,transcrib]'s note3 ,?is article has be5 extract$ & prep>$ f ,! ,geographical ,j\rnal1 v4 #gj1 #aibg4 ,,problems ,,( ,,! ,,libyan ,,des]t ,john ,ball1 ,o4,b4,e41 ,d4,sc41 ,m4,in/4,c4,e41 ,f4,g4,s41 ,director ( ,des]t ,surveys1 ,egypt ,map foll[+ p4 #if4 ,,!re c be few tracts ( ! ,e>?'s surface :i* pres5t su* a numb] & v>iety ( int]e/+ geographical problems z ! ,libyan ,des]t4 ,! foll[+ are "s ( ! m puzzl+ ( ! que/ions :i* h be5 ask$ at v>i\s times s9ce ,i made my fir/ acqua9tance ) ! des]t ?irty ye>s ago3 #a4 ,did ! ,nile1 or a bran* ( x1 ev] fl[ ?r\gh ! ,libyan ,des]t to ! we/ ( ! pres5t ,nile ,valley8 #b4 ,by :at natural ag5cy 7 ! great depressions 9 ! ,libyan ,des]t excavat$8 #c4 ,c o!r depressions 9 ?is region1 besides ! mu*-discuss$ ,wadi ,rayan1 be consid]$ possible ( utilization 9 connection ) irrigation1 flood-protection1 or dra9age ( ! ,nile ,delta1 or z s\rces ( wat]-p[] = ,egypt8 #d4 ,:5ce comes ! artesian wat] ( ! oases8 #e4 ,:y h n c]ta9 lakes1 su* z ,sittra & ,bahre91 situat$ z !y are 9 an almo/ ra9less region1 long s9ce dri$ up8 #f4 ,c ! pres5t scanty supplies ( dr9kable wat] d]iv$ f local ra9fall along ! ,egyptian ,m$it]ranean ,littoral be supplem5t$ by artesian bor+s8 #g4 ,are ! artesian wat]-supplies ( ! oases gradually dim9i%+8 #h4 ,:]e %all we look = ! my/]i\s 8,z]zura10 or 8,oasis ( ! ,blacks08 ,are !re any o!r 8lo/0 oases rema9+ to be discov]$8 #i4 ,c ! pres5t difficulties ( travel 9 ! ,libyan ,des]t1 :i* are *iefly due to sc>city ( wat]1 be alleviat$ by ! s9k+ ( new wells8 #aj4 ,:]e are ! 8,tortoise ,m>%es0 ( ,ptolemy8 #aa4 ,:at det]m9$ ! peculi> di/ribution ( ! s&-dunes1 "s ( :i* ext5d 9 /raight l9es = hundr$s ( kilometres8 #ab4 ,:at is ! full ext5t ( ! di/ribution ( fl9t implem5ts & pott]y1 ( :i* accumulations h be5 f\nd 9 :at are n[ among ! mo/ desolate p>ts ( ! des]t1 & :at light does ?is ?r[ on human hi/ory & climatic *anges8 ,to mo/ ( ^! que/ions no completely satisfactory answ]s h z yet be5 giv51 ?\gh sev]al ( !m h be5 ! subject ( pa9/ak+ inve/igations by _m work]s4 ,our k ( ! facts be>+ on !m has1 h[ev]1 increas$ consid]ably dur+ ! pa/ few ye>s2 & 9 ! pres5t pap] ,i propose briefly to review ! v>i\s que/ions 9 ! light ( ! late/ available data conc]n+ !m4 ,! problems1 ?\gh !y conc]n a v>iety ( subjects1 are v mu* int]wov5 ) ea* o!r1 & !y are mo/ly alike 9 t ! fir/ difficulty 9 any attempt to solve !m has always be5 ! incomplet5ess ( our geographical k conc]n+ ! ,libyan ,des]t z a :ole4 ,if we possess$ a s]ies ( detail$ cont\r-maps cov]+ ! :ole ( ! des]t1 mo/ ( ! problems w\ld be 9 a fair way ( solution4 ,b ! surveys :i* w\ld be necess>y = su* a result are impracticable at pres5t1 ma9ly on ! score ( exp5se4 ,"s fifte5 ye>s ago ,i essay$ ! con/ruction ( a cont\r$ map ( ! ,libyan ,des]t1 9 connection ) ! ,int]national 8,million0 map ( ! _w4 ,b ,i _h to give up ! task1 f9d+ t !re 7 va/ areas )\t a s+le obs]vation = altitude1 :ile su* altitude- data z did exi/ 7 op5 to errors ( a magnitude intol]able = ev5 a prelim9>y cont\r+ ( ! areas ) :i* !y 7 conc]n$4 ,! attempt h[ev] 0 n )\t its value1 because x drew att5tion to ! great lack ( reliable levels 9 ! ,libyan ,des]t1 & to ! necessity ( rem$y+ ?is defect be=e ev5 an approximately true picture ( ! g5]al relief c\ld be obta9$4 ,9 ! int]val :i* has elaps$ s9ce my =m] attempt1 a great deal ( survey+ has be5 c>ri$ out 9 ! ,libyan ,des]t1 & 9 mo/ ( ?is "w special att5tion has be5 paid to ! matt] ( levels4 ,two long *a9s ( triangulation h be5 run we/w>d f ! ,nile to ,siwa ,oasis1 "o vi ! ,wadi ,natrun & ! o!r vi ,bah>ia ,oasis2 ano!r *a9 has be5 complet$ along ! coa/ we/w>d f ,alex&ria to ,sollum1 & a triangulation is n[ be+ c>ri$ out :i* w ev5tually connect ! oases ( ,kh>ga & ,dakhla ) ! ,nile ,valley4 ,small local triangulations h also be5 made 9 c]ta9 areas1 su* z ,dakhla ,oasis & 9 ! neighb\rhood ( ,ow5at4 ,9 all ^! triangulations v]tical z well z horizontal angles h be5 obs]v$1 s t reliable trigonometric altitudes are n[ available = almo/ all ! po9ts occupi$ or sight$4 ,9 ! s\!rn & c5tral p>ts ( ! des]t1 we are /ill dep5d5t on b>ometric levels2 b ! old det]m9ations h mo/ly be5 replac$ by lat] & bett] ones2 & ! use ( motor-c> transport1 by facilitat+ exploratory j\rneys 9 hi!rto untrodd5 regions1 has p]mitt$ ( a l>ge numb] ( additional det]m9ations be+ made 9 tracts :]e no previ\s obs]vations ( any k9d exi/$4#a ,ow+ to ! me?ods employ$ 9 ! late/ b>ometric det]m9ations1 ! result+ altitudes are ( a mu* high] ord] ( accuracy ?an c\ld be hop$ = 9 previ\s measurem5ts ( ! k9d2 9 fact1 x is believ$ t !y are n s v f> beh9d trigonometric levels 9 precision4 ,?is increas$ accuracy has be5 r5d]$ possible by ! collaboration ( ! ,egyptian ,meteorological ,s]vice1 :i*1 ?anks to ! abundant wea!r-data n[ receiv$ & dealt ) by x1 is 9 a position to furni% a close approximation to ! sea-level pressure at any po9t 9 ! des]ts at any giv5 in/ant4 ,! an]oids us$ 9 ! des]t explorations are comp>$ ) ! /&>d b>omet] ( ! ,meteorological ,s]vice imm$iately be=e ! />t & imm$iately aft] ! return ( ea* exp$ition4 ,at ea* obs]vation 9 ! des]t1 ! date & "t1 & ! temp]ature ( ! air1 are not$1 z well z ! approximate latitude & longitude ( ! place4 ,on ! return ( ! exp$ition1 ! record$ read+ is correct$ = ! error ( ! an]oid on ! /&>d b>omet]1 & ! ,meteorological ,s]vice is ask$ to supply ! sea-level pressure at ! p>ticul> place & "t ( ! obs]vation4 ,f ! diff]5ce ( ! two pressures & ! air temp]ature1 ! height ( ! place above sea-level is !n work$ out directly ) ! aid ( ,jordan's table ( 8,b>ometris*e ,hh5/uf540 ,if ! place ( obs]vation lies consid]ably above sea-level1 a small correction is appli$ to ! obs]v$ air-temp]ature to all[ = ! temp]ature-gradi5t 9 ! air2 = ( c\rse x is ! mean temp]ature ( ! air column betwe5 ! place & sea-level :i* %\ld be employ$ 9 ! r$uction4 ,x is unlikely t ) ?is proc$ure ei!r ! sea-level pressures or ^? obs]v$ at ! places c be 9 error by mu* m ?an half a millimetre2 & consequ5tly1 ev5 ) "s slight unc]ta9ties 9 ! air-temp]atures1 ! result+ altitudes may be expect$ to be accurate to )9 ab\t #ae metres :5 only a s+le obs]vation has be5 made1 & to )9 mu* less ?an ?is :51 z is usually ! case at important po9ts1 ! altitudes h be5 calculat$ f a s]ies ( read+s spread ov] sev]al days4 ,t ?is degree ( accuracy is actually atta9$ seems evid5t on comp>ison ( ! levels obta9$ = ! same places on diff]5t dates4 ,?us1 = in/ance1 at a camp close to ,bir ,t]fawi1 dur+ a /ay ( five days 9 ,janu>y & two days 9 ,m>* #aibe1 ,i made 9 all eighte5 an]oid obs]vations4 ,! mean altitude f\nd 0 #bdd metres2 ! av]age deviation ( a s+le obs]vation f ?is mean 0 #h metres1 ! maximum be+ #af4 ,on ! assumption t all s\rces ( error are accid5tal1 ! probable error ( ! mean result+ altitude works out at less ?an #b metres1 :i* is ab\t ! same z might be expect$ 9 trigonometric levell+ to ! place f ! ,nile ,valley4 ,! po9ts 9 ! int]ior ( ! ,libyan ,des]t ( :i* ! levels are fairly accurately kn[n n[ numb] _m hundr$s1 & ) ! exception ( a l>ge blank 9 ! unexplor$ we/1 & ano!r betwe5 ! ,nile & ,m]ga 9 ! ,nor!rn ,sudan1 !y are fairly well di/ribut$ ov] ! c\ntry4 ,x !re=e seem$ to me to be wor? :ile to make ano!r attempt at a cont\r$ map2 = ! levels n[ kn[n 7?\gh /ill f> too few = cont\rs to be drawn ) any great precision7 might suffice = ! con/ruction ( a map on a comp>atively small scale & ) cont\rs at fairly wide int]vals2 x 0 felt t a provisional map ( ?is k9d1 besides af=d+ at lea/ an approximate g5]al view ( ! relief1 might help t[>ds ! solution ( "o or m ( ! problems :i* ,i h m5tion$ above4 ,i comm5c$ by sket*+ 9 ! cont\rs at #ajj-metre int]vals on ! orig9al %eets ( ! #a3ejj1jjj map ( ,egypt1 a revis$ edition ( :i* is 9 prep>ation4 ,!y 7 !n r$uc$ to ! million scale1 & aft]w>ds /ill fur!r r$uc$1 ) additions f my own rec5t obs]vations ar\nd ,ow5at1 ,gebel ,kissu1 & ,m]ga1 f ,hassane9 ,bey's records ( 8 j\rney betwe5 ,jalo & ,el ,fa%]1 f ! altitude-data obta9$ by ! ,anglo-,fr5* ,b\nd>y ,commission 9 ! ,sudan1#b & f ! maps ( ,pr9ce ,kemal el ,d9's exp$ition ( #aibf to ,ow5at & ,s>ra ,well1#c to =m ! map foll[+ p4 #if4 ,! attempt has1 ,i ?9k1 fully ju/ifi$ itself1 & m ?an fulfill$ ! hopes t 7 ent]ta9$ conc]n+ x4 ,probably ! feature t w fir/ /rike ! att5tion on glanc+ at ! cont\r$ map is ! s]ies ( depressions bel[ sea-level /ret*+ we/w>d f ne> ,cairo to ,j>abub4 ,! foll[+ table gives a li/ ( ! pr9cipal ( ^! depressions1 ) ! approximate ext5t ( ! areas ly+ bel[ sea-level1 & also 7:]e kn[n7 ! dep?s bel[ sea-level ( _! deepe/ p>ts4 ,:]e ! l[e/ "p is cov]$ by a lake1 ! dep? %[n is t ( ! wat]-surface1 n t ( ! lake-bottom3 ,approximate area ,approximate ,depression4 bel[ sea-level dep? ( deepe/ 9 sq4 kiloms4 "p bel[ sea 9 metres4 ,wadi ,natrun #bbj #bc ,faiyum #gjj #de ,wadi ,rayan #bhj #dc ,qatt>a 7includ+ ,mogh>a1 #ah1jjj #acd ,q>a1 & ,tebaghbagh7 ,sittra 7includ+ ,timata7 #cjj #cc ,bahre9 7includ+ ,nuemisa7 #aej #ae 787 ,watiya #gj #ae ,areg #fj #be ,siwa 7includ+ ,maasir & #hjj #ag ,m>aqi7 ,girba #bj 787 ,kheiba #dj 787 ,%iyata #bj 787 ,melfa 7includ+ ,%ebat1 #bfj #ah 787 ,exabia1 & ,gagub7 ,z w be se5 f ! figures 9 ! table1 by f> ! l>ge/ & deepe/ depression 9 ?is s]ies--sev5 times z l>ge1 9 fact1 z all ! o!rs put toge!r--is t ( ,qatt>a1 :i* has n be5 %[n on any previ\s map1 & ( :i* ! exi/5ce 0 prov$ only la/ ye>4 ,t s deep a depression %\ld exi/ comp>atively ne> to ! ,m$it]ranean & to ,cairo1 & h rema9$ undiscov]$ until n[1 is /rik+ evid5ce ( h[ mu* has /ill to be le>n$ conc]n+ ! configuration ( ! des]t surface4 ,! mann] 9 :i* ! discov]y 0 made is p]haps wor? plac+ on record4 ,he>+ e>ly 9 #aiag t a milit>y patrol 0 ab\t to op]ate 9 ! ,qatt>a region1 ,i l5t a small an]oid to ! (fic] 9 *>ge ( ! patrol1 & ask$ him to take read+s ) x at v>i\s po9ts along 8 r\te1 ) ! object ( gett+ "s idea ( ! altitudes 9 t region4 ,x 01 ( c\rse1 c]ta9 t l[-ly+ c\ntry exi/$ along ! foot ( ! ,qatt>a esc>pm5t1 b no idea 0 ent]ta9$ t any great "p ( x might be bel[ ! level ( ! sea4 ,! (fic] br\ght back an]oid-read+s :i* seem$ to indicate t ! spr+ at ! foot ( ! sc>p at ,qatt>a 0 ab\t #fj metres bel[ sea-level2 b ! an]oid itself 0 un=tunately lo/ dur+ ! return j\rney4 ,! result 0 too surpris+ = x to be accept$ )\t evid5ce t ! read+s ( ! in/rum5t _h n be5 vitiat$ by accid5tal r\gh treatm5t1 & consequ5tly ,i refra9$ at ! "t f plac+ ! levels f\nd on ! maps4 ,b ,i resolv$ to confirm !m or o!rwise on ! fir/ opportunity4 ,t opportunity came la/ ye>1 :5 x 0 possible to s5d a survey- p>ty to triangulate we/w>ds f ! ,wadi ,natrun s z to cross ! place :]e ! an]oid read+s _h be5 tak54 ,! "w 0 entru/$ to ,mr4 ,g4 ,f4 ,walpole1 ( ! ,survey ( ,egypt1 :o _h already di/+ui%$ himself by successfully c>ry+ out a difficult triangulation f ! ,nile across "s #ejj kilometres ( ! ,libyan ,des]t vi ,bah>ia to ,siwa4 ,! result ( ,mr4 ,walpole's "w 0 n only to confirm ! sub/antial accuracy ( ! previ\s e/imation ( ! level ( ! ,qatt>a spr+1 b to br+ to light ! exi/5ce ( a va/ holl[1 ?\s&s ( squ>e kilometres ( ^: floor lie at ev5 l[] levels1 & :i* at "o place desc5ds to a dep? ( no less ?an #acd metres bel[ ! level ( ! sea4#d ,?is la/-m5tion$ spot1 "s #ajj kilometres s\?-s\?-we/ ( ! ,qatt>a spr+1 is probably ! l[e/-ly+ po9t on ! l&-surface ( ! ,african cont95t4 ,a sc>cely less rem>kable feature is ! s]ies ( depressions compris+ ! oases ( ,kh>ga1 ,dakhla1 ,abu ,mung>1 & ,f>afra2 :ile ,bah>ia occupies a m isolat$ position ab\t midway betwe5 ,f>afra & ! ,faiyum4 ,9 contra/ ) ^? ( ! nor!rn s]ies1 ^! s\!rn depressions d n p5etrate to sea-level4 ,! l[e/ po9ts 9 ,kh>ga ,oasis are probably at or j bel[ ! level ( ! sea1 b ! g5]al level ( ! ,kh>ga floor is ab\t #gj metres above t datum4 ,! l[e/ po9t ( ,f>afra is some:at high] ?an ! g5]al level ( ,kh>ga2 ^? ( ,dakhla & ,abu ,mung> are high] /ill1 :ile 9 ,bah>ia1 ?\gh x lies f>!r to ! nor?1 ! floor-level av]ages "s #acj metres above sea4 ,a notewor?y ?+ ab\t ! l>g] depressions is t _! nor!rn b\nd>ies are =m$ by /eep esc>pm5ts1 :ile to ! s\? ! slope is m gradual4 ,?is is ! case1 = in/ance1 ) ! ,faiyum1 ,qatt>a1 ,siwa1 ,kh>ga1 ,dakhla1 ,abu ,mung>1 & ,f>afra4 ,bah>ia is unique 9 be+ entirely surr\nd$ by esc>pm5ts1 z also 9 conta9+ a l>ge numb] ( hills approximat+ 9 height to t ( ! b\nd+ sc>ps4 ,9 ! majority ( cases ! great] /eepness ( ! nor!rn b\nd>ies c be correlat$ ) geological /ructure2 ! nor!rn walls ( ! ,qatt>a & ,siwa depressions1 = in/ance1 m>k ! s\!rn limit ( c]ta9 ,mioc5e /rata1 :ile ^? ( ,kh>ga & ,dakhla co9cide ) ! s\!rn limit ( ! ,eoc5e lime/ones4 ,to ! s\?-we/ ( ,dakhla ,oasis !re /ret*es a broad tract ( ris+ gr\nd1 ) ! ,gilf ,kebir & ! peaks ( ,ark5u1 ,ow5at1 & ,kissu z conspicu\s features4 ,on ei!r side ( ?is tract ! g5]al level falls1 on ! "o h& t[>ds ,kufra & ,cyr5aica1 & on ! o!r h& t[>ds ! ,nile4 ,! ,gilf ,kebir is a great brok5 plateau ( s&/one1 ris+ v ne>ly to ! #ajjj-metre cont\r & /ret*+ 9 a direction a little we/ ( s\? = ov] #ajj kilometres4 ,i saw ! s\? end ( ?is plateau f sev]al po9ts on ! way f ,t]fawi to ,ow5at :5 travell+ ) ,pr9ce ,kemal el ,d9 9 #aibe1 & fix$ its position by compass-be>+s4 ,z se5 f ! s\?1 ! plateau appe>$ to be m]ely a flat-topp$ hill2 its true =m & ext5t 7 discov]$ by ! ,pr9ce 9 ! foll[+ ye>1 :5 he pass$ along its ea/]n foot on 8 way f ,pott]y ,hill to ,ow5at4 ,! highe/ peak ( ,ark5u1 9 ab\t lat4 #bb #ag1 long4 #bd #df1 ,i e/imat$ to rise to approximately #adaj metres above sea4 ,! altitude ( ,gebel ,ow5at f> transc5ds ,hassane9 ,bey's previ\s e/imate ( #aajj metres2 a trigonometric measurem5t :i* ,i made 9 #aibe gave its summit z #aijg metres above sea4 ,! g5]al level ( ! gr\nd at ! foot ( ,ow5at ,i f\nd to av]age ab\t #fjj metres1 ?us agree+ ) ,hassane9 ,bey's figures4 ,gebel ,kissu1 ?\gh l[] ?an ,gebel ,ow5at1 is consid]ably high] ?an ,gebel ,ark5u1 & be+ an isolat$ m\nta9 ) a s+le well-m>k$ peak1 x =ms a v conspicu\s l&m>k2 my trigonometric det]m9ation gave #agbf metres = ! altitude ( ! peak above sea4 ,to ! s\?-we/ ( ,kissu !re are a few widely scatt]$ isolat$ hills1 b ! c\ntry 9 g5]al =ms a roll+ pla9 ( s& ext5d+ to ! nor?-ea/ corn] ( ,fr5* ,equatorial ,africa ) a ne>ly uni=m level ( ab\t #gjj metres4 ,beyond ! ,fr5* b\nd>y ! gr\nd rises t[>ds ! ,erdi ,hills1 :i* atta9 ov] #ajjj metres4 ,f ! corn] ( ,fr5* ,equatorial ,africa to ,m]ga1 a di/ance ( #bej kilometres1 ! c\ntry consi/s ( alt]nations ( s&-pla9s & flat /ony tracts1 ) a v gradual fall to a level ( ab\t #efj metres at ! edge ( ! ,m]ga depression4 ,betwe5 ,m]ga & ,ow5at ! c\ntry consi/s ( s&y pla9s alt]nat+ ) brok5 /ony gr\nd & occasional hills4 ,! g5]al level rises f ab\t #egj metres on ! nor!rn side ( ! ,m]ga depression to ne>ly #hjj metres on ! p>allel ( #bj #cj1 !n gradually falls to ab\t #fjj metres at ! foot ( ,gebel ,ow5at4 #a4 ,did ! ,nile1 or a ,bran* ( x1 ev] fl[ ?r\gh ! ,libyan ,des]t to ! ,m$it]ranean8 ,! idea t a 8dry riv]0 exi/s 9 ! ,libyan ,des]t is a v old "o1 & seems to h _h its orig9 9 ! fact t b>r5 depressions 9 t region are sometimes call$ by ! local ,arabs ,bahr bela ma1 t is1 8sea )\t wat]40 ,b ! ,arabic ^w bahr1 :i* prop]ly means a sea or a lake1 z 9 ,bahr ,lot 7! ,dead ,sea71 is also appli$ 9 ,egypt & ! ,sudan to a riv]1 z 9 ,bahr el ,nil 7! ,nile7 & ,bahr el ,azrak 7! ,blue ,nile74#e ,! e>lie/ depression to be discov]$ be>+ ! "n ( ,bahr bela ma 0 ! "o n[ kn[n z ! ,wadi ,f>egh 7! 8empty wadi071 situat$ imm$iately s\? ( ! ,wadi ,natrun4 ,fa!r ,sic>d1 :o visit$ ! place 9 #agab1 correctly int]pret$ ! ,arabic "n to mean lit]ally 8sea )\t wat]10 & hav+ discov]$ fossil trees !re1 :i* he imag9$ to be ! petrifi$ rema9s ( ma/s ( %ips1 he inf]r$ t a n>r[ arm ( ! sea =m]ly ext5d$ f ! ,m$it]ranean into ! locality4#f ,at a lat] date ,sic>d's view t ! petrifi$ trees 7 ! ma/s ( %ips 0 recogniz$ to be a mi/ak5 "o2 b a great] error 0 committ$ by *ang+ ! translation ( ,bahr bela ma f 8sea )\t wat]0 into 8riv] )\t wat]10 z 0 done by ,d',anville 9 8 #8,mmoires sur l',egypte1' publi%$ at ,p>is 9 #agff4 ,d',anville 0 a great au?ority 9 8 day1 & 8 translation 0 adopt$ by subsequ5t writ]s = ov] a hundr$ ye>s4 ,g5]al ,&reossi1 :o comm&$ ,napoleon's artill]y 9 ! exp$ition ( #agih & #agii1 made a map ( ! 8,dry ,riv]0 & ! 8,valley ( ! ,natron ,lakes10 9 :i* bo? depressions are %[n op5 at bo? ends1 in/ead ( be+ clos$ 9 z !y really are4 ,&reossi consid]$ t ! 8,bahr bela ma0 0 ! dry b$ ( an anci5t bran* ( ! ,nile1 :i* left ! pres5t ,nile ,valley some:]e 9 ,middle ,egypt & ent]$ ! sea to ! we/ ( ,alex&ria2 he also ?\ght t ! anci5t ,lake ,mo]is 0 probably =m$ by ! damm+ ( ! 8dry riv]0 ne> its suppos$ (ftake f ! ,nile4#g ,lat] on1 aft] ,cailliaud & o!r e>ly n9ete5?-c5tury travell]s _h discov]$ t o!r depressions exi/$ 9 widely sep>at$ localities & bore ! same ,arabic "n ( ,bahr bela ma1 x seems to h be5 ha/ily assum$ by ! c>tograph]s ( ! "t t all ! depressions be>+ ! same "n 7 p>ts ( a s+le dri$-up riv]-*annel4 ,?us1 on a l>ge map ( ,egypt compil$ by ,colonel ,lap9 & publi%$ at ,p>is 9 #ahef1 ! 8,bah> ,belah-mah ou ,fleuve sans eau0 is /at$ to come f ! ,congo & is depict$ z c\rs+ ?r\gh ! ,oasis ( ,dakhla1 !nce pass+ to ! ea/ ( ,bah>ia ,oasis & on to ! ,m$it]ranean4 ,on ano!r l>ge map ( ,egypt1 compil$ by ,muzzi ,bey1 ! !n ,director-,g5]al ( ,egyptian ,po/s1 & publi%$ 9 ,flor5ce 9 #ahgf1 a cont9u\s valley is boldly %[n lead+ f ! m\nta9s ( ,d>fur to ! ,m$it]ranean1 :ile cross-valleys1 labell$ 8,old ,b$ ( ! ,nile10 are depict$ z connect+ ! ma9 valley ) t ( ! ,nile at ,korosko & ,dongola4 ,! ,rohlfs exp$ition ( #ahgd prov$ t !re 0 no su* *annel runn+ nor?w>ds ?r\gh ,dakhla ,oasis2 t ! 8,bahr bela ma0 :i* _h be5 cross$ by ,cailliaud betwe5 ,siwa & ,bah>ia 0 m]ely a clos$-9 local depression1 & t ! cont9u\s empty riv]-b$s :i* 7 %[n on ! maps ( t p]iod _h no real exi/5ce4#h ,b ?\gh ! cont9u\s 8,bahr bela ma0 *annel ( ! old c>tograph]s has be5 %[n to be purely imag9>y1 & has 9 consequ5ce disappe>$ f our mod]n maps1 ! belief t an old dry riv]-b$ may exi/ some:]e 9 ! ,libyan ,des]t has app>5tly p]si/$ 9 _m m9ds ev5 to our own day4 ,only a few ye>s ago ,i 0 call$ ^u by ! ,gov]nm5t to discuss a sugge/ion1 made 9 all s]i\sness by a p]son ( consid]able em95ce1 t ! ,nile1 or a bran* ( x1 mu/ at "o "t h foll[$ a c\rse f some:]e ne> ,dongola ?r\gh "s ( ! ,egyptian oases to ! ,m$it]ranean2 & t by trac+ out ! old *annel1 & deep5+ x artificially :]e necess>y1 a "p ( ! riv] might be tak5 (f along ?is pa? & its wat] utiliz$ to irrigate ! des]t c\ntry on ei!r side ( x4 ,to ! few sci5tific travell]s :o h j\rney$ ext5sively 9 ! ,libyan ,des]t1 x w d\btless appe> incr$ible t su* a sugge/ion z ?is c\ld be s]i\sly made4 ,b old traditions conc]n+ ! geography ( little-kn[n regions die h>d1 & ?is p>ticul> "o has probably be5 fo/]$ by ! circum/ance t on mo/ small-scale maps ! oasis-depressions h be5 %[n )\t any precise indication ( ! altitude ( ! int]v5+ gr\nd1 z well z by ! speculations ( geologi/s z to ! exi/5ce ( a riv] 9 ! region 9 pa/ geological ages4 ,blanck5horn1 = in/ance1 publi%$ 9 #aijb a s]ies ( small-scale maps depict+ ! c\rse ( a hypo!tical riv]1 ! 8,libyan ,ur-,nile10 runn+ nor?w>ds ?r\gh ! des]t 9 ! ,eoc5e & ,oligoc5e p]iods1 & ceas+ to exi/ 9 ! ,plioc5e4#i ,t a riv] fl[$ dur+ late ,eoc5e or e>ly ,oligoc5e times 9 a nor?-ea/]ly direction ?r\gh :at is n[ ! ,libyan ,des]t1 & ent]$ a sea ne> :at is n[ ! ,faiyum1 c sc>cely be que/ion$2 = x is only by ! exi/5ce ( su* a riv] t we c acc\nt = ! assemblage ( fossil trees & curi\s rema9s ( ,eoc5e animals 7includ+ ! ance/ors ( ! elephant7 discov]$ by ,mr4 ,beadnell nor? ( ! ,birket el ,q>un & describ$ by ! late ,dr4 ,&rews4#aj ,z to ! size1 l5g?1 & exact position ( ?is riv] ( ! geological pa/ !re is v great d\bt2 b nei!r its situation nor ev5 its exi/5ce c be consid]$ mat]ial to our pres5t inquiry4 ,no "o conv]sant ) ! kn[n facts be>+ on ! g5]al geological hi/ory ( ,egypt dur+ late ,t]ti>y & ,quat]n>y times c\ld make ! mi/ake ( ?9k+ t ! pa? ( an ,eoc5e riv] might be traceable 9 ! pres5t des]t relief4 ,! evid5ces ( su* pa? mu/ be purely ( a geological nature4 ,9 8 endeav\rs to sugge/ :at might possibly h be5 ! c\rse ( ! riv] on ^: banks ! ,eoc5e animals liv$1 ,mr4 ,beadnell f\nd 9 two localities deposits :i* might h be5 =m$ along ! anci5t dra9age-l9e2 & 9 bo? ^! localities ! deposits n[ occupy ! summits ( hills4 ,!re is insuffici5t evid5ce ( any k9d to ju/ify s exact a location ( ! ,eoc5e riv] z ,blanck5horn's maps w\ld sugge/1 & none :atev] ( any relation%ip1 ei!r betwe5 ! pres5t surface- configuration ( ! des]t & ,eoc5e or ,oligoc5e dra9age-l9es1 or betwe5 ^? dra9age-l9es & ! ,riv] ,nile4 ,! appellation 8,ur- ,nile0 ( ,blanck5horn is ?us a mislead+ "o4 ,if we are to look = traces ( a dri$-up riv] 9 ! surface-=ms ( ! ,libyan ,des]t ( to-day1 x mu/ be a riv] :i*1 l ! /reams :i* erod$ ! great wadis ( ! ,ea/]n ,des]t1 exi/$ 9 geologically rec5t times1 & :i* became dry ei!r ?r\gh ! cont9ual deep5+ ( ! pres5t ,nile-*annel1 ) consequ5t captur+ ( ! =m] dra9age1 or ?r\gh climatic *anges caus+ a dim9ution ( ra9fall 9 ! collect+ area4 ,a /udy ( ! cont\r$ map seems to me to be absolutely conclusive on ! "q at issue1 = x indicates cle>ly t at no "t )9 ! geologically rec5t p]iod c !re h be5 any su* cont9u\s *annel z has be5 suppos$4 ,let u try & trace1 ) ! aid ( ! map1 ! mo/ likely c\rse along :i* ! anci5t ,libyan ,nile-bran*1 if ev] "o exi/$1 mu/ h fl[$4 ,/>t+ f ! ,nile 9 ! neighb\rhood ( ,dongola1 we obs]ve t ! l[e/ possible track is ?r\gh ,%eb t[>ds ,kh>ga4 ,b ,%eb c only be rea*$ aft] pass+ ov] "s #cej kilometres ( des]t ly+ well above ! level ( ! ,nile at ,dongola1 :i* is ab\t #bcj metres above sea4 ,x is1 ( c\rse1 q conceivable t ev5 )9 geologically rec5t times ! ,nile at ,dongola may h be5 consid]ably above its pres5t level2 s t we may pass ov] ?is fir/ difficulty z n be+ def9itely conclusive aga9/ ! hypo!sis we are pursu+4 ,f ,%eb to ! s\? end ( ! ,kh>ga depression !re is a drop ( ab\t #afj metres1 s t up to ?is po9t a *annel may h be5 possible4 ,b f "h onw>ds1 difficulties come 94 ,if we turn nor?w>ds1 we soon enc\nt] ! nor? wall ( ,kh>ga1 ris+ to #cjj metres above sea-level2 s !re is no cont9u+ t way4 ,in/ead1 we turn we/w>ds t[>ds ,dakhla4 ,b f ,kh>ga we h to pass ov] gr\nd rea*+ v ne>ly to ! #bjj-metre cont\r = "s #hj kilometres to ent] ! ,dakhla depression1 :]e we drop to a level ( #aai metres4 ,we h to trav]se "s #ae kilometres at altitudes above #bjj metres to rea* ,abu ,mung>1 :]e ! level is #aag metres2 !n a fur!r #ab kilometres ly+ above ! #bjj-metre cont\r to get to ,f>afra1 :]e ! level desc5ds to ab\t #ij metres4 ,to get out ( ,f>afra1 we h aga9 to trav]se c\ntry ly+ ov] #bjj metres above sea1 :e!r we *oose a nor?-we/]ly r\te vi ,sittra to ! ,qatt>a depression1 or a nor?-we/]ly "o vi ,bah>ia to ! ,wadi ,rayan & ! ,faiyum4 ,tak+ ! fir/ ( ^! alt]natives z be+ ! %ort]1 we cross "s #cj kilometres ( gr\nd above ! #bjj-metre level1 & a fur!r #abj kilometres ly+ betwe5 ! #bjj- & #ajj-metre cont\rs1 be=e rea*+ ! ,qatt>a depression4 ,& once we are 9 ! ,qatt>a depression 7! l[e/ po9t ( :i* is #acd metres bel[ sea71 !re is no possibility ( rea*+ ! ,m$it]ranean except by cross+ a consid]able tract :]e ! gr\nd-level is ov] #ajj metres above sea-level4 ,a section ( ! r\te describ$1 ) a comp>ison-l9e pass+ ?r\gh ! pres5t ,nile-levels 9 correspond+ latitudes1 is %[n 9 ! figure bel[3 ,illu/ration3 ,pr(ile f ! ,nile at ,dongola ?r\gh ! depressions ( ,kh>ga1 ,dakhla1 ,abu ,mung>1 ,f>afra1 & ,qatt>a to ! ,m$it]ranean4 ,v]tical scale #ejj times ! horizontal4 ,total l5g? ( section ab\t #agjj kilometres2 = m ?an half ?is di/ance ! gr\nd-level is se5 to be above a l9e drawn ?r\gh ! levels ( ! ,nile 9 correspond+ latitudes4 ,c any "o believe t a c\rse ( ?is k9d 0 ev] t ( a riv]8 ,to make a ?r\gh *annel w\ld entail deep cutt+ = m ?an half ! total di/ance ( #agjj kilometres1 & w\ld consequ5tly involve excavat+ hundr$s ( millions ( cubic metres ( rock1 mu* ( x ( consid]able h>dness4 ,moreov]1 x _c be suppos$1 9 ! pres5t /ate ( our k ( ! region1 t m fav\rable l9es ( communication exi/ & h be5 ov]look$4 ,_m depressions d\btless /ill rema9 to be discov]$2 b ^!1 l ^? already kn[n1 w surely prove to be m]ely local bas9s4 ,s _m cross-c\ntry j\rneys h n[ be5 made 9 v>i\s directions ?r\gh ! des]t1 t x is inconceivable t any cont9u\s *annel ( ! k9d we are discuss+ c\ld possibly h escap$ notice4 ,nor c we reasonably suppose t rec5t e>?-movem5ts h produc$ ! pres5t sev]ance ( ! v>i\s depressions2 su* movem5ts might p]haps be invok$ to acc\nt = "o or two ( ! sep>at+ elevations1 b c]ta9ly n = all4 ,we c !re=e be absolutely c]ta9 t nei!r ! ,nile nor any bran* ( x ev] pass$ ?r\gh ! ,libyan ,des]t to ! sea4#aa ,ev5 *annels ( tribut>y /reams to ! ,nile1 su* z abundantly survive 9 ! great wadis :i* ent] ! ,nile valley on its ea/]n side1 are m>k$ly abs5t 9 ! ,libyan ,des]t4 ,! cont\rs on ! map sugge/ t possibly "o su* tribut>y /ream may =m]ly h dra9$ ! s\?-we/]n faces ( ! great plateau 9 :i* ! depressions ( ,dakhla & ,kh>ga =m bays1 & h ent]$ ! ,nile some:]e betwe5 ,aswan & ,halfa2 b if su* a *annel ev] did exi/1 its actual traces h long s9ce be5 oblit]at$ by ! smoo?+ action ( subsequ5t d5udation1 & only ! great esc>pm5t ( ! plateau :i* x half encircl$ rema9s to tell ! tale4 ,photographs by ,pr9ce ,kemal el ,d91 ,dr4 ,ball1 & ,mr4 ,walpole ,illu/ration3 #a4 ,des]t exploration ) ! ord9>y c>1 requir+ help ev5 on flat gr\nd 9 s&y tracts ,illu/ration3 #b4 ,cat]pill> c>s c>ry+ heavy loads ov] a s&- dune betwe5 ,bir ,t]fawi & ,gebel ,ow5at ,illu/ration3 #c4 ,! we/]n wall ( ! ,qatt>a depression #ce km4 ,s4,s4,w4 ( ,q>a ,oasis4 ,top ( cliff ab\t #c m4 & floor ( depression #acc m4 bel[ sea-level ,illu/ration3 #d4 ,a9 el ,hez1 ,bah>ia ,oasis1 irrigat$ by free fl[ f well ,illu/ration3 #e4 ,ov]look+ ! ,bah>ia depression f ! pass at its s\!rn end3 c>s b\nd = ,f>afra ,illu/ration3 #f4 ,nor!rn end ( ,f>afra depression3 s&y rav9e cutt+ back into *alk plateau #b4 ,orig9 ( ! ,depressions4 ,x is a m>k$ *>act]i/ic ( almo/ all true des]ts1 t !y are regions ( int]nal dra9age2 & ! ,libyan ,des]t is no exception to ! rule4 ,!re are a few gullies dra9+ f ! plateau to ! sea along its nor!rn edge1 & a few o!rs dra9+ into ! ,nile ,valley along its ea/]n bord]4 ,b none ( ^! ext]nal dra9age-l9es ext5ds = m ?an an insignificant di/ance backw>d into ! int]ior ( ! c\ntry4 ,!re is1 9 fact1 9 ! ,libyan ,des]t 7especially 9 its nor?-ea/]n portion7 a m>k$ abs5ce ( any di/9ct dra9age-l9es at all2 9 _! place we f9d ext5sive /ony pla9s1 :i* ei!r slope gradually d[n t[>ds ! v>i\s inl& depressions1 or t]m9ate 9 an abrupt fall :]e ! depressions are cut back into !m4 ,ev5 9 ! ,pluvial epo*1 :5 ,europe 0 glaciat$ & ! deep wadis ( ! ,ea/]n ,des]t on ei!r side ( ! ,nile 7 be+ cut out by great /reams1 !re _c h be5 any ext]nal dra9age f a l>ge "p ( ! ,libyan ,des]t2 = if !re _h be51 we %\ld f9d ! ,libyan plateau f> m int5sely dissect$ ?an is actually ! case4 ,! ra9fall 9 ?is region1 ev5 dur+ ! ,pluvial p]iod1 mu/ h be5 f> small] ?an 9 ! neighb\r+ tracts ( ! ,ea/]n ,des]t & ,s9ai2 su* ra9fall z did occur 0 dra9$ into ! s]ies ( depressions already describ$1 & !re ei!r =m$ lakes or 0 evaporat$4 ,b ! action ( infl[+ dra9age wat]s1 c>ry+ s& & mud 9 susp5sion & salts 9 solution1 is n to deep5 ! bas9s into :i* !y fl[1 b r to fill !m up4 ,h5ce we mu/ look to "s o!r cause = ! =mation ( ! bas9s !mselves4 ,a possible cause is to be f\nd 9 e>?- movem5ts--ei!r a local d[n-fold+ ( ! cru/1 or an uprais+ by fault+ ( ! surr\nd+ tracts4 ,x is n unlikely t su* e>?- movem5ts h to "s ext5t condition$ ! =mation ( ! depressions2 b t ! depressions are n simply fault$-d[n areas or subsid5ces is abundantly cle> f an exam9ation ( ! b\nd+ sc>ps & ! floors ( ! l>g] oases4 ,bah>ia is :olly1 & ,kh>ga & ,dakhla are p>tially1 b\nd$ by high esc>pm5ts1 9 _m places hundr$s ( metres high1 & ! floors ( ^! oases are compos$ ( ! same rocks z are expos$ at ! base ( ! b\nd+ esc>pm5ts4 ,x is1 9 fact1 obvi\s t ^! great holl[s are natural excavations1 n subsid5ces4 ,:at 0 ! excavat+ ag5cy8 ,to ! s\?-we/ ( ! lime/one plateau :i* lies we/ ( ! ,nile & ext5ds nor?w>ds f lat4 #bc #cj !re is fairly op5 s&/one c\ntry1 & ! oases ( ,dakhla & ,kh>ga occupy great bays 9 ! esc>pm5t :i* m>ks ! s\?-we/ t]m9ation ( ! plateau4 ,x seems reasonable to suppose t ! ,eoc5e & ,cretace\s seas _h _! s\!rn %ores some:]e ab\t ! latitude ( ,wadi ,halfa1 = ! lime/ones d n ext5d f> beyond ! tropic2 & to ! s\? ! h>d lime/ones ( ! plateau probably pass$ gradually into s(t] %all[-wat] deposits1 :i* w\ld l5d !mselves easily to degradation & to removal by ! ag5cy ( /reams ^: traces h long s9ce disappe>$4 ,we c ?us possibly invoke wat]-action to acc\nt = ! primitive =mation ( ! 8bays0 9 :i* ,kh>ga & ,dakhla are situat$4 ,b we _c 9 ?is way expla9 ei!r ! fur!r deep5+ ( ^! oases1 :i* conv]t$ !m into bas9s1 or ! excavation ( ! m nor!rly depressions su* z ,bah>ia & ,qatt>a4 ,! problem z to h[ ! excavation ( ! nor!rn depressions1 & ! deep5+ ( ! s\!rn ones1 7 accompli%$ is n altoge!r an easy "o4 ,!re is1 ( c\rse1 no d\bt t w9d has be5 ! ma9 excavat+ ag5cy4 ,all ! depressions occur 9 areas :]e s(t rocks are ov]la9 by h>d ones1 & once ! h>d ov]ly+ rock 0 remov$ at any place1 ! action ( ! w9d on ! s(t] b$s w\ld soon excavate a holl[4 ,! difficulty is only z to h[ ! h>d ov]ly+ rock 0 fir/ remov$4 ,to "s :o h /udi$ ! "q1 a suffici5t explanation is to be f\nd 9 ! g5]al degradation1 comb9$ ) ! tectonic di/urbances--folds & faults--:i* are kn[n to h affect$ "s ( ! oasis-areas1 z1 = in/ance1 t ( ,bah>ia4 ,giv5 an anticl9al fold1 a g5]al l[]+ ( ! surface by w9d-degradation w\ld ev5tually result 9 ! exposure ( ! s(t und]ly+ b$s at ! highe/ po9t ( ! fold1 & subsequ5tly cont9u$ w9d-erosion w\ld proce$ mo/ rapidly on ^! s(t und]ly+ rocks4 ,o!rs1 :ile admitt+ ?is explanation z p>tly acc\nt+ = ! =mation ( ! depressions1 consid] t o!r factors mu/ h assi/$ to cause p5etration ( ! h>d ov]ly+ b$s1 especially 9 ! areas :]e fold+ is n v m>k$4 ,( su* o!r factors1 ,i c only sugge/ :at appe>s to me to be a possible "o4 ,ab\t tw5ty ye>s ago ,i 0 mapp+ ! ,m$it]ranean coa/al area 9 ! neighb\rhood ( ,m]sa ,matruh1 "s #afj miles to ! we/ ( ,alex&ria4 ,9 ?is area1 :i* has ab\t #f in*es ( ra9fall 9 an av]age ye>1 !re are num]\s clos$-9 bas9s a few miles long & a few hundr$ y>ds 9 wid?1 desc5d+ to a dep? ( #aj metres or m 9 ! r s(t lime/one :i* =ms ! pr9cipal rock ( ! locality4 ,^! bas9s1 ?\gh call$ wadis on ! maps1 are m usually kn[n z 'ebs or 8bosoms0 by ! local ,arabs4 ,i 0 at fir/ mu* puzzl$ to acc\nt = !m2 b ev5tually ,i arriv$ at :at ,i believe to be ! true explanation ( _! =mation4 ,! bas9s1 ( c\rse1 collect ! surr\nd+ ra9fall 9 ! w9t]1 & ! dra9age c>ries ) x into ! holl[s a c]ta9 am\nt ( f9e loam & s& f ! dis9tegration ( ! surr\nd+ surfaces4 ,aft] ! ra9s1 ! bottoms ( ! bas9s become cov]$ ) vegetation1 :ile ! surr\nd+ rocky surfaces are left b>e4 ,9 ! hot dry summ]1 h[ev]1 ! vegetation p]i%es1 & mu* ( ! loam is remov$ by ! w9d4 ,we h only to assume 7:at seems inde$ likely7 t ! rootlets ( ! vegetation dis9tegrate ! lime/one floors ( ! holl[s to =m fre% soil1 & t m ( ?is is remov$ ea* ye> by ! w9d ?an is br\ght into ! holl[s by ! dra9age1 to f9d an adequate cause = a sl[ b progressive deep5+ ( ! bas9s4 ,! primitive holl[ may well be a v slight depression ( ! surface1 su* z might arise f great] local s(tness ( ! rock & consequ5t m rapid wea!r+4 ,once vegetation has comm5c$ to gr[ 9 a slight holl[ ( ?is k9d1 ! action w\ld lead to a sl[ yet /eadily progressive deep5+ ye> by ye>1 & s&-erosion w\ld t5d t[>ds an elongation ( ! depression 9 ! direction ( ! prevail+ w9d4 ,c su* a process z ?is1 cont9u$ ?r\gh va/ ages dur+ :i* ! climate 0 some:at less dry ?an at pres5t1 h be5 active 9 ! =mation ( ,bah>ia & simil> depressions8 #c4 ,! ,possibility ( ,utilization ( ,depressions 9 ! ,libyan ,des]t = ,irrigation1 ,dra9age1 or ,hydraulic ,p[]4 ,! depression kn[n z ! ,wadi ,rayan1 discov]$ by ,mr4 ,cope ,:iteh\se ab\t #ahhc1 situat$ "s #dj kilometres we/ ( ! ,nile ,valley & ab\t #ajj kilometres s\? ( ! latitude ( ,cairo1 has be5 s frequ5tly describ$ & consid]$ 9 reg>d to its utilization z a res]voir or z a flood-protection = ,l[] ,egypt1 t ,i ne$ say little ab\t x beyond a ref]5ce to ! pr9cipal publi%$ works deal+ ) x4#ab ,!re is no o!r depression 9 ! ,libyan ,des]t :i* has receiv$ s mu* att5tion f ! economic /&po9t4 ,t ! ,wadi ,rayan has n already be5 utiliz$ z a res]voir has dep5d$ on a v>iety ( consid]ations1 ( :i* "o ( ! pr9cipal has be5 ! d\bts always ent]ta9$ z to :e!r !re might n be an und]gr\nd leakage f x1 result+ n only 9 a loss ( "p ( ! /or$ wat]1 b also 9 damage to ! adjac5t fl\ri%+ prov9ce ( ! ,faiyum4 ,^! d\bts w n be less5$ by ! conv]g+ l9es ( evid5ce :i* ,i %all br+ =w>d 9 deal+ ) ! artesian wat]- supplies1 all lead+ to ! conclusion t !re is an und]gr\nd wat]-connection betwe5 ! v>i\s depressions ( ! ,egyptian ,libyan ,des]t4 ,( late ye>s1 h[ev]1 x has be5 propos$ to make use ( ! ,wadi ,rayan1 n z a res]voir1 b z a sump = dispos+ ( ! dra9age wat] ( ,upp] ,egypt by evaporation2 & to ?is1 9 consequ5ce ( ! l[] level at :i* ! wat] w\ld /& 9 ! depression1 !re is f> less objection f ! po9t ( view ( possible leakage or damage to ! ,faiyum4 ,! ,wadi ,natrun1 :i*1 ?\gh consid]ably less deep1 is n greatly less 9 ext5t ?an ! ,wadi ,rayan1 has receiv$ "s consid]ation z a possible sump = dispos+ ( "s ( ! dra9age wat] ( ! ,beh]a ,prov9ce ( ! ,delta4 ,s long ago z #ahie ! ,public ,works ,dep>tm5t /udi$ ! possibility ( dra9+ ,lake ,m>iut into ! ,wadi ,natrun1 & ?us do+ away ) ! increas+ co/ ( keep+ d[n ! level ( ! lake by pump+ wat] f x into ! sea4#ac ,! ,wadi ,natrun has an area1 ev5 at ! l[ cont\r ( #aj metres bel[ sea-level1 ( ne>ly #ajj squ>e kilometres1 & ! av]age rate ( evaporation f an op5 wat]-surface )9 x c\ld n well be mu* less ?an #c mm4 p] day2 ! wadi c\ld !re=e dispose annually ( at lea/ #ajj million cubic metres ( dra9age wat] by evaporation alone1 & probably ( a consid]able additional quantity by seepage4 ,b ! high co/ ( cutt+ a dra9age *annel f ,lake ,m>iut to ! wadi1 & ! heavy ma9t5ance *>ges :i* w\ld be incurr$ 9 keep+ ! dra9 op51 7 =midable objections to ! s*eme4 ,f a reconnaissance c>ri$ out 9 #ahif1 ,mr4 ,v]s*oyle f\nd t ! l5g? ( ! propos$ dra9 w\ld h to be almo/ #hj kilometres1 & = ! great] "p ( ! way x w\ld h to be a cutt+ #bj metres 9 dep? ?r\gh des]t & ridges ( drift+ s&2 he rem>k$ t if x 7 an easy matt] to make ! connection1 x w\ld be no easy matt] to ma9ta9 x2 & he conclud$ t ! s*eme 0 an impracticable "o4#ad ,a m rec5t proposal has be5 to con/ruct a p>tly tunnell$ *annel in/ead ( an op5 dra92 b ?is has likewise be5 f\nd to be impracticable1 z involv+ too heavy an outlay = ! b5efits :i* w\ld result4#ae ,illu/ration3 #g4 ,qasr ,f>afra1 se5 f ! nor? ov] pla9 ( p[d]y *alk =m+ floor ( depression ,illu/ration3 #h4 ,esc>pm5t & 8,hattia0 ( ,abu ,mung> f ! s\? ,illu/ration3 #i4 ,! 8,hattia0 ( ,abu ,mung> f ! ea/3 well on e>?-m\nd on "r ,illu/ration3 #aj4 ,! pr9cipal well at ,qasr ,f>afra ,illu/ration3 #aa4 ,a /reet 9 ,qasr ,f>afra1 ) irrigation canal ,:5 ! great ext5t & dep? ( ! ,qatt>a depression 7 e/abli%$ by ,mr4 ,walpole's explorations ( la/ ye>1 hopes 7 imm$iately ent]ta9$ t ! depression might be made to s]ve "s useful economic purpose4 ,any idea ( its be+ ( use z a ,nile res]voir 0 ( c\rse at once cut out1 n only by reason ( its position & di/ance f ! ,nile1 b also by its imm5se size2 = ev5 if we c\ld turn ! :ole ,nile flood into x1 "s tw5ty ye>s or m w\ld be occupi$ 9 fill+ x to a suffici5tly high level1#af & ! loss by evaporation f s l>ge an area w\ld be enorm\s4 ,z a receptacle = dra9age wat] f ! ,delta x appe>$ equally impossible ( consid]ation1 because ( ! great l5g? & dep? ( ! *annels :i* w\ld h to be cut to rea* x4 ,a m reasonable prospect seem$ to be t ( admitt+ sea-wat] f ! ,m$it]ranean into x by means ( a navigable canal f ! ,arabs ,gulf2 ?is w\ld only h to be ab\t #ef kilometres 9 l5g? to rea* ! ne>e/ po9t ( ! depression2 & once ! canal 0 made & ! depression fill$1 %ips might sail almo/ to ,siwa4 ,o!r advantages :i* might accrue f ! =mation ( ?is inl& sea 7 an increas$ humidity ( ! climate along ! ,m$it]ranean ,littoral ( ,egypt1 lead+ to increas$ ra9-crops 9 t region1 ! e/abli%m5t ( a valuable fi%]y1 & a possible increase 9 ! wat]-supply ( ! oases by ! caus+ ( a slight rise 9 ! /atic wat]-level !re4 ,ano!r idea1 :i* rapidly foll[$ ! fir/ "o1 0 to utilize ! depression z a s\rce ( electrical p[] = driv+ pumps by :i* ! dra9age ( ! nor!rn p>ts ( ! ,nile ,delta might be improv$4 ,at fir/ sight ?is latt] prospect look$ a v attractive "o4 ,! salt-m>% :i* cov]s mu* ( ! floor ( ! depression appe>$ sugge/ive ( a =m] sea-connection1 & if we c\ld trace out ?is old connection1 ! cutt+ ( a canal along x might n1 x 0 ?\ght1 be a v exp5sive matt]4 ,moreov]1 x 0 app>5t t evaporation f an inl& sea or lake ( s l>ge an area w\ld keep pace ) q a l>ge influx f ! ,m$it]ranean1 s t if ! influx 7 re/rict$ to su* a quantity z w\ld p]mit ( ! lake-surface be+ ma9ta9$ at a level consid]ably bel[ t ( ! ,m$it]ranean1 p[] c\ld be g5]at$ cont9u\sly4 ,suppose1 = in/ance1 t we ma9ta9$ a p]man5t wat]-level 9 ! lake ( #ej metres bel[ ! sea2 ! e/imat$ area ( ! lake at ?is level be+ #ijjj squ>e kilometres & ! mean evaporation assum$ to be #d millimetres a day1 an influx ( no less ?an #cf million cubic metres ( wat] p] day c\ld be pass$ into ! lake )\t alt]+ its level1 & ?is ) a fall ( #ej metres w\ld suffice !oretically = ! cont9u\s g5]ation ( ov] #bgj1jjj horse-p[]4 ,( c\rse ! lake w\ld gradually get m & m sal9e1 b ! p[] might cont9ue to be ma9ta9$ = v _m ye>s be=e ! lake became z ri* 9 salt z ! ,dead ,sea4 ,b alas6 an inve/igation ( ! nor!rn bord]s ( ! depression %[$ t ! hop$-= traces ( a =m] connection ) ! ,m$it]ranean ,sea d n exi/4 ,! depression is entirely %ut 9 f ! nor?1 ei!r by great cliffs1 or by gr\nd ly+ s high t ! cutt+ ( a canal to ! sea is utt]ly impracticable4 ,( ! #ef kilometres f ! sea-level cont\r ( ! depression to ! coa/ along a l9e runn+ #ab ea/ ( nor? f ,mogh>a ,lake 7:i* is ! easie/ l9e hi!rto f\nd = ! cutt+ ( a canal71 only #af kilometres are at less ?an #ej metres above sea2 #ca kilometres lie at altitudes betwe5 #ej & #ajj metres2 & #i kilometres are above ! #ajj-metre level4 ,an op5 *annel be+ ?us put out ( ! "q1 x 0 next natural to inquire :e!r a tunnel1 or a *annel p>tly op5 & p>tly tunnell$1 c\ld be excavat$ to s]ve = ! conveyance ( sea-wat] into ! depression4 ,b ev5 ?is1 ?\gh p]haps n impracticable1 w\ld be a v co/ly und]tak+4 ,! conveyance ( ! requisite quantity ( wat]1 ev5 at a relatively high velocity 7:i* ( c\rse implies a consid]able slope71 w\ld necessitate a tunnel or tunnels ( v l>ge dim5sions4 ,to convey #cf million cubic metres a day at a velocity ( #e kilometres p] h\r w\ld require a sectional area ( #cjj squ>e metres--i4e4 if two tunnels 7 made1 ea* w\ld h to be #ad metres 9 diamet]4 ,! co/ ( bor+ & l9+ su* tunnels w\ld c]ta9ly be v great1 & x is d\btful :e!r ! value ( ! p[] g5]at$ c\ld ju/ify ! capital exp5diture involv$ 9 ! excavation & o!r works :i* w\ld be necess>y4 ,until m is kn[n ( ! nature ( ! /rata ?r\gh :i* ! aqu$uct w\ld h to pass1 & z to ! l5g? ( tunnell+ & am\nt ( op5 cutt+ :i* w\ld be requir$1 x is impossible to =m ev5 an approximate e/imate ( ! exp5se :i* ! und]tak+ w\ld involve4 ,all t c be said at pres5t is t ! utilization ( ! ,qatt>a ,depression (f]s a possible means ( obta9+ p[] on a l>ge scale = ! dra9age ( ! l[-ly+ l&s ( ! ,delta1 ) at ! same "t a prospect ( improv+ 9 "s measure ! agricultural res\rces ( ! ,m$it]ranean ,littoral2 mu* fur!r inve/igation w h to be accompli%$ be=e any judgm5t c be =m$ z to :e!r su* a project w\ld be an economically s\nd "o4 ,9 =mulat+ any s*eme = improv+ ! dra9age ( ! ,delta1 x is ( c\rse important to consid]1 n only ! mann] 9 :i* ! dra9age wat] c\ld be ultimately dispos$ (1 b also ! modifications :i* w\ld h to be made 9 ! exi/+ dra9s & irrigation canals--modifications :i* w\ld ne$ to be c>ri$ out )\t s]i\s int]ruption to exi/+ agriculture4 ,a s*eme :i* o!rwise appe>$ attractive might easily prove to be impracticable by reason ( ! heavy exp5ses & inconv5i5ces ( ! subsidi>y works :i* w\ld be requir$ to make x effective4 ,illu/ration3 ,,! ,,libyan ,,des]t ,%[+ ,surface-,relief1 ,cont\rs ( ,/atic ,und]gr\nd ,wat]-levels1 ,di/ribution ( ,s&-dunes1 & ,r\tes ( ,pr9cipal ,exploratory ,exp$itions 9 ! ,we/ & ,s\? by ,dr4 ,,john ,,ball ,,! ,,geographical ,,j\rnal ,,july #aibg #d4 ,! ,artesian ,wat] ,supplies4 ,! orig9 ( ! artesian wat] supplies ( ! ,egyptian oases ( ,bah>ia1 ,kh>ga1 ,dakhla1 & ,f>afra 7,siwa seems hi!rto g5]ally to h be5 left out ( consid]ation7 has be5 a mu*-discuss$ "q4 ,"s geologi/s1 myself among/ !m1 h always reg>d$ ! wat] z be+ d]iv$ f ra9fall 9 ! we/]n ,sudan1 fl[+ und]gr\nd 9 p]meable b$s t[>ds ! ,m$it]ranean4 ,o!rs h held ! view t ! oasis wat]s are m]ely ,nile wat] :i* has p5etrat$ m or less lat]ally into ! adjo9+ des]ts4 ,! argum5ts t h be5 urg$ 9 support ( ! =m] view are1 fir/ly1 ! high temp]ature ( ! wat] 9 _m ( ! oasis wells2 &1 secondly1 t ! levels ( ! spr+s & wells are (t5 mu* high] ?an ^? ( ! ,nile 9 ! same latitudes4 ,to ^! argum5ts x has be5 ju/ly repli$ t nei!r ( !m is conclusive2 ! high temp]ature ( ! outfl[+ wat] m]ely te/ifies t x has desc5d$ to consid]able dep?s at "s "p ( its und]gr\nd pa? n v remote f ! po9t ( outfl[1 b really tells u no?+ z to its place ( orig92 & ! high level ( ! spr+s 9 ,bah>ia1 = in/ance1 z comp>$ ) t ( ! ,nile 9 ! same latitude1 might be acc\nt$ = by ! seepage f ! ,nile tak+ place fairly high up 9 ! riv]'s c\rse4 ,!re ! "q rema9$ until #aibe1 :5 ,i 0 able to visit & det]m9e ! positions & levels ( a numb] ( wat]-s\rces f>!r 9 ! int]ior ?an any ( ^? on :i* ! 8,nile0 argum5t 0 bas$4 ,among/ o!r level-det]m9ations1 ,i asc]ta9$ t ! ,%eb well is #bbh metres above sea-level1 & t ,m]ga ,lake1 ly+ f> to ! s\?-we/ 79 lat4 #ai #c1 long4 #bf #ah71 is at an altitude ( no less ?an #eji metres above ! sea4 ,%ortly be=e my t\r 9 ! ,sudan1 ,hassane9 ,bey _h confirm$ ,rohlfs' level ( #djj metres = ! ,kufra wat]-s\rces1 & ,i _h f\nd t ( ,abu ,mung> 7nor?-we/ ( ,dakhla ,oasis7 to be #aag metres4 ,at all ^! places ! wat]-supplies are d]iv$ f und]gr\nd s\rces 9 ! same rock--namely1 ! ,nubian s&/one1 :i* cov]s su* va/ areas 9 ! ,sudan & ,egypt4 ,i _h ?us f\r well-det]m9$ natural wat]-levels at ! corn]s ( a great quadrilat]al ^: sides av]ag$ ov] #ejj kilometres 9 l5g?1 & embrac$ m ?an #bj squ>e degrees ( ! ,e>?'s surface4 ,n[ j z 9 solid geometry ! levels ( any ?ree po9ts det]m9e ! incl9ation ( an oblique plane to ! horizontal1 s on ! ,e>? any ?ree levels w det]m9e ! incl9ation ( a surface to ! geoid 7( c\rse assum+ bo? geoid & surface to h ! same curvature74 ,& on mak+ ! calculation1 ,i f\nd t ,i obta9$ practically ! same degree & direction ( incl9ation = ! natural wat]-surface1 :i*ev] ?ree ( ! f\r kn[n po9ts ,i utiliz$ = ! calculation4 ,9 o!r words1 ,i f\nd t if ,i took1 say1 ! levels ( ,kufra1 ,abu ,mung>1 & ,%eb1 & d$uc$ f !m ! incl9ation ( ! wat]- surface to ! horizontal1 ,i c\ld calculate ! level ( ,m]ga pretty exactly4 ,ext5d+ ! trial1 ,i f\nd t ,i c\ld d ! same ) a fairly close approximation = ! o!r wells 9 ! ,%eb neighb\rhood1 & also = wells 9 ! oases ( ,dakhla & ,kh>ga4 ,! conclusion seem$ irresi/ible t all ! wells consid]$ 7 f$ f a cont9u\s %eet ( und]gr\nd wat]2 & x 0 evid5t t ?is wat] did n come f ! ,nile1 fir/ly because ( ! high level ( ,m]ga1 :i* is above t ( ! swamps ( ! ,bahr el ,ghazal & o!r we/]n fe$]s ( ! ,:ite ,nile1 & secondly because ( ! direction ( ! d[nw>d slope ( ! und]gr\nd /atic wat]-surface1 :i* is f ! s\?-we/1 in/ead ( f ! s\? z x w\ld h be5 _h ! wat] be5 d]iv$ f ! ,nile 9 ! ,bahr el ,ghazal region4 ,! true s\rce ( ! wat] mu/ be some:]e m or less ne>ly on a l9e drawn s\?-we/ f ,dakhla1 = ?is is ! direction ( upw>d slope ( ! und]gr\nd /atic wat]-surface2 & if su* a l9e be drawn on a map ( ,africa1 x w be f\nd to lead t[>ds ! ,erdi & ,enn$i region1 on ! bord]s ( ! ,*ad bas94 ,x is 9 ! highl&s ( ,ea/]n ,erdi & ,enn$i1 !re=e1 t we mu/ look = ! s\rce ( ! artesian wat] ( ! ,egyptian oases4 ,:at is kn[n ( ?is region f ! rec5t explorations ( ,colonel ,tilho l5ds good support to our conclusion4#ag ,x is a b>e & rugg$ s&/one c\ntry1 :]e1 9 spite ( a ra9fall :i* is by no means negligible1 p]man5t wat]-s\rces are scanty1 & :]e1 9 consequ5ce1 !re mu/ be a consid]able absorption ( moi/ure by ! rocks2 & x lies at s high an altitude z to give suffici5t 8head0 = ! absorb$ wat] to p]colate ?r\gh ! por\s s&/ones & ?us to rea* ,egypt4 ,be+ conv9c$ t ,i _h at la/ arriv$ at ! true orig9 ( ! artesian wat]1 ,i next began to ent]ta9 ! idea ( attempt+ to make a map :i* w\ld %[ ! cont\rs ( ! und]gr\nd wat]-%eet1 & f :i*1 9 conjunction ) ! cont\r-map ( ! surface :i* ,i _h already prep>$1 ,i might be able to pr$ict ! dep? ( bor+ requir$ to tap ! und]gr\nd wat] at any po9t 9 ! des]t4 ,b a little consid]ation %[$ t ?is idea 0 an impracticable "o1 by reason ( our ignorance ( ! und]gr\nd geological /ructure ov] ! great] "p ( ! des]t4 ,! und]gr\nd wat] w\ld naturally pass along p]meable s&/one b$s1 (t5 conf9$ betwe5 imp]meable clays above & bel[4 ,& al?\gh ! g5]al /ructure ( ! ,libyan ,des]t is d\btless "o ( simplicity z comp>$ ) t ( o!r p>ts ( ,egypt1 yet we "k1 f obs]vations 9 ! oases & 9 ! ,ow5at region1 t ! b$s are 9 "s places fold$ & fault$1 & t 9 o!rs !y h be5 uplift$ & entirely remov$ by d5udation1 ) ! exposure ( l>ge areas ( ! und]ly+ anci5t cry/all9e rocks4 ,! only p>ts ( ! des]t = :i* ! bor+-dep?s c\ld safely be pr$ict$ w\ld be c]ta9 small areas )9 ! oases1 :]e wells h be5 sunk 9 suffici5t numb]s to give u def9ite in=mation z to ! local und]gr\nd /ructure2 & und]gr\nd-wat] maps ( ^! small areas1 ?\gh !y might usefully sy/ematize our k conc]n+ !m1 w\ld n be ( any use = pr$ictions at po9ts situat$ else:]e 9 ! des]ts4 ,b :ile x 0 ?us impracticable to prep>e maps %[+ ev]y:]e ! dep? at :i* und]gr\nd wat] actually exi/s1 ,i conceiv$ t x might be q possible to prep>e maps %[+ /atic wat]- levels2 t is1 ! levels to :i* ! wat] w\ld any:]e rise hydro/atically :5 once x 0 tapp$ by bor+s4 ,= ! slope ( ! /atic wat]-surface betwe5 kn[n po9ts w be l>gely indep5d5t ( ! und]gr\nd /ructure ( ! int]v5+ c\ntry4 ,ap>t f any physical *anges :i* may /ill be go+ on 9 ! und]gr\nd rocks !mselves ?r\gh geological ag5cies--*anges :i*1 if tak+ place at all1 mu/ be s sl[ z to be negligible except 9 ! c\rse ( c5turies--! only factors :i* c affect ! slope ( ! /atic wat]-surface1 once x has be5 e/abli%$1 are v>iations ei!r 9 ! rate ( supply ( wat] to ! b$s1 or 9 ! rate at :i* x is remov$ f !m4 ,z reg>ds v>iations 9 ! rate ( supply1 x is obvi\s t v>iations 9 ! annual ra9fall ( ! ,erdi & ,enn$i region mu/ cause v consid]able v>iations f ye> to ye> 9 ! am\nt ( wat] receiv$ by ! und]gr\nd b$s4 ,b ! resi/ance ( friction to ! fl[ ( wat] ?r\gh ! pores ( ! s&/ones is s great1 t ! annual oscillations ( pressure mu/ be rapidly damp$ out z ! di/ance f ! place ( influx increases2 consequ5tly ! levels ( ! wat] 9 ! wells ( ! ,egyptian oases 7& ev51 s f> z is kn[n1 t ( ! lake at ,m]ga7 %[ little or no annual v>iation4 ,& ) reg>d to v>iations 9 ! rate ( removal ( wat] f ! b$s 7by outfl[ to ! ,nile1 or to ! sea1 or into lakes :]e9 x evaporates1 or by ! exploitation ( wells & spr+s = irrigation purposes71 ^! *anges1 ?\gh possibly 9 "s cases !y may be progressive1 & 9 re/rict$ localities v s5sible1 c ex]cise b little influ5ce f ye> to ye> on ! g5]al di/ribution ( wat]-pressure )9 ! und]gr\nd /rata4 ,we may !re=e conclude t ! gradi5t ( ! /atic wat]-surface w ev]y:]e h assum$ practically a /eady /ate4 ,unlike ! actual wat] be>+ b$s !mselves1 :i* may be mu* fold$1 ! /atic wat]-surface w 9 g5]al h simple g5tle slopes ev]y:]e 9 ! op5 des]t4 ,9 ! oases1 ( c\rse1 :]e numb]s ( wells yield+ l>ge outputs h be5 bor$ 9 proximity to ea* o!r1 ! /atic wat] surface w be wr9kl$2 b ov] ! va/ bulk ( ! des]t ! cont\rs may be expect$ to be smoo? curves4 ,! diagrammatic section bel[ w1 ,i ?9k1 make cle> ?is po9t ab\t ! g5]al non-dep5d5ce ( ! %ape ( ! /atic wat]-surface on ! geological /ructure3 ,illu/ration3 ,diagrammatic section %[+ t ! /atic wat]-level is l>gely indep5d5t ( ! und]gr\nd /ructure ,9 ! diagram1 ,,fekhg repres5ts a wat]-convey+ /ratum1 fold$ ?r\gh\t its c\rse & fault$ at ,,hk1 b hav+ a g5]al d[nw>d slope f ,f to ,g4 ,a & ,b repres5t po9ts at :i* ! wat] j rises to ! gr\nd-level1 ei!r ?r\gh natural fissures or 9 artificial bor+s4 ,! /raight l9e drawn ?r\gh ,a & ,b repres5ts v approximately ! /atic level at any po9t betwe5 ,a & ,b1 t is1 ! level to :i* ! wat] w\ld rise 9 bores c>ri$ d[n to ! wat]-be>+ b$4 ,a bor+ at ,c1 = in/ance1 w\ld h to g d[n to ,e to tap ! wat]1 b once ! b$ 0 tapp$ ! wat] w\ld rise 9 ! bore z f> z ,d4 ,at ! fault ,,hk1 ! pressure ( ! wat] at ,k w cause x to rise ?r\gh ! cru%$ rock at ! fault-plane & re-ent] ! por\s /ratum at ,h4 ,if !re is a consid]able outfl[ at ,b1 & ! fault-plane is a v n>r[ fissure1 we may expect "s drop 9 ! l9e ,,ab ov] ! fault1 by reason ( ! extra frictional absorption ( head at ?is place4 ,b unless ! ?ickness & degree ( p]meability ( ! fault-rock are m>k$ly diff]5t f ^? ( ! s&/one b$ itself1 ! drop ( pressure w n greatly di/urb ! g5]al slope ,,ab4 ,9 any case x is app>5t t ! level ( ! /atic wat]-surface at any place betwe5 ,a & ,b is capable ( be+ e/imat$ ) a f> clos] degree ( approximation ?an is ! level ( ! wat]-%eet itself4 ,we may !re=e ju/ifiably assume a uni=m gradi5t = ! /atic level betwe5 po9ts at :i* t level is kn[n1 disreg>d+ folds 9 ! /rata2 & ?\gh we _c entirely all[ 9 detail = unkn[n faults & v>iations 9 p]meability1 x mu/ be borne 9 m9d t ! total effect ( all ! unkn[n factors betwe5 any two kn[n po9ts is already automatically all[$ = 9 our data4 ,x is only ! v>iations f uni=mity1 due to ! unkn[n di/ribution ( ! faults & ( ! dep>tures f ! av]age p]meability1 :i* c affect u2 & ^! v>iations & dep>tures are probably b small 9 mo/ ( ! great unexplor$ areas ( ! s\?-we/ ( ,egypt1 :]e ! geological /ructure1 f all we "k ( x1 appe>s g5]ally to be rem>kably uni=m4 ,! fir/ requisite = ! con/ruction ( a map %[+ ! cont\rs ( ! /atic wat]-surface 01 ( c\rse1 a suffici5cy ( well- det]m9$ positions ( po9ts :]e ! /atic wat]-level ( ! artesian supply 0 fairly exactly kn[n4 ,su* po9ts are ! spr+s & wells ( ! v>i\s oases1 ! surfaces ( lakes occupy+ depressions & presumably f$ by und]gr\nd supplies1 & any places on ! ,nile :]e ! riv] taps artesian b$s4 ,9 reg>d to ! wells & spr+s1 x 0 obvi\s t only ^? kn[n to d]ive _! supplies f artesian s\rces c\ld be utiliz$ z giv+ po9ts on ! /atic wat]-surface4 ,?is consid]ation cut out f ! discussion ! spr+s ( ,ow5at & ,ark5u1 :i* are kn[n to be f$ by local ra9fall1 & also ! small wat]-s\rces ( ,kurkur1 ,dungul1 ,nakheil1 & ,a9 ,amur1 :i* occur 9 situations :]e p]colation f occasional local ra9fall seems to be ! only possible s\rce ( supply4 ,& = reasons :i* w appe> pres5tly1 none ( ! wells & spr+s situat$ to ! nor? ( ! ,siwa-,qatt>a-,faiyum *a9 ( depressions c\ld be consid]$ z ent]+ into ! problem4 ,) ^! exceptions1 e wat]-s\rce situat$ )9 ! area ( ! ,libyan ,des]t cov]$ by ! map1 & ^: level 0 kn[n1 0 utiliz$2 b 9 ! great] oases & 9 ! ,wadi ,natrun ! wells & spr+s are s num]\s & s close toge!r t 9 ^! localities x 0 necess>y to select "o or two wells z repres5tatives ( a gr\p4 ,i _h no hesitation 9 includ+ ! wells ( ,siwa & ! ,wadi ,natrun1 because ! temp]ature ( "s ( ! wells & spr+s ( ,siwa1 & ! quantity ( ! output ( wat] at bo? places1 seem to me to af=d conclusive evid5ce ( ! artesian "* ( _! supply4#ah ,! wells ( ! little oasis ( ,lageita1 to ! ea/ ( ! ,nile ne> ,q5a1 7 includ$1 = al?\gh !y are n 9 ! ,libyan ,des]t1 !y mo/ probably d]ive _! supplies f ! same und]gr\nd fl[ :i* fe$s ! we/]n oases4 ,i h includ$ only "o well 9 :i* ! wat] does n rise ne>ly to ! gr\nd-level4 ,t well is "o :i* 0 bor$ by ! ,briti% ,army dur+ ! ,great ,w>1 at a place call$ ,b#f1 "s #dj kilometres to ! ea/ ( ,bah>ia ,oasis4 ,! level ( ! gr\nd at ?is po9t is #aab metres above sea-level1 & z ! wat] 0 /ation>y 9 ! bore at #gh metres bel[ ! gr\nd1 ! /atic level "h is #cd metres above sea4#ai ,! well is said to h yield$ "s #hjj gallons p] h\r )\t ! wat]-level 9 ! bore be+ s5sibly *ang$4 ,! level ( ! well at ,s>ra has rec5tly be5 det]m9$ by ,pr9ce ,kemal el ,d92 b ,i h n includ$ x 9 my data1 because he in=ms me t ! wat]-level fluctuates by #bj metres or m 9 diff]5t ye>s1 :ile ! level ( ! artesian wat] ( ! ,egyptian oases & ,m]ga is v ne>ly con/ant4 ,! inf]5ce ,i draw f ! great fluctuations 9 ! wat]-level at ! ,s>ra well is t x is dep5d5t on p]colation f a m or less local ra9fall r ?an on ! same fl[ :i* fe$s ! ,egyptian oases4 #e4 ,p]man5ce ( ,lakes4 ,9 reg>d to lakes & salt-m>%es1 ! p]man5ce ( ^? occupy+ ! depressions ( ,areg1 ,bahre91 ,sittra1 & ,qatt>a c only1 ,i ?9k1 be adequately expla9$ by reg>d+ !m z f$1 at lea/ 9 "p1 by und]gr\nd supplies com+ into !m f ! s\?4 ,! total area ( ! lakes ( ,bahre91 ,nuemisa1 ,sittra1 & ,mogh>a is ne>ly #bj squ>e kilometres1 & t ( ! salt-m>%es 7sabakha7 is n less ?an #ejjj squ>e kilometres4 ,! depressions are situat$ 9 a region :i* is ne>ly ra9less2 9 ,siwa ! mean annual ra9fall is only ab\t a qu>t] ( an in*1 & t 9 ! depressions f>!r s\?1 su* z ,bahre9 & ,sittra1 is d\btless ev5 small]4 ,! mean daily evaporation f ! lake-surfaces _c well be less ?an "s #d mm41 :i* w\ld mean a l[]+ ( ! lake-levels by evaporation ( #a metres ea* ye> unless !re 0 "s infl[ to make up = ! loss4 ,& ?\gh ! rate ( evaporation f ! salt-m>%es1 area = area1 is d\btless v mu* small] ?an t ( ! lakes1 ! #bej-;fold great] ext5t ( ! m>%es makes x c]ta9 t ! total quantity ( wat] annually evaporat$ f !m mu/ f> exce$ t f ! lakes4 ,x appe>s unlikely t ! loss by evaporation 9 ! lakes & m>%es c be entirely made up f local ra9fall & by seepages f ! nor!rn slopes4 ,! rocks =m+ ! surface ( ! great ,mioc5e plateau1 #bjj metres high1 :i* sep>ates ! depressions f ! sea1 are *iefly lime/ones & clays2 ! b$s are ne>ly horizontal1 b su* slight dips z exi/ are believ$ to be t[>ds ! sea4 ,! av]age annual ra9fall on ! coa/al portion ( ! plateau is ab\t #f in*es2 b x falls (f rapidly inl&1 till x is only ab\t a qu>t] ( an in* ne> ,siwa4 ,! heavie/ ra9fall on ! plateau ?us occurs along a /rip p>allel to ! coa/1 :]e x is l>gely dra9$ (f t[>ds ! sea by ! gullies :i* ind5t ! plateau-edge4 ,( t :i* falls on ! plateau-surface f>!r inl& comp>atively little is absorb$1 ow+ to ! g5]ally non-por\s nature ( ! upp]mo/ rocks2 aft] a heavy %[]1 wat] lies on ! surface 9 %all[ pools = a few days & is soon evaporat$4 ,s imp]vi\s to wat] is ! lime/one 9 ?is region1 t ! ,romans excavat$ *amb]s 9 x to =m res]voirs1 ( :i* _m hundr$s /ill exi/4 ,nor c we ?9k t mu* surface dra9age-wat] f ! c\ntry to ! s\? ev] f9ds its way into ! depressions2 = !re is an almo/ complete abs5ce ( dra9age-l9es ent]+ !m4 ,at ! feet ( ! nor!rn sc>ps ( ! ,qatt>a depression1 & along ! nor?-ea/]n %ores ( ! lakes 9 ! ,wadi ,natrun1 !re are1 inde$1 small spr+s :i* %[ t "s ( ! ra9 fall+ on ! plateau does actually p5etrate ! rocks & escape by seepage into ! depressions4 ,b ! am\nt ( ?is seepage appe>s to be insignificant comp>$ ) ! volume ( wat] :i* mu/ annually disappe> f ! lakes & m>%es by evaporation4 ,a fur!r consid]ation be>+ on ?is po9t is t :atev] may h be5 ! ag5cy by :i* ! depression ( ,siwa 0 =m$1 t same ag5cy almo/ c]ta9ly op]at$ to produce ! o!r depressions ( ! nor!rn *a92 & x seems mo/ unlikely t a connection %\ld h be5 op5$ up ) ! und]gr\nd wat]-be>+ b$s 9 ,siwa1 & n also 9 ! l>g] & mu* deep] depression ( ,qatt>a4 ,x w\ld be a difficult matt] to e/imate ! relative proportion ( ! wat] ent]+ ! depressions by und]gr\nd fl[ f ! s\?1 to t contribut$ by local ra9fall & seepage f ! nor!rn slopes4 ,b t is n necess>y = our imm$iate purpose4 ,x is suffici5t to %[ t !re mu/ be "s influx into ! depressions f ! same s\rce z t :i* supplies ! wells ( ! great] oases1 to e/abli% ! exi/5ce ( t und]gr\nd wat]-connection :i* is all t we ne$ to ju/ify u 9 reg>d+ ! levels ( ! lakes & salt-m>%es z furni%+ u ) po9ts on ! /atic wat]- surface2 & f ! consid]ations m5tion$ above x seems to me c]ta9 t "s influx ( und]gr\nd wat] really does take place4 ,i h also ?\ght x ju/ifiable to include ! ,birket el ,q>un 9 my collection ( /atic wat]-level data1 because al?\gh t lake 0 probably fir/ =m$ by an ov]fl[ ( ! ,nile into ! ,faiyum1 & is ev5 n[ be+ f$ by ,nile wat] ?r\gh ! ,faiyum dra9s at ! rate ( "s #cej million tons a ye>1 !re is a c]ta9 am\nt ( evid5ce sugge/+ t x has "s und]gr\nd wat]-connection ) ! ,qatt>a depression4 ,t evid5ce1 to :i* att5tion 0 fir/ drawn by ,pr(essor ,s*we9fur?1#bj consi/s 9 ! relatively l[ sal9ity 7#ac p] c5t47 ( ! lake1 not)/&+ ! long p]iod ?r\gh :i* x has be5 subject to evaporation & ! fact ( its hav+ %runk to dim5sions v mu* small] ?an x possess$ 9 anci5t times4 ,unless !re has be5 a l>ge und]gr\nd efflux ( salt wat] f ! lake1 x appe>s impossible to acc\nt = its pres5t degree ( fre%ness4 ,9 ,pr(essor ,s*we9fur?'s day1 ( c\rse1 ! exi/5ce ( ! ,qatt>a depression 0 unkn[n1 & x 0 puzzl+ to sugge/ :]e ! salt wat] _h gone to4#ba ,an und]gr\nd leakage f ! ,birket el ,q>un into ! ,qatt>a depression is q conceivable1 = al?\gh ! two places are sep>at$ by "s #bjj kilometres1 !re is a v consid]able fall betwe5 !m4 ,?us ! salt 9 ! m>%es ( ! ,qatt>a depression may possibly h come 9 "p f ! ,birket el ,q>un4 ,! pres5t rate ( dis*>ge ( ! ,faiyum dra9s into ! lake is1 h[ev]1 j suffici5t to make up = an av]age daily evaporation f ! lake-surface ( a little ov] #d mm41 :i* is ab\t ! rate we might expect2 & al?\gh ! level ( ! lake-surface has fall5 "s #e metres s9ce obs]vations ( x 7 fir/ made 9 #ahhf1 x is n[ ne>ly /ation>y2 h5ce x does n appe> likely t !re is mu* und]gr\nd leakage at pres5t4 ,if ! =m] leakage f ! lake took place by lat]al fl[ into por\s /rata ne> its surface1 ( c\rse ! leakage may h be5 arre/$ by ! l[]+ ( ! lake-level uncov]+ ! por\s b$s into :i* x took place2 b ,i ?9k a m likely explanation is t ! leakage occurr$ at or ne> ! b$ ( ! lake1 & has gradually be5 r$uc$ by ! cont9u$ deposition ( ,nile mud on ! lake-bottom1 & by ! dim9ution ( head due to ! fall 9 ! wat]-level4 ,9 reg>d to ! tapp+ ( ! artesian wat]s by ! ,nile1 !re is only "o locality 9 :i* ?is is kn[n to take place2 b ! quantity ( und]gr\nd wat] :i* is !re )drawn by ! riv] is probably v consid]able4 ,:5 ,i 0 survey+ ! ,nile ,valley betwe5 ,aswan & ,korosko 9 ,decemb] #ahih1 ,i obs]v$ t 9 ! vic9ity ( ! temple ( ,dakka 7ab\t #aje kilometres s\? ( ,aswan7 ! l&s on ! we/ bank ( ! riv] 7 be+ irrigat$ ) w>m wat]1 drawn by 8sakias0 7wat]-rais+ ma*9es7 f pits sunk 9 ! alluvial flat :i* ext5ds betwe5 ! riv] & ! edge ( ! s&/one des]t4 ,! l5g? ( ! tract ov] :i* ! w>m wat] 0 be+ )drawn = irrigation 0 f\nd to be ab\t #af kilometres1 /ret*+ f #b kilometres nor? ( ,dakka temple s\?w>ds to ! temple ( ,mah>raga2 & ! wid? ( ! alluvial tract at ,dakka1 :]e x is wide/1 0 ab\t #acjj metres4 ,"s ( ! wat]-pits 7 m ?an a kilometre f ! riv]4 ,levell+ f ! ,nile 7! surface ( :i* 0 !n ab\t #ii metres above sea7 across ! cultivation to "o ( ! sakia- pits #gej metres we/ ( ! riv]1 ,i f\nd ! level ( ! gr\nd at ! sakia-pit to be #gi metres above t ( ! ,nile1 & ! wat]- surface 9 ! pit to be #hd metres bel[ ! gr\nd-level2 !re 0 #ab metres dep? ( wat] 9 ! pit4 ,! temp]ature ( ! wat] 9 ! pit ,i f\nd to be #hc ,f41 :ile t ( ! ,nile 0 #fj ,f4 & t ( ! air 0 #fg ,f4 ,! headman ( ,dakka told me t ! exploitation ( ?is w>m und]gr\nd wat] _h begun ab\t #ahhg2 !y dig out ! s&y mud1 & !n see ! wat] ooz+ rapidly into ! pit out ( ! s&/one bel[4 ,on cross+ to ! ea/ bank ( ! riv]1 ,i f\nd t !re also ! w>m wat] 0 be+ simil>ly rais$ = irrigation1 ?\gh to a small] ext5t1 because on t side ! s&/one des]t approa*es m closely to ! riv] & !re is mu* less cultivable l&4 ,! exploitation ( ! wat] on ! ea/ side ( ! ,nile ext5d$ only ov] a di/ance ( ab\t #e kilometres along ! bank1 ) a maximum wid? ( alluvial pla9 ( #fjj metres1 j at ! place :]e ! great ,wadi ,alagi deb\*es into ! ,nile ,valley4 ,z ! s&/one b$ f :i* ! w>m wat] issues is less ?an #b metres bel[ ! level ( ! wat]-surface ( ! ,nile1 & ! wat] occurs on bo? sides ( ! riv]1 x is c]ta9 t ! wat]-be>+ b$ is cut ?r\gh by ! ,nile *annel itself2 ! seepage into ! riv] along ! /ret* ( #af kilometres mu/ !re=e be v consid]able4 ,x seems evid5t t ! wat] is n d]iv$ f ! b$ ( ! ,wadi ,alagi1 great dra9age-*annel ?\gh t wadi is2 = we c\ld n !n acc\nt = ! temp]ature ( ! wat]1 nor = its appe>+ to a l>g] ext5t on ! we/ bank ?an on ! ea/1 ) ! riv] 9 betwe54 ,moreov]1 ! wat] appe>$ to be mu* m free f salts ?an we %\ld expect x to be if x 7 m]ely dra9age f ! ,wadi ,alagi4 ,x /rongly resembles1 9 fact1 bo? 9 temp]ature & "*1 ! artesian wat] ( ! great] oases1 & !re c h>dly be ! smalle/ d\bt t at ,dakka ! ,nile is n only cont9ually ab/ract+ artesian wat] f ! same und]gr\nd wat]-%eet t fe$s ! oases1 b is ab/ract+ x 9 f> l>g] quantities ?an ^? yield$ by all ! oasis wells & spr+s put toge!r4#bb ,x is c]ta9ly rem>kable t ! place :]e consid]able supplies ( w>m und]gr\nd wat] ent] ! ,nile %\ld co9cide ) ! emb\*ure ( :at is p]haps ! greate/ dra9age *annel ( ! ,ea/]n ,des]t ( ,egypt2 b ,i ?9k x is likely t ! explanation ( ! co9cid5ce may be a tectonic "o2 ! wat]-be>+ b$s may h be5 br\ght up by a local fold 9 ! /rata1 & ! same fold may 9 "s way h condition$ ! =mation ( ! primitive dra9age-l9e :i* 0 ultimately to become ! ,wadi ,alagi4 ,hav+ n[ indicat$ briefly ! gr\nds = _! acceptance1 ,i give bel[ a table %[+ ! v>i\s po9ts :i* ,i h adopt$ z furni%+ data = con/ruct+ ! cont\rs ( ! /atic wat]- surface und]ly+ ! ,libyan ,des]t1 toge!r ) ! altitudes ( ! po9ts above or bel[ sea1 & ! s\rces ( ^! level-data4 ,! levels are d\btless 9 "s cases slightly inaccurate2 b a few metres ( error are immat]ial to ! object 9 view1 & x is believ$ t ev5 ^? levels :i* re/ on b>ometric det]m9ations are suffici5tly accurate = our purpose4 ,,li/ ,,( ,,adopt$ ,,po9ts ,,on ,,! ,,/atic ,,wat]-,,surface ,place4 ,level ,det]m9$ by4 7metres74 ,wadi ,natrun1 surface ( #bc ,ball1 ,trigonometric lakes levell+1 #aiad4 ,birket el ,q>un1 surface ( #de ,survey ( ,egypt1 #aibf4 lake ,bas$ on _s levell+ f ,alex&ria4 ,mogh>a1 surface ( lake #bc ,walpole1 ,trigonometric levell+1 #aibd4 ,qatt>a ,depression1 v>i\s #hjbc po9ts on salt-m>%1 ! l[e/ be+ ,sittra1 surface ( lake #af ,areg1 surface ( lake #be ,siwa #ag ,jaghbub #cb ,hassane91 ,b>ometric obs]vations1 #aibc4 ,jalo #fa ,bir ,butaffal #ih ,el ,h>ra% #caj ,awadel 7,kufra ,oasis7 #dcd ,ezeila 7,kufra ,oasis7 #chi ,bawitti 7,bah>ia ,oasis7 #abi ,ball1 #aiag1 & ,walpole1 #aibd4 ,trigonometric levell+4 ,el ,hez 7,bah>ia ,oasis7 #acd ,b#f ,well 7wat] surface 97 #cd ,walpole1 ,trigonometric levell+1 #aibd1 & milit>y records ( dep?1 #aiaf4 ,lageita 7,ea/]n ,des]t7 #aba ,murray1 ,trigonometric levell+1 #aiba4 ,f>afra #ij ,ball1 ,b>ometric obs]vations1 #aibd4 ,abu ,mung> #aag ,mut 7,dakhla ,oasis7 #aai ,kh>ga 7av]age ( num]\s #gj ,beadnell1 ,_s levell+1 wells7 #aiji4 ,a9 ,ismail 7,kh>ga ,oasis7 #fg ,ball1 ,b>ometric obs]vations1 #aibe4 ,bir ,murr #aef ,bir ,abu ,husse9 #ahb ,bir ,kassaba #agf ,%eb ,well #bbh ,safsaf #bcj ,bir ,t]fawi #bdd ,m]ga1 surface ( lake #eji ,dakka1 wat]-surface 9 #ii ,ball1 ,_s levell+ f wells ! ,nile1 #ahih4 ,to prep>e a map %[+ ! cont\rs ( ! /atic wat]-surface1 ,i took a graticul$ %eet & plott$ ! above-s*$ul$ po9ts on x 9 _! asc]ta9$ geographical positions1 affix+ ! adopt$ level to ea*4 ,to get po9ts on ! v>i\s cont\rs at v]tical int]vals ( #ajj metres1 ,i jo9$ ea* pair ( po9ts on ! map by a p5cil l9e1 & !n1 by int]polation f ! t]m9al levels1 f\nd ! po9ts on ?is l9e :]e ! v>i\s cont\rs cross$ x1 on ! assumption ( a uni=m gradi5t betwe5 ! t]m9al po9ts4 ,_m ( ! l9es ?us drawn ( c\rse cross$ ea* o!r1 s t int]polation ( ! /atic level at ! po9t ( _! int]section gave two values = ! same place4 ,b _! agreem5t 0 wond]fully close1 consid]+ ! fewness & ! scatt]$ nature ( ! datum-po9ts1 & ?is w5t a long way to enc\rage me 9 ! belief t ! hypo!sis on :i* ,i _h be5 work+1 namely1 t ( an und]gr\nd wat]-connection betwe5 all ! po9ts includ$ 9 my li/1 0 correct4 ,i f\nd t ! cont\rs ( ! /atic wat]-surface c\ld be approximately repres5t$ by a s]ies ( smoo? curves1 z %[n 7on a r$uc$ scale7 9 ! outl9e map bel[4 ,ap>t f ! g5]al smoo?ness ( ! curves1 especially 9 ! s\?-we/1 :]e x may 9 "p be due to ! scant9ess ( control- po9ts1 ! mo/ /rik+ ?+ on ?is outl9e map is ! nor?- ea/w>d projection ( ! #ajj-metre /atic cont\r1 :]e x runs out s z to include ,bah>ia ,oasis4 ,! reasons = ?is projection are obvi\sly ! efflux ( wat]1 on ! "o h& nor?-we/w>ds into ! great ,qatt>a depression1 & on ! o!r h& into ! ,nile at ,dakka4 ,! ind5tation ( ! #djj-metre cont\r ne> ,kufra is likewise expla9$ by ! )drawal ( wat] f ! wells ( t oasis4 ,! g5]al p>allelism ( ! curves 9 ! s\?-we/]n "p ( ! map1 %[+ a gradual rise 9 a s\?-we/]ly direction t[>ds ! ,erdi & ,enn$i c\ntry 7:i*1 z ,i h already /at$1 is ! mo/ probable s\rce ( ! und]gr\nd wat]7 is /rik+ly app>5t4 ,i h n be5 able to ext5d ! cont\rs f> to ! ea/ & we/ ( ,m]ga1 = lack ( control-po9ts4 ,x is mu* to be hop$ t "s future travell] w det]m9e ! wat]-levels at ,selima & ,lagia1 :i* w\ld enable ! /atic cont\rs to be ext5d$ into ! region betwe5 ,m]ga & ,dongola2 provid$1 ( c\rse1 t an exam9ation ( ! wat]-s\rces at ^! places proves _! supplies to be artesian4 ,! mo/ effective way ( te/+ any work+ hypo!sis 9 natural sci5ce be+ ! pr$iction ( hi!rto unobs]v$ facts1 ,i v5ture to =eca/ t if1 z is mo/ likely1 ! wat]-s\rces ( ,selima & ,lagia are artesian1 _! levels :5 ev5tually det]m9$ w n be f\nd to diff] v mu* f #bgj & #cij metres above sea respectively4 ,^! are ! approximate levels d$uc$ by prolong+ ! /atic wat]-cont\rs ( my maps into ! localities ( ^! wells1 assum+ ! cont\rs to cont9ue z smoo? curves4 ,illu/ration3 ,outl9e map ( ! ,libyan ,des]t1 %[+ ! po9ts :]e ! /atic wat]-levels are kn[n1 & ! d$uc$ cont\rs ( ! und]gr\nd /atic wat]-surface1 on ! hypo!sis ( a cont9u\s hydraulic connection betwe5 ! po9ts ,ano!r int]e/+ progno/ication :i* ,i ?9k may fairly be d$uc$ f ! map is t if ev] ! well at ,s>ra is consid]ably deep5$1 a mu* m abundant wat]-supply w probably be obta9able4 ,! gr\nd-level at ,s>ra1 accord+ to obs]vations made by ,pr9ce ,kemal el ,d91 is #dfa metres above sea1 & ! wat]-level 9 ! well v>ies 9 diff]5t ye>s f ab\t #cij to ab\t #daj metres above sea4 ,b an exam9ation ( ! /atic cont\rs ( ! map %[s t ! /atic level ( ! true artesian wat] 9 ! neighb\rhood ( ! well is probably some:]e ab\t #ejj metres above sea1 ?\gh an exact e/imation ( ! /atic level at t spot is n possible because ( ! lack ( data f>!r we/4 ,z already rem>k$ on p4 #aaj1 ,i ?9k ! pres5t supply at ,s>ra is d]iv$ f m or less local ra9fall1 convey$ by high]-ly+ p]meable /rata ?an ^? :i* convey ! ma9 artesian supplies ( ,kufra & ! ,egyptian oases2 by deep5+ ! well consid]ably1 l[]-ly+ b$s might be rea*$ s z to tap ! ma9 supply1 & ! wat] might ev5 be expect$ to ov]fl[ at ! surface4 ,int]e/+ z are ! /atic wat]-level cont\rs 9 !mselves1 !y become va/ly m s :5 sup]pos$ on ! gr\nd-cont\rs1 z is done 9 ! l>g] map 7,g4,j41 ,july1 foll[+ p4 #if74 ,f ! two sets ( cont\rs on t map we c e/imate at any po9t ! approximate dep? ( ! /atic wat]-level bel[ ! gr\nd2 & ?is in=mation af=ds new light on "s ( ! mo/ int]e/+1 b hi!rto ! mo/ difficult1 ( ! problems connect$ ) ! ,libyan ,des]t4 ,illu/ration3 ,pott]y ,hill1 possible site ( 8,z]zura10 f ! s\?-we/ ,illu/ration3 ,bir ,kassaba1 a wat]+-place on ! ,d>b el ,arba'9 ,illu/ration3 ,! wat]less /ret* ( ! ,d>b el ,arba'9 betwe5 ,bir ,murr & ,kh>ga ,illu/ration3 ,bir ,%eb1 a well on ! ,d>b el ,ar ba'9 #f4 ,c ! ,pres5t ,wat]-supplies ( ! ,m$it]ranean ,littoral be supplem5t$ by ,artesian ,bor+s8 ,! pres5t wat]-supplies ( ! ,egyptian portion ( ! ,m$it]ranean littoral1 d]iv$ ma9ly f %all[ wells dep5d5t on ! local ra9fall1 are nei!r v abundant nor ( v good quality4 ,at "o or two ( ! mo/ important settlem5ts along ! coa/1 su* z ,matruh & ,sollum1 attempts h be5 made to improve ! supplies by s9k+ wells to a consid]able dep? 9 situations :]e x appe>$ likely t ! dra9age f ! inl& plateau w\ld be specially abundant4 ,b ^! h met ) little success2 ! yield has be5 f\nd to be v mod]ate 9 quantity1 & ( poor quality ow+ to dissolv$ salts4 ,! "q has (t5 be5 rais$ z to :e!r v deep bor+s1 c>ri$ d[n "r ?r\gh ! ,t]ti>y /rata & into ! ,nubian s&/one1 might result 9 ! procur+ ( an artesian supply ( ! same excell5t wat] z occurs 9 ! oases4 ,hi!rto x has n be5 possible to give a def9ite answ] to ?is "q1 & geologi/s h be5 reluctant to recomm5d deep bor+s1 :i* w\ld entail great exp5se1 )\t feel+ "s assurance t !y w\ld be successful4 ,! dep? to ! ,nubian s&/one is unkn[n1 b is c]ta9ly great2 & if bor+s 7 c>ri$ d[n into ! s&/one1 x 0 n kn[n :e!r ! wat] w\ld rise to any?+ l ! gr\nd-level4 ,f our new map1 h[ev]1 we obta9 a v decisive v]dict on ! matt]4 ,! ,nubian s&/one1 ev5 if rea*$1 w\ld n be f\nd to be *>g$ ) artesian wat] "u any?+ l ! pressure t x is 9 ! oases2 leakage into ! ,qatt>a & o!r depressions w h deplet$ ! b$s ( mu* ( ! wat] com+ f ! s\?-we/1 & w h l[]$ ! /atic head to su* an ext5t1 t ! wat] left 9 ! s&/one w h too little pressure to rise f> into ! bores4 ,any idea ( s9k+ deep artesian wells along ! coa/ to tap ! ,nubian s&/one c consequ5tly be def9itely ab&on$4 ,we are driv5 to ! conclusion t 9 any attempt to improve ! local wat]-supplies ( ! littoral settlem5ts1 we c c\nt only on local ra9fall = our prim>y s\rce1 & we mu/ d our be/ to collect ! run-(f be=e x has _h an opportunity to absorb mu* salt4 ,! ,romans evid5tly und]/ood ?is :5 !y excavat$ ! l>ge rock-ci/]ns on ! plateau1 ( :i* !re are hundr$s4 ,we _c d bett] ?an imitate _! example1 & arrange = ! collection & /orage ( a suffici5t volume ( ra9wat] z x runs f ! rocky surface ( ! plateau4 ,we may d ?is by re/or+ to use ! old res]voirs2 or we might possibly a*ieve our end by damm+ "s ( ! rocky gullies :i* br+ d[n ! run-(f f ! plateau to ! pla94 ,n[ t ! artesian idea is %[n to be out ( ! "q1 !re is ju/ification = a ?or\gh inve/igation z to ! be/ me?od ( collect+ & cons]v+ ! local ra9fall4 #g4 ,are ! ,artesian ,wat] ,supplies ( ! ,oases dim9i%+8 ,! native cultivators 9 c]ta9 p>ts ( ,kh>ga & ,dakhla h = "s ye>s pa/ f\nd t _! wells no long] dis*>ge at s high a level z =m]ly1 & 9 consequ5ce "s ( _! l& has gone out ( cultivation4 ,f ?is fact1 & f ! evid5ce ( ! =m] great] prosp]ity ( ! oases :i* is af=d$ by ! v>i\s ru9s ( temples1 =ts1 & villages1 by ! l>ge areas ( =m]ly cultivat$ l&s1 & by num]\s s&$-up wells1 x has sometimes be5 inf]r$ t ! total yield ( ! oasis-wells is n[ b a fraction ( :at x =m]ly 04 ,b1 z ,mr4 ,beadnell has po9t$ out1#bd ! rema9s ( ! pa/ :i* exi/ 9 ! oases belong to successive g5]ations1 s t we _c fairly draw su* a conclusion f !m2 & ! r$uction or cessation ( ! dis*>ge ( c]ta9 wells does n necess>ily imply any fall+-(f 9 ! total wat]-output ( ! oases4 ,mr4 ,beadnell's exp]im5ts on fl[+ wells 9 ,kh>ga h cle>ly %[n h[ ! op5+ ( a new well at a slightly l[] level w affect ! dis*>ge ( an old well1 ev5 "o at a di/ance ( a kilometre or m1 by l[]+ ! /atic head 9 its vic9ity4#be ,& s9ce a l>ge numb] ( wells h be5 bor$ 9 rec5t ye>s bo? 9 ,kh>ga & ,dakhla1 x is mo/ likely t ! dis*>ge f ^! wells has caus$ a fall+-(f 9 ! yield ( old] ones situat$ at slightly high] levels4 ,9 ?is connection x w be well to note t x is n ! m]e exi/5ce ( a new well t affects ! /atic head1 b ! dis*>ge f x4 ,if ! new well is securely clos$ s z to dis*>ge no?+1 x has !n no effect on ! /atic head & !re=e none on ! neighb\r+ wells4 ,b = ?is to hold1 x is important t ! well :i* is clos$ %\ld be clos$ ?r\gh\t its entire dep?2 x is n suffici5t m]ely to close its m\? s t x does n dis*>ge any wat] on to ! gr\nd1 = !re may /ill be rapid leakage some:]e 9 ! bore 7unless effectively cas$7 into por\s unsaturat$ und]gr\nd /rata4 ,ow+ to ! rapid rate at :i* iron pipes are corrod$ 9 ! wells ( ! oases1 leakage ( ?is k9d is m likely to happ5 ) an ab&on$ mod]n well cas$ ) iron pip+ & plugg$ ne> ! top1 ?an ) old wells :i* 7 fill$ up ) clay & s&4 ,^! factors1 ! mutual int]f]5ce ( wells & ! importance ( prev5t+ und]gr\nd leakage1 especially f ab&on$ wells1 be+ n[ ?or\ghly und]/ood1 /eps are be+ tak5 t[>ds ensur+ t future sites = new wells %all be judici\sly select$1 & t leakage & wa/e f ab&on$ wells %all be z f> z possible arre/$4 ,! me?od ( c>ry+ out measurem5ts ( well-dis*>ges 9 ! oases is s inaccurate1 & ! records ( ! pa/ output s defective1 t x is n possible to ga!r f !m :e!r ! total yield ( artesian wat] is at pres5t dim9i%+ or n4 ,mr4 ,beadnell consid]s x likely1 h[ev]1 t ! g5]al av]age wat]-pressure 9 ! oases has be5 v mu* r$uc$ )9 ! hi/orical p]iod1 ow+ to ! long-cont9u$ exploitation ( ! artesian supplies4#bf ,! g5]al /udy ( ! ,libyan ,des]t :i* ,i h made 9 ! la/ few ye>s sugge/s t a gradual r$uction 9 ! /atic wat]-pressure 9 ! oases may possibly h be5 br\ght ab\t by o!r ag5cies ?an ! exploitation ( ! wat] 9 ! oases !mselves4 ,! fir/ & mo/ important ( ^! o!r ag5cies is ! )drawal ( artesian wat] by ! ,nile 9 ! neighb\rhood ( ,dakka4 ,z m5tion$ on p4 #aac1 x is practically c]ta9 t s&/one b$s c>ry+ artesian wat] are cut ?r\gh by ! ,nile along a di/ance ( sev]al kilometres 9 t locality1 & ! influx ( artesian wat] into ! ,nile may f> transc5d 9 quantity t remov$ by ! wells & spr+s ( ! oases4 ,! ,nile has probably deep5$ its *annel 9 ?is region by a few metres )9 hi/orical times1 & ?us cut ?r\gh a great] section ( ! wat]-be>+ b$s4 ,an increase 9 ! sectional area ( ! b$s cut ?r\gh w\ld naturally mean an increase 9 ! quantity ( artesian wat] pass+ into ! ,nile1 & h5ce a l[]+ ( ! /atic wat]-surface ext5d+ p]haps to ! oases4 ,! second possible o!r cause op]at+ to dim9i% ! /atic head ( ! artesian wat] ( ! oases is ! progressive desiccation ( a lake :i* may once h occupi$ a "p ( ! ,qatt>a depression4 ,z m5tion$ on p4 #aaj1 ! floor ( ?is great depression1 l>ge areas ( :i* are #hj metres & m bel[ sea-level1 is p>tly cov]$ by a salt-m>%1 :i* is s s(t & wat]y t x c only be cross$ at a few places4 ,! hundr$s ( great wat]-ci/]ns cut 9 ! lime/ones ( ! plateau to ! nor? ( ! depression--ci/]ns mo/ ( :i* are n[ dry--z well z o!r ru9s along ! coa/ indicative ( a consid]able =m] population1 seem to %[ t ! ra9fall 9 ! littoral region has )9 ! hi/orical p]iod be5 great] ?an x is at ! pres5t day4 ,:5 ! ra9fall 9 ! coa/al region 0 great]1 !re mu/ h be5 m dra9age into ! ,qatt>a depression1 & :at is n[ salt-m>% 0 ?us possibly once a lake ( "s dep?4 ,assum+1 z ,i ?9k is likely1 t an und]gr\nd wat]-connection exi/s betwe5 ! m>% occupy+ ! bottom ( ! depression & ! artesian wat] ( ! oases1 x is obvi\s t any progressive l[]+ ( ! lake-level consequ5t on ! *ange ( climate mu/ h l[]$ ! /atic wat]-surface 9 ! c\ntry ext5d+ s\?w>ds t[>ds ! oases4 ,9 ! oases !mselves ! l[]+ ( ! /atic surface w\ld ( c\rse be mu* less ?an at ! lake2 b x is q conceivable t ev5 9 ! oases ! l[]+ may h am\nt$ to ! few metres :i* w\ld cause "s ( ! old] & high]-ly+ wells to cease to fl[4 ,z bo? ! deep5+ ( ! ,nile *annel 9 ,l[] ,nubia & ! desiccation ( ! ,qatt>a depression are probably /ill sl[ly progressive1 x is possible t ^! causes may to "s ext5t acc\nt = any sl[ l[]+ ( ! /atic wat]-surface 9 ! oases :i* may be /ill go+ on4 #h4 8,lo/0 ,oases--8,z]zura40 ,( all ! que/ions ask$ by int5d+ travell]s 9 ! ,libyan ,des]t1 none is m frequ5t ?an t z to ! mo/ likely :]eab\ts ( undiscov]$ oases1 & especially z to ! possibility ( f9d+ ! my/]i\s 8,z]zura1 or ,oasis ( ! ,blacks40 ,hi!rto ! only aid :i* ,i h be5 able to r5d] to su* inquir]s has be5 to acqua9t !m ) ! v>i\s /atem5ts :i* h be5 made by ,arabs at diff]5t times z to ! situation ( ,z]zura1 ) ! r\tes :i* h be5 foll[$ by o!rs 7includ+ myself7 :o h s\ght 9 va9 = x1 & ) ! indications ( old tracks :i* h be5 enc\nt]$ by ^! previ\s travell]s4 ,s contradictory h be5 ! v>i\s ,arab /atem5ts1 & s num]\s ! va9 attempts to f9d ! place1 t ,i h at times felt almo/ conv9c$ t 8,z]zura0 is a my?4 ,b ,ow5at & ,m]ga 7 little m ?an traditions until a ye> or two ago1 & ,i ?9k !re is a suffici5t possibility ( ! exi/5ce ( undiscov]$ spr+s or oases to enc\rage a fur!r look- out be+ kept = !m1 m especially z a consid]ation ( ! g5]al surface-cont\rs & /atic wat]-cont\rs :i* are n[ available may furni% a new aid 9 ! matt] by n>r[+ d[n ! field ( se>*4 ,z reg>ds ,arab traditions conc]n+ ,z]zura1 ! e>lie/ acc\nt ( !m :i* ,i h be5 able to trace is t ( ,sir ,g>dn] ,wilk9son 9 8 #8,topography ( ,!bes & ,g5]al ,view ( ,egypt1' publi%$ 9 #ahce1 p4 #cei4 ,wilk9son's book is n[ r sc>ce1 & 8 /atem5t conc]n+ ,z]zura is s %ort t ,i quote x 9 full3 8,ab\t five or six days we/ ( ! road f el ,hez to ,f>afra is ano!r ,oasis1 call$ ,wadee ,z]zoora1 ab\t ! size ( ! ,oasis ,p>va1 ab\nd+ 9 palms1 ) spr+s1 & "s ru9s ( unc]ta9 date4 ,x 0 discov]$ ab\t n9e ye>s ago 7i4e4 ab\t #ahbf7 by an ,arab 9 se>* ( a /ray camel1 & f ! foot/eps ( m5 & %eep he !re met )1 !y consid] x inhabit$4 ,g]babo1 ano!r ,wah1 lies six days beyond ?is to ! we/1 & twelve days f ,augela2 & ,taz]bo1 :i* is /ill f>!r to ! we/1 =ms "p ( ! same ,oasis2 & !y suppose t ,wadee ,z]zoora also communicates ) x4 ,! inhabitants are blacks1 & _m ( !m h be5 c>ri$ (f at diff]5t times by ! ,moghreb9s = slaves2 ?\gh ! 8,vallies ( ! ,blacks10 a s]ies ( simil> ,oases1 lie /ill f>!r to ! we/40 ,9 footnotes he adds3 8,x is suppos$ t ! blacks1 :o invad$ ,f>afreh "s ye>s ago1 & kidnapp$ a great numb] ( ! inhabitants1 7 f ?is ,oasis4 8,by ano!r acc\nt ,z]zoora is only two or ?ree days due we/ f ,dakhleh1 beyond :i* is ano!r wadee2 !n a second ab\nd+ 9 cattle2 !n ,gebabo & ,taz]bo2 & beyond ^! is ,wadee ,rebeeana2 ,gebabo is inhabit$ by two tribes ( blacks1 ! ,sim]tayn & ,ergezayn40 ,p>ticul> int]e/ atta*es to ,wilk9son's acc\nt ( ! tradition1 n only because ( its be+ ! e>lie/1 & !re=e less likely to be col\r$ by imag9ation ?an lat] v]sions1 b also because ( ! rem>kable fact t al?\gh "s ( ! o!r places nam$ 9 ! above extract 7 unkn[n to ,european geograph]s at ! "t1 !y h all s9ce be5 discov]$2 8,gebabo0 & 8,taz]bo0 by ,rohlfs1 & 8,rebeeana0 by ,mrs4 ,=bes & ,hassane9 ,bey4 ,! weak po9t 9 ! acc\nt is ! loose ,arab way ( /at+ directions4 ,gebabo 7,kufra7 is n we/ ( ! ,bah>ia-,f>afra road1 b s\?-we/4 ,if we am5d ! be>+ 9 ! fir/ acc\nt accord+ly1 & take ,z]zura at ab\t midway betwe5 ! ,bah>ia- ,f>afra road & ,kufra1 x mu/ lie ne> ! int]section ( ! p>allel ( #bf ) ! m]idian ( #bf1 or "s #bjj kilometres ea/ ( ! ,mehemsa ,hattia1 & "s #aej kilometres nor?-we/ ( ,rohlfs' 8,reg5feld0 camp4 ,9 ! second acc\nt giv5 by ,wilk9son1 ! words 8due we/0 sugge/ t ! be>+ is m c]ta94 ,two or ?ree days 7say #abj kilometres7 due we/ ( ,dakhla w\ld put ,z]zura 9 ab\t latitude #be1 longitude #bg1 or ab\t #cj kilometres nor? ( ,rohlfs' 8,reg5feld40 ,!re is ?us a diff]5ce ( m ?an #ajj kilometres betwe5 ! two positions indicat$ by ! acc\nts giv5 to ,wilk9son4 ,rohlfs evid5tly knew ( ! traditions reg>d+ ,z]zura be=e mak+ 8 attempt to rea* ,kufra f ,dakhla 9 #ahgd2 & s9ce su* an int]m$iate oasis1 if x really exi/$1 w\ld be an imm5se aid to him 9 atta9+ 8 objective1 he made c>eful inquiries conc]n+ x be=e leav+ ,dakhla4 ,b he f\nd t ?\gh e "o 9 ,dakhla knew ! names ( ,z]zura & ,kufra1 no "o c\ld tell him :]e ,z]zura 01 nor h[ f> x lay f ,dakhla4#bg ,evid5tly ,rohlfs plac$ little fai? 9 its exi/5ce1 = dur+ 8 j\rney he gave ! "n 8,z]zura0 to a locality1 ab\t #abj kilometres we/-s\?-we/ ( ,dakhla1 :]e he f\nd no?+ b v sp>sely scatt]$ vegetation4#bh ,as*]son1 :o 0 ) ,rohlfs 9 ,dakhla1 0 h[ev] in=m$ by ,hassan ,eff5di1 "o ( ! pr9cipal inhabitants ( ,mut1 t ab\t a hundr$ ye>s be=e 7i4e4 ab\t #aggj7 !re _h be5 frequ5t raids on ,dakhla by ,arabs f ! s\?-we/2 & t 9 ord] to /op ^! raids ! ,mameluke ,gov]nm5t ( ! "t in/all$ a milit>y colony1 call$ ! ,surbaghi1 9 ! village ( ,qalamun4 ,^! ,surbaghi w5t out & de/roy$ all ! wells = sev5 or eight days' di/ance along ! road by :i* ! raid]s _h come4 ,?is road1 :i* up to t "t _h s]v$ z a trade r\te f ,d>fur1 be=e ! road ?r\gh ,kh>ga 0 op5$1 0 9 consequ5ce =sak54 ,! road 0 said /ill to exi/1 & to lead into ! des]t f ,mut4 ,at a day-&-a-half's j\rney f ,mut !re 7 said to be two 8pill>s10 half an h\r's di/ance ap>t1 :i* ,hassan ,eff5di's gu>d lik5$ to m9>ets4 ,9 ! neighb\rhood !re 7 said to be l>ge /one-qu>ries1 & ab\t ?irty ye>s previ\sly an iron in/rum5t _h be5 f\nd2 ! in/rum5t 0 /ill 9 ,hassan ,eff5di's possession1 & "s wond]ful /ories _h gr[n up z to ! mann] ( its use4#bi ,as*]son himself believ$ ?is in=mation ( ,hassan ,eff5di's to be sub/antially correct1 b he ?\ght ! 8pill>s0 might be m]ely p9nacle-%ap$ natural rocks4 ,i am likewise incl9$ to believe x tru/wor?y1 b ,i ?9k ! 8wells0 may h be5 m]ely wat]- dumps--t is1 collections ( j>s ( wat]4 ,my reason = ?is view is tw(old4 ,9 ! fir/ place1 ! gr\nd-levels 9 t "p ( ! des]t are s f> above ! /atic wat]-level1 t if wells 7 bor$ !y w\ld h to be v deep1 & ! wat] w\ld n rise any:]e ne> ! surface 7!re c1 ( c\rse1 be "h no "q ( local ra9fall z a s\rce = ! wat] 9 ! 8wells072 & 9 ! second place1 x seems v probable t ! accumulation ( l>ge brok5 e>!nw>e j>s :i* ,i discov]$ 9 #aiaf at ! place ,i nam$ 8,pott]y ,hill0 7latitude #bd #bf #bg1 longitude #bg #ch #ed7 is "o ( ! dumps 9 "q4#cj ,! importance ( ! above int]pretation ( ,hassan ,eff5di's /ory1 if accept$1 z ,i ?9k x mu/ be1 lies 9 ! pro( :i* x furni%es t !re _c possibly be an undiscov]$ oasis any:]e ne> a po9t sev5 or eight days f ,dakhla 9 ! direction ( ,kufra4 ,a consid]ation ( ! /atic wat]-levels & ! cont\rs ( ! gr\nd w\ld lead u to ?is conclusion if ! views ,i h advanc$ on ! und]gr\nd-wat] "q are correct2 b ! fact t ! raid]s w\ld nev] h made su* a l>ge wat]-dump if wat] c\ld h be5 readily got f a well 9 ! vic9ity makes ! conclusion almo/ c]ta9 indep5d5tly ( my hypo!sis1 & ?us incid5tally t5ds to support ! latt]4 ,is x possible t 8,z]zura10 :i* ,i am told signifies 9 ,arabic a />l+ 7b is commonly appli$ to any small bird71 is "h a corruption ( "s o!r "n d]iv$ f 8ze]10 a wat]-j>1 & t in/ead ( 8! oasis ( ! blacks10 ,z]zura 0 really 8! wat]-dept ( ! blacks08 ,s*we9fur? has record$#ca a /ory :i* 0 told him 9 ,kh>ga ,oasis1 t 9 #ahgb "s ,arabs ( a ,d>fur c>avan1 :o _h miss$ ! road1 f\nd a small oasis ab\t #a days' j\rney to ! we/ ( ,b]is4 ,! oasis 0 said to conta9 pools full ( wild geese1 date-palms1 & a temple4 ,p :o w5t out ( ,b]is to f9d ! place return$ unsuccessful4 ,i am1 h[ev]1 r incl9$ to ?9k t ! /ory has "s basis ( fact1 because 9 #ahih ,i myself f\nd spr+s & traces ( anci5t conduits 9 ! dunes nor?-we/ ( ,b]is1 & !re may be o!r wat]-s\rces beyond ! po9t :i* ,i rea*$4#cb ,a writ] 9 ! n9? edition ( ,murray's #8,guide to ,egypt1' publi%$ 9 #ahif1 gives ! foll[+ diff]5t /atem5ts ( ,arabs z to ! position ( ,z]zura3 7#a7 ,"s days s\? ( ! ,dakhla ,oasis4 7#b7 ,five days we/ ( ,f>afra ,oasis4 7#c7 ,?ree days we/ ( ,dakhla ,oasis4 7#d7 ,two or ?ree days we/ ( ,selima ,oasis4 ,! fir/ ( ^! /atem5ts w\ld correspond suffici5tly well ) ,bir ,t]fawi1 :i* is ab\t #bhj kilometres due s\? ( ,dakhla ,oasis2 ! second w\ld place ,z]zura ab\t 9 latitude #bg1 longitude #bf2 ! ?ird corresponds ) ! second acc\nt giv5 to ,wilk9son2 :ile ! f\r? w\ld place x f> to ! s\?-we/1 9 latitude #ba1 longitude #bh1 or r less ?an halfway betwe5 ,bir ,t]fawi & ,m]ga4 ,mr4 ,h>d+ ,k+1 :o 9 8 j\rneys 9 ! ,libyan ,des]t 9 #aiji & #aiaa devot$ mu* att5tion to ! collect+ ( native in=mation reg>d+ its geography1 ?9ks t 8,z]zura0 may possibly be only a g5]ic "n appli$ to any my?ical or undiscov]$ oasis4#cc ,he he>d ! "n appli$ to ! foll[+ localities3 7#a7 ,rohlfs' 8,s]sura40 7#b7 ,! 8,egyptian ,oasis10 said to h be5 se5 by an ,arab f ! top ( a high black hill ly+ 9 ! dune-belt t5 long days by ord9>y c>avan f ,kh>ga4 ,ano!r ,arab is said to h se5 :at is possibly ! same place eight days some:]e to ! s\? ( ,dakhla4 ,bo? ^! acc\nts agree t ! place is a l>ge oasis ly+ at ! foot ( a sc>p & conta9+ olive trees4 7#c7 ,a /one temple eighte5 h\rs' j\rney we/ ( ,g$ida 9 ,dakhla ,oasis4 ,! in=mation giv5 to ,mr4 ,h>d+ ,k+ z to ! position ( ! 8,egyptian ,oasis0 w\ld place x ab\t 9 latitude #bc1 longitude #bh1 or only "s #ej kilometres nor?-we/ ( ,bir ,t]fawi2 9 fact1 consid]+ ! vagu5ess ( ! in=mation1 x might correspond fairly well ) ,t]fawi itself1 except t !re are no olive trees at ,t]fawi1 nor is t place ov]look$ by any esc>pm5t or hill4 ,! 8high black hill 9 ! dune-belt0 may j possibly be "o ( ^? ,i mapp$ ne> ! f>!/ po9t ,i rea*$ ) ,moore 9 #aiaf1 9 ab\t latitude #bd1 longitude #bf1 ?\gh ?is w\ld imply a direction s\?- we/ ( ,dakhla1 n s\?4 ,x is n likely to be ,gebel ,kamil1 z t m\nta9 lies mu* ne>] to ,m]ga ?an to ,dakhla4 ,! 8/one temple10 accord+ to ! in=mation giv5 to ,mr4 ,h>d+ ,k+1 w\ld be ab\t #hj kilometres we/ ( ,dakhla ,oasis1 9 ab\t latitude #be1 longitude #bh1 or n v f> f ! second ( ! positions indicat$ by ,wilk9son = ,z]zura4 ,b ,i ?9k x is likely t ! /atem5t really ref]s to ! 8,deir el ,hag>10 a well-kn[n temple ru9 mu* ne>] to ,dakhla4 ,s mu* = ! v>i\s /atem5ts z to :]eab\ts ,z]zura may be4 ,let u n[ inquire :]e x is n4 ,on ! map ,i h indicat$ ! pr9cipal r\tes foll[$ by explor]s ( ! s\!rn & we/]n p>ts ( ! ,libyan ,des]t dur+ ! la/ fifty ye>s4 ,we may be tol]ably c]ta9 t ,z]zura is n on1 or v close to1 any ( ^! r\tes2 = al?\gh a depression )9 a kilometre or s ( "o's track may easily be pass$ )\t notice1 ! exi/5ce ( a l>ge oasis1 su* z mo/ ( ! traditions make ,z]zura out to be1 w\ld almo/ c]ta9ly be betray$ to an explor] by animal-tracks lead+ to x f consid]able di/ances1 except possibly 9 places :]e ! gr\nd 0 v s&y4 ,illu/ration3 ,pott]y ,hill f ! s\?3 at foot1 petrol & wat] supplies ( ,pr9ce ,kemal el ,d9's ,exp$ition ( #aibc ,illu/ration3 ,j>s at nor!rn foot ( ,pott]y ,hill f\nd by ,pr9ce ,kemal el ,d9 9 #aibc ,illu/ration3 ,j>s1 worn away by s&-lad5 w9ds1 f\nd by ,dr4 ,ball 9 #aiag at s\!rn foot ( ,pott]y ,hill ,!re is ano!r me?od n[ available to u by :i* we may n>r[ d[n ! se>* = ,z]zura or o!r 8lo/0 oases2 & t is1 by a consid]ation ( ! g5]al cont\rs ( ! c\ntry & ^? ( ! /atic wat]-surface4 ,e oasis 9 ! ,libyan ,des]t mu/ owe its exi/5ce to "o or o!r ( two conditions3 ei!r x mu/ dep5d on spr+s f$ by local ra9fall1 9 :i* case1 l ! oases ( ,ow5at & ,ark5u1 x probably lies ne> to m\nta9s ( consid]able height2 or else x mu/ dep5d on und]gr\nd supplies1 & mu/ !re=e lie 9 a depression :]e9 ! gr\nd-level & /atic wat]-level are practically co9cid5t1 z 9 ! cases ( ,kh>ga1 ,dakhla1 ,f>afra1 ,bah>ia1 & ,siwa4 ,x is highly d\btful :e!r any m\nta9 masses at all approa*+ ! altitude ( ,ark5u & ,ow5at c rema9 undiscov]$ 9 any ( ! v>i\s areas 9 :i* ,z]zura has be5 traditionally plac$4 ,z]zura is imm5sely m likely to be 9 a depression1 & inde$ ,wilk9son's "n ,wadi ,z]zura almo/ conclusively po9ts to t view4 ,! depression is m likely to be %all[ ?an v deep1 = x presumably lies 9 ! s&/one c\ntry ( ! s\?-we/1 & all ! kn[n depressions ( great dep? are conf9$ to ! lime/one c\ntry ( ! nor?-ea/4 ,x is q easy to trace out on our new map ! areas :]e9 ! /atic wat]-surface w\ld be rea*$ by a depression ( say #ej or #ajj metres bel[ ! g5]al gr\nd-level z indicat$ by ! cont\rs4 ,we mu/1 h[ev]1 rememb] t our gr\nd-cont\rs are liable to be consid]ably 9 error 9 areas :]e obs]vations h be5 few--t is1 9 precisely ^? areas :]e ! depression1 if x exi/s1 is mo/ likely to be f\nd4 ,s we mu/ all[ a good m>g9 = our dep?1 & ,i h *os5 #ajj metres on ?is acc\nt4 ,on ! map ,i h drawn ! 8locus0 ( po9ts 9 ! s\!rn "p ( ! ,libyan ,des]t :]e ! /atic wat]-surface is #ajj metres l[] ?an ! g5]al gr\nd-surface#cd2 & ,i h edg$ ) r$ ! only areas 9 ! we/ & s\? 9 :i* ! two surfaces are )9 #ajj metres ( ea* o!r--t is to say1 ! areas )9 :i* comp>atively %all[ depressions ) und]gr\nd spr+s mu/ lie1 if our cont\rs ( ! two surfaces are drawn ) ev5 approximate correctness4 ,x w be se5 f ! map t ?is re/ricts ! se>* v consid]ably4 ,9 fact1 if ,z]zura is )9 ! fronti]s ( ,egypt1 x lies 9 all probability ei!r to ! we/ ( longitude #bf #bj & nor? ( latitude #bf1 or to ! ea/ ( longitude #bg & s\? ( latitude #bc #cj4 ,we may dismiss f our field ( se>* all ! broad tract ( ris+ gr\nd :i* ext5ds f ne> ! ,dakhla esc>pm5t s\?-we/w>ds t[>ds ,ow5at2 = unless ,z]zura is 9 a depression ( great dep?1 x _c possibly lie )9 ?is tract4 ,?is consid]ation %[s t mo/ ( ! previ\s r\gh e/imations ( ! possible position ( ,z]zura mu/ h be5 errone\s2 & inde$ x is rem>kable h[ few ( ! v>i\s explor]s' tracks trav]se ! only two areas )9 :i* ,z]zura1 if x exi/s1 almo/ c]ta9ly lies4 ,! nor!rn area has only be5 cross$ by ! ,rohlfs exp$ition ( #ahgd1 & t ne> its ea/]n edge2 :ile ! we/]n "p ( ! s\!rn area has only be5 cross$ by ,pr9ce ,kemal el ,d9's exp$ition ( #aibe4 ,( all ! ,arab traditions1 only ^? :i* w\ld place ,z]zura f> to ! we/ ( ,f>afra1 or f> to ! s\?-s\?-we/ ( ,dakhla1 c n[ h any high degree ( cr$ibility4 ,x c h>dly1 ,i ?9k1 be d\bt$ t ! v>i\s traditions ref] to m ?an "o place2 & 9 view ( ! almo/ totally unexplor$ /ate ( ! only two areas 9 :i* our new cont\r-maps indicate ! possibility ( unkn[n oases exi/+1 x is q conceivable t at lea/ "o oasis may lie )9 ea* ( !m4 ,9 ! nor!rn area1 ,siwa w\ld probably make ! be/ />t+- po9t f :i* to conduct a se>*1 z x is easily rea*$ by motor- c> f ,alex&ria or ,cairo1 & exploratory j\rneys s\?w>ds f ,siwa w\ld mo/ly lie along ! direction ( ! dune axes4 ,9 ! s\!rn area1 ,t]fawi or ,%eb w\ld =m ! be/ base = exploration2 at ea* ( ^! places !re is a good wat]-supply4 ,i am incl9$ to ?9k t ! s\!rn area is m likely to yield results ?an ! nor!rn "o4 ,wat] 0 f\nd to exi/ at a po9t "s #ae kilometres we/ ( ,t]fawi1 & ?\gh no o!r s\rce 0 obs]v$ on ! way to ,ow5at1 x 0 impossible to see v f> on ei!r side ( ! track1 s t su* a s\rce might h be5 pass$ )9 a few miles )\t notice4 ,t]fawi itself is v difficult to f9d1 be+ inconspicu\s ev5 :5 "o is fairly close to x2 & z x is 9 a s&y area1 tracks lead+ to x are soon oblit]at$4 ,ano!r factor :i* fav\rs ?is s\!rn area is t1 ! c\ntry be+ all ,nubian s&/one1 ! dep? f ! surface to ! actual wat]-be>+ b$s 7z di/+ui%$ f ! dep? to ! /atic wat]-surface7 is c]ta9 to be mu* less ?an 9 ! nor!rn area1 :]e ,cretace\s /rata may ov]lie ! s&/one2 h5ce !re is m likelihood ( ! exi/5ce ( natural spr+s 9 ! s\? ?an 9 ! nor?4 #i4 ,c ,travel 9 ! ,great ,tracts ( ! ,libyan ,des]t :i* are n[ ,wat]less be facilitat$ by ! ,s9k+ ( ,new ,wells8 ,x has sometimes be5 ask$ :e!r tracts :i* x is n[ difficult or impossible to cross by camel1 ow+ to ! non-exi/5ce ( wells or natural spr+s )9 !m1 might possibly be op5$ up to transport by ! s9k+ ( new wells along a propos$ r\te1 z has 9 fact be5 done by ! ,s5ussi 9 ! case ( ! well at ,s>ra1 betwe5 ,wanianga & ,kufra4 ,hi!rto !re h be5 no data f :i* to =m an op9ion on ?is "q2 b a /udy ( ! new map w enable at lea/ a qualifi$ answ] to be giv54 ,to be a practicable proposition1 a new well mu/ fulfil two conditions3 ! fir/1 t ! bor+ mu/ n be requir$ to desc5d to a v great dep?2 & ! second1 t ! wat] :5 /ruck mu/ rise 9 ! bore to )9 a reasonable di/ance1 say #bj to #cj metres1 ( ! gr\nd-level at ! place4 ,! fir/ ( ^! conditions w be satisfi$ if ! geological horizon ( ! wat]-be>+ b$ is comp>atively ne> ! surface2 ! second1 if ! level ( ! /atic wat]-surface 9 ! locality is )9 #bj or #cj metres ( ! gr\nd- level4 ,z reg>ds ! fir/ ( ^! conditions1 we "k t ! wat]-be>+ b$s und]ly+ ! ,libyan ,des]t are situat$ )9 ! geological =mation call$ ! ,nubian s&/one4 ,we may !re=e elim9ate f our consid]ation all tracts 9 :i* ?is =mation is ov]la9 by any great ?ickness ( y\ng] rocks1 = all ^! rocks w h to be pass$ ?r\gh 9 ord] to tap ! wat]-be>+ /rata4 ,conc]n+ ! second condition1 our map at once in=ms u t ! areas 9 :i* ! /atic wat]-surface is )9 #bj or #cj metres' dep? bel[ ! gr\nd-level are decid$ly limit$4 ,ev5 if we suppose t !re may be unkn[n depressions ext5d+ to a dep? ( #gj metres bel[ ! g5]al level ( ! c\ntry indicat$ by ! cont\r-l9es1 we see t ! localities 9 :i* ! second condition w\ld be fulfill$ are conf9$ to ! tracts edg$ ) r$ on ! map4 ,outside ^! tracts1 n only h we no *ance ( discov]+ oasis-depressions1 b we also h no prospect ( be+ able to bore wells 9 :i* wat] w\ld rise to )9 a reasonable di/ance ( ! gr\nd-level4 ,?us ! broad tract ( ris+ gr\nd :i* ext5ds s\?-we/w>ds f ,dakhla to ,ow5at mu/ always rema9 a wat]less wa/e4 ,9 ! ,egyptian portion ( ! ,libyan ,des]t !re are1 z w be se5 f ! map1 only two areas 9 ! we/ & s\? 9 :i* new wells might successfully be bor$3 "o ext5d+ = "s #cjj kilometres or s to ! s\? ( ,siwa ,oasis 9 ! neighb\rhood ( ! we/]n fronti]1 ! o!r ext5d+ = "s #djj kilometres we/w>d f ! ,nile 9 ! neighb\rhood ( ! s\!rn b\nd>y4 ,=tunately1 h[ev]1 ^! areas are s situat$ t wells sunk )9 !m might be ( consid]able use 9 op5+ up ! out]mo/ p>ts ( ,egypt to exploration4 ,a well ne> ! we/]n fronti] ab\t on ! p>allel ( #bf1 = in/ance1 might make x j possible = c>s or camels to rea* ,ow5at directly f ,siwa1 s9ce x w\ld divide ! pres5t wat]less /ret* ( #hcj kilometres into two /ret*es ( ab\t half t l5g?2 :ile "o or two wells ne> ! s\!rn fronti] betwe5 ! m]idians ( #bg & #bh w\ld r5d] ,ow5at fairly accessible to camels com+ f ! ,nile vi ,t]fawi or ,%eb1 & might be ( great use to explor]s or geologi/s desir\s ( mak+ a detail$ inve/igation ( ! ,ow5at & ,ark5u region4 ,f ! little ,i saw ( ,gebel ,ow5at dur+ my visit to x 9 #aibe1 its geology mu/ be ( extreme int]e/4 ,! s\?-we/]n "p ( ! m\nta9 appe>s to be entirely compos$ ( granite & o!r cry/all9e rocks1 :ile ! ea/]n "p pres5ts huge cliffs ( s&/one1 ) cry/all9e rocks %[+ only at ! foot2 !re is evid5tly a great fault trav]s+ ! m\nta9 mass1 ) a d[n?r[ to ! ea/2 & ! ?ickness ( ,nubian s&/one expos$ on ! ea/]n precipices is great] ?an t at any o!r place ,i h se54#ce ,gebel ,kissu probably resembles ! s\?-we/]n "p ( ,ow5at 9 be+ entirely compos$ ( cry/all9e rocks4#cf ,ea/ ( ,gebel ,ow5at !re are _m igne\s hills1 "s ( :i* exhibit b&s ( a d>k br[n col\r4 ,^! b&s1 ?\gh probably mo/ly igne\s dykes1 may possibly 9 "s cases be ! gozzany outcrops ( ve9s conta9+ metallic m9]als2 ,i _h no opportunity ( exam9+ !m at close qu>t]s1 b ,i saw en\gh to make me long = facilities to und]take su* an exam9ation4 ,x may be rem>k$ t a well 9 ! nor!rn area w\ld probably h to be v deep1 ow+ to ! comp>atively high geological horizon ( ! surface rocks !re2 b 9 ! s\!rn area1 :]e ! ,nubian s&/one =ms ! surface rock1 no great dep? ( bor+ w\ld be likely to be requir$ 9 ord] to tap ! wat]-be>+ b$s4 ,ap>t f ! "q ( new wells 9 ! m remote p>ts ( ! ,libyan ,des]t1 a /udy ( ! map gives u "s h9ts :i* may be ( value %\ld x ev] be desir$ to s9k wells 9 places ne>] to ! oases & to ! well-kn[n tracks4 ,!re appe>s1 = in/ance1 to be no reason :y wells %\ld n be successfully bor$ at int]vals along ! ,d>b el ,arba9 betwe5 ! s\? end ( ,kh>ga & ,lagia1 or on ! s\? side ( ! ,qatt>a & o!r depressions ( ! nor!rn "p ( ! des]t4 ,9 ! s\!rn area1 ! be/ sites to select = wells w be depressions :]e9 ! gr\nd-level approximates mo/ closely to ! /atic wat]-level2 & read+s ( an an]oid b>omet]1 comp>$ ) correspond+ read+s at a place ( kn[n altitude1 su* z ,%eb or ,t]fawi1 w\ld enable ! mo/ fav\rable sites to be det]m9$4 ,obs]vations ( ! geological /ructure w also be ( importance2 = anticl9al folds1 by br++ ! wat]-be>+ /rata ne>] to ! surface1 w\ld dim9i% ! dep? to :i* bor+s w\ld h to be c>ri$ to tap ! b$s2 :ile fault+ might likewise introduce fav\rable conditions by produc+ cracks & fissures along :i* ! wat] c\ld rise4 ,x may be rem>k$ t tam>isk-bu%es are g5]ally a sign t wat] exi/s at no great dep?4 ,9 reg>d to wells 9 or ne> ! nor!rn depressions1 x is obvi\s t sites on ! s\? side ( ! lakes & salt-m>%es %\ld be select$2 = n only is ! artesian /atic level high] 9 ! s\?1 b ! wat] is less likely to be contam9at$ by salts d]iv$ f ! lake & m>%es4 #aj4 ,! 8,tortoise ,m>%es0 ( ,ptolemy4 ,! passage 9 ,ptolemy's #8,geographia' 7lib4 ,,iv4 cap4 #f1 sect4 #d7 9 :i* ! position ( ! 8,tortoise ,m>%es0 is giv5 may be translat$ z foll[s#cg3 ,!re are two great riv]s runn+ into ! ,m$it]ranean2 "o ( !m is ! ,gir1 jo9+ ,m\nt ,us>gala ) ! ,g>amantic n>r[s1 f :i*1 *ang+ its c\rse1 ! riv] is locat$ 9 long4 #db1 lat4 #af1 & makes ! ,tortoise ,m>%es 7,*elonitides ,paludes71 ^: position is long4 #di1 lat4 #bj4 ,! in=mation on :i* ,ptolemy bas$ ?is /atem5t1 at lea/ z reg>ds ! riv] ,gir1 mu/ h be5 v fragm5t>y1 if n1 inde$1 grossly errone\s2 = !re is no riv] fl[+ to ! ,m$it]ranean any:]e ne> ! positions he gives4 ,b !re c be little d\bt t 8 names ,gir & ,*elonitides ,paludes ref] to real places1 & !re has be5 mu* speculation among geograph]s z to _! id5tification4 ,?us1 = in/ance1 ,knoetel#ch sugge/$ t ! ,gir may h be5 ! ,wadi ,dj$i1 to ! s\? ( ,biskra1 & ! ,tortoise ,m>%es ! mod]n ,lake ,melghir 7lat4 #cd1 long4 #f7ci ,dr4 ,william ,smi?#dj ?\ght t ! ,gir 0 really a bran* ( ! ,nig]1 & ! m>%es ! mod]n ,lake ,fittri 7lat4 #ac1 long4 #ah74 ,on ! map ( ,africa on ! #b-million scale publi%$ by ! ,geographical ,s]vice ( ! ,fr5* ,army 9 #ahii ! ,tortoise ,m>%es are depict$ z ly+ 9 ab\t latitude #ai #bj1 longitude #bg1 ) a note /at+ t ?is del9eation is tak5 f an e>li] map ( ! ,nile ,bas9 by ,miani1 :o reg>d$ ! m>%es z be+ connect$ ) ! 8dry riv]0 :i* 0 !n suppos$ to run nor?w>ds ?r\gh ,dakhla ,oasis4 ,colonel ,tilho has lately sugge/$#da ! l[l&s to ! nor?-ea/ ( ,lake ,*ad 7lat4 #ah1 long4 #ag7 z a possible site = ! m>%es2 :ile /ill m rec5tly ,mr4 ,h>d+ ,k+#db has ?\ght t !y might p]haps be id5tifi$ ) ! salt lake ( ,m]ga 7lat4 #ai #c1 long4 #bf #ai74 ,( ! v>i\s localities :i* h be5 sugge/$ z possible sites = ! ,tortoise ,m>%es1 ! only "o :i* ,i h visit$ is ,m]ga4 ,i f\nd ! salt lake at t place to be v small1 cov]+ only "s #aj acres2 x lies at an altitude ( #eji metres above sea-level1 & ?\gh x is situat$ 9 a r wide %all[ depression1 ,i saw no traces ( any ext5sive salt-m>%es ar\nd ! lake1 :ile ! configuration ( ! surr\nd+ c\ntry appe>$ to me to be su* t ! depression _c possibly h =m$ "p ( a cont9u\s dra9age-*annel4 ,i d n "k :e!r x has hi!rto be5 sugge/$ t ,kufra ,oasis may be ! site ( ! ,tortoise ,m>%es2 b on correct+ ,ptolemy's figures = ! errors 9 8 adopt$ position = ,alex&ria & 9 ! l5g? :i* he assum$ = a degree ( latitude1 ,i f9d ,kufra is 9 v mu* clos] agreem5t ) !m ?an any ( ! places nam$ above4 ,9 ,ptolemy's day1 ev5 ! latitudes ( b few places _h be5 a/ronomically obs]v$ 7,alex&ria1 :]e he himself resid$1 0 suppos$ by ,ptolemy to be 9 latitude #ca1 in/ead ( its true #cb #ab71 & z ! only me?od at t "t kn[n = a/ronomically det]m9+ diff]5ces ( longitude 0 by ! obs]vation ( eclipses1 ! numb] ( obs]v$ longitudes 0 small] /ill4 ,! process by :i* ,ptolemy d$uc$ 8 position = ! ,tortoise ,m>%es & o!r places 9 ! int]ior ( ,libya 0 mo/ probably t ( fir/ e/imat+ _! di/ances s\? & we/ ( ,alex&ria f travell]s' it9]>ies1 !n conv]t+ ^! di/ances into degrees ( latitude & longitude1 & f9ally subtract+ ! diff]5ces ?us f\nd f ! latitude & longitude ( ,alex&ria4 ,b ,ptolemy made ! great mi/ake ( assum+ ! l5g? ( a degree ( latitude 7or ( equatorial longitude7 to be #ejj /ades1 in/ead ( ! #gjj /ades :i* x really is4 ,?us all 8 dead-reckon+s result$ 9 diff]5ces ( latitude & longitude :i* 7 too great 9 ! proportion ( sev5 to five4 ,if we correct ,ptolemy's position = ,alex&ria1 & 8 dead-reckon+ = ! errone\s assumption :i* he made reg>d+ ! size ( ! ,e>?1 z foll[s3 ,latitude4 ,longitude4 ,ptolemy's position = ,alex&ria #ca #fj #cj ,ptolemy's position = ! ,tortoise ,m>%es #bj #di ------- ------- ,ptolemy's diff]5ce ( lat4 & long4 #aa #aa #cj ,^! diff]5ces r$uc$ 9 ! proportion ( #e to #g become1 9 true degrees & m9utes #g #eb #h #ab ,! true position ( ,alex&ria is #cb #ab #bi #ec ------- ------- ,:5ce ! correct$ position = ! m>%es 9 our mod]n sy/em becomes #bd #bj #ba #da we get lat4 #bd #bj & long4 #ba #da = ! ,tortoise ,m>%es4 ,comp>+ ?is position ) t ( ,kufra1 we f9d t ! latitude agrees v closely ) ! #bd #ad :i* ,hassane9 obs]v$ at ,taj1 ! pr9cipal village ( t oasis2 :ile ! longitude1 ?\gh m ?an a degree & a half we/ ( t ( ! pr9cipal village1 is almo/ exactly correct = ,tais]bo1 ! nor?-we/]n oasis ( ! ,kufra gr\p4 ,besides ?is rem>kably close agreem5t z reg>ds position1 ,kufra pres5ts sev]al natural features :i* w\ld t5d to support ! view t x may be ! locality :i* ,ptolemy meant by ! 8,tortoise ,m>%es40 ,n only is ,kufra an ext5sive tract ( relatively l[-ly+ gr\nd ) num]\s lakes & salt-m>%es ( v consid]able size1 b x has di/9ctly ! =m ( a valley4 ,hassane9 ,bey repeat$ly ref]s to x z a valley 9 8 description#dc2 ,mrs4 ,=bes also speaks ( ! 8,wadi ( ,kufra10 & m5tions t z a result ( a ride we/w>d f ,taj ! 8wadi0 0 f\nd to h no def9ite end+ to ! we/4#dd ,:at m natural1 !re=e1 ?an t "s anci5t travell] %\ld h imag9$ t ,kufra 0 a s]ies ( m>%es =m$ by a riv] com+ f ! s\?-we/8 ,& is x n possible t ,ptolemy1 9 endeav\r+ to piece toge!r ! scraps ( in=mation he c\ld get1 may h mi/ak5ly inf]r$ t ?is riv] ultimately rea*$ ! ,m$it]ranean1 & also h confus$ ! acc\nt ( x ) ^? ( o!r /reams fur!r s\?1 :i* may h be5 bran*es ( ! ,nig]8 #aa4 ,! ,s&-,dunes4 ,! s&-dunes ( ! ,libyan ,des]t h be5 ! subject ( c>eful /udies by ,mr4 ,beadnell & ,mr4 ,h>d+ ,k+1 & practically all t 0 kn[n conc]n+ !m up to ! outbreak ( ! ,great ,w> is conta9$ 9 ! excell5t pap]s by !m :i* 7 read & discuss$ at meet+s ( ! ,royal ,geographical ,society 9 #aiaj1 #aiaf1 & #aiah4de ,) ! comm5cem5t ( ! ,s5ussi campaign 9 ,egypt 9 #aiae1 ! s&-dunes sprang into new & unexpect$ importance1 f ! fact t !y =m$ "o ( ! pr9cipal h9drances to ! free movem5t ( troops1 & m especially ( milit>y motor transport1 across ! des]t4 ,e l9e ( dunes ( any consid]able ext5t _h n[ to be c>efully mapp$ & exam9$ = possible c>-passages ?r\gh x2 ! result 0 to add greatly to our k ( ! di/ribution & ext5t ( ^! rem>kable features ( ! ,libyan ,des]t1 especially 9 ! nor!rn p>ts1 :]e _m long l9es ( dunes 7 f\nd ( :i* ! exi/5ce 0 previ\sly unsuspect$ by geograph]s1 ?\gh !y are famili> l&m>ks to ! ,b$\9 ( ! region1 & all except ! smalle/ be> ,arabic names4 ,aft] ! cessation ( ho/ilities1 int]e/ 9 ! di/ribution ( dunes 0 ma9ta9$1 because ( ! increas+ use ( motor-c>s 9 place ( camels = transport across ! des]t2 & ?is circum/ance has rec5tly l$ to light be+ ?r[n on ! di/ribution ( ! dunes 9 ^? s\!rn regions :i* lay outside ! field ( op]ations dur+ ! w>4 ,x is r curi\s to note t al?\gh ! s&-dunes =m ! greate/ ob/acle to motor-c>s 9 ! des]t1 yet x is *iefly by means ( motor-c>s t ! true ext5t & di/ribution ( ! s&- dunes has be5 asc]ta9$4 ,dunes are ! mo/ difficult ( all des]t features to map prop]ly by ord9>y reconnaissance me?ods ) camel transport4 ,_! smoo? outl9es provide no po9ts on :i* int]sections c be made1 & no survey m>ks put on !m w rema9 9 place = m ?an a few h\rs1 or at mo/ a few days2 !y occur mo/ly 9 ne>ly level c\ntry1 :]e x is impossible to f9d a /ation :5ce !y c be ov]look$2 ! abs5ce ( %ad[s on !m r5d]s x impossible to say :e!r "o is look+ at a s+le l9e ( dunes1 or at sev]al l9es1 miles ap>t1 "o beh9d ! o!r 9 e*elon4 ,! only sure way ( mapp+ dunes is to trav]se _! sides along _! entire l5g?1 & ?is is impracticable ) camels ow+ to ! enorm\s di/ances :i* w\ld h to be cov]$ )\t wat]4 ,b ) motor- c>s "o c run alongside !m at #dj kilometres an h\r in/ead ( ! camel's #d1 & _! di/ribution c ?us be rapidly & easily asc]ta9$4 ,illu/ration3 ,pr9ce ,kemal el ,d9's exp$ition ( #aibe leav+ ,bir ,t]fawi = ! wat]less j\rney ( #bej miles to ,ow5at ,illu/ration3 ,! well at ,bir ,t]fawi 9 ! palm clump ,illu/ration3 ,jebel ,ow5at f ! s\?3 ! triple peak 7#efce feet7 9 c5tre ,illu/ration3 ,! lake at ,m]ga locat$ by ,pr9ce ,kemal el ,d9's exp$ition ( #aibe ,on ! map ,i h %[n ! di/ribution ( s&-dunes z f> z x is kn[n at ! pres5t day4 ,a comp>ison ( ?is late/ map ) t giv5 by ,mr4 ,beadnell 9 8 pap] ( #aiaj w %[ h[ l>ge is ! numb] ( dunes discov]$ & mapp$ 9 m rec5t ye>s2 & x w be notic$ h[ pron\nc$ is ! con/ancy ( ! g5]al direction ( all ! l9es 9 ! ,egyptian portion ( ! ,libyan ,des]t4 ,?\gh no new l9es ( dunes comp>able 9 ext5t ) ! great ,abu ,moh>ik belt h be5 add$ to ! map1 "s ( ! newly discov]$ l9es are ( v consid]able l5g?1 & !y %[ an ev5 m rem>kable ratio ( l5g? to bread? ?an ! ,abu ,moh>ik belt2 ! ,ramak dunes1 = in/ance1 ext5d f ne> ,mogh>a s\?-s\?-ea/w>ds 9 a /raight l9e = m ?an #ajj kilometres1 ) only a s+le small break1 & _! wid? n[h]e exce$s #a kilometre4 ,moreov]1 ! l9e> arrangem5t 9 a direction f ab\t #bj we/ ( nor? to ab\t #bj ea/ ( s\? is ev5 m pron\nc$ 9 ! newly discov]$ nor!rn dunes ?an x is 9 ^? previ\sly kn[n f>!r s\?4 ,9 ! 8great s& sea0 ( ,rohlfs1 to ! we/ ( ,dakhla & ,f>afra1 ! same directional arrangem5t ( ! individual l9es ( dunes is v noticeable1 at lea/ 9 ! "p ne> 8,reg5feld0 :]e ,i h p5etrat$ x2 & ,i f\nd ! same g5]al direction to hold = mo/ ( ! l9es ( dunes :i* ,i cross$ on ! way f ,t]fawi to ,ow5at4 ,dur+ ! w> ,i receiv$ a numb] ( reports ( dunes ext5d+ ne>ly ea/ & we/1 (t5 9 curv$ l9es2 b inve/igations on ! spot %[$ practically all ^! reports to be errone\s1 ! commone/ mi/ake hav+ be5 t ( sight+ diff]5t l9es ( dunes f diff]5t places "u ! impression t !y 7 a s+le "o4 ,! g5]al s\?-s\?-ea/]ly direction :i* is s m>k$ a feature ( ! l9es ( dunes 9 ! ,egyptian portion ( ! ,libyan ,des]t is n1 h[ev]1 pres]v$ 9 ! extreme s\?-we/ ( ! c\ntry4 ,! l9e ( dunes :i* crosses ! ,egypt-,sudan fronti] 9 longitude #be #dj has a direction ( ab\t #cj we/ ( s\?1 & ! dunes :i* ext5d f ne> ,gebel ,ark5u to ! we/ ( ,gebel ,ow5at run ab\t #ch we/ ( s\?4 ,f a po9t ab\t #hj kilometres to ! s\?-we/ ( ,gebel ,ow5at1 ! dune-l9es make a /ill fur!r di/9ct b5d to ! we/1 *ang+ _! direction by "s #bj to ab\t #eh we/ ( s\?1 & ?is latt] direction is ma9ta9$ by ! dunes 9 ! neighb\rhood ( ,s>ra ,well4#df ,accord+ to ! maps ( ,rohlfs & ,mrs4 ,=bes1 ! dunes 9 ! s&y tract to ! nor?-we/ ( ,kufra foll[ approximately ! same be>+ z ^? ne> ,s>ra4 ,! sugge/ion has be5 made t ?is sw++ r\nd ( ! direction ( ! dunes 9 ! s\?-we/]n "p ( ! ,libyan ,des]t may be due to a deflection ( ! prevail+ w9d by ! m\nta9-masses4 ,x is almo/ c]ta9ly caus$ by diff]5ces 9 ! prevail+ w9ds--9 fact1 ! dune-l9es probably af=d a v exact index to ! g5]al w9d- direction 9 ! areas :]e !y occur--b ,i am incl9$ to ?9k t ! regional di/ribution ( atmosph]ic pressure1 r ?an local deflection by m\nta9 masses1 is ! cause ( ! diff]5ces ( w9d- direction4 ,! dunes1 especially ^? ne> ,s>ra & ,kufra1 ext5d into localities :i* seem to me to be too di/ant f ! m\nta9s = ! influ5ce ( ! latt] on ! prevail+ w9ds to af=d a satisfactory explanation4#dg ,9 c]ta9 areas1 especially 9 ! s\?1 z = in/ance on ! ,arba9 road f ,kh>ga to ,%eb1 ar\nd ,pott]y ,hill1 & 9 ! region betwe5 ,ow5at & ,erdi1 !re are s&-cov]$ tracts :i* c h>dly be mapp$ z dunes1 because !re are no v m>k$ cre/s4 ,^! tracts are easily cross$ by camels1 b give mu* tr\ble ) c>s1 & w\ld probably be class$ z dunes by c>-driv]s2 b !y are really only va/ undulat+ fields ( drift$ s&4 ,on ! old ,d>b el ,arba9 slave road betwe5 ,kh>ga & ,%eb1 "o travels = m ?an #ajj miles ov] drifts ( ?is k9d4 ,no footpr9t ( camel m>ks ! track1 footpr9ts be+ oblit]at$ 9 an h\r or two4 ,b "o is nev] 9 d\bt ab\t ! way1 = x is indicat$ by ! skeletons ( c\ntless camels :i* h p]i%$ on ! we>y m>*4 ,z reg>ds ! relation betwe5 dunes & ! relief ( ! gr\nd1 !re is a gradual rise 9 ! g5]al level ( ! c\ntry f nor? to s\? ( ab\t #a 9 #ajjj1 s t ! s\!rn ends ( mo/ ( ! ,egyptian l9es ( dunes lie at high] altitudes ?an ! nor!rn ends2 b ,i am fairly c]ta9 t ?is slight g5]al incl9ation has no?+ to d ) dune-=mation4 ,! /raighte/ & cleane/-cut l9es ( dunes are f\nd 9 ! flatte/ & mo/ op5 p>ts ( ! des]t1 ov] :i* ! w9d c sweep )\t ob/ruction4 ,:5 a l9e ( dunes enc\nt]s 9 its s\?w>d progression a sudd5 fall 9 ! gr\nd- level1 z = in/ance at ! nor? end ( ,kh>ga ,oasis & to ! ea/ ( ,pott]y ,hill1 x usually breaks up2 sometimes1 z 9 ,kh>ga1 to resume its c\rse z dunes 9 m op5 =mation1 & at o!r times1 z ne> ,pott]y ,hill1 to spread out into a roll+ pla9 ( drift$ s&4 ,x is n uncommon = ! s\? end ( a l9e ( dunes to split up1 ev5 on level gr\nd1 & ! same c be se51 ?\gh less (t51 at ! nor!rn ends2 b ?is is probably m]ely because ! dunes are l[] ne> ! ends ( ! l9es1 & ! wid? ( ! individual dunes is consequ5tly less4 ,! great l9es ( dunes are probably ext5d+ s\?w>ds at a v sl[ rate1 "u ! action ( ! prevail+ w9d4 ,9 ?is connection x is int]e/+ to note t ,arabs always speak ( ! s\? end ( a l9e ( dunes z its head1 & ( ! nor? end z its tail2 j ! opposite ( :at "o w\ld in/9ctively call !m f m]ely look+ at ! map1 b correct if ! dunes are kn[n to h a s\?w>d progression4 ,mr4 ,beadnell f\nd ! av]age rate ( progression ( isolat$ small dunes 9 ,kh>ga to be #ae metres p] annum2 b he rem>ks t l>ge dunes move m sl[ly ?an small ones1 ow+ to ! great] mass ( mat]ial to be transport$4 ,x is int]e/+ to make a r\gh guess at ! antiquity ( ! dunes f ! l5g? ( c]ta9 ( ! long] l9es & an e/imat$ rate ( progression4 ,! ,abu ,moh>ik belt1 = in/ance1 has a total l5g? ( ab\t #cej kilometres2 h5ce1 if its tail has rema9$ /ation>y & its head has advanc$ s\?w>ds at an av]age rate ( #aj metres a ye>1 "s #ce1jjj ye>s mu/ h be5 occupi$ 9 its =mation4 ,z reg>ds ! height ( ! dunes1 ! only "o ,i h c>efully measur$ is t close we/ ( ,rohlfs' camp at ,reg5feld1 :i* ,i f\nd to be exactly #cj metres4 ,"s ( ! dunes 9 ! ,abu ,moh>ik & ,ramak belts are consid]ably high]2 b ,i d\bt if any ( ! ,libyan dunes rise mu* ov] #ej metres above gr\nd-level1 except 9 ! 8great s& sea0 :i* comm5ces "s #ad kilometres to ! nor?-we/ ( ,reg5feld1 :]e ,jordan e/imat$ !m to atta9 #ajj metres or m4#dh ,! top ( a high dune to ! s\? ( ,melfa ,oasis 0 f\nd by triangulation to be #agb metres above sea2 ! level ( ! gr\nd at its foot is unkn[n1 b is probably n v mu* above ! sea-level1 s t ,jordan's e/imate ( ! height ( "s ( ! dunes 9 ! 8s& sea0 is probably by no means an exagg]at$ "o4 ,mu* has be5 writt5 ab\t eddies 9 connection ) dune-=mation4 ,! only obs]vations ,i h made 9 ?is connection h be5 :5 endeav\r+ to place tempor>y m>ks on dunes to =m survey-po9ts2 & ! obs]vations seem to %[ t if eddies are artificially creat$ at ! top ( a dune1 a rapid l[]+ ( ! dune-cre/ results4 ,at fir/ ,i us$ to ?ru/ a walk+-/ick or rang+-rod "s #b or #c feet into ! dune-cre/4 ,! /ick or rod1 ?\gh q firmly fix$ at fir/1 0 always f\nd aft] a few h\rs ly+ halfway d[n ! dune-slope4 ,! same ?+ happ5$ ) tripods firmly press$ d[n into ! cre/4 ,:5 ,i 0 c>ry+ out a little triangulation to det]m9e ! position ( ,rohlfs' cairn at ,reg5feld 9 #aibd1 ,i _h occasion to place a m>k on a high dune1 & ,i us$ an empty #d-;gallon petrol t9 fill$ ) s&1 b$d+ x well into ! dune2 b ! next day x _h disappe>$ f ! cre/1 & 0 f\nd1 z ! o!r ?+s _h be51 halfway d[n ! slope4 ,^! obs]vations rem9d me ( ,mr4 ,b>clay's description ( ! me?od ( deal+ ) s&- accumulations 9 ! ,p]uvian ,des]t1#di :]e1 z soon z dunes appe> & ?reat5 to ob/ruct a railway1 ! local inhabitants turn out & scatt] pebbles & /ones on ! dunes1 & v soon ! dunes are gone1 hav+ be5 c>ri$ away ow+ to eddies produc$ by ! w9d ar\nd ! /ones4 ,?is device = remov+ incipi5t dunes might1 ,i ?9k1 be tri$ along ! ,kh>ga ,railway1 :]e ! scre5s put up at ,mr4 ,h>d+ ,k+'s sugge/ion h n be5 entirely successful 9 prev5t+ ! encroa*m5t ( s& on ! l9e4 ,i am ! m dispos$ to ?9k an exp]im5t ( ?is k9d might succe$1 because ,i h notic$ t at ! gap 9 ! ,ramak dunes 7discov]$ by ,col4 ,p>tridge & kn[n dur+ ! w> z 8,p>tridge ,gap07 !re are great lumps ( silicifi$ wood1 :i* may h l$ to ! gap be+ =m$4 ,!re are two o!r physical *>act]i/ics ( dunes :i* ,i d n rememb] hav+ se5 m5tion$4 ,"o ( ^! is _! l[ !rmal conductivity4 ,on a hot day1 "o has only to ?ru/ "o's h& a few in*es into a dune to enc\nt] cool s&1 :ile on a bitt]ly cold day "o c w>m "o's h&s by ! same action2 ?us :ile ! diurnal v>iation ( temp]ature ( ! surface ( a dune is frequ5tly v great1 x is almo/ ins5sible at q a %all[ dep? 9 ! int]ior4 ,x foll[s t diurnal expansion & contraction ( ! gra9s compos+ ! dune mu/ be conf9$ to ! sup]ficial lay]s1 & ! movem5ts !reby caus$ may mat]ially assi/ 9 consolidat+ ! dune1 by caus+ a 8pack+0 ( ! gra9s4 ,! o!r *>act]i/ic ,i h notic$ is t ! surfaces ( ! dunes always appe> to be h>d] und]foot 9 ! cool ( ! e>ly morn+ ?an 9 ! heat ( ! day4 ,?is may1 ( c\rse1 be an illusion ow+ to "o's be+ less s5sitive to fatigue 9 ! e>ly morn+1 :5 "o's body has be5 refre%$ by sleep1 & :5 ! air-temp]ature is com=table2 b ,i h (t5 fanci$ !re is m 9 x ?an t1 & x may easily be t ! 8pack+0 :i* has tak5 place ow+ to ! consid]able fall ( temp]ature 9 ! night1 to say no?+ ( ! action ( dew1 may h caus$ a real compact+ ( ! surface lay]s4 ,9 8 pap] ( #aiaj1 ,mr4 ,beadnell consid]$ t ! s& compos+ ! dunes mu/ be d]iv$ f ar5ace\s =mations to ! nor?1 & he made special m5tion ( ! ne$ = a c>eful survey ( ! region 9 :i* ! dunes orig9ate4 ,a ref]5ce to ! map w %[ h[ correct 0 ,mr4 ,beadnell's surmise1 at lea/ 9 ! case ( ! ,egyptian dunes :i* he _h /udi$4 ,! dunes orig9ate 9 ! great depressions :i* /ret* f ,siwa to ! ,wadi ,natrun1 :]e1 z ,mr4 ,beadnell inf]r$1 ! supply ( ar5ace\s mat]ial f ! loosely compact$ ,mioc5e & ,oligoc5e b$s is almo/ inexhau/ible4 ,we h "h "o ( ! cle>e/ possible examples ( ! p[]ful action ( ! w9d z an excavat+ & transport+ ag5t4 ,! ,qatt>a depression has be5 excavat$1 l>gely by ! w9d1 to a dep? ( ov] #acj metres bel[ sea- level1 & ! excavat$ mat]ial has be5 c>ri$ s\?w>ds1 "s ( x = ne>ly #ajjj kilometres 9 di/ance & m ?an #ejj metres 9 height1 to =m ! dunes4 ,we see n only ! excavat$ holl[1 b also ! transport$ & pil$-up ar5ace\s mat]ial remov$ f x2 & ! total quantity ( rock ?us remov$ & re-deposit$ mu/ am\nt to hundr$s ( cubic kilometres4#ej ,x is noticeable1 h[ev]1 t to ! nor? ( ! 8great s& sea0 ( ,rohlfs1 :i* ext5ds = an unkn[n di/ance we/w>d1 ! kn[n depressions are mu* %all[]2 & yet ! total quantity ( s& 9 ?is region is probably great] ?an t ( all ! m ea/]ly dunes put toge!r4 ,! c\ntry betwe5 ,j>abub & ,jalo is h[ev] almo/ unexplor$1 ?\gh x is kn[n to be v s&y & to be ! !atre ( viol5t s&/orms4 ,:e!r simil> deep depressions to t ( ,qatt>a h be5 excavat$ along1 or to ! nor? (1 ! ,j>abub-,jalo road1 or :e!r ! area ( loosely compact$ ar5ace\s rocks expos$ !re is s enorm\s z to h yield$ su* a quantity ( s& )\t any v deep depressions be+ =m$1 is a po9t t c only be cle>$ up by fur!r exploration 9 ,cyr5aica4 ,we n[ come to ! mo/ puzzl+ ( all ! que/ions connect$ ) ! s&-dunes1 namely1 :y d !y p]si/ z %>ply def9$ long n>r[ l9es 9 c]ta9 places1 :ile ! gr\nd else:]e is almo/ absolutely free f s&8 ,z ,mr4 ,h>d+ ,k+ puts x1 ! dunes appe> to h a curi\s p[] ( collect+ all ! s& 9 ! neighb\rhood4#ea ,! %elt]+ effect ( ripples 9 ! dunes1 & a 8%eph]d+0 effect ( w9ds a po9t or two (f ! normal1 h be5 sugge/$ z possibly af=d+ "s explanation ( ?is app>5t collect+-p[]1#eb b nei!r ( ^! sugge/ions appe>s to be adequate4 ,i ?9k a m direct explanation is possible2 namely1 t ! dunes really h ! p[] ( attract+ s& f ! air1 & t ?is p[] is due to ! well-kn[n law ( attraction betwe5 a conduct+ surface at z]o pot5tial & an electrically *>g$ body4 ,t air-borne s&-gra9s 9 ! neighb\rhood ( a dune may c>ry electric *>ges 0 %[n by ,mr4 ,h>d+ ,k+'s exp]im5ts1 record$ 9 8 pap] ( #aiaf1 & ,i h s9ce f\nd s&-gra9s to c>ry electric *>ges ( high voltage dur+ s&/orms 9 ,cairo4 ,z to h[ ! gra9s become *>g$ ) electricity1 x c h>dly be due to friction aga9/ ea* o!r1 = ! s& is rem>kably uni=m 9 composition1 consi/+ g5]ally ( r\nd$ gra9s ( almo/ pure silica coat$ ) a ?9 transluc5t film ( f]ric oxide1#ec & = ! production ( frictional electricity x is g5]ally reg>d$ z necess>y t ! two bodies :i* are rubb$ toge!r %\ld be ( dissimil> mat]ials2#ed nor is x likely t !y are electrifi$ by friction ) ! air1 = !y travel ) x4 ,b if ! p>ticles are c>ri$ into an upp] air-/ratum1 !y w gradually acquire its pot5tial by conduction2 & ! ord9>y pot5tial gradi5t 9 ! air 9 ,egypt is ov] #ajj volts p] metre ( height4#ee ,h5ce a s&-p>ticle c>ri$ along = "s "t 9 ! air at a height ( only #bj metres above ! gr\nd may become *>g$ to #g electro-/atic units ( pot5tial1 or #bajj volts2 & on approa*+ a dune ris+ to ab\t t height x w be attract$ to ! dune1 :i* be+ 9 connection ) ! e>? is at z]o pot5tial4 ,! attraction w be v small unless ! s&-gra9 approa*es v closely to ! surface ( ! dune2 f "s calculations ,i h made1 a sph]ical gra9 ( silica half a millimetre 9 diamet] *>g$ to a pot5tial ( #bajj volts w n be attract$ ) a =ce equal to half its own weight until x approa*es )9 a c5timetre ( ! conduct+ surface1 & ! gra9 w h to approa* )9 #g millimetres ( ! surface be=e ! attraction is equal to ! weight ( ! gra94 ,b ! important ?+ to note is t ! attraction is indep5d5t bo? ( ! sign ( ! *>ge & ( ! direction ( ! w9d :i* c>ries ! p>ticle4 ,ano!r way ( look+ at ! matt] is to consid] ! probable di/ribution ( ! equipot5tial surfaces 9 ! air ab\t a dune1 z 9 ! diagram bel[1 :i* repres5ts a section tak5 p]p5dicul> to ! direction ( ! w9d & to ! axis ( ! l9e ( dunes3 ,illu/ration3 ,diagram %[+ sugge/$ di/ribution ( equipot5tial surfaces & directions ( l9es ( =ce 9 ! neighb\rhood ( a s&-dune ,! equipot5tial surfaces1 %[n by dott$ l9es1 w be squeez$ toge!r ov] ! dune1 & ! electric =ces1 %[n by full l9es & arr[s1 be+ ev]y:]e normal to ^? surfaces1 w conv]ge to ! dune4 ,?us ! dune w attract any p>ticles c>ri$ by ! w9d & *>g$ by conduction to ! pot5tial ( ! air convey+ !m4 ,if ! w9d *anges its direction1 ! attraction w /ill p]si/ unalt]$4 ,ano!r po9t wor? m5tion is t a s&-dune may be f> m effective 9 its attraction ( s& ?an a rocky ridge ( ! same size & %ape w\ld be2 = :5 ! electrifi$ gra9s approa* ! surface ( ! dune1 un*>g$ gra9s c rise f ! dune to meet !m2 possibly "s ( ! danc+ ( s&-gra9s :i* is frequ5tly obs]vable along ! cre/ ( a dune dur+ a w9d may be due to ?is cause4#ef ,dur+ viol5t s&/orms1 p>ticles ( s& may be :irl$ al(t & kept susp5d$ at high altitudes = a p]iod long en\gh = !m to acquire v high pot5tials1 & a sudd5 cessation ( ! /orm may cause !m to fall s rapidly t mu* ( ! *>ge may be reta9$2 9 :i* case1 ( c\rse1 electrical attraction may cause a v consid]able deviation f ! purely gravitational pa?s ( ! fall+ p>ticles1 ! deviation be+ always t[>ds ! dune4 ,s9ce ! clos5ess ( ! equipot5tial surfaces & ! mobility ( ! s&-gra9s compos+ ! dune are greate/ along ! dune-cre/1 ! maximum rate ( deposition w be at ! top ( ! dune1 :i* is also ! place :]e ! w9d ex]ts its greate/ action 9 convey+ ! s& ( ! dunes s\?w>d4 ,t ! long axes ( ! l9es ( dunes d really correspond ) ! direction ( ! prevail+ w9d is c]ta91 bo? f ,mr4 ,beadnell's obs]vations ( w9d-direction#eg & f ! fact t ! scor+ ( ! lime/ones 9 ! ,libyan ,des]t by w9d- borne s& co9cides 9 direction ) ! l9es ( dunes4#eh ,?us ,i sugge/ t :ile ! ext5sion ( ! l9es ( dunes s\?w>ds is purely ! result ( ! prevail+ nor?-nor?-we/ w9d1 _! clean- cut "* & n>r[ wid? is ! consequ5ce ( lat]al attraction by ! dunes !mselves on ! fly+ electrifi$ p>ticles ( s&4 ,if ! hypo!sis ,i h v5tur$ to put =w>d is reg>d$ z likely to furni% ! true explanation = ! =mation ( ! l9es ( dunes1 ,i hope t someone possess+ ! necess>y skill 9 electro/atic measurem5t w put x to a crucial te/1 by c>ry+ out obs]vations on & ne> ! dunes !mselves4 ,if a numb] ( fly+ s&-gra9s c\ld be caught f diff]5t levels 9 an insulat$ metallic receiv] connect$ to an electromet]1 ! read+s ( ! electromet] w\ld enable ! pot5tials to :i* ! gra9s 7 *>g$ at diff]5t levels to be calculat$1 provid$ ! electro/atic capacity ( ! receiv+ sy/em & ! numb] & diamet] ( ! gra9s giv+ up _! *>ge 7 measur$4 ,all[ance w\ld1 ( c\rse1 h to be made = any electrification ( ! receiv] by ! air1 or by uncaught s&- gra9s1 z well z = losses by imp]fect insulation ( ! collect+ sy/em4 ,9 a prelim9>y exp]im5t ( ?is k9d :i* ,i made 9 ,cairo dur+ a s&/orm1 leakage f ! electromet] 0 f\nd to be s great z to vitiate any attempts at accurate measurem5t2 b ,i ?9k ?is p>ticul> difficulty c\ld be ov]come ) a specially design$ app>atus1 or t at lea/ ! leakage c\ld be measur$ & all[$ =4 ,! tak+ ( obs]vations ( any k9d 9 dune areas dur+ w9dy wea!r is a mo/ difficult & try+ op]ation1 & commonly means1 9 addition1 a soj\rn ( "s weeks 9 isolat$ & o!rwise un9t]e/+ areas4 ,b ! adv5t ( ! motor-c> has s greatly facilitat$ communication 9 ! des]ts t difficulties ( access are n[ n s great z !y us$ to be2 ! s\? end ( ! #ce-kilometre l9e ( ! ,kattania ,dunes1 = in/ance1 ab\t #ij kilometres we/- s\?-we/ ( ,cairo1 :i* w\ld =m an excell5t site = detail$ obs]vations1 c n[ be rea*$ by a f\r or five h\rs' c> j\rney f ,cairo in/ead ( ! sev]al days' j\rney by camel :i* 0 =m]ly necess>y4 #ab4 ,! ,di/ribution ( ,/one ,implem5ts4 ,i h n made any special /udy ( /one implem5ts1 nor h ,i be5 able to pay any consid]able att5tion to _! collection dur+ my j\rneys1 hav+ usually f\nd my "t v fully tak5 up ) o!r matt]s4 ,i h1 h[ev]1 be5 mu* /ruck by ! wide di/ribution ( /one implem5ts 9 ! des]t4 ,i h f\nd !m1 = in/ance1 n only 9 ,siwa ,oasis & ne> ! wells ( ,abu ,mung> & ,%eb1 b also on ! plateau betwe5 ,bah>ia & ,f>afra1 on ! op5 des]t betwe5 ,t]fawi & ,ow5at1 & to ! s\?-we/ ( ,ow5at ne> ! ,anglo- ,fr5* b\nd>y4 ,?is wide range ( occurr5ce1 c\pl$ ) ! f9ds ( ,s*we9fur? & o!rs on ! plateaux ne>] ! ,nile1 incl9es me to ?9k t !re is sc>cely any "p ( ! ,libyan ,des]t 9 :i* /one implem5ts might n be f\nd by an exp$ition :i* w\ld make ! collection & /udy ( !m "o ( its ma9 objects4 ,! likelie/ places 9 :i* to se>* 7besides ! neighb\rhood ( old wells & spr+s7 w\ld appe> to be ! %ores ( ! v>i\s lakes & ar\nd ! feet ( hills af=d+ %elt] f w9d & sun2 = x is 9 su* localities t ,i h mo/ usually come across implem5ts & pott]y4 ,gr9d+-/ones1 (t5 ) a sort ( /one roll+-p91 unpoli%$ celts1 knives1 & arr[-heads are ! pr9cipal =ms ( implem5t ,i h met )4 ,_m ( ! gr9d+-/ones mu/ h be5 c>ri$ = a consid]able di/ance1 = !y are made out ( rocks :i* d n naturally occur 9 ! localities 9 :i* !y are f\nd4 ,:e!r ?is wide di/ribution ( /one implem5ts w\ld ju/ify ! view ( ,blanckh5horn#ei t primitive man liv$ on ! des]t plateaux r ?an 9 ! ,nile ,valley mu/1 ,i ?9k1 be settl$ by ! fur!r collection ( specim5s & by a c>eful comp>ison ( ! =ms to be f\nd 9 ! two situations4 ,9 ?is connection ,i may rem>k t ! gr9d+-/ones ,i h se5 9 ! des]t seem$ to me to be v simil> to ^? us$ by ! ,nubians ( ! ,nile ,valley to-day = gr9d+ corn4 ,,app5dix ,,"s ,,rec5tly ,,det]m9$ ,,positions ,,9 ,,! ,,libyan ,,des]t ,9 view ( _! possible utility 9 connection ) future exploratory surveys1 ,i give bel[ a li/ ( "s ( ! m important ( ! positions :i* ,i h rec5tly det]m9$ by a/ronomical obs]vation 9 ! s\!rn "p ( ! ,libyan ,des]t1 ) brief descriptive notes on ! places to :i* !y ref]4 ,! obs]vations 7 made :il/ accompany+ ,h4,r4,h4 ,pr9ce ,kemal el ,d9 ,husse9 on 8 exploratory motor-c> exp$itions ( #aibc1 #aibd1 & #aibe4 ,! longitudes ( ,qasr ,f>afra1 ,abu ,mung>1 ,pott]y ,hill1 & ,reg5feld dep5d on ! transport ( a box-*ronomet]2 ! o!rs on wireless "t-signals f ,europe4 ,! positions ( peaks 7 f\nd by triangulation f ! actual obs]vation-spots4 ,! altitudes ( camps are f c>eful b>ometric det]m9ations2 ^? ( ! peaks dep5d on trigonometric levell+ f ! camps4 ,altitude ,place4 ,lat4 ,n4 ,long4 ,e4 above sea4 ,metres4 ,m]ga1 ,camp #fjj metres ,s4,w4 ( we/ #ai #b #bi #bf #ah #cb #ebf corn] ( salt lake4 7,lake surface is #ag metres bel[ level ( camp47 ,bir ,t]fawi1 ,camp close to well & #bb #ee #ab #bh #eb #ea #bdd palms ,gebel ,kissu1 ,summit #ba #cd #ei #be #h #bf #agbf ,gebel ,ow5at1 highe/ po9t #ba #ed #cd #be #j #dg #aijg rem>kable triple peak #ba #ec #ea #be #a #ci #agah ,s4,w4 peak #ba #di #ce #bd #ec #eb #adej ,*unk ,hill1 ,summit #ba #eb #df #be #ac #ef #ihe ,ow5at1 ,camp #fjj metres ,s4,e4 ( #ba #dh #ce #bd #ea #de #efh we/]n spr+1 on pla9 at foot ( m\nta9 ,ow5at1 ,camp at m\? ( gully1 #ba #ec #h #be #g #eh #fbf ab\t #hjj metres ,s4,s4,e4 ( ea/]n spr+s ,gebel ,kamil1 ,summit 7#afj metres #bb #af #ca #bf #ch #aa #hjj above pla97 ,reg5feld1 ,rohlfs' cairn ( #ahgd1 #be #aj #di #bg #bd #bb #dgj r$iscov]$ #aibd ,pott]y ,hill1 ,summit 7#ci metres #bd #bf #bg #bg #ch #ed #ejf above pla97 ,%eb1 ,camp close to well1 #bdj #bb #ai #dh #bi #de #df #bbh metres ,n4,n4,e4 ( =t ,mut1 ,gov]nm5t ,re/-h\se #be #bh #cg #bh #eh #bd #aai ,abu ,mung>1 ,camp at small ru91 #djj #bf #cj #bb #bg #ce #bi #aag metres ,e4,s4,e4 ( well ,qasr ,f>afra1 ,camp ,e4 ( village1 #bg #c #bf #bg #eg #eb #ij #bjj metres ,s4,w4 ( ,tomb ( ,%eikh ,dakhil ,m]ga4--,to geograph]s ! mo/ important ( ! new det]m9ations w und\bt$ly be t ( ! un9habit$ oasis ( ,m]ga1 :i* _h n previ\sly be5 visit$ by a ,european1 & ( :i* ! situation c\ld hi!rto only be guess$ at f ,arab /atem5ts4 ,m]ga lies 9 a %all[ depression ab\t #ej metres deep1 brok5 by s&/one hills & s&-dunes1 & ext5d+ = "s #bj kilometres nor?-ea/ & s\?- we/4 ,! salt-lake1 ne> ! c5tre ( ! depression1 measures ab\t #cjj metres by #aej4 ,x is surr\nd$ by tall ru%es & s&hills except at its s\?-we/]n end4 ,!re are num]\s date-palms1 bo? ne> ! lake & at consid]able di/ances f x1 z well z acacia trees & tam>isk bu%es4 ,! neighb\rhood ( ! lake sw>ms ) mosquitoes4 ,good & pl5tiful wat] 0 got by %all[ digg+ ab\t #a kilometre s\?-s\?-ea/ ( ! obs]vation-spot1 & c\ld probably be obta9$ by digg+ almo/ any:]e 9 ! depression4 ,x is possible t ! names ,bidi & ,tura el ,b$ai1 %[n ) a "q m>k on "s ,sudan maps1 may ref] to diff]5t spots )9 ! same depression4 ,ow+ to ! pl5tifully scatt]$ vegetation1 ! place _c easily be miss$1 ei!r by travell]s pass+ )9 sev]al miles ( ! lake1 or by aircraft2 b ! l&+ ( a]oplanes 9 ! neighb\rhood ( ! lake might be difficult ow+ to ! ext5sive s&-drifts4 ,bir ,t]fawi4--,sc>cely less important ?an ! accurate fixation ( ,m]ga is t ( ,bir ,t]fawi1 ! f>!/ s\?-we/ ( all ! ,egyptian artesian wat]-s\rces hi!rto kn[n4 ,x w be rem>k$ t ?is late/ det]m9ation places ! well "s #bb s\? & #ae ea/ ( ! position :i* ,i _h provisionally assign$ to x f ! r\gh data furni%$ by ,lieut4 ,moore's trav]se ( #aiaf4fj ,a k ( ! true position ( ,t]fawi is specially desirable = a travell] :o wi%es to rea* x1 ow+ to ! abs5ce ( any conspicu\s l&m>k ne> x & to ! s&y nature ( ! surr\nd+ c\ntry1 :i* causes tracks to be soon oblit]at$4 ,!re are num]\s s&hills cov]$ ) tam>isk bu%es ar\nd ,t]fawi1 & a little graz+ = camels4 ,besides ! well at :i* obs]vations 7 tak51 wat] 0 f\nd by digg+ 9 ! s& at ! foot ( "s tam>isk m\nds ab\t #ac kilometres f>!r we/1 & x is probable t good & pl5tiful supplies c\ld be obta9$ at %all[ dep?s ne> any ( ! o!r m\nds4 ,! palmtrees at ,t]fawi are few & small1 & are less conspicu\s ?an ! tam>isk-bu%es4 ,^! latt] %\ld enable ! place to be easily f\nd by aircraft2 b l&+ w\ld require "s caution ow+ to ! preval5ce ( drift$ s&4 ,gebels ,kissu & ,ow5at4--,! peaks ( ^! m\nta9s 7especially t ( ,kissu1 because ( its isolat$ "* & %>ply m>k$ summit7 w =m useful po9ts = ! connection ( future surveys1 be+ visible f v long di/ances4 ,*unk ,hill is a prom95t isolat$ hill1 ne>ly conical & ( d>k col\r1 :i* =ms a good l&m>k 9 ! brok5 c\ntry to ! ea/ ( ,gebel ,ow5at4 ,x rises "s #cce metres above ! gr\nd at its foot4 ,! ,spr+s ( ,ow5at4--,( ! two wat]-s\rces ( ,ow5at ^: positions are giv51 ! we/]n "o is ! bett]1 & is moreov] v easy ( access1 be+ practically on ! level ( ! pla9 :i* ext5ds s\?w>ds f ! m\nta9 mass1 & easily discov]able by ! num]\s animal-tracks conv]g+ to x2 x is a pool among great b\ld]s1 obvi\sly f$ by p]colation ?r\gh cracks & fissures 9 ! granite m\nta9 :i* t[]s above x4 ,! ea/]n wat]-s\rce is less easy ( access2 x lies ab\t #a kilometre up a /ony gully1 & consi/s ( a s]ies ( pools 9 ! rocky floor ( ! gully1 f$ by trickl+ spr+s at ! level :]e ! granite is ov]la9 by s&/one4 ,! pla9 to ! s\? ( ! we/]n spr+ w\ld =m an excell5t l&+-gr\nd = aircraft4 ,:il/ 9 ! neighb\rhood ( ,ow5at1 ,i _h hop$ to re-det]m9e ! longitude ( ,hassane9 ,bey's camp ( #aibc ) ! aid ( wireless "t- signals1 or at lea/ by triangulation-connection to "o ( my obs]vation-po9ts1 because a really accurate fixation ( ! longitude ( a po9t ab\t midway along 8 r\te f ,j>abub to ,furawia w\ld possibly furni% ! means ( slightly correct+ ! longitudes assign$ to ,hassane9's camps at ,kufra & ,erdi4 ,i 0 un=tunately prev5t$ f c>ry+ out ! desir$ connection2 b f a hurri$ c>-trav]se :i* ,i made1 skirt+ ! we/]n side ( ! m\nta9 mass1 9 ! c\rse ( :i* ,i mu/ h pass$ pretty close to ! site ( ,hassane9's ,ow5at camp1 ,i ?9k t camp really lies ab\t 9 longitude #bd #di1 or "s #e miles to ! we/ ( :]e ,i _h previ\sly calculat$ x to be f ,hassane9's trav]se data2#fa & t 9 consequ5ce ! longitudes assign$ to ,kufra & ,erdi f ! same data may be "s #b or #c miles too great4 ,gebel ,kamil is a %>ply po9t$ hill ( d>k s&/one1 =m+ a useful l&m>k betwe5 ,t]fawi & ,ow5at4 ,x 0 visible f ! ea/ = m ?an #dj1 or f ! we/ = ab\t #bj1 kilometres4 ,! "n 0 giv5 to x by ,pr9ce ,kemal el ,d9 9 hon\r ( 8 fa!r1 ,husse9 ,kamil1 ! late ,sultan ( ,egypt4 ,reg5feld1 x w be rem>k$1 0 f\nd to be v ne>ly 9 ! position assign$ to x by ,jordan 9 #ahgf1fb & my e/imation ( ! level ( ! place is only #bj metres high] ?an ,jordan's4 ,! neighb\r+ dune ,i f\nd to be #cj metres high1 agree+ exactly ) ,jordan's measurem5t ( fifty ye>s previ\sly2 & z f> z ,i c\ld judge1 ! situation ( ! dune relative to ! cairn seems to h rema9$ un*ang$ ?r\gh ?is long int]val1 %[+ t !re has be5 at any rate no great lat]al displacem5t ( ! dunes4 ,! iron tanks left by ,rohlfs 7 q intact1 9 spite ( _! half-c5tury ( exposure2 !y _h become cov]$ ) a h>d d>k br[n film1 app>5tly ( magnetic oxide ( iron1 n ru/$ 9 ! ord9>y way1 a fact :i* speaks /rongly = ! dryness ( ! region4 ,empty w9e- bottles left by ! ,rohlfs exp$ition 7 fro/$ by ! s&-bla/ :]ev] !y 7 expos$2 b surpris+ly little ( ! glass _h be5 remov$ 9 ?is way4 ,pott]y ,hill is ! nor!rn "o ( a pair ( ne>ly conical d>k s&/one hills ab\t #dj metres high1 :i* are conspicu\s f "s di/ance ow+ to _! situation on a ne>ly level s&-pla94 ,i gave x its "n f ! num]\s j>s :i* ,i discov]$ at its foot 9 #aiag4 ,%eb ,well is a pool ( fairly good wat] 9 a %all[ excavation at ! foot ( a clump ( dom palms1 half surr\nd$ by s&-dunes & tam>isk m\nds1 #bdj metres nor?-nor?-we/ ( a =t2 ! latt] occupies a fairly conspicu\s position on a s&/one hill ab\t #be metres high4 ,ano!r well1 situat$ at ! foot ( a clump ( dom palms ab\t #aa kilometres f>!r nor?1 is call$ ,bir ,t]fau2 ?is %\ld n1 ( c\rse1 be confus$ ) ! ,bir ,t]fawi already m5tion$4 ,to ! nor? ( ,%eb1 on ! ,arba9 road1 are ?ree o!r wat]-s\rces :i* ,i visit$ 9 #aibe & ( :i* ,i det]m9$ ! approximate positions by a c>efully controll$ c>-trav]se3 ,bir ,kassaba1 9 lat4 #bb #da1 long4 #bi #ee2 ,bir ,abu ,husse91 9 lat4 #bb #ec1 long4 #bi #ee2 & ,bir ,murr1 9 lat4 #bc #bb1 long4 #cj #e4 ,bir ,kassaba is a pool ( good wat] at ! foot ( a clump ( palms2 ,bir ,abu ,husse91 :i* likewise yields good wat]1 is a small excavation 9 s& ne> a hill ( p9k co>se-gra9$ granite1 & usually requires to be dug out afre% by ea* travell]2 ,bir ,murr1 on a little pla9 surr\nd$ by hills1 consi/s ( sev]al holes1 ) fairly good wat]1 excavat$ 9 ! s& at ! s\? side ( ! outcrop ( a /eeply incl9$ b$ ( speckl$ calc>e\s s&/one4 ,i h n myself trav]s$ ! des]t to ! ea/ ( ,%eb1 b accord+ to an intellig5t ,arab :o has rec5tly made ! j\rney f ,%eb to ,dungul1 !re are only two spots betwe5 ^! places at :i* wat] is obta9able2 !y are ,bir ,abu ,seifa1 a small well ( good wat] ab\t #ej kilometres f ,%eb1 & ,bir ,haleifa1 ab\t #ae kilometres f>!r on1 :]e !re is a =t & num]\s wells4 ,?is latt] place mu/1 ,i ?9k1 be ! same z ! ,bir ,nakhlai1 ( :i* ! position f\nd a/ronomically by ,colonel ,talbot 9 #ahic 0 lat4 #bb #bi #a1 long4 #cj #ai #cf4 ,! ,re/-h\se at ,mut1 9 ,dakhla ,oasis1 is a l[ :itewa%$ build+ ( ?ree rooms1 situat$ on op5 gr\nd a little to ! s\? ( ! village4 ,x is conspicu\s by reason ( its isolation f o!r build+s1 & =ms a conv5i5t />t+-po9t = c> j\rneys to ! s\?-we/ ( ,dakhla1 z well z = j\rneys to ,kh>ga by ! s\!rn track :i* 0 mo/ly foll[$ by c>s dur+ ! w>4 ,abu ,mung> prov$ to be v close to ! position giv5 by ,mr4 ,h>d+ ,k+'s obs]vations ( #aiab4fc ,to ! s\?-ea/ ( ,abu ,mung>1 ab\t halfway to ,dakhla1 are two rem>kable hills =m+ outli]s ( ! plateau4 ,^! hills1 :i* are good l&m>ks1 7 fir/ discov]$ by ,mr4 ,h>d+ ,k+1 & z !y appe> to possess no native "n1 ,i h call$ !m 8,k+'s ,hills0 on ! map4 ,! f>!r "o ( ! two f ! plateau lies 9 lat4 #be #eh1 long4 #bh #aa1 f compass-be>+s :i* ,i took to x on a c>eful c>-trav]se 9 #aibd4 ,! o!r is ab\t #b kilometres f>!r nor?4 ,qasr ,f>afra4--,! position ( ! village ( ,qasr ,f>afra :i* ,i gave 9 my pap] ( #aiaifd is prov$ to be v ne>ly correct by ! late/ obs]vations4 ,! ma9 advantage ( ! new det]m9ation is t x is ref]r$ to an easily id5tifiable l&m>k1 ! tomb ( ,%eikh ,dakhil1 in/ead ( to ! ill-def9$ c5tre ( ! village4 ,levels 9 or ne> ,bah>ia ,oasis4--,i ?9k x may be well to record "h t ow+ to an un=tunate mi/ake ab\t ! datum to :i* ! level ( ! old milit>y railway-t]m9us at ,b#f 0 ref]r$1 all ! altitudes giv5 on pp4 #aj & #aa ( my above-m5tion$ ,survey ,dep>tm5t ,pap] ( #aiai require a correction ( #ai metres to be add$4 ,at ! "t :5 ,i triangulat$ ,bah>ia 9 #aiag1 ,i 0 in=m$ t ! level f\nd by ! milit>y au?orities = ! t]m9us 0 measur$ f ! sea- level datum1 & ,i employ$ ! value giv5 = x z my fundam5tal level4 ,b ,i h s9ce asc]ta9$ t an arbitr>y datum 0 employ$ by ! milit>y eng9e]s2 & trigonometric levell+ f ! ,nile ,valley has %[n t ?is arbitr>y datum 0 ab\t #ai metres above sea-level4 ,,footnotes3 ,footnote #a3 ,! use ( motor c>s = reconnaissances 9 ! ,libyan ,des]t1 initiat$ by ! ,briti% ,army dur+ ! ,great ,w>1 has be5 consid]ably develop$ 9 ! la/ few ye>s by ,pr9ce ,kemal el ,d9 ,husse9 7son ( ! late ,sultan ( ,egypt71 :o has made sev]al long exploratory j\rneys ) a fleet ( c>s ( ! cat]pill> type1 specially con/ruct$ s z to be capable ( c>ry+ consid]able loads ov] s&y tracts :i* 7 impassable to ! ,army patrols4 ,! exp$itions organiz$ & l$ by ,8 ,royal ,highness h result$ 9 important additions to geographical k conc]n+ ! ,libyan ,des]t1 especially ! s\!rn & we/]n regions1 :i* h hi!rto be5 practically inaccessible4 ,i _h ! pleasure ( accompany+ ! ,pr9ce on ?ree ( 8 j\rneys1 & am mu* indebt$ to him = ?us af=d+ me ! opportunity ( mak+ _m obs]vations :i* w\ld o!rwise h be5 impossible to me4 ,i h fur!r to ?ank him = k9d p]mission to illu/rate ?is pap] by reproductions f a selection ( ! excell5t photographs tak5 by him dur+ our travels4 ,footnote #b3 ,k9dly communicat$ by ,mr4 ,a4 ,r4 ,boyce1 ,director ( ,sudan ,surveys4 ,footnote #c3 ,pr9t$ by ! ,fr5* ,army ,/aff1 #aibf4 ,footnote #d3 ,by way ( comp>ison1 x may be rem>k$ t ! area 7at ! sea-level cont\r7 ( ! ,qatt>a depression is ab\t f\r times t ( ! ,dead ,sea & ,jordan ,valley depression2 b ! latt] 7reckon+ to ! l[e/ po9t ( ! b$ ( ! ,dead ,sea7 is ab\t six times ! deep] ( ! two4 ,footnote #e3 ,a simil> ambiguity ( mean+ v commonly atta*es to ,arabic geographical t]ms4 ,?us bel$ may mean ei!r a village or an entire c\ntry2 gebel1 ei!r a m\nta9 or a des]t2 gezira1 ei!r an isl& or a "p ( ! ,nile bank :i* is alt]nately cov]$ & uncov]$ by ! rise & fall ( ! riv]2 bir1 :i* prop]ly signifies a well1 is (t5 us$ to d5ote a spr+ or a wat]-ci/]n2 wadi1 ?\gh its prop] mean+ is a valley1 is also sometimes employ$ = a clos$-9 depression1 z 9 ! case ( ! ,wadi ,natrun4 ,footnote #f3 #8,n\veaux ,mmoires des ,missions de la ,compagnie de ,jesus dans le ,levant1' ,tome ii4 7,p>is1 #agag71 p4 #gd4 ,?is little book conta9s a map ( ,egypt1 prep>$ by ,fa!r ,sic>d at ,cairo 9 #agae1 on :i* ! 8,m] sans eau0 is %[n ) 8bateaux petrifiez0 depict$ alongside x4 ,footnote #g3 ,see 8,p>ticul>s conc]n+ ! ,valley ( ! ,natron ,lakes1 & t ( ! ,old ,b$ ( ! ,riv]0 7) map7 by ,&reossi1 ,g5]al ( ,artill]y1 9 #8,memoirs ,relative to ,egypt1 ,writt5 9 ,t ,c\ntry dur+ ! ,campaigns ( ,g5]al ,bonap>te 9 ! ,ye>s #agih & #agii1 by ! ,le>n$ & ,sci5tific ,m5 :o ,accompani$ ! ,fr5* ,exp$ition' 7,london1 #ahjj71 p4 #bgj4 ,footnote #h3 ,see ,rohlfs' #8,drei ,monate 9 d] libys*5 ,w/e' 7,cassel1 #ahge71 p4 #ccg2 & ,jordan's #8,physis*e ,geographie und ,meteorologie d] libys*5 ,w/e' 7,cassel1 #ahgf71 p4 #bad4 ,footnote #i3 #8,zeits*rift d] ,gesells*aft fr ,erdkunde zu ,b]l91' #aijb1 ,tafel ,,iii4 ,footnote #aj3 #8,topography & ,geology ( ! ,fayum ,prov9ce ( ,egypt' 7,cairo1 #aije71 p4 #fg4 ,footnote #aa3 ,s9ce ! above 0 writt51 ! ,pr(essor ( ,geography at ! new ,univ]sity ( ,cairo1 ,m4 ,lor91 has sugge/$ t at a =m] geological epo* ! ,nile fl[$ ?r\gh ,siwa ,oasis & !nce we/w>d s z to make ,cyr5aica an isl&4 7,see #8,l',egypte1' publi%$ by ! ,in/itut ,franais d',ar*ologie ,ori5tale1 ,cairo1 9 ,decemb] #aibf1 p4 #aec47 ,m4 ,lor9 does n give any gr\nds = ?is sugge/ion1 & ,i "k ( no evid5ce t w\ld ju/ify its acceptance4 ,z w be ga!r$ f my rem>ks else:]e 9 ?is pap]1 x seems to me f> m likely t ,siwa & ! o!r nor!rn depressions 7 9 rec5t geological times regions ( c5tral dra9age1 ?an t !y 7 local deep5+s ( a s+le *annel2 & 9 reg>d to e>li] geological p]iods we possess too little in=mation to =m any conclusions z to dra9age- l9es4 ,footnote #ab3 ,see ,sir ,hanbury ,br[n's #8,! ,fayum & ,lake ,mo]is' 7,london1 #ahib71 pp4 #dj-dh & #ajf-aaj2 ,sir ,w4 ,willcocks' #8,p]5nial ,irrigation & ,flood ,protection = ,egypt' 7,cairo1 #ahid71 also ! same au?or's #8,egyptian ,irrigation1' ?ird edition 7,london1 #aiac72 ,sir ,w4 ,e4 ,g>/9's #8,report on ! ,bas9 ( ! ,upp] ,nile' 7,cairo1 #aijd71 ,app5dix ,i4 pp4 #f-i2 & ,beadnell's #8,topography & ,geology ( ! ,fayum ,prov9ce ( ,egypt' 7,cairo1 #aije71 pp4 #af-bd4 ,footnote #ac3 #8,report on ! ,adm9i/ration ( ! ,irrigation ,dep>tm5t = #ahie' 7,cairo1 #ahif71 pp4 #be-bh4 ,footnote #ad3 #8,report on ! ,adm9i/ration ( ! ,irrigation ,dep>tm5t = #ahif' 7,cairo1 #ahig71 p4 #abi4 ,footnote #ae3 ,x may be rem>k$ t ev5 if a *annel 7 to be cut f ,lake ,m>iut to ! ,wadi ,natrun x w\ld /ill be necess>y to cont9ue pump+ on a l>ge scale f ,lake ,m>iut to ma9ta9 ! pres5t-day dra9age4 ,accord+ to returns communicat$ to me by ! ,irrigation ,s]vice1 ! quantity ( wat] remov$ f ,lake ,m>iut by pump+ dur+ ! la/ n9e ye>s has av]ag$ ne>ly #fej million cubic metres a ye>2 & ev5 on ! mo/ optimi/ic e/imate ( evaporation & seepage ! ,wadi ,natrun c\ld n be expect$ to dispose ( half ?is quantity aft] ! lapse ( ! few ye>s :i* w\ld be occupi$ 9 fill+ x to a p]man5t level4 ,footnote #af3 ,! av]age dis*>ge ( ! ,nile pa/ ,cairo 9 a ye> is #fg cubic kilometres4 ,! capacity ( ! ,qatt>a depression at ! sea- level cont\r is r\ghly ab\t #ijj cubic kilometres4 ,footnote #ag3 ,geogr4 ,j\rn41 #ef 7#aibj71 pp4 #ig-ii & #afa-afc4 ,footnote #ah3 ,lucas 7#8,natural ,soda ,deposits 9 ,egypt' 7,cairo1 #aiab71 p4 #ae7 reg>d$ ! wat] ( ! ,wadi ,natrun z be+ due to infiltration f ! ,nile1 ma9ly on ! gr\nd t he suppos$ ! wat] visibly ent]+ ! wadi 7z small /reams & trickles7 to d s on ! nor?- ea/]n side4 ,b ?is1 ,i ?9k1 is a mi/ake1 z x neglects consid]ation ( ! l>ge spr+s ( w>m fre% wat] :i* bubble up 9 & ne> ! salt-lakes !mselves1 & also ! l>ge wells ( ! mona/]ies :i* lie s\?-we/ ( ! lakes4 ,footnote #ai3 ,a f> bett] site = ?is well w\ld h be5 9 ! depression "s #bj kilometres to ! nor?-we/1 ! gr\nd-level at ! l[e/ po9t ( :i* is only #fa metres above sea2 b at ! "t :5 ! site ( ! well 0 *os5 ! exi/5ce ( ?is depression 0 unkn[n4 ,footnote #bj3 ,see ,app5dix ,,iii4 9 ! second volume ( ,willcocks' #8,egyptian ,irrigation1' #aiac1 p4 #hec4 ,footnote #ba3 ,s*we9fur? 7op4 cit41 p4 #hee7 ev5 ?\ght x might h p]colat$ to ,siwa1 9 spite ( ! great di/ance 7#dhj kilometres7 & ! adv]se direction ( ! diff]5ce ( level betwe5 ! two places4 ,footnote #bb3 ,mr4 ,beadnell 7#8,an ,egyptian ,oasis' 7,london1 #aiji71 p4 #aci7 e/imates ! total dis*>ge ( ! ,kh>ga wells at #ec1jjj cubic metres p] diem4 ,we %\ld n1 ,i ?9k1 err greatly 9 e/imat+ ! total output ( all ! ,egyptian oasis wells & spr+s at ab\t five times ?is1 or #bej1jjj cubic metres p] diem4 ,z ! av]age daily dis*>ge ( ! ,nile 9 ,egypt is ab\t #ahj million tons1 a local influx into x sev]al times z great z ! total dis*>ge ( ! oasis wells w\ld repres5t b an insignificant addition to ! dis*>ge ( ! ,nile4 ,i ?9k1 h[ev]1 t ! influx ( ! w>m wat] might possibly be detect$ by c>eful !rmometric measurem5ts made 9 ! riv] ne> its banks at ! "t ( l[ ,nile4 ,footnote #bc3 ,?is 0 ! l[e/ level record$ at ! "t :5 ,i drew ! /atic cont\rs on ! map4 ,! fur!r exploration ( ! depression by ,mr4 ,walpole 9 ! pres5t ye> has %[n t at "o place ! gr\nd-level desc5ds to #acd metres2 ! situation ( ?is local deep5+ is1 h[ev]1 su* t ! draw+ ( ! /atic cont\rs w\ld n h be5 affect$ by its inclusion 9 ! li/ ( adopt$ data4 ,footnote #bd3 #8,an ,egyptian ,oasis' 7,london1 #aiji71 p4 #aef4 ,footnote #be3 8,! ,mutual ,int]f]5ce ( ,artesian ,wells10 ,geol4 ,mag4 7,london1 #aiji71 pp4 #bc-bf2 also #8,an ,egyptian ,oasis1' *ap4 x4 ,footnote #bf3 #8,an ,egyptian ,oasis1' p4 #adj4 ,footnote #bg3 #8,drei ,monate 9 d] libys*5 ,w/e' 7,cassel1 #ahge71 p4 #aae4 ,footnote #bh3 ,see ,jordan's #8,physis*e ,geographie und ,meteorologie d] libys*5 ,w/e' 7,cassel1 #ahge71 p4 #bjd4 ,jordan /ates t !re 7 on an av]age n m ?an two to f\r plants p] acre 9 ?is locality4 ,footnote #bi3 ,see ,rohlfs' #8,drei ,monate 9 d] libys*5 ,w/e' 7*apt] ix4 ( :i* 0 writt5 by ,as*]son71 p4 #bej4 ,footnote #cj3 ,9 my pap] on #8,rec5t ,det]m9ations ( ,geographical ,positions 9 ! ,libyan ,des]t1' publi%$ 9 ,cairo 9 #aiai1 ,i gave 7p4 #ad7 ! approximate position ( ! hill z lat4 #bd #bh1 long4 #bg #de4 ,! position giv5 above is f a m accurate det]m9ation :i* ,i made on revisit+ ! place ) ,pr9ce ,kemal el ,d9 9 #aibc4 ,on ?is latt] occasion ! excavations made by ! ,pr9ce 9 ! s& ar\nd ! foot ( ! hill reveal$ ! exi/5ce ( hundr$s ( additional j>s1 _m ( !m intact1 set 9 regul> ord] 9 ! s& & obvi\sly =m+ a wat]-dump4 ,! majority ( ! j>s bore cle>ly incis$ m>k+s1 :i* ,pr9ce ,kemal el ,d9 0 able lat] to id5tify positively z be+ tribal m>ks ( ! ,tebus1 ! m>ks be+ 9 fact exactly ! same z ^? us$ by ! ,tebu tribes ( ! pres5t day4 ,footnote #ca3 ,see ,jordan's #8,physis*e ,geographie und ,meteorologie d] libys*5 ,w/e' 7,cassel1 #ahgf71 p4 #bae4 ,footnote #cb3 #8,kh>ga ,oasis' 7,cairo1 #aijj71 p4 #eg4 ,footnote #cc3 ,geogr4 ,j\rn41 #db 7#aiac71 p4 #bhc2 also #8,my/]ies ( ! ,libyan ,des]t' 7,london1 #aibe71 p4 #cjd4 ,footnote #cd3 ,?is 8locus0 is ( c\rse giv5 by jo9+ up ! po9ts ( int]section ( ! #ajj-metre wat]-cont\r ) ! #bjj-metre gr\nd-cont\r1 ! #bjj-metre wat]-cont\r ) ! #cjj-metre gr\nd- cont\r1 & s on4 ,footnote #ce3 ,x is possible t ! 8,nubian ,s&/one0 9 ,s\?- ,we/]n ,egypt may repres5t m ?an "o geological sy/em1 z has be5 f\nd to be ! case 9 ,s9ai4 ,col4 ,tilho 7,geogr4 ,j\rn41 #ef 7#aibj71 p4 #bei7 records t ! s&/one at ,enn$i prov$ to be ( ,upp] ,silurian age by ! fossils f\nd 9 x4 ,footnote #cf3 ,?\gh we pass$ )9 #aj kilometres to ! ea/ ( ,kissu on ! return j\rney f ,m]ga to ,ow5at 9 #aibe1 ,i c\ld n obs]ve ! /ructure ( ! m\nta9 v cle>ly1 ow+ to haze4 ,b x appe>$ to me to be a huge granitic intrusion ris+ ?r\gh ! gneiss :i* cov]s a l>ge area h]eab\ts4 ,! s&/one ov] :i* we _h be5 travell+ s9ce leav+ ,m]ga gave place to gneiss ab\t #dj kilometres be=e we came abrea/ ( ,kissu1 & gneiss cont9u$ to be ! c\ntry rock until we came )9 #cj kilometres ( ,gebel ,ow5at1 :5 s&/one reappe>$1 at fir/ capp+ l[ hills & !n =m+ ! pla94 ,footnote #cg3 ,!re exi/s no publi%$ ,engli% translation ( ,ptolemy's #8,geographia4' ,mll]'s edition 7,p>is1 #ahhc-aija7 gives ,greek & ,lat9 texts1 ) a ,lat9 comm5t>y4 ,footnote #ch3 ,quot$ 9 ,mll]'s edition ( ,ptolemy's ,geography1 p4 #gci4 ,footnote #ci3 ,= ! late/ del9eation ( ?is region1 see ! hypsometric map 9 ! #8,atlas de l',algrie et de ,tunisie1' publi%$ by ! ,c>tographic ,s]vice ( ! ,alg]ian ,gov]nm5t1 ,p>is1 #aibd1 ,fascicule ,,ii4 ,footnote #dj3 #8,manual ( ,anci5t ,geography' 7,london1 #ahia71 p4 #caa4 ,footnote #da3 ,geogr4 ,j\rn41 #ef 7#aibj71 p4 #id4 ,footnote #db3 #8,my/]ies ( ! ,libyan ,des]t' 7,london1 #aibe71 p4 #cjc4 ,footnote #dc3 ,geogr4 ,j\rn41 #dc 7#aibd71 p4 #bhh4 ,footnote #dd3 #8,! ,secret ( ! ,sah>a--,kuf>a' 7,london1 #aiba71 p4 #bce4 ,footnote #de3 ,beadnell1 8,! ,s&-dunes ( ! ,libyan ,des]t10 ,geogr4 ,j\rn4 #be 7#aiaj71 pp4 #cgi-cie2 ,h>d+ ,k+1 8,! ,nature & ,=mation ( ,s& ,ripples & ,dunes10 ,geogr4 ,j\rn41 #dg 7#aiaf71 pp4 #ahi-bji2 ,h>d+ ,k+1 8,/udy ( a ,dune ,belt10 ,geogr4 ,j\rn41 #ea 7#aiah71 pp4 #af-cc1 & ,discussion1 pp4 #bej-beh4 ,footnote #df3 ,= p>ticul>s ( ! di/ribution & directions ( ! dune-l9es ne> ,gebel ,ow5at1 & betwe5 t m\nta9 & ,s>ra1 ,i am indebt$ to ,pr9ce ,kemal el ,d9 ,husse91 :o devot$ special att5tion to obs]v+ !m on 8 exp$ition to ,s>ra 9 #aibf4 ,footnote #dg3 ,x may be rem>k$ t 9 ! des]t ( ,nor!rn ,s9ai ! dune-l9es foll[ curv$ directions1 in/ead ( ma9ta9+ a s+le direction z 9 ! ,egyptian ,libyan ,des]t1 & ?is curvature ( ! ,s9ai dune-l9es has g5]ally be5 ?\ght to be caus$ by ! deflect+ influ5ce ( ! m\nta9 masses ( ,mogh>a1 ,yelleg1 & ,hellal on ! w9d2 b "h also !re may be regional influ5ces at "w4 ,footnote #dh3 ,see ,jordan's notes on pp4 #bjd & #bjh ( 8 8,physis*e ,geographie und ,meteorologie d] libys*5 ,w/e10 ,cassel1 #ahgf4 ,footnote #di3 ,geogr4 ,j\rn41 #di 7#aiag71 p4 #ee4 ,footnote #ej3 ,! /eep ,mioc5e esc>pm5t1 "s #bjj metres high1 :i* b\nds ! ,qatt>a depression on ! nor?1 probably produces a great eddy 9 ! w9d sweep+ s\?w>ds ov] x1 caus+ a :irl r\nd a horizontal axis s\? ( ! sc>p2 ?is may be v influ5tial bo? 9 excavation & 9 lift+ ! s& to great heights4 ,z a l>ge portion ( ! floor ( ! depression at ! foot ( ! esc>pm5t is =m$ ( a salty sludge 7sabakha71 x is likely t a consid]able quantity ( f9ely divid$ salt cry/als are c>ri$ up ) ! s& 9 summ]1 & ?is may conceivably act to "s ext5t z a b9d+ mat]ial 9 ! dunes4 ,footnote #ea3 ,geogr4 ,j\rn41 #dg 7#aiaf71 p4 #bji4 ,footnote #eb3 ,geogr4 ,j\rn41 #ea 7#aiah71 p4 #beb4 ,footnote #ec3 ,_m ye>s ago ,i discov]$ t ?is film c be remov$ by treat+ ! s& ) w>m hydro*loric acid1 leav+ ! gra9s ( sn[y :it5ess 7see my #8,description ( ! ,fir/ or ,aswan ,cat>act' 7,cairo1 #aijg71 p4 #eg74 ,b ,i ?9k traces ( iron 9 ! f]r\s /ate mu/ exi/ ev5 )9 ! :ite silice\s body ( ! gra9s1 & be cont9ually pass+ into ! f]ric /ate at ! expos$ surface2 = o!rwise x is difficult to acc\nt = ! p]si/5ce ( ! yell[ col\r 9 spite ( ! mutual rubb+ ( ! gra9s dur+ _! transport by w9d4 ,! film may possibly h an influ5ce on ! electrical behavi\r ( ! gra9s1 by reason ( its possess+ a high] conductivity ?an ! silica ( ! int]ior4 ,footnote #ed3 ,pr(essor ,p4 ,e4 ,%aw has q rec5tly 7,nature1 vol4 #aah 7#aibf71 p4 #fei7 sugge/$ f laboratory exp]im5ts t ! mutual impact ( p>ticles ( id5tical nature 9 s&/orms may result 9 electrification ( ! p>ticles & ( ! air ) *>ges ( opposite signs4 ,i ?9k mo/ travell]s 9 ! des]t h1 l myself1 =m$ ! impression t electrification ( ! air is m pron\nc$ dur+ s&/orms 7khams9s7 ?an at o!r times4 ,b x is curi\s t obs]vations made ) ! electrograph at ,helwan ,obs]vatory d n seem altoge!r to be> ?is out4 ,dr4 ,hur/ f9ds t khams9s are usually accompani$ by l[ values ( pot5tial gradi5t 9 _! e>ly /ages1 ) great di/urbances bo? positive & negative = "s h\rs be=e & aft] ! break ( ! khams9 7#8,obs]vations on ,atmosph]ic ,electricity at ,helwan ,obs]vatory1' ,survey ,dept4 ,pap] ,no4 #aj 7,cairo1 #aiji71 pp4 #eb-fb74 ,footnote #ee3 ,hur/1 op4 cit41 p4 #bh4 ,footnote #ef3 ,mr4 ,h>d+ ,k+ records 9 8 pap] ( #aiae t he f\nd s&-gra9s bl[n (f ! dunes to be electrifi$2 c ^! h be5 gra9s :i* became *>g$ by jump+ up f ! dune to meet o!rs & ?us %>+ _! *>ge8 ,footnote #eg3 ,geogr4 ,j\rn41 #ce 7#aiaj71 p4 #che4 ,footnote #eh3 ,! hot & disagreeable s\? & s\?-we/ w9ds :i* bl[ dur+ ! 8khams9s0 ( ! spr+ are mu* less con/ant 9 _! direction2 & ?\gh !y produce ?ick s&/orms1 !y h always seem$ to me to occasion r a g5]al scatt]+ ( ! s& ov] ! surface ( ! des]t ?an to =m deposits hav+ v def9ite arrangem5t2 moreov]1 !y bl[ on relatively few days ( ! ye>4 ,! f> m preval5t cool /rong nor?-nor?-we/ w9ds1 on ! o!r h&1 (t5 sc>cely v>y a po9t 9 direction = days toge!r1 & _! co9cid5ce ) ! direction ( ! dune-l9es is v pron\nc$4 ,footnote #ei3 ,zeits*rift d] ,gesells*aft fr ,erdkunde zu ,b]l91 #aijb1 p4 #gfj4 ,footnote #fj3 #8,rec5t ,det]m9ations ( ,geographical ,positions 9 ! ,libyan ,des]t1' ,survey ,dep>tm5t ,pap] ,no4 #cd1 #aiai4 p4 #ad4 ,footnote #fa3 ,geogr4 ,j\rn41 #dd 7#aibd71 p4 #cgg4 ,footnote #fb3 ,jordan's f9al position 0 lat4 #be #aa #g1 long4 #bg #be #j4 ,! longitude giv5 on ! docum5t f\nd 9 ! bottle at ! cairn 7#ad #db ,e4 ( ,b]l91 or #bh #c #db ,e4 ( ,gre5wi*7 0 a prelim9>y value result+ f ,jordan's obs]vation ( a lun> di/ance1 & 0 wrong by "s #dj miles4 ,footnote #fc3 ,mr4 ,h>d+ ,k+'s obs]vations gave ! latitude z #bf #cj #df1 & ! longitude z #bg #cf 7see 8 map 9 ,geogr4 ,j\rn41 #db 7#aiac71 p4 #eaf74 ,footnote #fd3 #8,rec5t ,det]m9ations ( ,positions 9 ! ,libyan ,des]t1' ,survey ,dep>tm5t ,pap] ,no4 #cd 7,cairo1 #aiai71 p4 #ab4