layers with small1 well-rounded pebbles of white quartz1 like the bed at the ,r4 ,claro at ,coquimbo2 others of a greenish1 fine-grained1 less siliceous stone1 somewhat resembling the pseudo-honestones lower down the valley2 and lastly1 others of a black calcareous shale-rock4 ,in one of the layers of conglomerate1 there was embedded a fragment of mica-slate1 of which this is the first instance2 hence perhaps1 it is from a formation of mica-slate1 that the numerous small pebbles of quartz1 both here and at ,coquimbo1 have been derived4 ,not only does the siliceous sandstone include layers of the black1 thinly stratified1 not fissile1 calcareous shale-rock1 but in one place the whole mass1 especially the upper part1 was1 in a marvellously short horizontal distance1 after frequent alternations1 replaced by it4 ,when this occurred1 a mountain-mass1 several thousand feet in thickness was thus composed2 the black calcareous shale-rock1 however1 always included some layers of the pale-yellowish siliceous sandstone1 of the red conglomerate1 and of the greenish jaspery and pseudo-honestone varieties4 ,it likewise included three or four widely separated layers of a brown limestone1 abounding with shells immediately to be described4 ,this pile of strata was in parts traversed by many veins of gypsum4 ,the calcareous shale-rock1 though when freshly broken quite black1 weathers into an ash- colour3 in which respect and in general appearance1 it perfectly resembles those great fossiliferous beds of the ,peuquenes range1 alternating with gypsum and red sandstone1 described in the last chapter4 ,the shells out of the layers of brown limestone1 included in the black calcareous shale-rock1 which latter1 as just stated1 replaces the white siliceous sandstone1 consist of3-- ,pecten ,dufreynoyi1 d',orbigny1 8,voyage10 ,part ,pal4 ,turritella ,andii1 d',orbigny1 8,voyage10 ,part ,pal4 ,astarte ,darwinii1 ,e4 ,forbes4 ,grypha ,darwinii1 ,e4 ,forbes4 ,an intermediate form between ,g4 gigantea and ,g4 incurva4 ,grypha nov4 spec481 ,e4 ,forbes4 ,perna ,americana1 ,e4 ,forbes4 ,avicula1 nov4 spec4 ,considered by ,mr4 ,g4 ,b4 ,sowerby as the ,a4 echinata1 by ,m4 d',orbigny as certainly a new and distinct species1 having a ,jurassic aspect4 ,the specimen has been unfortunately lost4 ,terebratula nigma1 d',orbigny1 7var4 of do4 ,e4 ,forbes47 ,this is the same variety1 with that from ,guasco1 considered by ,m4 ,d',orbigny to be a distinct species from his ,t4 nigma1 and related to ,t4 obsoleta4 ,plagiostoma and ,ammonites1 fragments of4 ,the lower layers of the limestone contained thousands of the ,grypha2 and the upper ones as many of the ,turritella1 with the ,grypha 7nov4 species7 and ,serpul adhering to them2 in all the layers1 the ,terebratula and fragments of the ,pecten were included4 ,it was evident1 from the manner in which species were grouped together1 that they had lived where now embedded4 ,before making any further remarks1 ,i may state1 that higher up this same valley we shall again meet with a similar association of shells2 and in the great ,despoblado ,valley1 which branches off near the town from that of ,copiapo1 the ,pecten ,dufreynoyi1 some ,gryphites 7,i believe ,g4 ,darwinii71 and the true ,terebratula nigma of d',orbigny were found together in an equivalent formation1 as will be hereafter seen4 ,a specimen also1 ,i may add1 of the true ,t4 nigma1 was given me from the neighbourhood of the famous silver mines of ,chanuncillo1 a little south of the valley of the ,copiapo1 and these mines1 from their position1 ,i have no doubt1 lie within the great gypseous formation3 the rocks close to one of the silver veins1 judging from fragments shown me1 resemble those singular metamorphosed deposits from the mining district of ,arqueros near ,coquimbo4 ,i will reiterate the evidence on the association of these several shells in the several localities4 ,coquimbo4 ,in the same bed1 ,rio ,claro3 ,pecten ,dufreynoyi4 ,ostrea hemispherica4 ,terebratula nigma4 ,spirifer linguiferoides4 ,same bed1 near ,arqueros3 ,hippurites ,chilensis4 ,grypha orientalis4 ,collected by ,m4 ,domeyko from the same locality1 apparently near ,arqueros3 ,terebratula nigma and ,terebratula ignaciana1 in same block of limestone4 ,pecten ,dufreynoyi4 ,ostrea hemispherica4 ,hippurites ,chilensis4 ,turritella ,andii4 ,nautilus ,domeykus4 ,guasco4 ,in a collection from the ,cordillera1 given me3 the specimens all in the same condition3 ,pecten ,dufreynoyi4 ,turritella ,andii4 ,terebratula ignaciana4 ,terebratula nigma1 var4 ,spirifer ,chilensis4 ,copiapo4 ,mingled together in alternating beds in the main valley of ,copiapo near ,las ,amolanas1 and likewise higher up the valley3 ,pecten ,dufreynoyi4 ,turritella ,andii4 ,terebratula nigma1 var41 as at ,guasco4 ,astarte ,darwinii4 ,grypha ,darwinii4 ,grypha nov4 species8 ,perna ,americana4 ,avicula1 nov4 species4 ,main valley of ,copiapo1 apparently same formation with that of ,amolanas3 ,terebratula nigma 7true74 ,in the same bed1 high up the great lateral valley of the ,despoblado1 in the ravine of ,maricongo3 ,terebratula nigma 7true74 ,pecten ,dufreynoyi4 ,grypha ,darwinii8 ,considering this table1 ,i think it is impossible to doubt that all these fossils belong to the same formation4 ,if1 however1 the species from ,las ,amolanas1 in the ,valley of ,copiapo1 had1 as in the case of those from ,guasco1 been separately examined1 they would probably have been ranked as oolitic2 for1 although no ,spirifers were found here1 all the other species1 with the exception of the ,pecten1 ,turritella1 and ,astarte1 have a more ancient aspect than cretaceous forms4 ,on the other hand1 taking into account the evidence derived from the cretaceous character of these three shells1 and of the ,hippurites1 ,grypha orientalis1 and ,ostrea1 from ,coquimbo1 we are driven back to the provisional name already used of cretaceo-oolitic4 ,from geological evidence1 ,i believe this formation to be the equivalent of the ,neocomian beds of the ,cordillera of ,central ,chile4 ,to return to our section near ,las ,amolanas3--,above the yellow siliceous sandstone1 or the equivalent calcareous slate-rock1 with its bands of fossil-shells1 according as the one or other prevails1 there is a pile of strata1 which cannot be less than from two to three thousand feet in thickness1 in main part composed of a coarse1 bright red conglomerate1 with many intercalated beds of red sandstone1 and some of green and other coloured porcelain-jaspery layers4 ,the included pebbles are well-rounded1 varying from the size of an egg to that of a cricket-ball1 with a few larger2 and they consist chiefly of porphyries4 ,the basis of the conglomerate1 as well as some of the alternating thin beds1 are formed of a red1 rather harsh1 easily fusible sandstone1 with crystalline calcareous particles4 ,this whole great pile is remarkable from the thousands of huge1 embedded1 silicified trunks of trees1 one of which was eight feet long1 and another eighteen feet in circumference3 how marvellous it is1 that every vessel in so thick a mass of wood should have been converted into silex6 ,i brought home many specimens1 and all of them1 according to ,mr4 ,r4 ,brown1 present a coniferous structure4 ,above this great conglomerate1 we have from two to three hundred feet in thickness of red sandstone2 and above this1 a stratum of black calcareous slate-rock1 like that which alternates with and replaces the underlying yellowish-white1 siliceous sandstone4 ,close to the junction between this upper black slate-rock and the upper red sandstone1 ,i found the ,grypha ,darwinii1 the ,turritella ,andii1 and vast numbers of a bivalve1 too imperfect to be recognised4 ,hence we see that1 as far as the evidence of these two shells serves--and the ,turritella is an eminently characteristic species--the whole thickness of this vast pile of strata belongs to the same age4 ,again1 above the last-mentioned upper red sandstone1 there were several alternations of the black1 calcareous slate-rock2 but ,i was unable to ascend to them4 ,all these uppermost strata1 like the lower ones1 vary extremely in character in short horizontal distances4 ,the gypseous formation1 as here seen1 has a coarser1 more mechanical texture1 and contains much more siliceous matter than the corresponding beds lower down the valley4 ,its total thickness1 together with the upper beds of the porphyritic conglomerate1 ,i estimated at least at #h1jjj feet2 and only a small portion of the porphyritic conglomerate1 which on the eastern flank of the fourth axis of elevation appeared to be from fifteen hundred to two thousand feet thick1 is here included4 ,as corroborative of the great thickness of the gypseous formation1 ,i may mention that in the ,despoblado ,valley 7which branches from the main valley a little above the town of ,copiapo7 ,i found a corresponding pile of red and white sandstones1 and of dark1 calcareous1 semi-jaspery mudstones1 rising from a nearly level surface and thrown into an absolutely vertical position2 so that1 by pacing1 ,i ascertained their thickness to be nearly two thousand seven hundred feet2 taking this as a standard of comparison1 ,i estimated the thickness of the strata above the porphyritic conglomerate at #g1jjj feet4 ,the fossils before enumerated1 from the limestone-layers in the whitish siliceous sandstone1 are now covered1 on the least computation1 by strata from #e1jjj to #f1jjj feet in thickness4 ,professor ,e4 ,forbes thinks that these shells probably lived at a depth of from about #cj to #dj fathoms1 that is from #ahj to #bdj feet2 anyhow1 it is impossible that they could have lived at the depth of from #e1jjj to #f1jjj feet4 ,hence in this case1 as in that of the ,puente del ,inca1 we may safely conclude that the bottom of the sea on which the shells lived1 subsided1 so as to receive the superincumbent submarine strata3 and this subsidence must have taken place during the existence of these shells2 for1 as ,i have shown1 some of them occur high up as well as low down in the series4 ,that the bottom of the sea subsided1 is in harmony with the presence of the layers of coarse1 well-rounded pebbles included throughout this whole pile of strata1 as well as of the great upper mass of conglomerate from #b1jjj to #c1jjj feet thick2 for coarse gravel could hardly have been formed or spread out at the profound depths indicated by the thickness of the strata4 ,the subsidence1 also1 must have been slow to have allowed of this often-recurrent spreading out of the pebbles4 ,moreover1 we shall presently see that the surfaces of some of the streams of porphyritic lava beneath the gypseous formation1 are so highly amygdaloidal that it is scarcely possible to believe that they flowed under the vast pressure of a deep ocean4 ,the conclusion of a great subsidence during the existence of these cretaceo-oolitic fossils1 may1 ,i believe1 be extended to the district of ,coquimbo1 although owing to the fossiliferous beds there not being directly covered by the upper gypseous strata1 which in the section north of the valley are about #f1jjj feet in thickness1 ,i did not there insist on this conclusion4 ,the pebbles in the above conglomerates1 both in the upper and lower beds1 are all well rounded1 and1 though chiefly composed of various porphyries1 there are some of red sandstone and of a jaspery stone1 both like the rocks intercalated in layers in this same gypseous formation2 there was one pebble of mica-slate and some of quartz1 together with many particles of quartz4 ,in these respects there is a wide difference between the gypseous conglomerates and those of the porphyritic-conglomerate formation1 in which latter1 angular and rounded fragments1 almost exclusively composed of porphyries1 are mingled together1 and which1 as already often remarked1 probably were ejected from craters deep under the sea4 ,from these facts ,i conclude1 that during the formation of the conglomerates1 land existed in the neighbourhood1 on the shores of which the innumerable pebbles were rounded and thence dispersed1 and on which the coniferous forests flourished--for it is improbable that so many thousand logs of wood should have drifted from any great distance4 ,this land1 probably islands1 must have been mainly formed of porphyries1 with some mica-slate1 whence the quartz was derived1 and with some red sandstone and jaspery rocks4 ,this latter fact is important1 as it shows that in this district1 even previously to the deposition of the lower gypseous or cretaceo-oolitic beds1 strata of an analogous nature had elsewhere1 no doubt in the more central ranges of the ,cordillera1 been elevated2 thus recalling to our minds the relations of the ,cumbre and ,uspallata chains4 ,having already referred to the great lateral valley of the ,despoblado1 ,i may mention that above the #b1gjj feet of red and white sandstone and dark mudstone1 there is a vast mass of coarse1 hard1 red conglomerate1 some thousand feet in thickness1 which contains much silicified wood1 and evidently corresponds with the great upper conglomerate at ,las ,amolanas3 here1 however1 the conglomerate consists almost exclusively of pebbles of granite1 and of disintegrated crystals of reddish feldspar and quartz firmly recemented together4 ,in this case1 we may conclude that the land whence the pebbles were derived1 and on which the now silicified trees once flourished1 was formed of granite4 ,the mountains near ,las ,amolanas1 composed of the cretaceo-oolitic strata1 are interlaced with dikes like a spider's web1 to an extent which ,i have never seen equalled1 except in the denuded interior of a volcanic crater3 north and south lines1 however1 predominate4 ,these dikes are composed of green1 white1 and blackish rocks1 all porphyritic with feldspar1 and often with large crystals of hornblende4 ,the white varieties approach closely in character to andesite1 which composes as we have seen1 the injected axes of so many of the lines of elevation4 ,some of the green varieties are finely laminated1 parallel to the walls of the dikes4 ,sixth axis of elevation 7,valley of ,copiapo74--,this axis consists of a broad mountainous mass ,o of andesite1 composed of albite1 brown mica1 and chlorite1 passing into andesitic granite1 with quartz3 on its western side it has thrown off1 at a considerable angle1 a thick mass of stratified porphyries1 including much epidote ,,nn1 and remarkable only from being divided into very thin beds1 as highly amygdaloidal on their surfaces as subaerial lava-streams are often vesicular4 ,this porphyritic formation is conformably covered1 as seen some way up the ravine of ,jolquera1 by a mere remnant of the lower part of the cretaceo-oolitic formation ,,mm1 which in one part encases1 as represented in the coloured section1 the foot of the andesitic axis ,l1 of the already described fifth line1 and in another part entirely conceals it3 in this latter case1 the gypseous or cretaceo-oolitic strata falsely appeared to dip under the porphyritic conglomerate of the fifth axis4 ,the lowest bed of the gypseous formation1 as seen here ,m1 is of yellowish siliceous sandstone1 precisely like that of ,amolanas1 interlaced in parts with veins of gypsum1 and including layers of the black1 calcareous1 non-fissile slate-rock3 the ,turritella ,andii1 ,pecten ,dufreynoyi1 ,terebratula nigma1 var41 and some ,gryphites were embedded in these layers4 ,the sandstone varies in thickness from only twenty to eighty feet2 and this variation is caused by the inequalities in the upper surface of an underlying stream of purple claystone porphyry4 ,hence the above fossils here lie at the very base of the gypseous or cretaceo-oolitic formation1 and hence they were probably once covered up by strata about seven thousand feet in thickness3 it is1 however1 possible1 though from the nature of all the other sections in this district not probable1 that the porphyritic claystone lava may in this case have invaded a higher level in the series4 ,above the sandstone there is a considerable mass of much indurated1 purplish-black1 calcareous claystone1 allied in nature to the often-mentioned black calcareous slate-rock4 ,eastward of the broad andesitic axis of this sixth line1 and penetrated by many dikes from it1 there is a great formation ,p of mica-schist1 with its usual variations1 and passing in one part into a ferruginous quartz-rock4 ,the folia are curved and highly inclined1 generally dipping eastward4 ,it is probable that this mica-schist is an old formation1 connected with the granitic rocks and metamorphic schists near the coast2 and that the one fragment of mica-slate1 and the pebbles of quartz low down in the gypseous formation at ,las ,amolanas1 have been derived from it4 ,the mica-schist is succeeded by stratified porphyritic conglomerate ,q of great thickness1 dipping eastward with a high inclination3 ,i have included this latter mountain-mass in the same anticlinal axis with the porphyritic streams ,,nn2 but ,i am far from sure that the two masses may not have been independently upheaved4 ,seventh axis of elevation4--,proceeding up the ravine1 we come to another mass ,r of andesite2 and beyond this1 we again have a very thick1 stratified porphyritic formation ,s1 dipping at a small angle eastward1 and forming the basal part of the main ,cordillera4 ,i did not ascend the ravine any higher2 but here1 near ,castano1 ,i examined several sections1 of which ,i will not give the details1 only observing1 that the porphyritic beds1 or submarine lavas1 preponderate greatly in bulk over the alternating sedimentary layers1 which have been but little metamorphosed3 these latter consist of fine-grained red tuffs and of whitish volcanic grit-stones1 together with much of a singular1 compact rock1 having an almost crystalline basis1 finely brecciated with red and green fragments1 and occasionally including a few large pebbles4 ,the porphyritic lavas are highly amygdaloidal1 both on their upper and lower surfaces2 they consist chiefly of claystone porphyry1 but with one common variety1 like some of the streams at the ,puente del ,inca1 having a grey mottled basis1 abounding with crystals of red hydrous oxide of iron1 green ones apparently of epidote1 and a few glassy ones of feldspar4 ,this pile of strata differs considerably from the basal strata of the ,cordillera in ,central ,chile1 and may possibly belong to the upper and gypseous series3 ,i saw1 however1 in the bed of the valley1 one fragment of porphyritic breccia-conglomerate1 exactly like those great masses met with in the more southern parts of ,chile4 ,finally1 ,i must observe1 that though ,i have described between the town of ,copiapo and the western flank of the main ,cordillera seven or eight axes of elevation1 extending nearly north and south1 it must not be supposed that they all run continuously for great distances4 ,as was stated to be the case in our sections across the ,cordillera of ,central ,chile1 so here most of the lines of elevation1 with the exception of the first1 third1 and fifth1 are very short4 ,the stratification is everywhere disturbed and intricate2 nowhere have ,i seen more numerous faults and dikes4 ,the whole district1 from the sea to the ,cordillera1 is more or less metalliferous2 and ,i heard of gold1 silver1 copper1 lead1 mercury1 and iron veins4 ,the metamorphic action1 even in the lower strata1 has certainly been far less here than in ,central ,chile4 ,valley of the ,despoblado4--,this great barren valley1 which has already been alluded to1 enters the main valley of ,copiapo a little above the town3 it runs at first northerly1 then ,n4,e41 and more easterly into the ,cordillera2 ,i followed its dreary course to the foot of the first main ridge4 ,i will not give a detailed section1 because it would be essentially similar to that already given1 and because the stratification is exceedingly complicated4 ,after leaving the plutonic hills near the town1 ,i met first1 as in the main valley1 with the gypseous formation1 having the same diversified character as before1 and soon afterwards with masses of porphyritic conglomerate1 about one thousand feet in thickness4 ,in the lower part of this formation there were very thick beds composed of fragments of claystone porphyries1 both angular and rounded1 with the smaller ones partially blended together and the basis rendered porphyritic2 these beds separated distinct streams1 from sixty to eighty feet in thickness1 of claystone lavas4 ,near ,paipote1 also1 there was much true porphyritic breccia-conglomerate3 nevertheless1 few of these masses were metamorphosed to the same degree with the corresponding formation in ,central ,chile4 ,i did not meet in this valley with any true andesite1 but only with imperfect andesitic porphyry1 including large crystals of hornblende3 numerous as have been the varieties of intrusive porphyries already mentioned1 there were here mountains composed of a new kind1 having a compact1 smooth1 cream-coloured basis1 including only a few crystals of feldspar1 and mottled with dendritic spots of oxide of iron4 ,there were also some mountains of a porphyry with a brick-red basis1 containing irregular1 often lens-shaped1 patches of compact feldspar1 and crystals of feldspar1 which latter to my surprise ,i find to be orthite4 ,at the foot of the first ridge of the main ,cordillera1 in the ravine of ,maricongo1 and at an elevation which1 from the extreme coldness and appearance of the vegetation1 ,i estimated at about ten thousand feet1 ,i found beds of white sandstone and of limestone including the ,pecten ,dufreynoyi1 ,terebratula nigma1 and some ,gryphites4 ,this ridge throws the water on the one hand into the ,pacific1 and on the other1 as ,i was informed1 into a great gravel-covered1 basin-like plain1 including a salt-lake1 and without any drainage-exit4 ,in crossing the ,cordillera by this ,pass1 it is said that three principal ridges must be traversed1 instead of two1 or only one as in ,central ,chile4 ,the crest of this first main ridge and the surrounding mountains1 with the exception of a few lofty pinnacles1 are capped by a great thickness of a horizontally stratified1 tufaceous deposit4 ,the lowest bed is of a pale purple colour1 hard1 fine-grained1 and full of broken crystals of feldspar and scales of mica4 ,the middle bed is coarser1 and less hard1 and hence weathers into very sharp pinnacles2 it includes very small fragments of granite1 and innumerable ones of all sizes of grey vesicular trachyte1 some of which were distinctly rounded4 ,the uppermost bed is about two hundred feet in thickness1 of a darker colour and apparently hard3 but ,i had not time to ascend to it4 ,these three horizontal beds may be seen for the distance of many leagues1 especially westward or in the direction of the ,pacific1 capping the summits of the mountains1 and standing on the opposite sides of the immense valleys at exactly corresponding heights4 ,if united they would form a plain1 inclined very slightly towards the ,pacific2 the beds become thinner in this direction1 and the tuff 7judging from one point to which ,i ascended1 some way down the valley7 finer-grained and of less specific gravity1 though still compact and sonorous under the hammer4 ,the gently inclined1 almost horizontal stratification1 the presence of some rounded pebbles1 and the compactness of the lowest bed1 though rendering it probable1 would not have convinced me that this mass had been of subaqueous origin1 for it is known that volcanic ashes falling on land and moistened by rain often become hard and stratified2 but beds thus originating1 and owing their consolidation to atmospheric moisture1 would have covered almost equally every neighbouring summit1 high and low1 and would not have left those above a certain exact level absolutely bare2 this circumstance seems to me to prove that the volcanic ejections were arrested at their present1 widely extended1 equable level1 and there consolidated by some other means than simple atmospheric moisture2 and this no doubt must have been a sheet of water4 ,a lake at this great height1 and without a barrier on any one side1 is out of the question2 consequently we must conclude that the tufaceous matter was anciently deposited beneath the sea4 ,it was certainly deposited before the excavation of the valleys1 or at least before their final enlargement2#d and ,i may add1 that ,mr4 ,lambert1 a gentleman well acquainted with this country1 informs me1 that in ascending the ravine of ,santandres 7which branches off from the ,despoblado7 he met with streams of lava and much erupted matter capping all the hills of granite and porphyry1 with the exception of some projecting points2 he also remarked that the valleys had been excavated subsequently to these eruptions4 #d ,i have endeavoured to show in my 8,journal10 etc4 7#bnd edit471 p4 #cee1 that this arid valley was left by the retreating sea1 as the land slowly rose1 in the state in which we now see it4 ,this volcanic formation1 which ,i am informed by ,mr4 ,lambert extends far northward1 is of interest1 as typifying what has taken place on a grander scale on the corresponding western side of the ,cordillera of ,peru4 ,under another point of view1 however1 it possesses a far higher interest1 as confirming that conclusion drawn from the structure of the fringes of stratified shingle which are prolonged from the plains at the foot of the ,cordillera far up the valleys1--namely1 that this great range has been elevated in mass to a height of between eight and nine thousand feet2#e and now1 judging from this tufaceous deposit1 we may conclude that the horizontal elevation has been in the district of ,copiapo about ten thousand feet4 #e ,i may here mention that on the south side of the main valley of ,copiapo1 near ,potrero ,seco1 the mountains are capped by a thick mass of horizontally stratified shingle1 at a height which ,i estimated at between fifteen hundred and two thousand feet above the bed of the valley4 ,this shingle1 ,i believe1 forms the edge of a wide plain1 which stretches southwards between two mountain ranges4 ,no4 #dj ,illustration ,in the valley of the ,despoblado1 the stratification1 as before remarked has been much disturbed1 and in some points to a greater degree than ,i have anywhere else seen4 ,i will give two cases3 a very thick mass of thinly stratified red sandstone1 including beds of conglomerate1 has been crushed together 7as represented in figure no4 #bd7 into a yoke or urn-formed trough1 so that the strata on both sides have been folded inwards3 on the right hand the properly underlying porphyritic claystone conglomerate is seen overlying the sandstone1 but it soon becomes vertical1 and then is inclined towards the trough1 so that the beds radiate like the spokes of a wheel3 on the left hand1 the inverted porphyritic conglomerate also assumes a dip towards the trough1 not gradually1 as on the right hand1 but by means of a vertical fault and synclinal break2 and a little still further on towards the left1 there is a second great oblique fault 7both shown by the arrow-lines71 with the strata dipping to a directly opposite point2 these mountains are intersected by infinitely numerous dikes1 some of which can be seen to rise from hummocks of greenstone1 and can be traced for thousands of feet4 ,in the second case1 two low ridges trend together and unite at the head of a little wedge-shaped valley3 throughout the right-hand ridge1 the strata dip at #de to the east2 in the left-hand ridge1 we have the very same strata and at first with exactly the same dip2 but in following this ridge up the valley1 the strata are seen very regularly to become more and more inclined until they stand vertical1 they then gradually fall over 7the basset edges forming symmetrical serpentine lines along the crest71 till at the very head of the valley they are reversed at an angle of #de3 so that at this point the beds have been turned through an angle of #ace2 and here there is a kind of anticlinal axis1 with the strata on both sides dipping to opposite points at an angle of #de1 but those on the left hand upside down4 ,on the eruptive sources of the porphyritic claystone and greenstone lavas4--,in ,central ,chile1 from the extreme metamorphic action1 it is in most parts difficult to distinguish between the streams of porphyritic lava and the porphyritic breccia-conglomerate1 but here1 at ,copiapo1 they are generally perfectly distinct1 and in the ,despoblado1 ,i saw for the first time1 two great strata of purple claystone porphyry1 after having been for a considerable space closely united together1 one above the other1 become separated by a mass of fragmentary matter1 and then both thin out2--the lower one more rapidly than the upper and greater stream4 ,considering the number and thickness of the streams of porphyritic lava1 and the great thickness of the beds of breccia-conglomerate1 there can be little doubt that the sources of eruption must originally have been numerous3 nevertheless1 it is now most difficult even to conjecture the precise point of any one of the ancient submarine craters4 ,i have repeatedly observed mountains of porphyries1 more or less distinctly stratified towards their summits or on their flanks1 without a trace of stratification in their central and basal parts3 in most cases1 ,i believe this is simply due either to the obliterating effects of metamorphic action1 or to such parts having been mainly formed of intrusive porphyries1 or to both causes conjoined2 in some instances1 however1 it appeared to me very probable that the great central unstratified masses of porphyry were the now partially denuded nuclei of the old submarine volcanoes1 and that the stratified parts marked the points whence the streams flowed4 ,in one case alone1 and it was in this ,valley of the ,despoblado1 ,i was able actually to trace a thick stratum of purplish porphyry1 which for a space of some miles conformably overlay the usual alternating beds of breccia-conglomerates and claystone lavas1 until it became united with1 and blended into1 a mountainous mass of various unstratified porphyries4 ,the difficulty of tracing the streams of porphyries to their ancient and doubtless numerous eruptive sources1 may be partly explained by the very general disturbance which the ,cordillera in most parts has suffered2 but ,i strongly suspect that there is a more specific cause1 namely1 that the original points of eruption tend to become the points of injection4 ,this in itself does not seem improbable2 for where the earth's crust has once yielded1 it would be liable to yield again1 though the liquified intrusive matter might not be any longer enabled to reach the submarine surface and flow as lava4 ,i have been led to this conclusion1 from having so frequently observed that1 where part of an unstratified mountain-mass resembled in mineralogical character the adjoining streams or strata1 there were several other kinds of intrusive porphyries and andesitic rocks injected into the same point4 ,as these intrusive mountain-masses form most of the axes-lines in the ,cordillera1 whether anticlinal1 uniclinal1 or synclinal1 and as the main valleys have generally been hollowed out along these lines1 the intrusive masses have generally suffered much denudation4 ,hence they are apt to stand in some degree isolated1 and to be situated at the points where the valleys abruptly bend1 or where the main tributaries enter4 ,on this view of there being a tendency in the old points of eruption to become the points of subsequent injection and disturbance1 and consequently of denudation1 it ceases to be surprising that the streams of lava in the porphyritic claystone conglomerate formation1 and in other analogous cases1 should most rarely be traceable to their actual sources4 ,iquique1 ,southern ,peru4--,differently from what we have seen throughout ,chile1 the coast here is formed not by the granitic series1 but by an escarpment of the porphyritic conglomerate formation1 between two and three thousand feet in height4#f ,i had time only for a very short examination2 the chief part of the escarpment appears to be composed of various reddish and purple1 sometimes laminated1 porphyries1 resembling those of ,chile2 and ,i saw some of the porphyritic breccia-conglomerate2 the stratification appeared but little inclined4 ,the uppermost part1 judging from the rocks near the famous silver mine of ,huantajaya1#g;consists of laminated1 impure1 argillaceous1 purplish-grey limestone1 associated1 ,i believe1 with some purple sandstone4 ,in the limestone shells are found3 the three following species were given me3-- ,lucina ,americana1 ,e4 ,forbes4 ,terebratula inca1 ,e4 ,forbes4 ,terebratula nigma1 ,d',orbigny4 #f ,the lowest point1 where the road crosses the coast-escarpment1 is #a1ijj feet by the barometer above the level of the sea4 #g ,mr4 ,bollaert has described 78,geolog4 ,proceedings10 vol4 ii1 p4 #eih1 a singular mass of stratified detritus1 gravel1 and sand1 eighty-one yards in thickness1 overlying the limestone1 and abounding with loose masses of silver ore4 ,the miners believe that they can attribute these masses to their proper veins4 ,this latter species we have seen associated with the fossils of which lists have been given in this chapter1 in two places in the valley of ,coquimbo1 and in the ravine of ,maricongo at ,copiapo4 ,considering this fact1 and the superposition of these beds on the porphyritic conglomerate formation2 and1 as we shall immediately see1 from their containing much gypsum1 and from their otherwise close general resemblance in mineralogical nature with the strata described in the valley of ,copiapo1 ,i have little doubt that these fossiliferous beds of ,iquique belong to the great cretaceo-oolitic formation of ,northern ,chile4 ,iquique is situated seven degrees latitude north of ,copiapo2 and ,i may here mention1 that an ,ammonites1 nov4 species1 and an ,astarte1 nov4 species1 were given me from the ,cerro ,pasco1 about ten degrees of latitude north of ,iquique1 and ,m4 ,d',orbigny thinks that they probably indicate a ,neocomian formation4 ,again1 fifteen degrees of latitude northward1 in ,colombia1 there is a grand fossiliferous deposit1 now well known from the labours of ,von ,buch1 ,lea1 d',orbigny1 and ,forbes1 which belongs to the earlier stages of the cretaceous system4 ,hence1 bearing in mind the character of the few fossils from ,tierra del ,fuego1 there is some evidence that a great portion of the stratified deposits of the whole vast range of the ,south ,american ,cordillera belongs to about the same geological epoch4 ,proceeding from the coast escarpment inwards1 ,i crossed1 in a space of about thirty miles1 an elevated undulatory district1 with the beds dipping in various directions4 ,the rocks are of many kinds1--white laminated1 sometimes siliceous sandstone1--purple and red sandstone1 sometimes so highly calcareous as to have a crystalline fracture1--argillaceous limestone1--black calcareous slate-rock1 like that so often described at ,copiapo and other places1--thinly laminated1 fine-grained1 greenish1 indurated1 sedimentary1 fusible rocks1 approaching in character to the so-called pseudo-honestone of ,chile1 including thin contemporaneous veins of gypsum1--and lastly1 much calcareous1 laminated porcelain jasper1 of a green colour1 with red spots1 and of extremely easy fusibility3 ,i noticed one conformable stratum of a freckled-brown1 feldspathic lava4 ,i may here mention that ,i heard of great beds of gypsum in the ,cordillera4 ,the only novel point in this formation1 is the presence of innumerable thin layers of rock-salt1 alternating with the laminated and hard1 but sometimes earthy1 yellowish1 or bright red and ferruginous sandstones4 ,the thickest layer of salt was only two inches1 and it thinned out at both ends4 ,on one of these saliferous masses ,i noticed a stratum about twelve feet thick1 of dark-brown1 hard brecciated1 easily fusible rock1 containing grains of quartz and of black oxide of iron1 together with numerous imperfect fragments of shells4 ,the problem of the origin of salt is so obscure1 that every fact1 even geographical position1 is worth recording4#h ,with the exception of these saliferous beds1 most of the rocks as already remarked1 present a striking general resemblance with the upper parts of the gypseous or cretaceo-oolitic formation of ,chile4 #h ,it is well known that stratified salt is found in several places on the shores of ,peru4 ,the island of ,san ,lorenzo1 off ,lima1 is composed of a pile of thin strata1 about eight hundred feet in thickness1 composed of yellowish and purplish1 hard siliceous1 or earthy sandstones1 alternating with thin layers of shale1 which in places passes into a greenish1 semi-porcellanic1 fusible rock4 ,there are some thin beds of reddish mudstone1 and soft ferruginous rotten-stones1 with layers of gypsum4 ,in nearly all these varieties1 especially in the softer sandstones1 there are numerous thin seams of rock-salt3 ,i was informed that one layer has been found two inches in thickness4 ,the manner in which the minutest fissures of the dislocated beds have been penetrated by the salt1 apparently by subsequent infiltration1 is very curious4 ,on the south side of the island1 layers of coal and of impure limestone have been discovered4 ,hence we here have salt1 gypsum1 and coal associated together4 ,the strata include veins of quartz1 carbonate of lime1 and iron pyrites2 they have been dislocated by an injected mass of greenish-brown feldspathic trap4 ,not only is salt abundant on the extreme western limits of the district between the ,cordillera and the ,pacific1 but1 according to ,helms1 it is found in the outlying low hills on the eastern flank of the ,cordillera4 ,these facts appear to me opposed to the theory1 that rock-salt is due to the sinking of water1 charged with salt1 in mediterranean spaces of the ocean4 ,the general character of the geology of these countries would rather lead to the opinion1 that its origin is in some way connected with volcanic heat at the bottom of the sea3 see on this subject ,sir ,r4 ,murchison's 8,anniversary ,address to ,geolog4 ,soc41 #ahdc10 p4 #fe47 ,metalliferous ,veins4 ,i have only a few remarks to make on this subject3 in nine mining districts1 some of them of considerable extent1 which ,i visited in ,central ,chile1 ,i found the principal veins running from between ,n4 and ,n4,w4 to ,s4 and ,s4,e42#i at the ,c4 de los ,hornos 7further northward71 it is ,n4,n4,w4 and ,s4,s4,e42 at ,panuncillo1 it is ,n4,n4,w4 and ,s4,s4,e42 and1 lastly1 at ,arqueros1 the direction is ,n4,w4 and ,s4,e473 in some other places1 however1 their courses appeared quite irregular1 as is said to be generally the case in the whole valley of ,copiapo3 at ,tambillos1 south of ,coquimbo1 ,i saw one large copper vein extending east and west4 ,it is worthy of notice1 that the foliation of the gneiss and mica-slate1 where such rocks occur1 certainly tend to run like the metalliferous veins1 though often irregularly1 in a direction a little westward of north4 ,at ,yaquil1 ,i observed that the principal auriferous veins ran nearly parallel to the grain or imperfect cleavage of the surrounding granitic rocks4 ,with respect to the distribution of the different metals1 copper1 gold1 and iron are generally associated together1 and are most frequently found 7but with many exceptions1 as we shall presently see7 in the rocks of the lower series1 between the ,cordillera and the ,pacific1 namely1 in granite1 syenite1 altered feldspathic clay-slate1 gneiss1 and as near ,guasco mica-schist4 ,the copper-ores consist of sulphurets1 oxides1 and carbonates1 sometimes with lamin of native metal3 ,i was assured that in some cases 7as at ,panuncillo ,s4,e4 of ,coquimbo71 the upper part of the same vein contains oxides1 and the lower part sulphurets of copper4#aj ,gold occurs in its native form2 it is believed that1 in many cases1 the upper part of the vein is the most productive part3 this fact probably is connected with the abundance of this metal in the stratified detritus of ,chile1 which must have been chiefly derived from the degradation of the upper portions of the rocks4 ,these superficial beds of well-rounded gravel and sand1 containing gold1 appeared to me to have been formed under the sea close to the beach1 during the slow elevation of the land3 ,schmidtmeyer#aa remarks that in ,chile gold is sought for in shelving banks at the height of some feet on the sides of the streams1 and not in their beds1 as would have been the case had this metal been deposited by common alluvial action4 ,very frequently the copper-ores1 including some gold1 are associated with abundant micaceous specular iron4 ,gold is often found in iron-pyrites3 at two gold mines at ,yaquil 7near ,nancagua71 ,i was informed by the proprietor that in one the gold was always associated with copper-pyrites1 and in the other with iron-pyrites3 in this latter case1 it is said that if the vein ceases to contain iron-pyrites1 it is yet worth while to continue the search1 but if the iron-pyrites1 when it reappears1 is not auriferous1 it is better at once to give up working the vein4 ,although ,i believe copper and gold are most frequently found in the lower granitic and metamorphic schistose series1 yet these metals occur both in the porphyritic conglomerate formation 7as on the flanks of the ,bell of ,quillota and at ,jajuel71 and in the superincumbent strata4 ,at ,jajuel ,i was informed that the copper-ore1 with some gold1 is found only in the greenstones and altered feldspathic clay-slate1 which alternate with the purple porphyritic conglomerate4 ,several gold veins and some of copper-ore are worked in several parts of the ,uspallata range1 both in the metamorphosed strata1 which have been shown to have been of probably subsequent origin to the ,neocomian or gypseous formation of the main ,cordillera1 and in the intrusive andesitic rocks of that range4 ,at ,los ,hornos 7,n4,e4 of ,illapel71 likewise1 there are numerous veins of copper-pyrites and of gold1 both in the strata of the gypseous formation and in the injected hills of andesite and various porphyries4 #i ,these mining districts are ,yaquil near ,nancagua1 where the direction of the chief veins1 to which only in all cases ,i refer1 is north and south2 in the ,uspallata range1 the prevailing line is ,n4,n4,w4 and ,s4,s4,e42 in the ,c4 de ,prado1 it is ,n4,n4,w4 and ,s4,s4,e42 near ,illapel1 it is ,n4 by ,w4 and ,s4 by ,e42 at ,los ,hornos the direction varies from between ,n4 and ,n4,w4 to ,s4 and ,s4,e44 #aj ,the same fact has been observed by ,mr4 ,taylor in ,cuba3 ,london ,phil4 ,journ41 vol4 xi1 p4 #ba4 #aa 8,travels in ,chile10 p4 #bi4 ,silver1 in the form of a chloride1 sulphuret1 or an amalgam1 or in its native state1 and associated with lead and other metals1 and at ,arqueros with pure native copper1 occurs chiefly in the upper great gypseous or cretaceo-oolitic formation which forms probably the richest mass in ,chile4 ,we may instance the mining districts of ,arqueros near ,coquimbo1 and of nearly the whole valley of ,copiapo1 and of ,iquique 7where the principal veins run ,n4,e4 by ,e4 and ,s4,w4 by ,w471 in ,peru4 ,hence comes ,molina's remark1 that silver is born in the cold and solitary deserts of the ,upper ,cordillera4 ,there are1 however1 exceptions to this rule3 at ,paral 7,s4,e4 of ,coquimbo7 silver is found in the porphyritic conglomerate formation2 as ,i suspect is likewise the case at ,s4 ,pedro de ,nolasko in the ,peuquenes ,pass4 ,rich argentiferous lead is found in the clay-slate of the ,uspallata range2 and ,i saw an old silver-mine in a hill of syenite at the foot of the ,bell of ,quillota3 ,i was also assured that silver has been found in the andesitic and porphyritic region between the town of ,copiapo and the ,pacific4 ,i have stated in a previous part of this chapter1 that in two neighbouring mines at ,arqueros the veins in one were productive when they traversed the singular green sedimentary beds1 and unproductive when crossing the reddish beds2 whereas at the other mine exactly the reverse takes place2 ,i have also described the singular and rare case of numerous particles of native silver and of the chloride being disseminated in the green rock at the distance of a yard from the vein4 ,mercury occurs with silver both at ,arqueros and at ,copiapo3 at the base of ,c4 de los ,hornos 7,s4,e4 of ,coquimbo1 a different place from ,los ,hornos1 before mentioned7 ,i saw in a syenitic rock numerous quartzose veins1 containing a little cinnabar in nests3 there were here other parallel veins of copper and of a ferrugino-auriferous ore4 ,i believe tin has never been found in ,chile4 ,from information given me by ,mr4 ,nixon of ,yaquil1#ab and by others1 it appears that in ,chile those veins are generally most permanently productive1 which1 consisting of various minerals 7sometimes differing but slightly from the surrounding rocks71 include parallel strings rich in metals2 such a vein is called a veta real4 ,more commonly the mines are worked only where one1 two1 or more thin veins or strings running in a different direction1 intersect a poor 8veta real30 it is unanimously believed that at such points of intersection 7cruceros71 the quantity of metal is much greater than that contained in other parts of the intersecting veins4 ,in some cruceros or points of intersection1 the metals extend even beyond the walls of the main1 broad1 stony vein4 ,it is said that the greater the angle of intersection1 the greater the produce2 and that nearly parallel strings attract each other2 in the ,uspallata range1 ,i observed that numerous thin auri-ferruginous veins repeatedly ran into knots1 and then branched out again4 ,i have already described the remarkable manner in which rocks of the ,uspallata range are indurated and blackened 7as if by a blast of gunpowder7 to a considerable distance from the metallic veins4 #ab ,at the ,durazno mine1 the gold is associated with copper-pyrites1 and the veins contain large prisms of plumbago4 ,crystallised carbonate of lime is one of the commonest minerals in the matrix of the ,chilean veins4 ,finally1 ,i may observe1 that the presence of metallic veins seems obviously connected with the presence of intrusive rocks1 and with the degree of metamorphic action which the different districts of ,chile have undergone4#ac ,such metamorphosed areas are generally accompanied by numerous dikes and injected masses of andesite and various porphyries3 ,i have in several places traced the metalliferous veins from the intrusive masses into the encasing strata4 ,knowing that the porphyritic conglomerate formation consists of alternate streams of submarine lavas and of the debris of anciently erupted rocks1 and that the strata of the upper gypseous formation sometimes include submarine lavas1 and are composed of tuffs1 mudstones1 and mineral substances1 probably due to volcanic exhalations1--the richness of these strata is highly remarkable when compared with the erupted beds1 often of submarine origin1 but not metamorphosed1 which compose the numerous islands in the ,pacific1 ,indian1 and ,atlantic ,oceans2 for in these islands metals are entirely absent1 and their nature even unknown to the aborigines4 #ac ,sir ,r4 ,murchison and his fellow travellers have given some striking facts on this subject in their account of the ,ural ,mountains 78,geolog4 ,proc410 vol4 iii1 p4 #gdh4 ,summary of the ,geological ,history of the ,chilean ,cordillera1 and of the ,southern ,parts of ,south ,america4 ,we have seen that the shores of the ,pacific1 for a space of #a1bjj miles from ,tres ,montes to ,copiapo1 and ,i believe for a very much greater distance1 are composed1 with the exception of the tertiary basins1 of metamorphic schists1 plutonic rocks1 and more or less altered clay-slate4 ,on the floor of the ocean thus constituted1 vast streams of various purplish claystone and greenstone porphyries were poured forth1 together with great alternating piles of angular and rounded fragments of similar rocks ejected from the submarine craters4 ,from the compactness of the streams and fragments1 it is probable that1 with the exception of some districts in ,northern ,chile1 the eruptions took place in profoundly deep water4 ,the orifices of eruption appear to have been studded over a breadth1 with some outliers1 of from fifty to one hundred miles3 and closely enough together1 both north and south1 and east and west1 for the ejected matter to form a continuous mass1 which in ,central ,chile is more than a mile in thickness4 ,i traced this mould-like mass1 for only #dej miles2 but judging from what ,i saw at ,iquique1 from specimens1 and from published accounts1 it appears to have a manifold greater length4 ,in the basal parts of the series1 and especially towards the flanks of the range1 mud1 since converted into a feldspathic slaty rock1 and sometimes into greenstone1 was occasionally deposited between the beds of erupted matter3 with this exception the uniformity of the porphyritic rocks is very remarkable4 ,at the period when the claystone and greenstone porphyries nearly or quite ceased being erupted1 that great pile of strata which1 from often abounding with gypsum1 ,i have generally called the gypseous formation was deposited1 and feldspathic lavas1 together with other singular volcanic rocks1 were occasionally poured forth3 ,i am far from pretending that any distinct line of demarcation can be drawn between this formation and the underlying porphyries and porphyritic conglomerate1 but in a mass of such great thickness1 and between beds of such widely different mineralogical nature1 some division was necessary4 ,at about the commencement of the gypseous period1 the bottom of the sea here seems first to have been peopled by shells1 not many in kind1 but abounding in individuals4 ,at the ,p4 del ,inca the fossils are embedded near the base of the formation2 in the ,peuquenes range1 at different levels1 halfway up1 and even higher in the series2 hence1 in these sections1 the whole pile of strata belongs to the same period3 the same remark is applicable to the beds at ,copiapo1 which attain a thickness of between seven and eight thousand feet4 ,the fossil shells in the ,cordillera of ,central ,chile1 in the opinion of all the palontologists who have examined them1 belong to the earlier stages of the cretaceous system2 whilst in ,northern ,chile there is a most singular mixture of cretaceous and oolitic forms3 from the geological relations1 however1 of these two districts1 ,i cannot but think that they all belong to nearly the same epoch1 which ,i have provisionally called cretaceo-oolitic4 ,the strata in this formation1 composed of black calcareous shaly-rocks of red and white1 and sometimes siliceous sandstone1 of coarse conglomerates1 limestones1 tuffs1 dark mudstones1 and those singular fine-grained rocks which ,i have called pseudo-honestones1 vast beds of gypsum1 and many other jaspery and scarcely describable varieties1 vary and replace each other in short horizontal distances1 to an extent1 ,i believe1 unequalled even in any tertiary basin4 ,most of these substances are easily fusible1 and have apparently been derived either from volcanoes still in quiet action1 or from the attrition of volcanic products4 ,if we picture to ourselves the bottom of the sea1 rendered uneven in an extreme degree1 with numerous craters1 some few occasionally in eruption1 but the greater number in the state of solfataras1 discharging calcareous1 siliceous1 ferruginous matters1 and gypsum or sulphuric acid to an amount surpassing1 perhaps1 even the existing sulphureous volcanoes of ,java1#ad we shall probably understand the circumstances under which this singular pile of varying strata was accumulated4 ,the shells appear to have lived at the quiescent periods when only limestone or calcareo-argillaceous matter was depositing4 ,from ,dr4 ,gillies' account1 this gypseous or cretaceo-oolitic formation extends as far south as the ,pass of ,planchon1 and ,i followed it northward at intervals for #ejj miles3 judging from the character of the beds with the ,terebratula nigma1 at ,iquique1 it extends from four to five hundred miles further3 and perhaps even for ten degrees of latitude north of ,iquique to the ,cerro ,pasco1 not far from ,lima3 again1 we know that a cretaceous formation1 abounding with fossils1 is largely developed north of the equator1 in ,colombia3 in ,tierra del ,fuego1 at about this same period1 a wide district of clay-slate was deposited1 which in its mineralogical characters and external features1 might be compared to the ,silurian regions of ,north ,wales4 ,the gypseous formation1 like that of the porphyritic breccia-conglomerate on which it rests1 is of inconsiderable breadth2 though of greater breadth in ,northern than in ,central ,chile4 #ad ,von ,buch's 8,descript4 ,physique des ,iles ,canaries10 p4 #dbh4 ,as the fossil shells in this formation are covered1 in the ,peuquenes ridge1 by a great thickness of strata2 at the ,puente del ,inca1 by at least five thousand feet2 at ,coquimbo1 though the superposition there is less plainly seen1 by about six thousand feet2 and at ,copiapo1 certainly by five or six thousand1 and probably by seven thousand feet 7the same species there recurring in the upper and lower parts of the series71 we may feel confident that the bottom of the sea subsided during this cretaceo-oolitic period1 so as to allow of the accumulation of the superincumbent submarine strata4 ,this conclusion is confirmed by1 or perhaps rather explains1 the presence of the many beds at many levels of coarse conglomerate1 the well-rounded pebbles in which we cannot believe were transported in very deep water4 ,even the underlying porphyries at ,copiapo4 with their highly amygdaloidal surfaces1 do not appear to have flowed under great pressure4 ,the great sinking movement thus plainly indicated1 must have extended in a north and south line for at least four hundred miles1 and probably was co-extensive with the gypseous formation4 ,the beds of conglomerate just referred to1 and the extraordinarily numerous silicified trunks of fir-trees at ,los ,hornos1 perhaps at ,coquimbo and at two distant points in the valley of ,copiapo1 indicate that land existed at this period in the neighbourhood4 ,this land1 or islands1 in the northern part of the district of ,copiapo1 must have been almost exclusively composed1 judging from the nature of the pebbles of granite3 in the southern parts of ,copiapo1 it must have been mainly formed of claystone porphyries1 with some mica-schist1 and with much sandstone and jaspery rocks exactly like the rocks in the gypseous formation1 and no doubt belonging to its basal series4 ,in several other places also1 during the accumulation of the gypseous formation1 its basal parts and the underlying porphyritic conglomerate must likewise have been already partially upheaved and exposed to wear and tear2 near the ,puente del ,inca and at ,coquimbo1 there must have existed masses of mica-schist or some such rock1 whence were derived the many small pebbles of opaque quartz4 ,it follows from these facts1 that in some parts of the ,cordillera the upper beds of the gypseous formation must lie unconformably on the lower beds2 and the whole gypseous formation1 in parts1 unconformably on the porphyritic conglomerate2 although ,i saw no such cases1 yet in many places the gypseous formation is entirely absent2 and this1 although no doubt generally caused by quite subsequent denudation1 may in others be due to the underlying porphyritic conglomerate having been locally upheaved before the deposition of the gypseous strata1 and thus having become the source of the pebbles of porphyry embedded in them4 ,in the porphyritic conglomerate formation1 in its lower and middle parts1 there is very rarely any evidence1 with the exception of the small quartz pebbles at ,jajuel near ,aconcagua1 and of the single pebble of granite at ,copiapo1 of the existence of neighbouring land3 in the upper parts1 however1 and especially in the district of ,copiapo1 the number of thoroughly well-rounded pebbles of compact porphyries make me believe1 that1 as during the prolonged accumulation of the gypseous formation the lower beds had already been locally upheaved and exposed to wear and tear1 so it was with the porphyritic conglomerate4 ,hence in following thus far the geological history of the ,cordillera1 it may be inferred that the bed of a deep and open1 or nearly open1 ocean was filled up by porphyritic eruptions1 aided probably by some general and some local elevations1 to that comparatively shallow level at which the cretaceo-oolitic shells first lived4 ,at this period1 the submarine craters yielded at intervals a prodigious supply of gypsum and other mineral exhalations1 and occasionally1 in certain places poured forth lavas1 chiefly of a feldspathic nature3 at this period1 islands clothed with fir-trees and composed of porphyries1 primary rocks1 and the lower gypseous strata had already been locally upheaved1 and exposed to the action of the waves2--the general movement1 however1 at this time having been over a very wide area1 one of slow subsidence1 prolonged till the bed of the sea sank several thousand feet4 ,in ,central ,chile1 after the deposition of a great thickness of the gypseous strata1 and after their upheaval1 by which the ,cumbre and adjoining ranges were formed1 a vast pile of tufaceous matter and submarine lava was accumulated1 where the ,uspallata chain now stands2 also after the deposition and upheaval of the equivalent gypseous strata of the ,peuquenes range1 the great thick mass of conglomerate in the valley of ,tenuyan was accumulated3 during the deposition of the ,uspallata strata1 we know absolutely1 from the buried vertical trees1 that there was a subsidence of some thousand feet2 and we may infer from the nature of the conglomerate in the valley of ,tenuyan1 that a similar and perhaps contemporaneous movement there took place4 ,we have1 then1 evidence of a second great period of subsidence2 and1 as in the case of the subsidence which accompanied the accumulation of the cretaceo-oolitic strata1 so this latter subsidence appears to have been complicated by alternate or local elevatory movement-- for the vertical trees1 buried in the midst of the ,uspallata strata1 must have grown on dry land1 formed by the upheaval of the lower submarine beds4 ,presently ,i shall have to recapitulate the facts1 showing that at a still later period1 namely1 at nearly the commencement of the old tertiary deposits of ,patagonia and of ,chile1 the continent stood at nearly its present level1 and then1 for the third time1 slowly subsided to the amount of several hundred feet1 and was afterwards slowly re-uplifted to its present level4 ,the highest peaks of the ,cordillera appear to consist of active or more commonly dormant volcanoes1--such as ,tupungato1 ,maypu1 and ,aconcagua1 which latter stands #bc1jjj feet above the level of the sea1 and many others4 ,the next highest peaks are formed of the gypseous and porphyritic strata1 thrown into vertical or highly inclined positions4 ,besides the elevation thus gained by angular displacements1 ,i infer1 without any hesitation--from the stratified gravel-fringes which gently slope up the valleys of the ,cordillera from the gravel-capped plains at their base1 which latter are connected with the plains1 still covered with recent shells on the coast--that this great range has been upheaved in mass by a slow movement1 to an amount of at least #h1jjj feet4 ,in the ,despoblado ,valley1 north of ,copiapo1 the horizontal elevation1 judging from the compact1 stratified tufaceous deposit1 capping the distant mountains at corresponding heights1 was about ten thousand feet4 ,it is very possible1 or rather probable1 that this elevation in mass may not have been strictly horizontal1 but more energetic under the ,cordillera1 than towards the coast on either side2 nevertheless1 movements of this kind may be conveniently distinguished from those by which strata have been abruptly broken and upturned4 ,when viewing the ,cordillera1 before having read ,mr4 ,hopkins's profound 8,researches on ,physical ,geology10 the conviction was impressed on me1 that the angular dislocations1 however violent1 were quite subordinate in importance to the great upward movement in mass1 and that they had been caused by the edges of the wide fissures1 which necessarily resulted from the tension of the elevated area1 having yielded to the inward rush of fluidified rock1 and having thus been upturned4 ,the ridges formed by the angularly upheaved strata are seldom of great length3 in the central parts of the ,cordillera they are generally parallel to each other1 and run in north and south lines2 but towards the flanks they often extend more or less obliquely4 ,the angular displacement has been much more violent in the central than in the exterior main lines2 but it has likewise been violent in some of the minor lines on the extreme flanks4 ,the violence has been very unequal on the same short lines2 the crust having apparently tended to yield on certain points along the lines of fissures4 ,these points1 ,i have endeavoured to show1 were probably first foci of eruption1 and afterwards of injected masses of porphyry and andesite4#ae ,the close similarity of the andesitic granites and porphyries1 throughout ,chile1 ,tierra del ,fuego1 and even in ,peru1 is very remarkable4 ,the prevalence of feldspar cleaving like albite1 is common not only to the andesites1 but 7as ,i infer from the high authority of ,professor ,g4 ,rose1 as well as from my own measurements7 to the various claystone and greenstone porphyries1 and to the trachytic lavas of the ,cordillera4 ,the andesitic rocks have in most cases been the last injected ones1 and they probably form a continuous dome under this great range3 they stand in intimate relationship with the modern lavas2 and they seem to have been the immediate agent in metamorphosing the porphyritic conglomerate formation1 and often likewise the gypseous strata1 to the extraordinary extent to which they have suffered4 #ae ,sir ,r4 ,murchison and his companions state 78,geolog4 ,proc410 vol4 iii1 p4 #gdg71 that no true granite appears in the higher ,ural ,mountains2 but that syenitic greenstone--a rock closely analogous to our andesite--is far the most abundant of the intrusive masses4 ,with respect to the age at which the several parallel ridges composing the ,cordillera were upthrown1 ,i have little evidence4 ,many of them may have been contemporaneously elevated and injected in the same manner#af as in volcanic archipelagoes lavas are contemporaneously ejected on the parallel lines of fissure4 ,but the pebbles apparently derived from the wear and tear of the porphyritic conglomerate formation1 which are occasionally present in the upper parts of this same formation1 and are often present in the gypseous formation1 together with the pebbles from the basal parts of the latter formation in its upper strata1 render it almost certain that portions1 we may infer ridges1 of these two formations were successively upheaved4 ,in the case of the gigantic ,portillo range1 we may feel almost certain that a preexisting granitic line was upraised 7not by a single blow1 as shown by the highly inclined basaltic streams in the valley on its eastern flank7 at a period long subsequent to the upheavement of the parallel ,peuquenes range4#ag ,again1 subsequently to the upheavement of the ,cumbre chain1 that of ,uspallata was formed and elevated2 and afterwards1 ,i may add1 in the plain of ,uspallata1 beds of sand and gravel were violently upthrown4 ,the manner in which the various kinds of porphyries and andesites have been injected one into the other1 and in which the infinitely numerous dikes of various composition intersect each other1 plainly show that the stratified crust has been stretched and yielded many times over the same points4 ,with respect to the age of the axes of elevation between the ,pacific and the ,cordillera1 ,i know little3 but there are some lines which must--namely1 those running north and south in ,chiloe1 those eight or nine east and west1 parallel1 far-extended1 most symmetrical uniclinal lines at ,p4 ,rumena1 and the short ,n4,w4-,s4,e4 and ,n4,e4-,s4,w4 lines at ,concepcion--have been upheaved long after the formation of the ,cordillera4 ,even during the earthquake of #ahce1 when the linear north and south islet of ,st4 ,mary was uplifted several feet above the surrounding area1 we perhaps see one feeble step in the formation of a subordinate mountain-axis4 ,in some cases1 moreover1 for instance1 near the baths of ,cauquenes1 ,i was forcibly struck with the small size of the breaches cut through the exterior mountain-ranges1 compared with the size of the same valleys higher up where entering the ,cordillera2 and this circumstance appeared to me scarcely explicable1 except on the idea of the exterior lines having been subsequently upthrown1 and therefore having been exposed to a less amount of denudation4 ,from the manner in which the fringes of gravel are prolonged in unbroken slopes up the valleys of the ,cordillera1 ,i infer that most of the greater dislocations took place during the earlier parts of the great elevation in mass3 ,i have1 however1 elsewhere given a case1 and ,m4 de ,tschudi#ah has given another1 of a ridge thrown up in ,peru across the bed of a river1 and consequently after the final elevation of the country above the level of the sea4 #af 8,volcanic ,islands10 etc47 #ag ,i have endeavoured to show in my 8,journal0 7#bnd edit41 p4 #cba71 that the singular fact of the river1 which drains the valley between these two ranges1 passing through the ,portillo and higher line1 is explained by its slow and subsequent elevation4 ,there are many analogous cases in the drainage of rivers3 see ,edinburgh ,new ,phil4 ,journal1 vol4 xxviii1 pp4 #cc and #dd4 #ah 8,reise in ,peru10 ,band #b1 s4 #h3 ,author's 8,journal10 #bnd edit41 p4 #cei4 ,ascending to the older tertiary formations1 ,i will not again recapitulate the remarks already given at the end of the ,fifth ,chapter1--on their great extent1 especially along the shores of the ,atlantic--on their antiquity1 perhaps corresponding with that of the eocene deposits of ,europe1--on the almost entire dissimilarity1 though the formations are apparently contemporaneous1 of the fossils from the eastern and western coasts1 as is likewise the case1 even in a still more marked degree1 with the shells now living in these opposite though approximate seas1--on the climate of this period not having been more tropical than what might have been expected from the latitudes of the places under which the deposits occur2 a circumstance rendered well worthy of notice1 from the contrast with what is known to have been the case during the older tertiary periods of ,europe1 and likewise from the fact of the southern hemisphere having suffered at a much later period1 apparently at the same time with the northern hemisphere1 a colder or more equable temperature1 as shown by the zones formerly affected by ice-action4 ,nor will ,i recapitulate the proofs of the bottom of the sea1 both on the eastern and western coast1 having subsided seven or eight hundred feet during this tertiary period2 the movement having apparently been co-extensive1 or nearly co-extensive1 with the deposits of this age4 ,nor will ,i again give the facts and reasoning on which the proposition was founded1 that when the bed of the sea is either stationary or rising1 circumstances are far less favourable than when its level is sinking1 to the accumulation of conchiferous deposits of sufficient thickness1 extension1 and hardness to resist1 when upheaved1 the ordinary vast amount of denudation4 ,we have seen that the highly remarkable fact of the absence of any extensive formations containing recent shells1 either on the eastern or western coasts of the continent1--though these coasts now abound with living mollusca1--though they are1 and apparently have always been1 as favourable for the deposition of sediment as they were when the tertiary formations were copiously deposited1--and though they have been upheaved to an amount quite sufficient to bring up strata from the depths the most fertile for animal life--can be explained in accordance with the above proposition4 ,as a deduction1 it was also attempted to be shown1 first1 that the want of close sequence in the fossils of successive formations1 and of successive stages in the same formation1 would follow from the improbability of the same area continuing slowly to subside from one whole period to another1 or even during a single entire period2 and secondly1 that certain epochs having been favourable at distant points1 in the same quarter of the world for the synchronous accumulation of fossiliferous strata1 would follow from movements of subsidence having apparently1 like those of elevation1 contemporaneously affected very large areas4 ,there is another point which deserves some notice1 namely1 the analogy between the upper parts of the ,patagonian tertiary formation1 as well as of the upper possibly contemporaneous beds at ,chiloe and ,concepcion1 with the great gypseous formation of ,cordillera2 for in both formations1 the rocks1 in their fusible nature1 in their containing gypsum1 and in many other characters1 show a connection1 either intimate or remote1 with volcanic action2 and as the strata in both were accumulated during subsidence1 it appears at first natural to connect this sinking movement with a state of high activity in the neighbouring volcanoes4 ,during the cretaceo-oolitic period this certainly appears to have been the case at the ,puente del ,inca1 judging from the number of intercalated lava-streams in the lower #c1jjj feet of strata2 but generally1 the volcanic orifices seem at this time to have existed as submarine solfataras1 and were certainly quiescent compared with their state during the accumulation of the porphyritic conglomerate formation4 ,during the deposition of the tertiary strata we know that at ,s4 ,cruz1 deluges of basaltic lava were poured forth2 but as these lie in the upper part of the series1 it is possible that the subsidence may at that time have ceased3 at ,chiloe1 ,i was unable to ascertain to what part of the series the pile of lavas belonged4 ,the ,uspallata tuffs and great streams of submarine lavas1 were probably intermediate in age between the cretaceo-oolitic and older tertiary formations1 and we know from the buried trees that there was a great subsidence during their accumulation2 but even in this case1 the subsidence may not have been strictly contemporaneous with the great volcanic eruptions1 for we must believe in at least one intercalated period of elevation1 during which the ground was upraised on which the now buried trees grew4 ,i have been led to make these remarks1 and to throw some doubt on the strict contemporaneousness of high volcanic activity and movements of subsidence1 from the conviction impressed on my mind by the study of coral formations1#ai that these two actions do not generally go on synchronously2--on the contrary1 that in volcanic districts1 subsidence ceases as soon as the orifices burst forth into renewed action1 and only recommences when they again have become dormant4 #ai 8,the ,structure1 etc41 of ,coral ,reefs40 ,at a later period1 the ,pampean mud1 of estuary origin1 was deposited over a wide area1--in one district conformably on the underlying old tertiary strata1 and in another district unconformably on them1 after their upheaval and denudation4 ,during and before the accumulation1 however1 of these old tertiary strata1 and1 therefore1 at a very remote period1 sediment1 strikingly resembling that of the ,pampas1 was deposited2 showing during how long a time in this case the same agencies were at work in the same area4 ,the deposition of the ,pampean estuary mud was accompanied1 at least in the southern parts of the ,pampas1 by an elevatory movement1 so that the ,m4 ,hermoso beds probably were accumulated after the upheaval of those round the ,s4 ,ventana2 and those at ,p4 ,alta after the upheaval of the ,m4 ,hermoso strata2 but there is some reason to suspect that one period of subsidence intervened1 during which mud was deposited over the coarse sand of the ,barrancas de ,s4 ,gregorio1 and on the higher parts of ,banda ,oriental4 ,the mammiferous animals characteristic of this formation1 many of which differ as much from the present inhabitants of ,south ,america1 as do the eocene mammals of ,europe from the present ones of that quarter of the globe1 certainly co-existed at ,b4 ,blanca with twenty species of mollusca1 one balanus1 and two corals1 all now living in the adjoining sea3 this is likewise the case in ,patagonia with the ,macrauchenia1 which co-existed with eight shells1 still the commonest kinds on that coast4 ,i will not repeat what ,i have elsewhere said1 on the place of habitation1 food1 wide range1 and extinction of the numerous gigantic mammifers1 which at this late period inhabited the two ,americas4 ,the nature and grouping of the shells embedded in the old tertiary formations of ,patagonia and ,chile show us1 that the continent at that period must have stood only a few fathoms below its present level1 and that afterwards it subsided over a wide area1 seven or eight hundred feet4 ,the manner in which it has since been rebrought up to its actual level1 was described in detail in the ,first and ,second ,chapters4 ,it was there shown that recent shells are found on the shores of the ,atlantic1 from ,tierra del ,fuego northward for a space of at least #a1ahj nautical miles1 and at the height of about #ajj feet in ,la ,plata1 and of #djj feet in ,patagonia4 ,the elevatory movements on this side of the continent have been slow2 and the coast of ,patagonia1 up to the height in one part of #iej feet and in another of #a1bjj feet1 is modelled into eight great1 step-like1 gravel-capped plains1 extending for hundreds of miles with the same heights2 this fact shows that the periods of denudation 7which1 judging from the amount of matter removed1 must have been long continued7 and of elevation were synchronous over surprisingly great lengths of coasts4 ,on the shores of the ,pacific1 upraised shells of recent species1 generally1 though not always1 in the same proportional numbers as in the adjoining sea1 have actually been found over a north and south space of #b1jge miles1 and there is reason to believe that they occur over a space of #b1dhj miles4 ,the elevation on this western side of the continent has not been equable2 at ,valparaiso1 within the period during which upraised shells have remained undecayed on the surface1 it has been #a1cjj feet1 whilst at ,coquimbo1 #bjj miles northward1 it has been within this same period only #beb feet4 ,at ,lima1 the land has been uplifted at least #hj feet since ,indian man inhabited that district2 but the level within historical times apparently has subsided4 ,at ,coquimbo1 in a height of #cfd feet1 the elevation has been interrupted by five periods of comparative rest4 ,at several places the land has been lately1 or still is1 rising both insensibly and by sudden starts of a few feet during earthquake-shocks2 this shows that these two kinds of upward movement are intimately connected together4 ,for a space of #gge miles1 upraised recent shells are found on the two opposite sides of the continent2 and in the southern half of this space1 it may be safely inferred from the slope of the land up to the ,cordillera1 and from the shells found in the central part of ,tierra del ,fuego1 and high up the ,river ,santa ,cruz1 that the entire breadth of the continent has been uplifted4 ,from the general occurrence on both coasts of successive lines of escarpments1 of sand-dunes and marks of erosion1 we must conclude that the elevatory movement has been normally interrupted by periods1 when the land either was stationary1 or when it rose at so slow a rate as not to resist the average denuding power of the waves1 or when it subsided4 ,in the case of the present high sea-cliffs of ,patagonia and in other analogous instances1 we have seen that the difficulty in understanding how strata can be removed at those depths under the sea1 at which the currents and oscillations of the water are depositing a smooth surface of mud1 sand1 and sifted pebbles1 leads to the suspicion that the formation or denudation of such cliffs has been accompanied by a sinking movement4 ,in ,south ,america1 everything has taken place on a grand scale1 and all geological phenomena are still in active operation4 ,we know how violent at the present day the earthquakes are1 we have seen how great an area is now rising1 and the plains of tertiary origin are of vast dimensions2 an almost straight line can be drawn from ,tierra del ,fuego for #a1fjj miles northward1 and probably for a much greater distance1 which shall intersect no formation older than the ,patagonian deposits2 so equable has been the upheaval of the beds1 that throughout this long line1 not a fault in the stratification or abrupt dislocation was anywhere observable4 ,looking to the basal1 metamorphic1 and plutonic rocks of the continent1 the areas formed of them are likewise vast2 and their planes of cleavage and foliation strike over surprisingly great spaces in uniform directions4 ,the ,cordillera1 with its pinnacles here and there rising upwards of twenty thousand feet above the level of the sea1 ranges in an unbroken line from ,tierra del ,fuego1 apparently to the ,arctic circle4 ,this grand range has suffered both the most violent dislocations1 and slow1 though grand1 upward and downward movements in mass2 ,i know not whether the spectacle of its immense valleys1 with mountain-masses of once liquified and intrusive rocks now bared and intersected1 or whether the view of those plains1 composed of shingle and sediment hence derived1 which stretch to the borders of the ,atlantic ,ocean1 is best adapted to excite our astonishment at the amount of wear and tear which these mountains have undergone4 ,the ,cordillera from ,tierra del ,fuego to ,mexico1 is penetrated by volcanic orifices1 and those now in action are connected in great trains4 ,the intimate relation between their recent eruptions and the slow elevation of the continent in mass1#bj appears to me highly important1 for no explanation of the one phenomenon can be considered as satisfactory which is not applicable to the other4 ,the permanence of the volcanic action on this chain of mountains is1 also1 a striking fact2 first1 we have the deluges of submarine lavas alternating with the porphyritic conglomerate strata1 then occasionally feldspathic streams and abundant mineral exhalations during the gypseous or cretaceo-oolitic period3 then the eruptions of the ,uspallata range1 and at an ancient but unknown period1 when the sea came up to the eastern foot of the ,cordillera1 streams of basaltic lava at the foot of the ,portillo range2 then the old tertiary eruptions2 and lastly1 there are here and there amongst the mountains1 much worn and apparently very ancient volcanic formations without any craters2 there are1 also1 craters quite extinct1 and others in the condition of solfataras1 and others occasionally or habitually in fierce action4 ,hence it would appear that the ,cordillera has been1 probably with some quiescent periods1 a source of volcanic matter from an epoch anterior to our cretaceo-oolitic formation to the present day2 and now the earthquakes1 daily recurrent on some part of the western coast1 give little hope that the subterranean energy is expended4 #bj ,on the ,connection of certain ,volcanic ,phenomena in ,south ,america3 8,geolog4 ,transact410 vol4 v1 p4 #fji4 ,recurring to the evidence by which it was shown that some at least of the parallel ridges1 which together compose the ,cordillera1 were successively and slowly upthrown at widely different periods2 and that the whole range certainly once1 and almost certainly twice1 subsided some thousand feet1 and being then brought up by a slow movement in mass1 again1 during the old tertiary formations1 subsided several hundred feet1 and again was brought up to its present level by a slow and often interrupted movement2 we see how opposed is this complicated history of changes slowly effected1 to the views of those geologists who believe that this great mountain-chain was formed in late times by a single blow4 ,i have endeavoured elsewhere to show1#ba that the excessively disturbed condition of the strata in the ,cordillera1 so far from indicating single periods of extreme violence1 presents insuperable difficulties1 except on the admission that the masses of once liquified rocks of the axes were repeatedly injected with intervals sufficiently long for their successive cooling and consolidation4 ,finally1 if we look to the analogies drawn from the changes now in progress in the earth's crust1 whether to the manner in which volcanic matter is erupted1 or to the manner in which the land is historically known to have risen and sunk3 or again1 if we look to the vast amount of denudation which every part of the ,cordillera has obviously suffered1 the changes through which it has been brought into its present condition1 will appear neither to have been too slowly effected1 nor to have been too complicated4 #ba 8,geolog4 ,transact410 vol4 v1 p4 #fbf4 ,,note4--,as1 both in ,france and ,england1 translations of a passage in ,professor ,ehrenberg's ,memoir1 often referred to in the ,fourth ,chapter of this volume1 have appeared1 implying that ,professor ,ehrenberg believes1 from the character of the infusoria1 that the ,pampean formation was deposited by a sea-debacle rushing over the land1 ,i may state1 on the authority of a letter to me1 that these translations are incorrect4 ,the following is the passage in question3--8,durch ,beachtung der mikroscopischen ,formen hat sich nun feststellen lassen1 das die ,mastodonten-,lager am ,la ,plata und die ,knochen-,lager am ,monte ,hermoso1 who wie die der ,riesen-,grtelthiere in den ,dnenhgeln bei ,bahia ,blanca1 beides in ,patagonien1 unvernderte brakische ,ssswasserbildungen sind1 die einst wohl smmtlich zum obersten ,fluthgebiethe des ,meeres im tieferen ,festlande gehrten40--,monatsberichten der knigl4 ,akad41 etc41 zu ,berlin vom ,april #ahde4 ,,index ,,to ,,coral-,,reefs4 ,the names in italics are all names of places1 and refer exclusively to the ,appendix3 in well-defined archipelagoes1 or groups of islands1 the name of each separate island is not given4 ,abrolhos1 ,brazil1 coated by corals #ej ,abrolhos 7,australia7 #acj ,absence of coral-reefs from certain coasts #ea ,acaba1 gulf of #adg ,admiralty group #abd ,africa1 east coast1 fringing-reef of #dh ---- ,madreporitic rock of #aja ,africa1 east coast #ada ,age of individual corals #eg1 #fd ,aiou #abh ,aitutaki #aad ,aldabra #aci ,alert reef #abc ,alexander1 ,grand ,duke1 island #aae ,allan1 ,dr41 on ,holuthuri feeding on corals #ba ---- on quick growth of corals at ,madagascar #fb ---- on reefs affected by currents #i ,alloufatou #aai ,alphonse #aci ,amargoura 7,amargura7 #aai ,amboina #abh ,america1 west coast #aaa ,amirantes #ach ,anachorites #abe ,anambas #acc ,anamouka1 description of #ii ,anamouka #aai ,anadaman islands #acb ,antilles #aec ,appoo reef #acd ,arabia ,felix #adc ,areas1 great extent of1 interspersed with low islands ---- of subsidence and of elevation #ajf ---- of subsidence appear to be elongated #ajf ---- of subsidence alternating with areas of elevation #ajh ,arru group #abh ,arzobispo #abg ,ascidia1 depth at which found #fg ,assomption #aci ,astova #aci ,atlantic islands #aba ,atolls1 breaches in their reefs #ca1 #ha ---- dimensions of #be ---- dimensions of groups of #ga ---- not based on craters or on banks of sediment1 or of ck #fi1 #ga1 #gb1 #gc1 #ajh ---- of irregular forms #be1 #hd ---- steepness of their flanks #bf ---- width of their reef and islets #be ---- their lowness #gj ---- lagoons #bi ---- general range #id ---- with part of their reef submerged1 and theory of #bi1 #ha ,augustine1 ,st4 #abj ,aurora island1 an upraised atoll #fd1 #ga1 #ajd ,aurora #aab ,austral islands1 recently elevated #ii ,austral islands #aad ,australia1 ,n4,w4 coast #acj ,australian barrier-reef #db1 #ic ,australian barrier #abc ,babuyan group #acd ,bahama banks #adi1 #aej ,balahac #acc ,bally #aca ,baring #aba ,barrier-reef of ,australia #db1 #ic ---- of ,new ,caledonia #dd ,barrier-reefs1 breaches through #gg ---- not based on worn down margin of rock #dc ---- on banks of sediment #dc ---- on submarine craters #dd ---- steepness of their flanks #ci ---- their probable vertical thickness #dc1 #gf ---- theory of their formation #gf1 #gh ,bampton shoal #abc ,banks islands #abb ,banks in the ,west ,indies #adg ,bashee islands #acd ,bass island #aae ,batoa #aai ,beaupr reef #abc ,beechey1 ,captain1 obligations of the author to #bf ---- on submerged reefs #bg ---- account of ,matilda island #fj ,belcher1 ,captain1 on boring through coral-reef #ei ,belize reef1 off #aea ,bellinghausen #aac ,bermuda islands #aec ,beveridge reef #aah ,bligh #abb ,bolabola1 view of #ab ,bombay shoal #acf ,bonin ,bay #aca ,bonin group #abg ,borings through coral-reefs #ei ,borneo1 ,w4 coast1 recently elevated #aja ,borneo1 ,e4 coast #aca ---- ,s4,w4 and ,w4 coast #acc ---- ,n4 coast #acc ---- western bank #acf ,boscawen #aai ,boston #aba ,bouka #abd ,bourbon #ach ,bourou #abh ,bouton #acb ,brazil1 fringing-reefs on coast of #dh ,breaches through barrier-reefs #ga ,brook #aae ,bunker #aae ,bunoa #acc ,byron #aba ,cagayanes #acc ,candelaria #abd ,cargados ,carajos #ach ,caroline archipelago #abe ,caroline island #aae ,carteret shoal #abh ,caryophyllia1 depth at which it lives #ff ,cavilli #acc ,cayman island #aeb ,celebes #abi ,ceram #abh ,ceylon1 recently elevated #aja ,ceylon #acg ,chagos ,great ,bank1 description and theory of #cg1 #he ,chagos group #hf ,chagos group #acg ,chama-shells embedded in coral-rock #fh ,chamisso1 on corals preferring the surf #eb ,changes in the state of ,keeling atoll #ba ---- of atolls #gd ,channels leading into the lagoons of atolls #cj1 #hb ---- ---- into the ,maldiva atolls #cc1 #ce ---- through barrier-reefs #gg ,chase #abj ,china sea #ace ,christmas atoll #fj1 #ig ,christmas atoll #aaf ,christmas island 7,indian ,ocean7 #acg ,clarence #aaf ,clipperton rock #aaa ,cocos1 or ,keeling atoll #ae ,cocos 7or ,keeling7 #acg ,cocos island 7,pacific7 #aaa ,cochin ,china1 encroachments of the sea on the coast #ie ,cochin ,china #ahc ,coetivi #aci ,comoro group #aci ,composition of coral-formations #hh ,conglomerate coral-rock on ,keeling atoll #bj ---- on other atolls #bh ---- coral-rock #hh ,cook islands1 recently elevated #ih1 #ajc ,cook islands #aad ,coral-blocks bored by vermiform animals #ba1 #hh ,coral-reefs1 their distribution and absence from certain areas #ej ---- destroyed by loose sediment #ec ,coral-rock at ,keeling atoll #bj ---- ,mauritius #dg ---- organic remains of #hh ,corals dead but upright in ,keeling lagoon #bb ---- depths at which they live #fd ---- off ,keeling atoll #ag ---- killed by a short exposure #af ---- living in the lagoon of ,keeling atoll #bj ---- quick growth of1 in ,keeling lagoon #ba ---- merely coating the bottom of the sea #ej ---- standing exposed in the ,low archipelago #if ,corallian sea #id ,corallian sea #abc ,cornwallis #aba ,cosmoledo #aci ,couthouy1 ,mr41 alleged proofs of recent elevation of the ,low archipelago #if ---- on coral-rock at ,mangaia and ,aurora islands #fd ---- on external ledges round coral-islands #hj ---- remarks confirmatory of the author's theory #if ,crescent-formed reefs #hd ,cuba #aej ,cuming1 ,mr41 on the recent elevation of the ,philippines #aja ,dangerous1 or ,low archipelago #aaa ,danger islands #aaf ,depths at which reef-building corals live #fc ---- at ,mauritius1 the ,red ,sea1 and in the ,maldiva archipelago #ff ---- at which other corals and corallines can live #fg ,dhalac group #add ,diego ,garcia1 slow growth of reef #ef ,dimensions of the larger groups of atolls #ga ,disseverment of the ,maldiva atolls1 and theory of #cg1 #hb ,distribution of coral-reefs #ej ,domingo1 ,st4 #aeb ,dory1 ,port1 recently elevated #ajj ,dory1 ,port #abg ,duff islands #abb ,durour #abe ,eap #abf arthquakes at ,keeling atoll #bc ---- in groups of atolls #ge ---- in ,navigator archipelago #ajj ast ,indian archipelago1 recently elevated #ajj ,easter #aaa ,echequier #abe hrenberg1 on the banks of the ,red ,sea #di1 #adc ---- on depths at which corals live in the ,red ,sea #ff ---- on corals preferring the surf #ec ---- on the antiquity of certain corals #eg ,eimeo #aab levated reef of ,mauritius #dg levations1 recent proofs of #ih ---- immense areas of #ajf ,elivi #abf lizabeth island #ei ---- recently elevated #ih1 #ajd ,elizabeth island #aab ,ellice group #abj ncircled islands1 their height #da ---- geological composition #db1 #dd ua1 description of #ii ,eoua #aah upted matter probably not associated with thick masses of coral-rock #hi ,fais1 recently elevated #ajj1 #ajd ,fais #abf ,fanning #aaf ,farallon de ,medinilla #abg ,farson group #add ,fataka #abb ,fiji archipelago #aai ,fish1 feeding on corals #ba ---- killed in ,keeling lagoon by heavy rain #bd ,fissures across coral-islands #ge ,fitzroy1 ,captain1 on a submerged shed at ,keeling atoll #bc ---- on an inundation in the ,low archipelago #gd ,flint #aae ,flores #acj ,florida #adi ,folger #abg ,formosa #ace ,forster1 theory of coral-formations #gc ,frederick reef #abc ,freewill #abh ,friendly group recently elevated #ii1 #aje ,friendly archipelago #aah ,fringing-reefs1 absent where coast precipitous #e ---- breached in front of streams #ed ---- described by ,,mm4 ,quoy and ,gaimard #ih ---- not closely attached to shelving coasts #df ---- of east coast of ,africa ---- of ,cuba #dh ---- of ,mauritius #de ---- on worn down banks of rock #i ---- on banks of sediment #di ---- their appearance when elevated #g ---- their growth influenced by currents #di ---- by shallowness of sea #di ,galapagos archipelago #aaa ,galega #aci ,gambier islands1 section of #dc ,gambier islands #aab ,gardner #aaf ,gaspar rico #aba ,geological composition of coral-formations ,gilbert archipelago #abj ,gilolo #abi ,glorioso #aci ,gloucester island #gd ,glover reef #aeb ,gomez #aaa ,gouap #abf ,goulou #abf ,grampus #abg ,gran ,cocal #abj ,great ,chagos ,bank1 description and theory of #cg1 #he ,grey1 ,captain1 on sandbars #df ,grouping of the different classes of reefs #ic ,guedes #abh ,hall1 ,captain ,b41 on ,loo ,choo #aja ,harvey islands1 recently elevated #ajd ,height of encircled islands #da ,hermites #abe ,hervey or ,cook islands #aad ,hogoleu #abe ,holothuri 7,holuthuri7 feeding on coral #ba ,houden island1 height of #ga ,honduras1 reef off #aea ,horn #aai ,houtman ,abrolhos #acj ,huaheine2 alleged proofs of its recent elevation #ajc ,huaheine #aac ,humphrey #aae ,hunter #aai ,hurricanes1 effects of1 on coral-islands #gd ,immaum #adc ,independence #abj ,india1 west coast1 recently elevated #aja ,india #adc ,irregular reefs in shallow seas #di ,islets of coral-rock1 their formation #ai ---- their destruction in the ,maldiva atolls #cf ,jamaica #aeb ,jarvis #aae ,java1 recently elevated #ajj ,java #aca ,johnston island #aaf ,juan de ,nova #aci ,juan de ,nova 7,madagascar7 #adj ,kalatoa #aca ,kamtschatka1 proofs of its recent elevation #aje ,karkalang #abi ,keeling atoll1 section of reef #ae ,keeling1 south atoll #acg ---- north atoll #acg ,keffing #abh ,kemin #aae1 #aaf ,kennedy #abc ,keppel #aai ,kumi #ace ,laccadive group #acg ,ladrones1 or ,marianas1 recently elevated #ajj ,ladrones archipelago #abg ,lagoon of ,keeling atoll #bj ,lagoons bordered by inclined ledges and walls1 and theory of their formation #cb1 #gi ---- of small atolls filled up with sediment #cb ,lagoon-channels within barrier-reefs #dj ,lagoon-reefs1 all submerged in some atolls1 and rising to the surface in others #ee ,lancaster reef #aae ,latte #aai ,lauglan islands #abc ,ledges round certain lagoons #cb1 #gi ,lette #abi ,lighthouse reef #aeb ,lloyd1 ,mr41 on corals refixing themselves #fb ,loo ,choo1 recently elevated #aja ,loo ,choo #ace ,louisiade #abc ,low archipelago1 alleged proofs of its recent elevation #if ,low archipelago #aaa ,lowness of coral-islands #gj ,loyalty group #abc ,lucepara #acc ,lutk1 ,admiral1 on fissures across coral-islands #ge ,luzon1 recently elevated #aja ,luzon #acd ,lyell1 ,mr41 on channels into the lagoons of atolls #ca ---- on the lowness of their leeward sides #hb ---- on the antiquity of certain corals #eh ---- on the apparent continuity of distinct coral-islands #hi ---- on the recently elevated beds of the ,red ,sea #ajb ---- on the outline of the areas of subsidence #ajf ,macassar strait #aca ,macclesfield bank #acf ,madagascar1 quick growth of corals at #fb ---- madreporitic rock of #aja ,madagascar #adj ,madjiko-sima #ace ,madura 7,java7 #aca ,madura 7,india7 #acg ,mahlos ,mahdoo1 theory of formation #hh ,malacca1 recently elevated #ajj ,malacca #acc ,malcolmson1 ,dr41 on recent elevation of ,w4 coast of ,india #ajj ---- on recent elevation of ,camaran island #ajb ,malden #aae ,maldiva atolls1 and theory of their formation #cc1 #hj1 #hb ---- steepness of their flanks #bf ---- growth of coral at #fb ,maldiva archipelago #acg ,mangaia island #fd ---- recently elevated #ii1 #ajd ,mangaia #aad ,mangs #abg ,marianas1 recently elevated #ajj ,mariana archipelago #abg ,mariere #abf ,marquesas archipelago #aac ,marshall archipelago #aba ,marshall island #abg ,martinique #aec ,martires #abf ,mary's ,st4 in ,madagascar1 harbour made in reefs #ed ,mary island #aaf ,matia1 or ,aurora #aab ,matilda atoll #fj ,mauritius1 fringing-reefs of #de ---- depths at which corals live there #fd ---- recently elevated #aja ,mauritius #ach ,maurua1 section of #dc ,maurua #aac ,menchikoff atoll #be1 ,mendana archipelago #aac ,mendana isles #abb ,mexico1 gulf of #adi ,millepora complanata at ,keeling atoll #af ,mindoro #acd ,mohilla 7,mohila7 #aci ,molucca islands1 recently elevated #ajj ,mopeha #aac ,moresby1 ,captain1 on boring through coral-reefs #ei ,morty #abi ,mosquito coast #aeb ,musquillo atoll #hd ,mysol #abi ,namourrek group #hd ,natunas #acc ,navigator archipelago1 elevation of #ii ,navigator archipelago #aag ,nederlandisch #abj ,nelson1 ,lieutenant1 on the consolidation of coral-rocks under water #ei ---- theory of coral-formations #gc ---- on the ,bermuda islands #aed ,new ,britain #abd ,new ,caledonia1 steepness of its reefs #ci ---- ---- barrier-reef of 1 #gi1 #hc1 #ic ,new ,caledonia #abc ,new ,guinea 7,e4 end7 #abd ,new ,guinea 7,w4 end7 #abg ,new ,hanover #abd ,new ,hebrides1 recently elevated #ajj ,new ,hebrides #aba ,new ,ireland1 recently elevated #ajj ,new ,ireland #abd ,new ,nantucket #aaf ,nicobar islands #acb ,niouha #aai ,nullipor at ,keeling atoll #ah ---- on the reefs of atolls #bh ---- on barrier-reefs #ci ---- their wide distribution and abundance #fh ,objections to the theory of subsidence #g ,ocean islands #aag1 #aba ,ono #abj ,onouafu 7,onouafou7 #aai ,ormuz #adc ,oscar group #abj ,oscillations of level #ajc1 #ajh ,ouallan1 or ,ualan 7,oualan7 #abe ,ouluthy atoll #fj ,outong ,java #abd ,palawan1 ,s4,w4 coast #acc ---- ,n4,w4 coast #acd ---- western bank #acf ,palmerston #aad ,palmyra #aaf ,paracells #acf ,paraquas #acf ,patchow #ace ,pelew islands #abf ,pemba island1 singular form of #ajb ,pemba #adb ,penrhyn #aae ,peregrino #aae ,pernambuco1 bar of sandstone at #dg ,persian gulf1 recently elevated #ajb ,persian gulf #adc ,pescado #aae ,pescadores #ace ,peyster group #abj ,philip #abf ,philippine archipelago1 recently elevated #aja ,philippine archipelago #acd ,phnix #aaf ,piguiram #abf ,pitcairn #aab ,pitt's bank #hf ,pitt island #abj ,platte #aci ,pleasant #aba ,porites1 chief coral on margin of ,keeling atoll #af ,postillions #aca ,pouynipte #ie ---- its probable subsidence #ie ,pouynipte #abe ,pratas shoal #ace ,proby #aai ,providence #aci ,puerto ,rico #aeb ,pulo ,anna #abf ,pumice floated to coral-islands #hh ,pylstaart #aah ,pyrard de ,laval1 astonishment at the atolls in the ,indian ,ocean #aa ,quoy and ,gaimard1 depths at which corals live #ff ---- description of reefs applicable only to fringing-reefs #ih ,range of atolls #id ,rapa #aae ,rearson #aae ,red ,sea1 banks of rock coated by reefs #di ---- proofs of its recent elevation #ajb ---- supposed subsidence of #ajc ,red ,sea #adc ,reefs1 irregular in shallow seas #di ---- rising to the surface in some lagoons and all submerged in others #ee ---- their distribution #ej ---- their absence from some coasts #ea ,revilla-gigedo #aaa ,ring-formed reefs of the ,maldiva atolls1 and theory of 1 #hj ,rodriguez #ach ,rosario #abg ,rose island #aah ,rotches #abj ,rotoumah #abj ,roug #abe ,rowley shoals #acj ,rppell1 ,dr41 on the recent deposits of ,red ,sea #ajb ,sable1 ile de #ach ,sahia de ,malha #acg ,st4 ,pierre #aci ,sala #aaa ,salomon 7,solomon7 archipelago #abc ,samoa1 or ,navigator archipelago1 elevation of #ii ,samoa archipelago #aag ,sand-bars parallel to coasts #df ,sandal-wood #abi ,sandwich archipelago1 recently elevated #ih ,sandwich archipelago #aag ,sanserot #abf ,santa-,cruz group #abb ,savage island1 recently elevated #ei1 #ii1 #ajd ,savage #aah ,savu #abi ,saya1 or ,sahia de ,malha #acg ,scarborough shoal #acf ,scarus feeding on corals #ba ,schouten #abd ,scilly #aac ,scori floated to coral-islands #hi ,scott's reef #acj ,sections of islands encircled by barrier-reefs #dc1 #agf ---- of ,bolabola #gf ,sediment in ,keeling lagoon #ba ---- in other atolls #bi1 #ce ---- injurious to corals #ec ---- transported from coral-islands far seaward #hi ,seniavine #abe ,serangani #abi ,seychelles #ach ,ship-bottom quickly coated with coral #fb ,smyth island #aaf ,society archipelago1 stationary condition of #if ---- alleged proofs of recent elevation #ajc ,society archipelago #aab ,socotra #adc ,solor #acj ,sooloo islands1 recently elevated #aja ,sooloo islands #acc ,souvaroff #aae ,spanish #abf ,sponge1 depths at which found #fg ,starbuck 7,slarbuck7 #aae ,stones transported in roots of trees #hi ,storms1 effects of1 on coral-islands #gd ,stutchbury1 ,mr41 on the growth of an ,agaricia #fc ---- on upraised corals in ,society archipelago #ajc ,subsidence of ,keeling atoll #bh ---- extreme slowness of #hg1 #ajh ---- areas of1 apparently elongated #ajf ---- areas of immense #ajf ---- great amount of #ajh ,suez1 gulf of #adg ,sulphur islands #abg ,sumatra1 recently elevated #ajj ,sumatra #acb ,sumbawa #acj ,surf favourable to the growth of massive corals #eb ,swallow shoal #acf ,sydney island #aaf ,tahiti1 alleged proofs of its recent elevation #ajc ,tahiti #aab ,temperature of the sea at the ,galapagos archipelago #ea ,tenasserim #acc ,tenimber island #abh ,teturoa #aac ,theories on coral-formations #fi1 #gc ,theory of subsidence1 and objections to #gb1 #hf ,thickness1 vertical1 of barrier-reefs #dc1 #gf ,thomas1 ,st4 #aec ,tikopia #abb ,timor1 recently elevated #ajj ,timor #abi ,timor-laut #abh ,tokan-,bessees #aca ,tongatabou #aah ,tonquin #acg ,toubai #aac ,toufoa 7,toofoa7 #aai ,toupoua #abb ,traditions of change in coral-islands #gc ,tridacn embedded in coral-rock #fc ---- left exposed in the ,low archipelago #if ,tubularia1 quick growth of #fc ,tumbelan #acc ,turneffe reef #aeb ,turtle #aai ,ualan #abe ,vanikoro1 section of #dc ---- its state and changes in its reefs #ie ,vanikoro #abb ,vine reef #abe ,virgin ,gorda #aec ,viti archipelago #aai ,volcanic islands1 with living corals on their shores #ea ---- matter1 probably not associated with thick masses of coral-rock #hh ,volcanoes1 authorities for their position on the map #ij ---- their presence determined by the movements in progress #ajd ---- absent or extinct in the areas of subsidence #aje ,waigiou #abh ,wallis island #aai ,washington #aaf ,wells' reef #abc ,wellstead1 ,lieutenant1 account of a ship coated with corals #fb ,west ,indies1 banks of sediment fringed by reefs #di ---- recently elevated #ajb ,west ,indies #adg ,whitsunday island1 view of #ab ---- changes in its state #gd ,williams1 ,rev4 ,j41 on traditions of the natives regarding coral-islands #gd ---- on antiquity of certain corals #fd ,wolchonsky #aaa ,wostock #aae ,xulla islands #abh ,york island #aaf ,yucutan1 coast of #aea ,zones of different kinds of corals outside the same reefs #ee1 #fj ,,index ,,to ,,volcanic ,,islands4 ,abel1 ,m41 on calcareous casts at the ,cape of ,good ,hope #bfa ,abingdon island #bcd ,abrolhos islands1 incrustation on #ahh ,aeriform explosions at ,ascension #aia ,albatross1 driven from ,st4 ,helena #bbe ,albemarle island #bcd ,albite1 at the ,galapagos archipelago #bcd ,amygdaloidal cells1 half filled #ahd ,amygdaloids1 calcareous origin of #agf ,ascension1 arborescent incrustation on rocks of #ahh ---- absence of dikes1 freedom from volcanic action1 and state of lava-streams #bbf ,ascidia1 extinction of #beh ,atlantic ,ocean1 new volcanic focus in #bbf ,augite1 fused #bci ,australia #bea ,azores #ahb1 #bdh ,bahia in ,brazil1 dikes at #bdg ,bailly1 ,m41 on the mountains of ,mauritius #ahe ,bald ,head #bfj ,banks' ,cove #bcd1 #bcf ,barn1 ,the1 ,st4 ,helena #baf ,basalt1 specific gravity of #bde ,basaltic coast-mountains at ,mauritiu #ahe ---- at ,st4 ,helena #bah ---- at ,st4 ,jago #agh ,beaumont1 ,m4 ,elie de1 on circular subsidences in lava #bcc ---- on dikes indicating elevation #bbh ---- on inclination of lava-streams #bbg ---- on laminated dikes #bab ,bermuda1 calcareous rocks of #bfj1 #bfb ,beudant1 ,m41 on bombs #aia ---- on jasper #aig ---- on laminated trachyte #baa ---- on obsidian of ,hungary #bjg ---- on silex in trachyte #bgj1 #aig ,bole #beg ,bombs1 volcanic #ahi ,bory ,st4 ,vincent1 on bombs #aij ,boulders1 absence in ,australia and ,cape of ,good ,hope #bfe ,brattle island #bch ,brewster1 ,sir ,d41 on a calcareo-animal substance #bja ---- on decomposed glass #beb ,brown1 ,mr4 ,r41 on extinct plants from ,van ,diemen's land #beg ---- on sphrulitic bodies in silicified wood #bjg ,buch1 ,von1 on cavernous lava #bcc ---- on central volcanoes #bdi ---- on crystals sinking in obsidian #bdc ---- on laminated lava #bji ---- on obsidian streams #bjh ---- on olivine in basalt #bcd ---- on superficial calcareous beds in the ,canary islands #bbd ,calcareous deposit at ,st4 ,jago affected by heat #afi1 #aga ---- fibrous matter1 entangled in streaks in scori #agd ---- freestone at ,ascension #aih ---- incrustations at ,ascension #aii ---- sandstone at ,st4 ,helena #bbb ---- superficial beds at ,king ,george's sound #bfj ,cape of ,good ,hope #bfc ,carbonic acid1 expulsion of1 by heat #aga1 #agf ,carmichael1 ,capt41 on glassy coatings to dikes #baf ,casts1 calcareous1 of branches #bfa ,chalcedonic nodules #beg ,chalcedony in basalt and in silicified wood #aif ,chatham island #bca1 #bce1 #bda1 #bdh1 #bei ,chlorophite #beg ,clarke1 ,rev4 ,w41 on the ,cape of ,good ,hope #beh1 #bfc ,clay-slate1 its decomposition and junction with granite at the ,cape of ,good ,hope #bfd ,cleavage of clay-slate in ,australia #beb ,cleavage1 cross1 in sandstone #bec ,coast denudation at ,st4 ,helena #bbf ,columnar basalt #agc 8,comptes ,rendus10 account of volcanic phenomena in the ,atlantic #bbf ,concepcion1 earthquake of #bbh1 #bdi ,concretions in aqueous and igneous rocks compared #bjf ---- in tuff #aig ---- of obsidian #bjf1 #bjh ,conglomerate1 recent1 at ,st4 ,jago #aha ,coquimbo1 curious rock of #bfa ,corals1 fossil1 from ,van ,diemen's ,land #bef ,crater1 segment of1 at the ,galapagos #bch ---- great central one at ,st4 ,helena #bai ---- internal ledges round1 and parapet on #bbj ,craters1 basaltic1 at ,ascension #ahi ---- form of1 affected by the trade wind #ahi ---- of elevation #bbg ---- of tuff at ,terceira #ahb ---- of tuff at the ,galapagos archipelago #bcj1 #bca1 #bce1 #bcg ---- their breached state #bdj ---- small basaltic at ,st4 ,jago #agg ---- ---- at the ,galapagos archipelago #bcb ,crystallisation favoured by space #baa ,dartigues1 ,m41 on sphrulites #bjg ,daubeny1 ,dr41 on a basin-formed island #bcg ---- on fragments in trachyte #aic ,d',aubuisson on hills of phonolite #bbb ---- on the composition of obsidian #bjf ---- on the lamination of clay-slate #baj ,de la ,beche1 ,sir ,h41 on magnesia in erupted lime #agd ---- on specific gravity of limestones #aih ,denudation of coast at ,st4 ,helena #bbf ,diana's ,peak1 ,st4 ,helena #bbj ,dieffenbach1 ,dr41 on the ,chatham ,islands #bei ,dikes1 truncated1 on central crateriform ridge of ,st4 ,helena #bai ---- at ,st4 ,helena2 number of2 coated by a glossy layer2 uniform thickness of #baf ---- great parallel ones at ,st4 ,helena #bbb ---- not observed at ,ascension #bbf ---- of tuff #bca ---- of trap in the plutonic series #bdg ---- remnants of1 extending far into the sea round ,st4 ,helena #bbf ,dislocations at ,ascension #aib ---- at ,st4 ,helena #bag1 #bba ,distribution of volcanic islands #bdh ,dolomieu1 on decomposed trachyte #ahb ---- on laminated lava #baj1 #baa ---- on obsidian #bjh ,dre1 ,m41 on crystals sinking in lava #bdc ,dufrenoy1 ,m41 on the composition of the surface of certain lava-streams #bji1 #bdc ---- on the inclination of tuff-strata #bcf ,eggs of birds embedded at ,st4 ,helena #bbd ---- of turtle at ,ascension #aih ,ejected fragments at ,ascension #aib ---- at the ,galapagos archipelago #bci ,elevation of ,st4 ,helena #bbe ---- the ,galapagos archipelago #bda ---- ,van ,diemen's ,land1 ,cape of ,good ,hope1 ,new ,zealand1 ,australia1 and ,chatham island #beh ---- of volcanic islands #bej ,ellis1 ,rev4 ,w41 on ledges within the great crater at ,hawaii #bbj ---- on marine remains at ,otaheite #ahd ,eruption1 fissures of #bbd1 #bdi1 #bej ,extinction of land-shells at ,st4 ,helena #bbd ,faraday1 ,mr41 on the expulsion of carbonic acid gas #aga ,feldspar1 fusibility of #bdf ---- in radiating crystals #bfc ---- ,labrador1 ejected #aic ,feldspathic lavas #agi ---- at ,st4 ,helena #bai ---- rock1 alternating with obsidian #bjb ---- lamination1 and origin of #bji ,fernando ,noronha #aha1 #baj ,ferruginous superficial beds #bei ,fibrous calcareous matter at ,st4 ,jago #agd ,fissures of eruption #bdb1 #bdi1 #bej ,fitton1 ,dr41 on calcareous breccia #bfb ,flagstaff ,hill1 ,st4 ,helena #baf ,fleurian de ,bellevue on sphrulites #bjg ,fluidity of lavas #bcd1 #bce ,forbes1 ,professor1 on the structure of glaciers #bab ,fragments ejected at ,ascension #aib ---- at the ,galapagos archipelago #bci ,freshwater ,bay #bch1 #bdc ,fuerteventura 7,feurteventura71 calcareous beds of #bbd ,galapagos archipelago #bbi ---- parapets round craters #bbj ,gay ,lussac1 on the expulsion of carbonic acid gas #aga ,glaciers1 their structure #bab ,glossiness of texture1 origin of #bjf ,gneiss1 derived from clay-slate #bfd ---- with a great embedded fragment #beb ,gneiss-granite1 form of hills of #bei ,good ,hope1 ,cape of #bfc ,gorges1 narrow1 at ,st4 ,helena #bbe ,granite1 junction with clay-slate1 at the ,cape of ,good ,hope #bfc ,granitic ejected fragments #aib1 #bci ,gravity1 specific1 of lavas #bdc-h ,gypsum1 at ,ascension #bja ---- in volcanic strata at ,st4 ,helena #bae ---- on surface of the ground at ditto #bbc ,hall1 ,sir ,j41 on the expulsion of carbonic acid gas #aga ,heat1 action of1 on calcareous matter #agj ,hennah1 ,mr41 on ashes at ,ascension #ahi ,henslow1 ,prof41 on chalcedony #aig ,hoffmann1 on decomposed trachyte #ahb ,holland1 ,dr41 on ,iceland #bbh ,horner1 ,mr41 on a calcareo-animal substance #bja ---- on fusibility of feldspar #bdf ,hubbard1 ,dr41 on dikes #bdg ,humboldt on ejected fragments #aic ---- on obsidian formations #bjg1 #bji ---- on parapets round craters #bbj ---- on sphrulites #baj ,hutton on amygdaloids #agf ,hyalite in decomposed trachyte #ahb ,iceland1 stratification of the circumferential hills #bbh ,islands1 volcanic1 distribution of #bdh ---- their elevation #bej ,incrustation1 on ,st4 ,paul's rocks #ahg ,incrustations1 calcareous1 at ,ascension #aii ,jago1 ,st4 #afg ,james island #bcd1 #bcg1 #bdb ,jasper1 origin of #aif ,jonns1 ,m4 ,moreau de1 on craters affected by wind #ahi ,juan ,fernandez #bej ,keilhau1 ,m41 on granite #bfd ,kicker ,rock #bcb ,king ,george's sound #bei ,labrador feldspar1 ejected #aic ,lakes at bases of volcanoes #bbi ,lamination of volcanic rocks #bji ,land-shells1 extinct1 at ,st4 ,helena #bbd ,lanzarote1 calcareous beds of #bbc ,lava1 adhesion to sides of a gorge #agg ---- feldspathic #agi ---- with cells semi-amygdaloidal #ahd ,lavas1 specific gravity of #bdc1 #bdg ,lava-streams blending together at ,st4 ,jago #agg ---- composition of surface of #bjh ---- differences in the state of their surfaces #bdd ---- extreme thinness of #bch ---- heaved up into hillocks at the ,galapagos archipelago #bcc ---- their fluidity #bcd1 #bce ---- with irregular hummocks at ,ascension #ahi ,lead1 separation from silver #bdd ,lesson1 ,m41 on craters at ,ascension #ahi ,leucite #bcd ,lime1 sulphate of1 at ,ascension #bjj ,lonsdale1 ,mr41 on fossil-corals from ,van ,diemen's land #bef ,lot1 ,st4 ,helena #bba ,lyell1 ,mr41 on craters of elevation #bbg ---- on embedded turtles' eggs #aih ---- on glossy coating to dikes #baf ,macaulay1 ,dr41 on calcareous casts at ,madeira #bfb ,mac,culloch1 ,dr41 on an amygdaloid #ahd ---- on chlorophite #bhg ---- on laminated pitchstone #bji ,mackenzie1 ,sir ,g41 on cavernous lava-streams #bcc ---- on glossy coatings to dikes #baf ---- on obsidian streams #bjh ---- on stratification in ,iceland #bbh ,madeira1 calcareous casts at #bfb ,magazine1 ,nautical1--account of volcanic phenomena in the ,atlantic #bbf ,marekanite #bjf ,mauritius1 crater of elevation of #ahd1 #bbg ,mica1 in rounded nodules #afh ---- origin in metamorphic slate #bfd ---- radiating form of #bfc ,miller1 ,prof41 on ejected ,labrador feldspar #aic ---- on quartz crystals in obsidian beds #bjb ,mitchell1 ,sir ,t41 on bombs #aia ---- on the ,australian valleys #bed ,mud streams at the ,galapagos archipelago #bcf ,narborough island #bcd ,nelson1 ,lieut41 on the ,bermuda islands #bfj1 #bfb ,new ,caledonia #bdh ,new ,red sandstone1 cross cleavage of #bec ,new ,south ,wales #bea ,new ,zealand #bei ,nullipor 7fossil71 resembling concretions #afi ,obsidian1 absent at the ,galapagos archipelago #bda ---- bombs of #aia ---- composition and origin of #bjg1 #bjh ---- crystals of feldspar sink in #bdc ---- its irruption from lofty craters #bdf ---- passage of beds into #bjb ---- specific gravity of #bdc1 #bdf ---- streams of #bjh ,olivine decomposed at ,st4 ,jago #agh ---- at ,van ,diemen's land #beg ---- in the lavas at the ,galapagos archipelago #bcd ,oolitic structure of recent calcareous beds at ,st4 ,helena #bbc ,otaheite #ahc ,oysters1 extinction of #beh ,panza islands1 laminated trachyte of #bji ,pattinson1 ,mr41 on the separation of lead and silver #bdd ,paul's1 ,st41 rocks of #ahg ,pearlstone #bjf ,peperino #bcb ,pron1 ,m41 on calcareous rocks of ,australia #bfb1 #bfc ,phonolite1 hills of #agi1 #aha1 #bba ---- laminated #baj ---- with more fusible hornblende #bdf ,pitchstone #bjd ---- dikes of #bji ,plants1 extinct #beg ,plutonic rocks1 separation of constituent parts of1 by gravity #bdf ,porto ,praya #afg ,prevost1 ,m4 ,c41 on rarity of great dislocations in volcanic islands #bag ,prosperous hill1 ,st4 ,helena #bah ,pumice1 absent at the ,galapagos archipelago #bda ---- laminated #bji1 #baj1 #baa ,puy de ,dome1 trachyte of #aic ,quail island1 ,st4 ,jago #afh1 #agj1 #agc ,quartz1 crystals of1 in beds alternating with obsidian #bjb ---- crystallised in sandstone #beb ---- fusibility of #bdf ---- rock1 mottled from metamorphic action with earthy matter #agj ,red hill #agc ,resin-like altered scori #aga ,rio de ,janeiro1 gneiss of #beb ,robert1 ,m41 on strata of ,iceland #bbh ,rogers1 ,professor1 on curved lines of elevation #bdi ,salses1 compared with tuff craters #bdj ,salt deposited by the sea #bjj ---- in volcanic strata #bja1 #bae ---- lakes of1 in craters #bdj ,sandstone of ,brazil #bfe ---- of the ,cape of ,good ,hope #bfe ---- platforms of1 in ,new ,south ,wales #beb1 #bfe ,schorl1 radiating #bfc ,scrope1 ,mr4 ,p41 on laminated trachyte #bji1 #baj1 #bab ---- on obsidian #bjh ---- on separation of trachyte and basalt #bdd ---- on silex in trachyte #agf ---- on sphrulites #baj ,seale1 ,mr41 geognosy of ,st4 ,helena #bae ---- on dikes #bbf ---- on embedded birds' bones #bbe ,seale1 on extinct shells of ,st4 ,helena #bbd ,sedgwick1 ,professor1 on concretions #bjf ,septaria1 in concretions in tuff #aih ,serpul on upraised rocks #ahe ,seychelles #bdh ,shells1 colour of1 affected by light #bja ---- from ,van ,diemen's land #bef ---- land1 extinct1 at ,st4 ,helena #bbd ---- particles of1 drifted by the wind at ,st4 ,helena #bbc ,shelly matter deposited by the waves #bjj ,siau1 ,m41 on ripples #bed ,signal ,post ,hill #afh1 #age1 #agf ,silica1 deposited by steam #ahb ---- large proportion of1 in obsidian #bjf1 #bjh ---- specific gravity of #bdf ,siliceous sinter #aif ,smith1 ,dr4 ,a41 on junction of granite and clay-slate #bfd ,spallanzani on decomposed trachyte #ahb ,specific gravity of recent calcareous rocks and of limestone #aih ---- of lavas #bde ,sphrulites in glass and in silicified wood #bjg ---- in obsidian #bjd1 #baj ,sowerby1 ,mr4 ,g4 ,b41 on fossil-shells from ,van ,diemen's land #bef ---- from ,st4 ,jago #afi ---- land-shells from ,st4 ,helena #bbd ,st4 ,helena #bad ---- crater of elevation of #bbg ,st4 ,jago1 crater of elevation of #bbg ---- effects of calcareous matter on lava #bca ,st4 ,paul's rocks #ahg1 #bdh ,stokes1 ,mr41 collections of sphrulites and of obsidians #bjg1 #bab ,stony-top1 ,little #bah1 #bbb ---- ,great #bah ,stratification of sandstone in ,new ,south ,wales #bec1 #bee ,streams of obsidian #bjh ,stutchbury1 ,mr41 on marine remains at ,otaheite #ahd ,subsided space at ,ascension #aib ,tahiti #ahc ,talus1 stratified1 within tuff craters #bcf ,terceira #ahb ,tertiary deposit of ,st4 ,jago #afi ,trachyte1 absent at the ,galapagos archipelago #bda ---- at ,ascension #aic ---- at ,terceira #ahb ---- decomposition of1 by steam #ahb ---- its lamination #bjj1 #baj ---- its separation from basalt #bdd ---- softened at ,ascension #aid ---- specific gravity of #bde ---- with singular veins #aie ,trap-dikes in the plutonic series #bdg ---- at ,king ,george's sound #bei ,travertin at ,van ,diemen's land #beg ,tropic-bird1 now rare1 at ,st4 ,helena #bbe ,tuff1 craters of #bca1 #bce1 #bcf ---- their breached state #bdj ---- peculiar kind of #bca ,turner1 ,mr41 on the separation of molten metals #bdd ,tyerman and ,bennett on marine remains at ,huaheine #ahd ,valleys1 gorge-like1 at ,st4 ,helena #bbe ---- in ,new ,south ,wales #bed ---- in ,st4 ,jago #ahj ,van ,diemen's land #bef ,veins in trachyte #aie ---- of jasper #aie ,vincent1 ,bory ,st41 on bombs #aij ,volcanic bombs #ahi ---- island in process of formation in the ,atlantic #bbf ---- islands1 their distribution #bdh ,wacke1 its passage into lava #ahc1 #beg ,wackes1 argillaceous #afh1 #agh ,webster1 ,dr41 on a basin-formed island #bcg ---- on gypsum at ,ascension #bja ,white1 ,martin1 on soundings #bed ,wind1 effects of1 on the form of craters ,,index ,,to ,,south ,,american ,,geology4 ,abich1 on a new variety of feldspar #ddf ,abrolhos islands #dae ,absence of recent formations on the ,s4 ,american coasts #dji ,aguerros on elevation of ,imperial #cje ,albite1 constituent mineral in andesite #ddf ---- in rocks of ,tierra del ,fuego #dbg ---- in porphyries #ddd ---- crystals of1 with orthite #ddg ,alison1 ,mr41 on elevation of ,valparaiso #cjg1 #caj ,alumina1 sulphate of #dci ,ammonites from ,concepcion #djj1 #dje ,amolanas1 ,las #dic ,amygdaloid1 curious varieties of #ddd ,amygdaloids of the ,uspallata range #dga ---- of ,copiapo #dih ,andesite of ,chile #ddf ---- in the valley of ,maypu #ddi1 #dej ---- of the ,cumbre pass #dfj1 #dff ---- of the ,uspallata range #dge ---- of ,los ,hornos #dhj ---- of ,copiapo #dhh1 #dia ,anhydrite1 concretions of #dej1 #dfc ,araucaria1 silicified wood of #cid1 #dgd ,arica1 elevation of #cbc ,arqueros1 mines of #dha ,ascension1 gypsum deposited on #cbh ---- laminated volcanic rocks of #dci1 #ddj ,augite in fragments1 in gneiss #dad ---- with albite1 in lava #cdg ,austin1 ,mr4 ,r4 ,a4 ,c41 on bent cleavage lamina #dcd ,austin1 ,captain1 on sea-bottom #cjb ,australia1 foliated rocks of #dch ,azara labiata1 beds of1 at ,san ,pedro #bgg1 #ceb ,baculites vagina #djj ,bahia ,blanca1 elevation of #bhj ---- formations near #cee ---- character of living shells of #djh ,bahia 7,brazil71 elevation near #bhj ---- crystalline rocks of #dad ,ballard1 ,m41 on the precipitation of sulphate of soda #cdi ,banda ,oriental1 tertiary formations of #cfe ---- crystalline rocks of #dah ,barnacles above sea-level #caa ---- adhering to upraised shells #cjf ,basalt of ,s4 ,cruz #chi ---- streams of1 in the ,portillo range #def ---- in the ,uspallata range #dgb ,basin chains of ,chile #ccc ,beagle ,channel #dbg1 #dcj ,beaumont1 ,elie de1 on inclination of lava-streams #cij1 #deg ---- on viscid quartz-rocks #dge ,beech-tree1 leaves of fossil #cia ,beechey1 ,captain1 on sea-bottom #bii ,belcher1 ,lieutenant1 on elevated shells from ,concepcion #cjf ,bella ,vista1 plain of #cbe ,benza1 ,dr41 on decomposed granite #dag ,bettington1 ,mr41 on quadrupeds transported by rivers #cgd ,blake1 ,mr41 on the decay of elevated shells near ,iquique #cbb ---- on nitrate of soda #cdf ,bole #ddd ,bollaert1 ,mr41 on mines of ,iquique #ejc ,bones1 silicified #djb ---- fossil1 fresh condition of #cff ,bottom of sea off ,patagonia #bib1 #bih ,bougainville1 on elevation of the ,falkland islands #bij ,boulder formation of ,s4 ,cruz #bhe1 #bie ---- of ,falkland islands #bij ---- anterior to certain extinct quadrupeds #cga ---- of ,tierra del ,fuego #cia ,boulders in the ,cordillera #cci1 #cda ---- transported by earthquake-waves #cdd ---- in fine-grained tertiary deposits #dja ,brande1 ,mr41 on a mineral spring #dfa ,bravais1 ,m41 on elevation of ,scandinavia #cbj ,brazil1 elevation of #bgi ---- crystalline rocks of #dad1 #dah ,broderip1 ,mr41 on elevated shells from ,concepcion #cjf ,brown1 ,mr4 ,r41 on silicified wood of ,uspallata range #dgd ,brown1 on silicified wood #die ,bucalema1 elevated shells near #cjg ,buch1 ,von1 on cleavage #dch ---- on cretaceous fossils of the ,cordillera #dec1 #dfe ---- on the sulphureous volcanoes of ,java #eji ,buenos ,ayres #ceb ,burchell1 ,mr41 on elevated shells of ,brazil #bgi ,byron1 on elevated shells #cjc ,cachapual1 boulders in valley of #cci1 #cda ,caldcleugh1 ,mr41 on elevation of ,coquimbo #cad ---- on rocks of the ,portillo range #def ,callao1 elevation near #cbc ---- old town of #cbg ,cape of ,good ,hope1 metamorphic rocks of #dci ,carcharias megalodon #djb ,carpenter1 ,dr41 on microscopic organisms #ceb ,castro 7,chiloe71 beds near #cid ,cauquenes ,baths1 boulders near #cci1 #cda ---- pebbles in porphyry near #ddc ---- volcanic formation near #ddg ---- stratification near #ddi ,caves above sea-level #cjc1 #cjg1 #cbb ,cervus pumilus1 fossil-horns of #cjd ,chevalier1 ,m41 on elevation near ,lima #cbc ,chile1 structure of country between the ,cordillera and the ,pacific #ccc ---- tertiary formations of #ccg ---- crystalline rocks in #dce ---- central1 geology of #dda ---- northern1 geology of #dgi ,chiloe1 gravel on coast #bid ---- elevation of #cjc ---- tertiary formation of #ccg1 #dje ---- crystalline rocks of #dcc ,chlorite-schist1 near ,m4 ,video #dai ,chonos archipelago1 tertiary formations of #cic ---- crystalline rocks of #dcj ,chupat1 ,rio1 scori transported by #bhj ,claro1 ,rio1 fossiliferous beds of #dhe ,clay-shale of ,los ,hornos #dhj ,clay-slate1 formation of1 ,tierra del ,fuego #dbd ---- of ,concepcion #dcc ---- feldspathic1 of ,chile #ddb1 #ddd1 #ddh ---- ---- of the ,uspallata range #dfh1 #dgj ---- black siliceous1 band of1 in porphyritic formations of ,chile #dde ,claystone porphyry1 formation of1 in ,chile #ddb ---- origin of #dde ---- eruptive sources of #ddd ,cleavage1 definition of #dad ---- at ,bahia #dae ---- ,rio de ,janeiro #dae ---- ,maldonado #dah ---- ,monte ,video #dbj ---- ,s4 ,guitru-gueyu #dba ---- ,falkland ,i4 #dbd ---- ,tierra del ,fuego #dbh ---- ,chonos ,i4 #dcd ---- ,chiloe #dce ---- ,concepcion #dcd ---- ,chile #dce ---- discussion on #dcf ,cleavage-lamin superficially bent #dcd ,cliffs1 formation of #cja ,climate1 late changes in #cde ---- of ,chile during tertiary period #djh ,coal of ,concepcion #cii ---- ,s4 ,lorenzo #ejd ,coast-denudation of ,st4 ,helena #cja ,cobija1 elevation of #cbb ,colombia1 cretaceous formation of #ejd ,colonia del ,sacramiento1 elevation of #bgh ---- ,pampean formation near #cee ,colorado1 ,rio1 gravel of #bie ---- sand-dunes of #bha1 #bid ---- ,pampean formation near #cee ,combarbala #dgi1 #dha ,concepcion1 elevation of #cje ---- deposits of #cii1 #dje ---- crystalline rocks of #dcc ,conchalee1 gravel-terraces of #caa ,concretions of gypsum1 at ,iquique #cde ---- in sandstone at ,s4 ,cruz #chg ---- in tufaceous tuff of ,chiloe #chg ---- in gneiss #dad ---- in claystone-porphyry at ,port ,desire #dba ---- in gneiss at ,valparaiso #dce ---- in metamorphic rocks #dcf ---- of anhydrite #dej ---- relations of1 to veins #dgc ,conglomerate claystone of ,chile #ddc1 #dde ---- of ,tenuyan #ded1 #deh1 #dgh ---- of the ,cumbre ,pass #dfb1 #dff ---- of ,rio ,claro #dhe ---- of ,copiapo #dif1 #dii ,cook1 ,captain1 on form of sea-bottom #cjj ,copiapo1 elevation of #cba ---- tertiary formations of #djc ---- secondary formations of #dhi ,copper1 sulphate of #dhi ---- native1 at ,arqueros #dhb ---- mines of1 at ,panuncillo #dha ---- veins1 distribution of #eje ,coquimbo1 elevation and terraces of #cab ---- tertiary formations of #djd ---- secondary formations of #dhb ,corallines living on pebbles #bii ,cordillera1 valleys bordered by gravel fringes #ccg ---- basal strata of #ddb ---- fossils of #dec1 #dfe1 #dhf1 #dhg1 #dic1 #ejc ---- elevation of #ddb1 #dei1 #dgd1 #dgf1 #ejj1 #ejb1 #eaj1 #eab1 #eag ---- gypseous formations of #dej1 #deb1 #dfa1 #dfc1 #dgi1 #dhc1 #dhi1 #dia1 #ejc ---- claystone-porphyries of #ddb ---- andesitic rocks of #ddf ---- volcanoes of #ddg1 #eaa1 #eag ,coste1 ,m41 on elevation of ,lemus #cjc ,coy inlet1 tertiary formation of #cij ,crassatella ,lyellii #cib ,cruickshanks1 ,mr41 on elevation near ,lima #cbg ,crystals of feldspar1 gradual formation of1 at ,port ,desire #dbb ,cumbre1 ,pass of1 in ,cordillera #ejb ,cuming1 ,mr41 on habits of the ,mesodesma #caj ---- on range of living shells on west coast #djg ,dana1 ,mr41 on foliated rocks #dch ---- on amygdaloids #ddd ,darwin1 ,mount #dbg ,d',aubuisson1 on concretions #cig ---- on foliated rocks #dch ,decay1 gradual1 of upraised shells #cbc1 #cbg ,decomposition of granite rocks #dag ,de la ,beche1 ,sir ,h41 his theoretical researches in geology #bii ---- on the action of salt on calcareous rocks #cbg ---- on bent cleavage-lamin #dcd ,denudation on coast of ,patagonia #bib1 #cjj1 #dji ---- great powers of #daj ---- of the ,portillo range #def1 #deh ,deposits1 saline #cdd ,despoblado1 valley of #dif1 #dig1 #dii ,detritus1 nature of1 in ,cordillera #cch ,devonshire1 bent cleavage in #dcd ,dikes1 in gneiss of ,brazil #dad1 #dah ---- near ,rio de ,janeiro #dag ---- pseudo1 at ,port ,desire #dbc ---- in ,tierra del ,fuego #dbf ---- in ,chonos archipelago1 containing quartz #dcb ---- near ,concepcion1 with quartz #dcd ---- granitic-porphyritic1 at ,valparaiso #dce ---- rarely vesicular in ,cordillera #cdg ---- absent in the central ridges of the ,portillo pass #deb ---- of the ,portillo range1 with grains of quartz #def ---- intersecting each other often #dff ---- numerous at ,copiapo #dih ,domeyko1 ,m41 on the silver mines of ,coquimbo #dhb ---- on the fossils of ,coquimbo #dhf ,d',orbigny1 ,m4 ,a41 on upraised shells of ,monte ,video #bgh ---- on elevated shells at ,st4 ,pedro #bgh ---- on elevated shells near ,b4 ,ayres #bgi ---- on elevation of ,s4 ,blas #bha ---- on the sudden elevation of ,la ,plata #bic ---- on elevated shells near ,cobija #cbb ---- on elevated shells near ,arica #cbb ---- on the climate of ,peru #cbd ---- on salt deposits of ,cobija #cde ---- on crystals of gypsum in salt-lakes #cdi ---- on absence of gypsum in the ,pampean formation #cec ---- on fossil remains from ,bahia ,blanca #cei1 #cfj ---- on fossil remains from the banks of the ,parana #cfb ---- on the geology of ,st4 ,f #cfc ---- on the age of ,pampean formation #cfg1 #cgf ---- on the ,mastodon ,andium #cgi ---- on the geology of the ,rio ,negro #cha ---- on the character of the ,patagonian fossils #cia ---- on fossils from ,concepcion #cii ---- ---- from ,coquimbo #djd ---- ---- from ,payta #dje ---- on fossil tertiary shells of ,chile #djf ---- on cretaceous fossils of ,tierra del ,fuego #dbf ---- ---- from the ,cordillera of ,chile #dec1 #dfe1 #dhf1 #dhh1 #dic1 #ejd ,earth1 marine origin of #cjd1 #cjh ,earthenware1 fossil #cbf ,earthquake1 effect of1 at ,s4 ,maria #bic ---- elevation during1 at ,lemus #cjc ---- of #ahbb1 at ,valparaiso #caj ---- effects of1 in shattering surface #cbe ---- fissures made by #cbe ---- probable effects on cleavage #cbe ,earthquakes in ,pampas #bij ,earthquake-waves1 power of1 in throwing up shells #caj ---- effects of1 near ,lima #cbg ---- power of1 in transporting boulders #cdd ,edmonston1 ,mr41 on depths at which shells live at ,valparaiso #cji ,ehrenberg1 ,professor1 on infusoria in the ,pampean formation #cee1 #cei1 #cfb ---- on infusoria in the ,patagonian formation #chc1 #chd1 #chf1 #cia1 #cib ,elevation of ,la ,plata #bgh ---- ,brazil #bgi ---- ,bahia ,blanca #bhj1 #ceg ---- ,san ,blas #bha ---- ,patagonia #bha1 #bia1 #bic ---- ,tierra del ,fuego #bhh ---- ,falkland islands #bij ---- ,pampas #bhi1 #cgg ---- ,chonos archipelago #cjc ---- ,chiloe #cjd ---- ,chile #cjd ---- ,valparaiso #cjg1 #caj ---- ,coquimbo #cab1 #cbj ---- ,guasco #cbj ---- ,iquique #cbb ---- ,cobija #cbb ---- ,lima #cbc ---- sudden1 at ,s4 ,maria #bic ---- ---- at ,lemus #cjc ---- insensible1 at ,chiloe #cjd ---- ---- at ,valparaiso #caa ---- ---- at ,coquimbo #cad ---- axes of1 at ,chiloe #cih1 #dje ---- ---- at ,p4 ,rumena #cih1 #dje ---- ---- at ,concepcion #cih1 #dje ---- unfavourable for the accumulation of permanent deposits #daj ---- lines of1 parallel to cleavage and foliation #daf1 #dag1 #dbd1 #dbh1 #dcb1 #dcd1 #dch ---- lines of1 oblique to foliation #dca ---- areas of1 causing lines of elevation and cleavage #dda ---- lines of1 in the ,cordillera #ddb ---- slow1 in the ,portillo range #dge ---- two periods of1 in ,cordillera of ,central ,chile #dgf ---- of the ,uspallata range #dgd ---- two periods of1 in ,cumbre ,pass #dgf ---- horizontal1 in the ,cordillera of ,copiapo #ejj ---- axes of1 coincident with volcanic orifices #ejc ---- of the ,cordillera1 summary on #eaj1 #eac1 #eag ,elliott1 ,captain1 on human remains #bgi ,ensenada1 elevated shells of #bgh ,entre ,rios1 geology of #cfc ,equus curvidens #cfd1 #cgi ,epidote in ,tierra del ,fuego #dbf ---- in gneiss #dce ---- frequent in ,chile #dde ---- in the ,uspallata range #dge ---- in porphyry of ,coquimbo #dhb ,erman1 ,m41 on andesite #cdg ,escarpments1 recent1 of ,patagonia #cja ,extinction of fossil mammifers #cgj ,falkland islands1 elevation of #bij ---- pebbles on coast #big1 #bii ---- geology of #dbd ,falkner1 on saline incrustations #cdg ,faults1 great1 in ,cordillera #dfa1 #dfi ,feldspar1 earthy1 metamorphosis of1 at ,port ,desire #dbb ---- albitic #cdg ---- crystals of1 with albite #cdg ---- orthitic1 in conglomerate of ,tenuyan #ded ---- in granite of ,portillo range #dee ---- in porphyries in the ,cumbre ,pass #dff ,feuille on sea-level at ,coquimbo #cad ,fissures1 relations of1 to concretions #cig ---- upfilled1 at ,port ,desire #dbd ---- in clay-slate #dgj ,fitton1 ,dr41 on the geology of ,tierra del ,fuego #dbg ,fitzroy1 ,captain1 on the elevation of the ,falkland islands #dbg ---- on the elevation of ,concepcion #cje ,foliation1 definition of #dad ---- of rocks at ,bahia #dad ---- ,rio de ,janeiro #dae ---- ,maldonado #dah ---- ,monte ,video #dbj ---- ,s4 ,guitru-gueyu #dba ---- ,falkland ,i4 #dbd ---- ,tierra del ,fuego #dbg ---- ,chonos archipelago #dcj ---- ,chiloe #dcc ---- ,concepcion #dcd ---- ,chile #dce ---- discussion on #dce ,forbes1 ,professor ,e41 on cretaceous fossils of ,concepcion #djj ---- on cretaceous fossils and subsidence in ,cumbre ,pass #dfe ---- on fossils from ,guasco #dhh ---- ---- from ,coquimbo #dhc1 #dhg ---- ---- from ,copiapo #dic ---- on depths at which shells live #dji1 #dif ,formation1 ,pampean #ceb ---- ---- area of #cga ---- ---- estuary origin #cgc ---- tertiary of ,entre ,rios #cfc ---- of ,banda ,oriental #cfe ---- volcanic1 in ,banda ,oriental #cfg ---- of ,patagonia #cha ---- summary on #cia ---- tertiary of ,tierra del ,fuego #cia ---- ---- of the ,chonos archipelago #cic ---- ---- of ,chiloe #cid ---- ---- of ,chile #cid ---- ---- of ,concepcion #cih1 #djd ---- ---- of ,navidad #djj ---- ---- of ,coquimbo #djb ---- ---- of ,peru #djd ---- ---- subsidence during #djb ---- volcanic1 of ,tres ,montes #cic ---- ---- of ,chiloe #cid ---- ---- old1 near ,maldonado #dah ---- ---- with laminar structure #ddj ---- ---- ancient1 in ,tierra del ,fuego #dbf ---- recent1 absent on ,s4 ,american coast #dji ---- metamorphic1 of claystone-porphyry of ,patagonia #dba1 #ddj ---- foliation of #dcf ---- plutonic1 with laminar structure #ddj ---- palaeozoic1 of the ,falkland ,i4 #dbd ---- claystone1 at ,concepcion #dcc ---- ,jurassic1 of ,cordillera #eab ---- ,neocomian1 of the ,portillo ,pass #dec ---- volcanic1 of ,cumbre ,pass #dfe ---- gypseous1 of ,los ,hornos #dgi1 #dhg ---- ---- of ,coquimbo #dhb ---- ---- of ,guasco #dhg ---- ---- of ,copiapo #dhh ---- ---- of ,iquique #ejc ---- cretaceo-oolitic1 of ,coquimbo #dhf1 #die ---- ---- of ,guasco #dhg1 #did ---- ---- of ,copiapo #die ---- ---- of ,iquique #ejd ,fossils1 ,neocomian1 of ,portillo ,pass #dec ---- ---- of ,cumbre ,pass #dfe ---- secondary1 of ,coquimbo #dhe ---- ---- of ,guasco #dhg ---- ---- of ,copiapo #did ---- ---- of ,iquique #ejc ---- palozoic1 from the ,falklands #dbd ,fragments of hornblende-rock in gneiss #dad ---- of gneiss in gneiss #daf ,freyer1 ,lieutenant1 on elevated shells of ,arica #cbc ,frezier on sea-level at ,coquimbo #cad ,galapagos archipelago1 pseudo-dikes of #dbd ,gallegos1 ,port1 tertiary formation of #cij ,garnets in gneiss #dae ---- in mica-slate #dbg ---- at ,panuncillo #dha ,gardichaud1 ,m41 on granites of ,brazil #dag ,gay1 ,m41 on elevated shells #cjf ---- on boulders in the ,cordillera #cci1 #cda ---- on fossils from ,cordillera of ,coquimbo #dhg ,gill1 ,mr41 on brickwork transported by an earthquake-wave #cbg ,gillies1 ,dr41 on heights in the ,cordillera #ddh ---- on extension of the ,portillo range #deh ,glen ,roy1 parallel roads of #cai ---- sloping terraces of #cdj ,gneiss1 near ,bahia #dad ---- of ,rio de ,janeiro #dae ---- decomposition of #dag ,gold1 distribution of #ejf ,gorodona1 formations near #cfb ,granite1 axis of oblique1 to foliation #dca ---- andesitic #ddf ---- of ,portillo range #dee ---- veins of1 quartzose #dcb1 #dge ---- pebble of1 in porphyritic conglomerate #dic ---- conglomerate #dig ,grauwacke of ,uspallata range #dfh ,gravel at bottom of sea #bic1 #bih ---- formation of1 in ,patagonia #bie ---- means of transportation of #bih ---- strata of1 inclined #dfg ,gravel-terraces in ,cordillera #ccg ,greenough1 ,mr41 on quartz veins #dcg ,greenstone1 resulting from metamorphose hornblende-rock #dai ---- of ,tierra del ,fuego #dbf ---- on the summit of the ,campana of ,quillota #ddb ---- porphyry #ddc ---- relation of1 to clay-slate #ddc ,grypha orientalis #dhc ,guasco1 elevation of #cba ---- secondary formation of #dhg ,guitru-gueyu1 ,sierra #dba ,guyana1 gneissic rocks of #dae ,gypsum1 nodules of1 in gravel at ,rio ,negro #bif ---- deposited from sea-water #cbg ---- deposits of1 at ,iquique #cde ---- crystals of1 in salt lakes #cdf ---- in ,pampean formation #cec ---- in tertiary formation of ,patagonia #chb1 1 1 ---- great formation of1 in the ,portillo ,pass #dfa1 #dfc ---- ---- in the ,cumbre ,pass #dfa1 #dfc ---- ---- near ,los ,hornos #dgi ---- ---- at ,coquimbo #dhb ---- ---- at ,copiapo #dij1 #dib ---- ---- near ,iquique #ejd ---- of ,san ,lorenzo #ejd ,hall1 ,captain1 on terraces at ,coquimbo #caf ,hamilton1 ,mr41 on elevation near ,tacna #cbc ,harlan1 ,dr41 on human remains #bgi ,hayes1 ,mr4 ,a41 on nitrate of soda #cdf ,henslow1 ,professor1 on concretions #dcg ,herbert1 ,captain1 on valleys in the ,himalaya #cce ,herradura ,bay1 elevated shells of #cae ---- tertiary formations of #djb ,himalaya1 valleys in #cce ,hippurites ,chilensis #dhc1 #dhf ,hitchcock1 ,professor1 on dikes #dad ,honestones1 pseudo1 of ,coquimbo #dhc ---- of ,copiapo #dhi ,hooker1 ,dr4 ,j4 ,d41 on fossil beech-leaves #cia ,hopkins1 ,mr41 on axes of elevation oblique to foliation #dcb ---- on origin of lines of elevation #ddj1 #eab ,hornblende-rock1 fragments of1 in gneiss #dad ,hornblende-schist1 near ,m4 ,video #dbj ,hornos1 ,los1 section near #dgi ,hornstone1 dike of #dcc1 #dcd ,horse1 fossil tooth of #ceh1 #cfd ,huafo island #cic1 #djd ---- subsidence at #daa ,huantajaya1 mines of #ejc ,humboldt1 on saline incrustations #cdg ---- on foliations of gneiss #dae ---- on concretions in gneiss #dce ,icebergs1 action on cleavage #dcd1 #dcf ,illapele1 section near #dgi ,imperial1 beds of shells near #cje ,incrustations1 saline #cdg ,infusoria in ,pampean formation #ceb1 #cee1 #cfj1 #cfc ---- in ,patagonian formation #chb1 #chc1 #chd1 #cia ,iodine1 salts of #cdg1 #cdh ,iquique1 elevation of #cbb ---- saliferous deposits of #cdd ---- cretaceo-oolitic formation of #ejc ,iron1 oxide of1 in lavas #dfc1 #dii ---- in sedimentary beds #dhj1 #dhb ---- tendency in1 to produce hollow concretions #cih ---- sulphate of #dhi ,isabelle1 ,m41 on volcanic rocks of ,banda ,oriental #cfh ,joints in clay-slate #dbh ,jukes1 ,mr41 on cleavage in ,newfoundland #dcg ,kamtschatka1 andesite of #cdg ,kane1 ,dr41 on the production of carbonate of soda #cbh ,king ,george's sound1 calcareous beds of #cab ,lakes1 origin of #cjj ---- fresh-water1 near salt lakes #cej ,lava1 basaltic1 of ,s4 ,cruz #chi ---- claystone-porphyry1 at ,chiloe #cie ---- ---- ancient submarine #ddf ---- basaltic1 of the ,portillo range #deg ---- feldspathic1 of the ,cumbre ,pass #dfc ---- submarine1 of the ,uspallata range #dga1 #dgc1 #dgf ---- basaltic1 of the ,uspallata range #dge ---- submarine1 of ,coquimbo #dhd1 #dhf ---- of ,copiapo #dij1 #dif1 #dii ,lemus island #cic1 #djd ,lemuy islet #cid ,lignite of ,chiloe #cie ---- of ,concepcion #cih ,lima1 elevation of #cbc ,lime1 muriate of #cbh1 #cdd1 #cdg ,limestone of ,cumbre ,pass #dfb ---- of ,coquimbo #dhc1 #dhe ---- of ,copiapo #dic ,lund and ,clausen on remains of caves in ,brazil #cgh1 #chj ,lund1 ,m41 on granites of ,brazil #dag ,lyell1 ,m41 on upraised shells retaining their colours #bhi ---- on terraces at ,coquimbo #cae ---- on elevation near ,lima #cbg ---- on fossil horse's tooth #cfd ---- on the boulder-formation being anterior to the extinction of ,north ,american mammifers #cga ---- on quadrupeds washed down by floods #cgd ---- on age of ,american fossil mammifers #cgi ---- on changes of climate #dji ---- on denudation #daj ---- on foliation #dch ,mac,culloch1 ,dr41 on concretions #dcg ---- on beds of marble #ddj ,maclaren1 ,mr41 letter to1 on coral-formations #dac ,macrauchenia ,patachonica #ceh1 #cgj ,madeira1 subsidence of #cjb ,magellan1 ,strait1 elevation near1 of #bhh ,magnesia1 sulphate of1 in veins #chg ,malcolmson1 ,dr41 on trees carried out to sea #dge ,maldonado1 elevation of #bgg ---- ,pampean formation of #cfe ---- crystalline rocks of #dah ,mammalia1 fossil1 of ,bahia ,blanca #cef1 #cfd ---- ---- near ,st4 ,f #cfc ---- ---- of ,banda ,oriental #cff ---- ---- of ,st4 ,julian #cfi ---- ---- at ,port ,gallegos #cia ---- washed down by floods #cgc ---- number of remains of1 and range of1 in ,pampas #cgf ,man1 skeletons of 7,brazil7 #bgi ---- remains of1 near ,lima #cbe ---- ,indian1 antiquity of #cbe ,marble1 beds of #dah ,maricongo1 ravine of #ejj ,marsden1 on elevation of ,sumatra #cje ,mastodon ,andium1 remains of #cfb ---- range of #cgh ,maypu1 ,rio1 mouth of1 with upraised shells #cjg ---- gravel fringes of #cci ---- debouchement from the ,cordillera #ddi ,megalonyx1 range of #cgi ,megatherium1 range of #cgi ,miers1 ,mr41 on elevated shells #caa ---- on the height of the ,uspallata plain #cce ,minas1 ,las #dah ,mocha ,island1 elevation of #cje ---- tertiary form of #cih ---- subsidence at #daa ,molina1 on a great flood #cda ,monte ,hermoso1 elevation of #bhj ---- fossils of #cee ,monte ,video1 elevation of #bgh ---- ,pampean formation of #cfe ---- crystalline rocks of #dai ,morris and ,sharpe1 ,messrs41 on the palozoic fossils of the ,falklands #dbd ,mud1 ,pampean #ceb ---- long deposited on the same area #cgf ,murchison1 ,sir ,r41 on cleavage #dcf ---- on waves transporting gravel #bii ---- on origin of salt formations #eje ---- on the relations of metalliferous veins and intrusive rocks #ejg ---- on the absence of granite in the ,ural #eab ,nautilus d',orbignyanus #djj1 #dje ,navidad1 tertiary formations of1 subsidence of #djj1 #daa ,negro1 ,rio1 pumice of pebbles of #bha ---- gravel of #bie ---- salt lakes of #bie ---- tertiary strata of #chd ,north ,america1 fossil remains of #cgi ,north ,wales1 sloping terraces absent in #cdj ---- bent cleavage of #dcd ,neuvo ,gulf1 plains of #bhb ---- tertiary formation of #chd ,owen1 ,professor1 on fossil mammiferous remains #cef1 #ceh1 #cfd1 #cff1 #cgj ,palmer1 ,mr41 on transportation of gravel #cjj ,pampas1 elevation of #bij ---- earthquakes of #bij ---- formation of #bie1 #cej ---- localities in which fossil mammifers have been found #chj ,panuncillo1 mines of #dha ,parana1 ,rio1 on saline incrustations #cdg ---- ,pampean formations near #cfa ---- on the ,s4 ,tandil #dbj ,parish1 ,sir ,w41 on elevated shells near ,buenos ,ayres #bgh1 #bgi ---- on earthquakes in the ,pampas #bij ---- on fresh-water near salt lakes #cej ---- on origin of ,pampean formation #cgc ,patagonia1 elevation and plains of #bha ---- denudation of #bia ---- gravel-formation of #bie ---- sea-cliffs of #cja ---- subsidence during tertiary period #daa ---- crystalline rocks of #dba ,payta1 tertiary formations of #djd ,pebbles of pumice #bhj ---- decrease in size on the coast of ,patagonia #bic ---- means of transportation #bih ---- encrusted with living corallines #bii ---- distribution of1 at the eastern foot of ,cordillera #ccg ---- dispersal of1 in the ,pampas #ced ---- zoned with colour #ddc ,pentland1 ,mr41 on heights in the ,cordillera #dfj ---- on fossils of the ,cordillera #dfe ,pernambuco #bgi ,peru1 tertiary formations of #djc ,peuquenes1 ,pass of1 in the ,cordillera #ddh ---- ridge of #deb ,pholas1 elevated shells of #cjc ,pitchstone of ,chiloe #cie ---- of ,port ,desire #dba ---- near ,cauquenes #ddh ---- layers of1 in the ,uspallata range #dgb ---- of ,los ,hornos #dhj ---- of ,coquimbo #dhc ,plains of ,patagonia #bhb1 #bia ---- of ,chiloe #cjd ---- of ,chile #ccc ---- of ,uspallata #cce ---- on eastern foot of ,cordillera #ccf ---- of ,iquique #cdf ,plata1 ,la1 elevation of #bgg ---- tertiary formation of #bie1 #cec ---- crystalline rocks of #dah ,playfair1 ,professor1 on the transportation of gravel #cjj ,pluclaro1 axis of #dhc ,pondicherry1 fossils of #djj ,porcelain rocks of ,port ,desire #dbb ---- of the ,uspallata range #dga1 #dgc1 #dgf ,porphyry1 pebbles of1 strewed over ,patagonia #bif ,porphyry1 claystone1 of ,chiloe #cie ---- ---- of ,patagonia #dba ---- ---- of ,chile #ddb1 #dde ---- greenstone1 of ,chile #ddd ---- doubly columnar #ddh ---- claystone1 rare1 on the eastern side of the ,portillo ,pass #ded ---- brick-red and orthitic1 of ,cumbre ,pass #deh1 #dfg ---- intrusive1 repeatedly injected #dfg ---- claystone of the ,uspallata range #dfh ---- ---- of ,copiapo #dhi1 #dii ---- ---- eruptive sources of #ejb ,port ,desire1 elevation and plains of #bhc ---- tertiary formation of #chc ---- porphyries of #dba ,portillo ,pass in the ,cordillera #ddh ,portillo chain #ded1 #deh ---- compared with that of the ,uspallata #dgh ,prefil or sea-wall of ,valparaiso #caj ,puente del ,inca1 section of #dfa ,pumice1 pebbles of #bcj ---- conglomerate of ,r4 ,negro #chb ---- hills of1 in the ,cordillera #cdg ,punta ,alta1 elevation of #bhj ---- beds of #cef ,quartz-rock of the ,s4 ,ventana #dba ---- ,c4 ,blanco #dba ---- ,falkland islands #dbd ---- ,portillo range #dee ---- viscidity of #dge ---- veins of1 near ,monte ,video #dbj ---- ---- in dike of greenstone #dbf ---- grains of1 in mica slate #dcj ---- ---- in dikes #dcb1 #dcd ---- veins of1 relations to cleavage #dcg ,quillota1 ,campana of #ddb ,quintero1 elevation of #caa ,quiriquina1 elevation of #cjf ---- deposits of #cii ,rancagua1 plain of #ccd ,rapel1 ,r41 elevation near #cjg ,reeks1 ,mr4 ,t41 his analysis of decomposed shells #cbh ---- his analysis of salts #cdd ,remains1 human #cbd ,rio de ,janeiro1 elevation near #bgi ---- crystalline rocks of #dae ,rivers1 small power of transporting pebbles #bih ---- small power of1 in forming valleys #cdc ---- drainage of1 in the ,cordillera #ddi1 #eac ,roads1 parallel1 of ,glen ,roy #cai ,rocks1 volcanic1 of ,banda ,oriental #cfg ---- ,tres ,montes #cic ---- ,chiloe #cid ---- ,tierra del ,fuego #dbf ---- with laminar structure #ddj ,rodents1 fossil1 remains of #cef ,rogers1 ,professor1 address to ,association of ,american ,geologists #dab ,rose1 ,professor ,g41 on sulphate of iron at ,copiapo #dhi ,s4 ,blas1 elevation of #bha ,s4 ,cruz1 elevation and plains of #bhd ---- valley of #bhe ---- nature of gravel in valley of #bif ---- boulder formation of #cga ---- tertiary formation of #chf ---- subsidence at #dab ,s4 ,f ,bajada1 formations of #cfc ,s4 ,george's bay1 plains of #bhb ,s4 ,helena island1 sea-cliffs1 and subsidence of #cja ,s4 ,josef1 elevation of #bha ---- tertiary formation of #chc ,s4 ,juan1 elevation near #bgh ,s4 ,julian1 elevation and plains of #bhd ---- salt lake of #cdh ---- earthy deposit with mammiferous remains #cfi ---- tertiary formations of #chd ---- subsidence at #daa ,s4 ,lorenzo1 elevation of #cbc ---- old salt formation of #ejd ,s4 ,mary1 island of1 elevation of #cje ,s4 ,pedro1 elevation of #bgh ,salado1 ,r41 elevated shells of #bgi ---- ,pampean formation of #cec ,salines #cdh ,salt1 with upraised shell #cbd1 #cbg ---- lakes of #cdh ---- purity of1 in salt lakes #cdi ---- deliquescent1 necessary for the preservation of meat #cdi ---- ancient formation of1 at ,iquique #ejd ---- ---- at ,s4 ,lorenzo #ejd ---- strata of1 origin of #eje ,salts1 superficial deposits of #cdd ,sand-dunes of the ,uruguay #bgi ---- of the ,pampas #bha ---- near ,bahia ,blanca #bha1 #bic ---- of the ,colorado #bha1 #bid ---- of ,s4 ,cruz #bhf ---- of ,arica #cbc ,sarmiento1 ,mount #dbg ,schmidtmeyer on auriferous detritus #ejf ,schomburghk1 ,sir ,r41 on sea-bottom #bii ---- on the rocks of ,guyana #dae ,scotland1 sloping terraces of #cdj ,sea1 nature of bottom of1 off ,patagonia #bib ---- power of1 in forming valleys #cdc ,sea cliffs1 formation of #cja ,seale1 ,mr41 model of ,st4 ,helena #cja ,sebastian ,bay1 tertiary formation of #cia ,sedgwick1 ,professor1 on cleavage #ccf ,serpentine of ,copiapo #dhi ,serpul1 on upraised rocks #cbe ,shale-rock1 of the ,portillo ,pass #deb ---- of ,copiapo #dic ,shells1 upraised state of1 in ,patagonia #bhh ---- elevated1 too small for human food #cjh ---- transported far inland1 for food #cji ---- upraised1 proportional numbers varying #cab1 #cbd ---- ---- gradual decay of #cbc1 #cbd1 #cbg ---- ---- absent on high plains of ,chile #cce ---- ---- near ,bahia ,blanca #ceh ---- preserved in concretions #cid1 #cig ---- living and fossil range of1 on west coast #djf1 #djh ---- living1 different on the east and west coast #daa ,shingle of ,patagonia #bie ,siau1 ,m41 on sea-bottom #bii ,silver mines of ,arqueros #dca ---- of ,chanuncillo #did ---- of ,iquique #ejc ---- distribution of #ejf ,slip1 great1 at ,s4 ,cruz #chg ,smith1 ,mr41 of ,jordan ,hill1 on upraised shells retaining their colours #bhi ---- on ,madeira #cjb ---- on elevated seaweed #cbe ---- on inclined gravel beds #dfg ,soda1 nitrate of #cdf ---- sulphate of1 near ,bahia ,blanca #cdh1 #cdi ---- carbonate of #cdg ,soundings off ,patagonia #bic1 #bii ---- in ,tierra del ,fuego #cjj ,spirifers #dhf1 #dhh ,spix and ,martius on ,brazil #dag ,sprengel on the production of carbonate of soda #cbh ,springs1 mineral1 in the ,cumbre ,pass #dfa ,stratification of sandstone in metamorphic rocks #dad ---- of clay-slate in ,tierra del ,fuego #dbh ---- of the ,cordillera of ,central ,chile #ddb1 #ddh1 #dfa ---- little disturbed in ,cumbre ,pass #dfj1 #dff ---- disturbance of1 near ,copiapo #eja ,streams of lava at ,s4 ,cruz1 inclination of #cij ---- in the ,portillo range #deg ,string of cotton with fossil-shells #cbe ,struthiolaria ornata #cib ,studer1 ,m41 on metamorphic rocks #dch ,subsidence during formation of sea-cliffs #cja ---- near ,lima #cbg ---- probable1 during ,pampean formation #cgf ---- necessary for the accumulation of permanent deposits #daa ---- during the tertiary formations of ,chile and ,patagonia #dac ---- probable during the ,neocomian formation of the ,portillo ,pass #dec ---- probable during the formation of conglomerate of ,tenuyan #dei ---- during the ,neocomian formation of the ,cumbre ,pass #dfe ---- of the ,uspallata range #dgd1 #dgg ---- great1 at ,copiapo #dif ---- ---- during the formation of the ,cordillera #eaj ,sulphur1 volcanic exhalations of #eji ,sumatra1 promontories of #cje ,summary on the recent elevatory movements #bei1 #cbi1 #ead ---- on the ,pampean formation #cga1 #eae ---- on the tertiary formations of ,patagonia and ,chile #cia1 #djd1 #eac ---- on the ,chilean ,cordillera #ejh ---- on the cretaceo-oolitic formation #ejh ---- on the subsidences of the ,cordillera #eji ---- on the elevation of the ,cordillera #eaa1 #eag ,tacna1 elevation of #cbc ,tampico1 elevated shells near #cbi ,tandil1 crystalline rocks of #dbj ,tapalguen1 ,pampean formation of #cec ---- crystalline rocks of #dbj ,taylor1 ,mr41 on copper veins of ,cuba #ejf ,temperature of ,chile during the tertiary period #djh ,tension1 lines of1 origin of1 axes of elevation and of cleavage #ddj ,tenuy ,point1 singular section of #cie ,tenuyan1 valley of #ded1 #dgh ,terraces of the valley of ,s4 ,cruz #bhf ---- of equable heights throughout ,patagonia #bij ---- of ,patagonia1 formation of #bid ---- of ,chiloe #cjd ---- at ,conchalee #caa ---- of ,coquimbo #caf ---- not horizontal at ,coquimbo #cag ---- of ,guasco #cbj ---- of ,s4 ,lorenzo #cbc ---- of gravel within the ,cordillera #ccg ,theories on the origin of the ,pampean formation #cgb ,tierra ,amarilla #dhi ,tierra del ,fuego1 form of sea-bottom #cjj ---- tertiary formations of #cia ---- clay-slate formation of #dbd ---- cretaceous formation of #dbf ---- crystalline rocks of #dbf ---- cleavage of clay-slate #dbg1 #dcf ,tosca rock #ceb ,trachyte of ,chiloe #cid ---- of ,port ,desire #dba ---- in the ,cordillera #cdg ,traditions of promontories having been islands #cje ---- on changes of level near ,lima #cbg ,trees buried in plain of ,iquique #cdf ---- silicified1 vertical1 of the ,uspallata range #dgc ,tres ,montes1 elevation of #cjc ---- volcanic rocks of #cic ,trigonocelia insolita #cib ,tristan ,arroyo1 elevated shells of #bgh ,tschudi1 ,mr41 on subsidence near ,lima #cbg ,tuff1 calcareous1 at ,coquimbo #cac ---- on basin-plain near ,st4 ,jago #ccd ---- structure of1 in ,pampas #ceb ---- origin of1 in ,pampas #cgd ---- pumiceous1 of ,r4 ,negro #chb ---- ,nuevo ,gulf #chc ---- ,port ,desire #chc ---- ,s4 ,cruz #chf ---- ,patagonia1 summary on ,chiloe #cia ---- formation of1 in ,portillo chain #cie ---- great deposit of1 at ,copiapo #deg ,tuffs1 volcanic1 metamorphic1 of ,uspallata #dga ---- of ,coquimbo #dhd ,ulloa1 on rain in ,peru #cbd ---- on elevation near ,lima #cbg ,uruguay1 ,rio1 elevation of country near #bgh ,uspallata1 plain of #cce1 #eae ---- pass of #dei ---- range of #cfh ---- concluding remarks on #dgf ,valdivia1 tertiary beds of #cih ---- mica-slate of #dcc ,valley of ,s4 ,cruz1 structure of #bhe ---- ,coquimbo #cad ---- ,guasco1 structure of #cbj ---- ,copiapo1 structure of #cba ---- ,s4 ,cruz1 tertiary formations of #chf ---- ,coquimbo1 geology of #dhb ---- ,guasco1 secondary formations of #dhg ---- ,copiapo1 secondary formations of #dhh ---- ,despoblado #dif1 #dig1 #dii ,valleys in the ,cordillera bordered by gravel fringes #ccg ---- formation of #cch ---- in the ,cordillera #ddi ,valparaiso1 elevation of #cjg ---- gneiss of #dce ,vein of quartz near ,monte ,video #dai ---- in mica-slate #dcj ---- relations of1 to cleavage #dcg ---- in a trap dike #dbf ---- of granite1 quartzose #dcb1 #dge ---- remarkable1 in gneiss1 near ,valparaiso #dce ,veins1 relations of1 to concretions #cif ---- metalliferous1 of the ,uspallata range #dge ---- metalliferous1 discussion on #eje ,venezuela1 gneissic rocks of #dae ,ventana1 ,sierra1 ,pampean formation near #cec ---- quartz-rock of #dba ,villa ,vincencio ,pass #dfh ,volcan1 ,rio1 mouth of #ddi ---- fossils of #dec ,volcanoes of the ,cordillera #cib1 #ddg1 #eaa ---- absent1 except near bodies of water #deg ---- ancient submarine1 in ,cordillera #ejb ---- action of1 in relation to changes of level #ead ---- long action of1 in the ,cordillera #eag ,wafer on elevated shells #cbb ,waves caused by earthquakes1 power of1 in transporting boulders #cbf1 #cdd ---- power of1 in throwing up shells #cji ,weaver1 ,mr41 on elevated shells #cbi ,white1 ,martin1 on sea-bottom #bii ,wood1 silicified1 of ,entre ,rios #cfd ---- ,s4 ,cruz #chh ---- ,chiloe #cid1 #cif ---- ,uspallata range #dgc ---- ,los ,hornos #dgi ---- ,copiapo #die1 #dig ,yeso1 ,rio1 and plain of #dej ,ypun ,island1 tertiary formation of #cic ,zeagonite #dbf