,transcrib]'s ,notes3 ,und]scores 80 be=e & aft] a ^w or phrase indicate italics 9 ! orig9al text4 ,equal signs 80 be=e & aft] a ^w or phrase indicate bold 9 ! orig9al text4 ,small capitals h be5 conv]t$ to ,,solid capitals4 ,illu/rations h be5 mov$ s !y d n break up p>agraphs4 ,typographical & punctuation errors h be5 sil5tly correct$4 ,a ,,!ory ,,( ,,pure ,,design ,h>mony1 ,balance1 ,rhy?m ,,) ,,illu/rations ,,& ,,diagrams ,,by ,,d5man ,w4 ,,ross1 ,,ph4 ,d4 ,,lectur] ,,on ,,! ,,!ory ,,( ,,design ,,9 ,,h>v>d ,,univ]sity1 ,,fell[ ,,( ,,! ,,am]ican ,,academy ,,( ,,arts ,,& ,,sci5ces ,,bo/on ,,& ,,new ,,york ,,h\ghton1 ,,miffl9 ,,& ,,company ,,mdccccvii ,,copyright #aijg ,,by ,,d5man ,w4 ,,ross ,,all ,,rights ,,res]v$ ,publi%$ ,april1 #aijg ,,preface ,my purpose 9 ?is book is to elucidate1 s f> z ,i c1 ! pr9ciples :i* und]lie ! practice ( draw+ & pa9t+ z a ,f9e ,art4 ,art is g5]ally reg>d$ z ! expression ( feel+s & emotions :i* h no explanation except p]haps 9 su* a ^w z inspiration1 :i* is expletive r ?an explanatory4 ,art is reg>d$ z ! "o activity ( man :i* has no sci5tific basis1 & ! appreciation ( ,art is said to be a matt] ( ta/e 9 :i* no two p]sons c be expect$ to agree4 ,x is my purpose 9 ?is book to %[ h[1 9 ! practice ( ,art1 z 9 all o!r practices1 we use c]ta9 t]ms & foll[ c]ta9 pr9ciples4 ,be+ def9$ & expla9$1 ^! t]ms & pr9ciples may be kn[n & und]/ood by ev]ybody4 ,!y are1 s to speak1 ! =m ( ! language4 ,:ile an und]/&+ ( ! t]ms & pr9ciples ( ,art w n1 9 itself1 enable any "o to produce important works1 su* works are n produc$ )\t x4 ,x mu/ be und]/ood1 h[ev]1 t ! und]/&+ ( t]ms & pr9ciples is n1 necess>ily1 an und]/&+ 9 words4 ,x may lie 9 te*nical processes & 9 visual images & may nev] rise1 or %all ,i say fall1 to any =mulation 9 words1 ei!r spok5 or writt54 ,! t]ms & pr9ciples ( ,art h1 z a rule1 be5 und]/ood by ! arti/ 9 ! =m ( te*nical processes & visual images1 n 9 words4 ,x is 9 words t !y w become g5]ally und]/ood4 ,x is 9 words t ,i propose to expla9 !m 9 ?is book4 ,i want to br+ to def9ition :at1 until n[1 has n be5 cle>ly def9$ or exactly measur$4 ,9 a s5se ?is book is a contribution to ,sci5ce r ?an to ,art4 ,x is a contribution to ,sci5ce made by a pa9t]1 :o has us$ 8 ,art 9 ord] to und]/& 8 ,art1 n to produce ,works ( ,art4 ,9 a passage ( ,plato 7,philebus1 #ee7 ,socrates says3 8,if ari?metic1 m5suration1 & weigh+ be tak5 out ( any art1 t :i* rema9s w n be mu*48 8,n mu*1 c]ta9ly10 0 ! reply4 ,! only ?+ :i* rema9s 9 ,art1 beyond measurable quantities & qualities1 is ! p]sonality1 ! peculi> ability or g5ius ( ! arti/ himself4 ,t1 ,i believe1 admits ( no explanation4 ,! elem5t ( p]sonality is :at we mean :5 we speak ( ! elem5t ( inspiration 9 a ,"w ( ,art4 ,und]ly+ ?is elem5t ( p]sonality are ! t]ms & pr9ciples ( ! art4 ,9 !m ! arti/ has f\nd ! possibility ( expression2 9 !m 8 inspiration is convey$ to 8 fell[m54 ,i propose to expla91 n ! arti/1 b ! mode ( expression :i* ! arti/ uses4 ,my purpose1 9 sci5tific language1 is to def9e1 classify1 & expla9 ! ph5om5a ( ,design4 ,9 try+ to d t1 ,i h (t5 be5 at a loss = t]ms & h be5 oblig$1 9 c]ta9 in/ances1 to use t]ms ) new mean+s1 mean+s :i* diff]1 m or less1 f ^? ( common usage & f ^? ( writ]s 9 o!r bran*es ( le>n+4 ,9 all su* cases ,i h tak5 pa9s to def9e my t]ms & to make my mean+s p]fectly cle>4 ,i d n expect any "o to read ?is book :o is n will+ to all[ to my t]ms ! mean+s ,i h giv5 !m4 ,^? :o are unwill+ to accept my def9itions w c]ta9ly n foll[ me to my conclusions4 ,i am giv+ ?is book to ! ,public ) great reluctance4 ,?\gh ,i h _h x 9 m9d = _m ye>s & h put no end ( ?\ght & "w into x1 x seems to me inadequate & unsatisfactory4 ,x w h>dly be publi%$ be=e ,i %all discov] 9 x errors bo? ( omission & commission4 ,! book pres5ts a new def9ition ( pr9ciples1 a new association ( ideas4 ,x is inconceivable t ?is1 my fir/ publi%$ /atem5t1 %\ld be ei!r consi/5t or complete4 ,x w be a long "t1 ,i am sure1 be=e x c be br\ght to a satisfactory %ape4 ,x is simply ! be/ /atem5t t ,i c make at ?is "t4 ,"o reason1 p]haps my be/ reason1 = publi%+ ?is ,!ory1 be=e x is completely work$ out1 is to br+ o!r /ud5ts into ! inve/igation4 ,i ne$ _! cop]ation1 _! sugge/ions1 & _! criticisms4 ,)\t assi/ance f o!rs ! book w\ld n be z good z x is4 ,i am indebt$ to a numb] ( p]sons = helpful sugge/ions4 ,i am p>ticul>ly indebt$ to ?ree m51 :o h be5 associat$ ) me 9 my tea*+3 ,william ,lu!r ,m[ll1 ,h5ry ,hunt ,cl>k1 & ,edg> ,osc> ,p>k]4 ,ea* ( !m has _h a %>e 9 ! "w4 ,i am indebt$ to ,pr(essor ,m[ll = v important contributions to ! doctr9e ( ,rhy?m1 z x is pres5t$ 9 ?is book1 & he has k9dly help$ me 9 ! revision ( ! "w = ! press4 ,my fri5d ,dean ,briggs has k9dly read my pro( %eets1 & ,i am indebt$ to him = _m sugge/ions4 ,,d5man ,w4 ,,ross4 ,,h>v>d ,,univ]sity1 ,febru>y #af1 #aijg4 ,,cont5ts ,,introduction #a ,,positions ,,9 ,,h>mony1 ,,balance1 ,,& ,,rhy?m #i ,,l9es ,,9 ,,h>mony1 ,,balance1 ,,& ,,rhy?m #cg ,,outl9es ,,9 ,,h>mony1 ,,balance1 ,,& ,,rhy?m #if ,,tones ,,& ,,tone-,,relations #aca ,,sequ5ces ,,( ,,values ,,& ,,colors #adc ,,tone-,,h>mony #aeh ,,tone-,,balance #agb ,,tone-,,rhy?m #ahb ,,composition1 ,,g5]al ,,rules #ahf ,,! ,,/udy ,,( ,,ord] ,,9 ,,nature ,,& ,,9 ,,works ,,( ,,art #aij ,,conclusion #aib ,,p>agraph ,,index #aie ,,introduction ,,! ,,mean+ ,,( ,,design #a4 ,by ,design ,i mean ,ord] 9 human feel+ & ?\ght & 9 ! _m & v>i$ activities by :i* t feel+ or t ?\ght is express$4 ,by ,ord] ,i mean1 p>ticul>ly1 ?ree ?+s1--,h>mony1 ,balance1 & ,rhy?m4 ,^! are ! pr9cipal modes 9 :i* ,ord] is reveal$ 9 ,nature &1 ?r\gh ,design1 9 ,works ( ,art4 ,,! ,,ord] ,,( ,,h>mony #b4 ,:5ev] two or m impressions or ideas h some?+ 9 common t is appreciable1 !y are 9 h>mony1 9 ! measure ( :at !y h 9 common4 ,! h>mony increases z ! common elem5t increases2 or ! common elem5ts4 ,x dim9i%es 9 ! measure ( e diff]5ce or contra/4 ,by ! ,ord] ( ,h>mony ,i mean "s recurr5ce or repetition1 "s correspond5ce or lik5ess4 ,! lik5ess may be 9 s\nds or 9 sights1 9 muscul> or o!r s5se-impressions4 ,x may lie 9 s5sations1 9 p]ceptions1 9 ideas1 9 sy/ems ( ?\ght4 ,,! ,,ord] ,,( ,,balance #c4 ,by ! ,ord] ( ,balance ,i mean "s equal opposition & consequ5t equilibrium1 z x occurs at "s mom5t ( ,"t or at "s po9t ( ,space2 an equilibrium :i* induces1 = ! mom5t & 9 its place1 a susp5sion ( all *ange or movem5t1 & causes a pause or a re/4 ,! equilibrium may be "o ( physical =ces 7=ces ( weight or resi/ance7 or =ces ( w4 ,x may be an equilibrium ( s5se-impressions or attractions1 ( int]e/s1 ( alt]native propositions or ideas4 ,x may be ! equilibrium ( a p]fect anti!sis4 ,c]ta9 mom5ts ( ,"t1 c]ta9 po9ts ( ,space1 are di/+ui%$ f o!rs by in/ances ( equilibrium or balance4 ,! balance be+ lo/1 9 any case1 we h at once "s movem5t4 ,if ?is movem5t is regul>1 & m>k$ 9 its regul>ity1 we get1 in/ead ( ,balance1 ,rhy?m4 ,,! ,,ord] ,,( ,,rhy?m #d4 ,by ! ,ord] ( ,rhy?m ,i mean *anges ( s5sation2 *anges 9 muscul> impressions z we feel !m1 9 s\nds z we he> !m1 9 sights z we see !m2 *anges 9 objects1 p1 or ?+s z we "k !m & ?9k ( !m1 *anges :i* induce ! feel+ or idea ( movem5t1 ei!r 9 ! duration ( ,"t or 9 ! ext5sion ( ,space2 provid$ t ! *anges take place at regul> & m>k$ int]vals ( "t or 9 regul> & m>k$ measures ( space4 ,by regul> int]vals & measures ,i mean equal or lawfully v>y+ int]vals & measures4 ,i d n mean1 by ,rhy?m1 *anges simply1 induc+ ! s5se or idea ( movem5t3 ,i mean1 by ,rhy?m1 a regul>ity ( *anges 9 a regul>ity ( measures1 ) ! effect ( movem5t ^u our m9ds4 ,rhy?ms 9 ,"t diff] f ,rhy?ms 9 ,space1 inasmu* z ! movem5t 9 ,"t is 9 "o direction1 inevitably4 ,x lies 9 ! duration & pass+ ( "t1 f :i* no?+ escapes4 ,! movem5t 9 space1 on ! contr>y1 may be 9 any "o ( _m possible directions4 ,a movem5t 9 diff]5t directions1 p>ticul>ly 9 contr>y directions1 am\nts to a negation ( movem5t4 ,9 any space-rhy?m1 !re=e1 ! direction 9 :i* ! rhy?m leads u1 ! direction 9 :i* we foll[ x1 mu/ be unmi/akable4 #e4 ,( ^! ?ree pr9ciples ( ,ord]1 ! fir/ & =emo/1 ! mo/ f>-rea*+ & compreh5sive1 is ! pr9ciple ( ,h>mony4 ,we h ,h>mony 9 all balances1 & we h x also 9 all rhy?ms4 ,x is1 !re=e1 undesirable to ?9k ( ! ?ree pr9ciples z cord9ate4 ,x w be bett] to ?9k ( ! pr9ciple ( ,h>mony fir/1 & !n ( two o!r pr9ciples1 ^? ( ,balance & ( ,rhy?m1 z ly+ )9 ! range ( ,h>mony b n coext5sive ) x4 ,we might express ! idea 9 a logical diagram4 ,illu/ration3 ,fig4 #a ,)9 ! field ( ,h>mony we h two di/9ct modes ( ,ord]--,balance & ,rhy?m2 b we h ,h>mony beyond ! range ( ,balance & beyond ! range ( ,rhy?m4 ,9 cases :]e rhy?ms1 correspond+ 9 "* & 9 direction ( movem5t1 are set side by side1 "o on ! "r1 ! o!r on ! left1 ( a v]tical axis1 s t !y balance1 "o aga9/ ! o!r1 & ! v]tical axis ( ! balance is ! l9e ( ! movem5t1 we h ! union ( all ?ree pr9ciples4 ,?is idea1 also1 may be express$ 9 a logical diagram4 ,illu/ration3 ,fig4 #b ,examples ( ?is union ( ! ?ree pr9ciples ( ,ord] w be giv5 f>!r on4 ,,beauty ,a ,,supreme ,,in/ance ,,( ,,ord] #f4 ,i refra9 f any ref]5ce to ,beauty z a pr9ciple ( ,design4 ,x is n a pr9ciple1 b an exp]i5ce4 ,x is an exp]i5ce :i* defies analysis & has no explanation4 ,we di/+ui% x f all o!r exp]i5ces4 ,x gives u pleasure1 p]haps ! highe/ pleasure t we h4 ,at ! same "t x is idle to talk ab\t x1 or to write ab\t x4 ,! less said ab\t x ! bett]4 8,x is beautiful10 y say4 ,!n somebody asks1 8,:y is x beautiful80 ,!re is no answ] to t "q4 ,y say x is beautiful because x gives y pleasure3 b o!r ?+s give y pleasure :i* are n beautiful4 ,pleasure is1 !re=e1 no crit]ion ( ,beauty4 ,:at is ! pleasure :i* ,beauty gives8 ,x is ! pleasure :i* y h 9 ! s5se ( ,beauty4 ,t is all y c say4 ,y _c expla9 ei!r ! exp]i5ce or ! k9d ( pleasure :i* x gives y4 ,:ile ,i am q unable to give any def9ition or explanation ( ,beauty1 ,i "k :]e to look = x1 :]e ,i am sure to f9d x4 ,! ,beautiful is reveal$1 always1 s f> z ,i "k1 9 ! =ms ( ,ord]1 9 ! modes ( ,h>mony1 ( ,balance1 or ( ,rhy?m4 ,:ile !re are _m in/ances ( ,h>mony1 ,balance1 & ,rhy?m :i* are n p>ticul>ly beautiful1 !re is1 ,i believe1 no?+ really beautiful :i* is n ord]ly 9 "o or ! o!r1 9 two1 or 9 all ?ree ( ^! modes4 ,9 seek+ ! ,beautiful1 !re=e1 we look = x 9 in/ances ( ,ord]1 9 in/ances ( ,h>mony1 ,balance1 & ,rhy?m4 ,we %all f9d x 9 :at may be call$ supreme in/ances4 ,?is is p]haps our ne>e/ approa* to a def9ition ( ,beauty3 t x is a supreme in/ance ( ,ord]1 intuitively felt1 in/9ctively appreciat$4 ,,! ,,arts ,,z ,,diff]5t ,,modes ,,( ,,expression #g4 ,! ,arts are diff]5t =ms or modes ( expression3 modes ( feel+1 modes ( ?\ght1 modes ( action4 ,!re are _m ,arts 9 :i* diff]5t t]ms ( expression1 diff]5t mat]ials1 diff]5t me?ods are us$4 ,! pr9cipal ,arts are 7#a7 ,gymna/ics1 includ+ ,danc+1 7#b7 ,music1 7#c7 ,spee*1 spok5 & writt51 7#d7 ,con/ruction ) all k9ds ( mat]ials1 7#e7 ,sculpture1 includ+ ,model+ & ,c>v+1 7#f7 ,draw+ & ,pa9t+4 ,^! are ! pr9cipal ,arts1 b !re are _m o!rs1 m or less connect$ ) !m4 ,design comes into all ( ^! ,arts1 br++ ,ord]1 9 "s cases ,beauty4 ,,! ,,art ,,( ,,draw+ ,,& ,,pa9t+ #h4 ,! ,art :i* ,i h /udi$ & practic$1 ! ,art 9 :i* ,i am giv+ in/ruction1 is t ( ,draw+ & ,pa9t+4 ,by ! ,art ( ,draw+ & ,pa9t+ ,i mean expression 9 pigm5t-tones 7values1 colors1 int5sities ( color7 spread 9 diff]5t measures or quantities & 9 diff]5t %apes3 %apes be+ diff]5ces ( "* giv5 to a l9e by its /raightness or curvature1 to a spot or area by its outl9e or cont\r4 ,by ,draw+ & ,pa9t+ ,i mean1 !re=e1 expression by l9es & spots ( pa9t4 ,,two ,,modes ,,( ,,draw+ ,,& ,,pa9t+ #i4 ,!re are two modes ( ,draw+ & ,pa9t+1 ! mode ( ,pure ,design & ! mode ( ,repres5tation4 ,,pure ,,design #aj4 ,by ,pure ,design ,i mean simply ,ord]1 t is to say1 ,h>mony1 ,balance1 & ,rhy?m1 9 l9es & spots ( pa9t1 9 tones1 measures1 & %apes4 ,pure ,design appeals to ! eye j z absolute ,music appeals to ! e>4 ,! purpose 9 ,pure ,design is to a*ieve ,ord] 9 l9es & spots ( pa9t1 if possible1 ! p]fection ( ,ord]1 a supreme in/ance ( x1 ! ,beautiful3 ?is ) no o!r1 no fur!r1 no high] motive2 j = ! satisfaction1 ! pleasure1 ! delight ( x4 ,9 ! practice ( ,pure ,design we aim at ,ord] & hope = ,beauty4 ,ev5 ! motive ( giv+ pleasure to o!rs lies beyond ! prop] purpose ( ,pure ,design1 ?\gh x con/antly happ5s t 9 pleas+ ourselves we give o!rs pleasure4 ,,applications ,,( ,,design #aa4 ,! application ( ,design 9 ! v>i\s ,arts & ,crafts is well und]/ood & appreciat$4 ,we h _m in/ances & examples 9 ! ,art ( ! ,pa/4 ,! possibility ( ,pure ,design1 pure ,art1 foll[$ = ! sake ( ,ord] & ,beauty1 ) no purpose ( s]vice or ( utility1 is n at all und]/ood or appreciat$4 ,i ?9k ( ,pure ,design z ,i ?9k ( ,music4 ,music is ! arrangem5t & composition ( s\nds = ! sake ( ,ord] & ,beauty1 to give pleasure to ! e> ( ! compos]4 ,pure ,design is ! arrangem5t & composition ( l9es & spots ( pa9t = ! sake ( ,ord] & ,beauty1 to give pleasure to ! eye ( ! design]4 ,i am prep>$ to admit1 h[ev]1 t z ,music1 once creat$1 be+ appropriate to ! occasion :5 x is p]=m$ & to ! mood ( ! li/5]s1 may give pleasure to _m p]sons1 s ,pure ,design1 once a*iev$1 be+ appropriate to ! "t & ! place ( exhibition & to ! mood ( ! behold]s1 may give pleasure to o!rs besides ! design]4 ,9 t s5se1 ,i am will+ to all[ t ,pure ,design may be ,appli$ ,art1--,art appli$ 9 ! s]vice ( ,humanity1 its purpose be+ to br+ pleasure into human exp]i5ce4 ,! und]ly+ motive ( x1 h[ev]1 is f\nd n 9 ! s]vice ( humanity1 b 9 ! ideal ( ! arti/4 ,he aims at ,ord] & hopes = ,beauty1 z ! highe/ rew>d ( 8 ef=t4 ,john ,seba/ian ,ba* said ( ,music3 8,x is = ! glory ( ,god & a pleasant recreation40 ,t is :at ,i mean4 ,! design]1 l ! musician1 seeks fir/ ( all to a*ieve ,ord] & ,beauty = ! sake ( ,ord] & ,beauty4 ,t is 8 religion1--! wor%ip ( ! ,ideal4 ,:5 ! ,ideal is realiz$1 ! object :i* has be5 produc$ may s]ve a useful purpose 9 giv+ pleasure1 & p]haps inspiration1 to o!rs4 ,! pr9ciples ( ,pure ,design :i* are def9$ & illu/rat$ 9 ?is book are ! pr9ciples :i* %\ld be foll[$ 9 all applications ( ,design 9 ! ,arts & ,crafts4 ,9 su* applications1 h[ev]1 ! ideals ( design are (t5 obscur$ by ! consid]ation ( mat]ials & te*nical processes on ! "o h&1 & ( s]vice or utility on ! o!r4 ,x w be wor? :ile1 !re=e1 = ^? :o wi% to br+ ,design into _! "w1 :atev] t is1 to /udy ,design 9 ! ab/ract1 ,pure ,design1 s t !y may "k1 be=e !y und]take to use x1 :at ,design is4 ,x is ! purpose ( ?is book to expla9 :at x is4 ,,repres5tation #ab4 ,ord]1 :i* 9 ,pure ,design is an end1 becomes 9 ,repres5tation a means to an end2 ! end be+ ! ,tru? ( ,repres5tation4 ,9 ,repres5tation we are no long] deal+1 z 9 ,pure ,design1 ) mean+less t]ms1 or1 if ! t]ms h mean+s1 ) no reg>d = !m4 ,9 ,repres5tation we are putt+ l9es & spots ( pa9t toge!r = ! sake ( _! mean+s4 ,design 9 ,repres5tation means ,ord] 9 ! composition or arrangem5t ( mean+s4 ,:at we aim at is ! ,tru? ( ,repres5tation 9 a =m ( expression :i* w be simple1 cle>1 reasonable1 & consi/5t1 z well z true4 ,! att5tion mu/ be direct$ to :at is important1 away f :at is unimportant4 ,objects1 p1 & ?+s repres5t$ mu/ be br\ght out & emphasiz$ or suppress$ & subord9at$1 accord+ to ! ,idea or ,tru? :i* ! arti/ wi%es to express4 ,! irrelevant mu/ be elim9at$4 ,! inconsi/5t & ! incongru\s mu/ be avoid$4 ,t is :at ,i mean by ,design 9 ,repres5tation1 ! k ( ,nature & ,life pres5t$ 9 a sy/ematic1 logical1 & ord]ly way4 ,,repres5tation ,,9 ,,=ms ,,( ,,design #ac4 ,x sometimes happ5s t we h ! ,tru? ( ,repres5tation 9 a =m ( ,pure ,design or ,pure ,design comb9$ ) ,repres5tation4 ,9 ,poetry we h ! mean+ ( ! words 9 ! measures ( ! v]se4 ,s 9 ,repres5tation x is sometimes possible to a*ieve ! ,tru? 9 =ms ( ,h>mony1 ,balance1 & ,rhy?m4 ,9 su* cases ! appeal is simultane\sly to ! love ( ,k & to ! s5se ( ,ord] & ( ,beauty1 s t we h an /hetic pleasure 9 ! /atem5t ( ! ,tru?4 ,9 ?is book ,i %all expla9 :at ,i mean by ,draw+ & ,pa9t+ 9 ,pure ,design4 ,lat]1 ,i hope to write ano!r book on ,design 9 ,repres5tation4 ,,draw+ ,,& ,,pa9t+ ,,9 ,,pure ,,design ,,positions ,,def9ition ,,( ,,positions #ad4 ,take a p5cil & a piece ( pap]4 ,) ! p5cil1 on ! pap]1 m>k a dot or po9t4 ,illu/ration3 ,fig4 #c ,by ?is dot 7,a7 ?ree ideas are express$3 an idea ( ,tone1 ! tone ( lead 9 ! p5cil2 an idea ( ,measure1 ! ext5t ( ! space cov]$ by ! dot2 & an idea ( ,%ape1 ! "* giv5 to ! dot by its outl9e or cont\r4 ,! dot is s small t its tone1 its measure1 & its %ape w n be s]i\sly consid]$4 ,!re is ano!r idea1 h[ev]1 :i* is express$ by ! dot or po9t1--an idea ( ,position4 ,t is its prop] mean+ or signification4 ,!re is presumably a reason = giv+ ! dot "o position r ?an ano!r4 ,,positions ,,det]m9$ ,,by ,,directions ,,& ,,di/ances #ae4 ,put ano!r dot 7,b7 on y\r pap]1 n f> f dot 8,a40 ,illu/ration3 ,fig4 #d ,we h n[ a relation ( two positions1--! relation ( position 8,a0 to position 8,b40 ,! relation is "o ( ,directions & ( ,di/ances4 ,proce$+ f 8,a0 9 a c]ta9 direction to a c]ta9 di/ance we rea* 8,b40 ,proce$+ f 8,b0 9 a c]ta9 direction & to a c]ta9 di/ance we rea* 8,a40 ,e position means two ?+s2 a direction & a di/ance tak5 f "s po9t :i* may be describ$ z ! premise-po9t4 ,,directions #af4 ,directions may be ref]r$ ei!r to ! ,horizontal or to ! ,v]tical4 ,ref]r+ !m to ! horizontal1 we say ( a c]ta9 direction1 t x is up-to-!-left1 or up-to-!-"r1 or d[n-to-!-left1 or d[n-to-!-"r1 a c]ta9 numb] ( degrees4 ,x may be ?irty 7#cj71 x may be =ty-five 7#de71 x may be sixty 7#fj71--;any numb] ( degrees up to n9ety 7#ij71 9 :i* case we say simply t ! direction is up or d[n4 ,directions on ! horizontal may be describ$ by ! t]ms1 to ! "r or to ! left4 ,illu/ration3 ,fig4 #e ,! me?od ( describ+ & def9+ diff]5t directions f any po9t1 z a c5t]1 is cle>ly expla9$ by ?is diagram4 ,,di/ances #ag4 ,! def9ition ( ,di/ances 9 any direction is well und]/ood4 ,9 def9+ position 8,b10 9 ,fig4 #d1 we say t x is1 9 a direction f 8,a10 ! premise-po9t1 d[n-to-!-"r =ty-five degrees 7#de71 t x is at a di/ance f 8,a0 ( "o in*4 ,di/ances are always tak5 f premise-po9ts4 ,,positions ,,det]m9$ ,,by ,,triangulations #ah4 ,if we m>k a ?ird dot1 8,c10 on our pap] & wi% to def9e its position1 we may give ! direction & ! di/ance f 8,a10 or f 8,b10 or1 if we pref]1 we may foll[ ! pr9ciple ( ,triangulation & give two directions1 "o f 8,a0 & ! o!r f 8,b40 ,no di/ances ne$ be giv5 9 t case4 ,! position ( 8,c0 w be f\nd at ! int]section ( ! two directions4 ,illu/ration3 ,fig4 #f ,! pr9ciple ( ,triangulation is illu/rat$ 9 ! above diagram4 ,,int]vals #ai4 ,we %all h occasion to speak n only ( ,di/ances1 b ( ,int]vals4 ,!y may be def9$ z int]m$iate spaces4 ,! spaces betwe5 ! po9ts 8,a0 & 8,b10 8,a0 & 8,c10 8,b0 & 8,c10 9 ,fig4 #f1 are ,int]vals4 ,,scale ,,9 ,,relations ,,( ,,positions #bj4 ,giv5 any relation ( positions1 ! scale may be *ang$ by *ang+ ! int]vals1 provid$ we make no *ange ( directions4 ,t is well und]/ood4 ,be=e proce$+ to ! consid]ations :i* foll[1 ,i mu/ ask ! read] to ref] to ! def9itions ( ,h>mony1 ,balance1 & ,rhy?m :i* ,i h giv5 9 ! ,introduction4 ,,! ,,ord] ,,( ,,h>mony ,,9 ,,positions3 ,,directions1 ,,di/ances1 ,,int]vals #ba4 ,all ,positions ly+ 9 ! same direction & at ! same di/ance f a giv5 po9t1 tak5 z a premise-po9t1 are "o4 ,!re is no su* ?+1 !re=e1 z a ,h>mony ( ,positions4 ,positions 9 ,h>mony are id5tical positions4 ,two or m positions may1 h[ev]1 lie 9 ! same direction f or at ! same di/ance f a giv5 po9t tak5 z a premise-po9t4 ,9 t case1 ! two or m positions1 hav+ a direction or a di/ance 9 common1 are1 to t ext5t1 9 h>mony4 #bb4 ,:at d we mean by ,h>mony ( ,directions8 ,illu/ration3 ,fig4 #g ,?is is an example ( ,direction-,h>mony4 ,all ! po9ts or positions lie 9 "o & ! same direction f ! premise-po9t 8,a40 ,! di/ances f 8,a0 v>y4 ,!re is no ,h>mony ( ,int]vals4 ,directions be+ def9$ by angles ( div]g5ce1 we may h a ,h>mony ( ,directions 9 ! repetition ( simil> angles ( div]g5ce3 9 o!r words1 :5 a c]ta9 *ange ( direction is repeat$4 ,illu/ration3 ,fig4 #h ,9 ?is case ! angles ( div]g5ce are equal4 ,!re is a ,h>mony1 n only 9 ! repetition ( a c]ta9 angle1 b 9 ! correspond5ce ( ! int]vals4 ,illu/ration3 ,fig4 #i ,?is is an example ( ,h>mony produc$ by ! repetition ( a c]ta9 alt]nation ( directions4 ,illu/ration3 ,fig4 #aj ,9 ?is case we h ,h>mony 9 ! repetition ( a c]ta9 relation ( directions 7angles ( div]g5ce74 ,9 ^! cases1 ,fig4 #i & ,fig4 #aj1 !re is ,h>mony also1 9 ! repetition ( a c]ta9 relation ( int]vals4 #bc4 ,two or m positions may lie at ! same di/ance f a giv5 po9t tak5 z a premise-po9t4 ,9 t case ! positions1 hav+ a c]ta9 di/ance 9 common1 are1 to t ext5t1 9 ,h>mony4 ,illu/ration3 ,fig4 #aa ,?is is an example ( ,di/ance-,h>mony4 ,all ! po9ts are equally di/ant f ! premise-po9t 8,a40 ,! directions v>y4 ,we may h ,di/ance-,h>mony1 also1 9 ! repetition ( a c]ta9 relation ( di/ances4 ,illu/ration3 ,fig4 #ab ,?is is an illu/ration ( :at ,i h j describ$4 ,! ,h>mony is ( a c]ta9 relation ( di/ances repeat$4 #bd4 ,int]vals1 t is to say int]m$iate spaces1 are 9 ,h>mony :5 !y h ! same measure4 ,illu/ration3 ,fig4 #ac ,9 ?is case we h1 n only a ,h>mony ( ,direction1 z 9 ,fig4 #g1 b also a ,h>mony ( ,int]vals4 ,illu/ration3 ,fig4 #ad ,9 ?is case ! po9ts are 9 a gr\p & we h1 z 9 ,fig4 #aa1 a ,h>mony ( ,di/ances f ! premise-po9t 8,a40 ,we h also a ,h>mony ( ,int]vals1 ! di/ances betwe5 adjac5t po9ts be+ equal4 ,we h a ,h>mony ( ,int]vals1 n only :5 ! int]vals are equal1 b :5 a c]ta9 relation ( int]vals is repeat$4 ,illu/ration3 ,fig4 #ae ,! repetition ( ! ratio "o to ?ree 9 ^! int]vals is di/9ctly appreciable4 ,9 ! repetition we h ,h>mony1 ?\gh we h no ,h>mony 9 ! t]ms ( ! ratio itself1 t is to say1 no ,h>mony t is appreciable 9 ! s5se ( vision4 ,! fact t "o & ?ree are bo? multiples ( "o means t "o & ?ree h some?+ 9 common1 b inasmu* z ! common divisor1 "o1 _c be visually appreciat$1 z su* 7,i feel sure t x _c71 x has no int]e/ or value 9 ,pure ,design4 ,illu/ration3 ,fig4 #af ,! relation ( int]vals is1 9 ?is case1 ! relation ( ?ree-"o-five4 ,we h ,h>mony 9 ! repetition ( ?is relation ( int]vals ?\gh !re is no ,h>mony 9 ! relation itself1 :i* is repeat$4 ,illu/ration3 ,fig4 #ag ,9 ?is case1 also1 we h ,int]val-,h>mony1 b z ! int]vals 9 ! v]tical & horizontal directions are %ort] ?an ! int]vals 9 ! diagonal directions1 ! ,h>mony is t ( a relation ( int]vals repeat$4 #be4 ,9 mov+ f po9t to po9t 9 any s]ies ( po9ts1 x w be f\nd easi] to foll[ ! s]ies :5 ! int]vals are %ort ?an :5 !y are long4 ,9 ,fig4 #ag x is easi] to foll[ ! v]tical or horizontal s]ies ?an x is to foll[ a diagonal s]ies1 because 9 ! v]tical & horizontal directions ! int]vals are %ort]4 ,illu/ration3 ,fig4 #ah ,9 ?is case x is easi] to move up or d[n on ! v]tical ?an 9 any o!r directions1 because ! %ort int]vals lie on ! v]tical4 ,! horizontal int]vals are long]1 ! diagonal int]vals long] /ill4 ,illu/ration3 ,fig4 #ai ,9 ?is case ! s]ies :i* lies on ! diagonal up-left-d[n-"r is ! m easily foll[$4 ,x is possible 9 ?is way1 by means ( %ort] int]vals1 to keep ! eye on c]ta9 l9es4 ,! applications ( ?is pr9ciple are v int]e/+4 #bf4 ,9 ea* position1 z indicat$ by a po9t 9 ^! arrangem5ts1 may be plac$ a composition ( dots1 l9es1 outl9es1 or areas4 ,! dots indicate positions 9 :i* any ( ! possibilities ( design may be develop$4 ,!y are po9ts f :i* all ?+s may em]ge & become visible4 ,,! ,,ord] ,,( ,,balance ,,9 ,,positions3 ,,directions1 ,,di/ances1 ,,& ,,int]vals #bg4 ,directions balance :5 !y are opposite4 ,illu/ration3 ,fig4 #bj ,! opposite directions1 "r & left1 balance on ! po9t f :i* !y are tak54 #bh4 ,equal di/ances 9 opposite directions balance on ! po9t f :i* ! directions are tak54 ,illu/ration3 ,fig4 #ba ,! equal di/ances ,,ab & ,,ac1 tak5 9 ! directions ,,ab & ,,ac respectively1 balance on ! po9t 8,a0 f :i* ! directions are tak54 #bi4 ,two directions balance :51 tak5 f any po9t1 !y div]ge at equal angles f any axis1 v]tical1 horizontal1 or diagonal4 ,illu/ration3 ,fig4 #bb ,! directions ,,ab & ,,ac balance on ! v]tical axis ,,ad f :i* !y div]ge equally1 t is to say1 at equal angles4 #cj4 ,equal di/ances balance 9 directions :i* div]ge equally f a giv5 axis4 ,illu/ration3 ,fig4 #bc ,! equal di/ances ,,ab & ,,ac balance 9 ! directions ,,ab & ,,ac :i* div]ge equally f ! axis ,,ad1 mak+ ! equal angles ,,cad & ,,dab4 ,bo? directions & di/ances balance on ! v]tical axis ,,ad4 #ca4 ,! positions ,b & ,c 9 ,fig4 #bc1 dep5d+ on balanc+ directions & di/ances1 balance on ! same axis4 ,we %\ld feel ?is balance ( ! positions ,a & ,b on ! v]tical axis ev5 )\t any indication ( ! axis4 ,we h s def9ite an image ( ! v]tical axis t :5 x is n drawn we imag9e x4 ,illu/ration3 ,fig4 #bd ,9 ?is case ! two positions ,c & ,b _c be said to balance1 because !re is no sugge/ion1 no indication1 & no visual image ( any axis4 ,x is only ! v]tical axis :i* w be imag9$ :5 n drawn4 #cb4 ,p]fect v]ticality 9 relations ( position sugge/s /ability & balance4 ,! relation ( positions ,c-,b 9 ,fig4 #bd is "o ( in/ability4 ,illu/ration3 ,fig4 #be ,^! two positions are felt to balance because !y lie 9 a p]fectly v]tical relation1 :i* is a relation ( /ability4 ,horizontality 9 relations1 ( position is also a relation ( /ability4 ,see ,fig4 #bh1 p4 #ba4 #cc4 ,all ^! consid]ations lead u to ! def9ition ( ,symmetry4 ,by ,symmetry ,i mean opposite directions or incl9ations1 opposite & equal di/ances1 opposite positions1 & 9 ^? positions equal1 correspond+1 & oppos$ attractions on a v]tical axis4 ,briefly1 ,symmetry is "r & left balance on a v]tical axis4 ,?is axis w be imag9$ :5 n drawn4 ,9 ,symmetry we h a balance :i* is p]fectly obvi\s & in/9ctively felt by ev]ybody4 ,all o!r =ms ( ,balance are comp>atively obscure4 ,"s ( !m may be describ$ z occult4 ,illu/ration3 ,fig4 #bf ,9 ?is case we h a symmetry ( positions :i* means opposite directions1 opposite & equal di/ances1 & simil> & opposite attractions 9 ^? positions4 ,! attractions are black dots correspond+ 9 tone1 measure1 & %ape4 ,illu/ration3 ,fig4 #bg ,9 ?is case we h a balance ( positions 7directions & di/ances7 & attractions 9 ^? positions1 n only on ! v]tical axis b on a c5t]4 ,t means ,symmetry reg>d+ ! v]tical axis1 ,balance reg>d+ ! c5t]4 ,if we turn ! figure1 slightly1 f ! v]tical axis1 we %all /ill h ,balance ^u a c5t] & axial ,balance2 b ,symmetry1 :i* dep5ds ^u ! v]tical axis1 w be lo/4 #cd4 ,! c5tral v]tical axis ( ! :ole composition %\ld pr$om9ate 9 symmetrical balances4 ,illu/ration3 ,fig4 #bh ,9 ?is case we d n feel ! balance ( attractions cle>ly or satisfactorily1 because ! v]tical axis ( ! :ole arrangem5t does n pr$om9ate suffici5tly ov] ! six axes ( adjac5t attractions4 ,x is necess>y1 9 ord] t symmetrical balance %all be in/9ctively felt1 t ! c5tral v]tical axis pr$om9ate4 ,illu/ration3 ,fig4 #bi ,! c5tral v]tical axis is cle>ly indicat$ 9 ?is case4 ,illu/ration3 ,fig4 #cj ,"h1 also1 ! c5tral v]tical axis is cle>ly indicat$4 #ce4 ,all relations ( position 7directions1 di/ances1 int]vals71 z indicat$ by dots or po9ts1 :e!r ord]ly or n1 be+ inv]t$ on ! v]tical axis1 give u an obvi\s symmetrical balance4 ,illu/ration3 ,fig4 #ca ,?is is a relation ( positions to be inv]t$4 ,illu/ration3 ,fig4 #cb ,"h ! same relation is repeat$1 ) its inv]sion to ! "r on a v]tical axis4 ,! result is an obvi\s symmetrical balance4 ,if ?is inv]sion 7 made on any o!r ?an ! v]tical axis1 ! result w\ld be ,balance b n ,symmetry4 ,! balance w\ld /ill be axial1 b ! axis1 n be+ v]tical1 ! balance w\ld n be symmetrical4 #cf4 ,9 ! case ( any unsymmetrical arrangem5t ( dots1 ! dots become equal attractions 9 ! field ( vision1 provid$ !y are ne> en\gh toge!r to be se5 toge!r4 ,to be satisfactorily se5 z a s+le composition or gr\p !y ought to lie1 all ( !m1 )9 a visual angle ( ?irty degrees4 ,we may1 )9 ^! limits1 disreg>d ! fact t visual attractions lose _! =ce z !y are remov$ f ! c5t] ( ! field ( vision4 ,z equal attractions 9 ! field ( vision1 ! dots 9 any unsymmetrical arrangem5t may be br\ght into a balance by weigh+ ! sev]al attractions & indicat+ :at ,i might call ! c5t] ( equilibrium4 ,?is is be/ done by means ( a symmetrical inclosure or frame4 ,9 asc]ta9+ j :]e ! c5t] is1 9 any case1 we dep5d ^u visual s5sitiv5ess or visual feel+1 guid$ by an und]/&+ ( ! pr9ciple ( balance3 t equal attractions1 t5sions or pulls1 balance at equal di/ances f a giv5 c5t]1 t unequal attractions balance at di/ances inv]sely proportional to !m4 ,giv5 c]ta9 attractions1 to f9d ! c5t]1 we weigh ! attractions toge!r 9 ! field ( vision & obs]ve ! position ( ! c5t]4 ,9 simple cases we may be able to prove or disprove our visual feel+ by calculations & reason+4 ,9 cases1 h[ev]1 :]e ! attractions v>y 9 _! tones1 measures1 & %apes1 & :]e !re are qualities z well z quantities to be consid]$1 calculations & reason+ become difficult if n impossible1 & we h to dep5d ^u visual s5sitiv5ess4 ,all balances ( positions1 z indicat$ by dots correspond+ 9 tone1 measure1 & %ape1 are balances ( equal attractions1 & ! calculation to f9d ! c5t] is a v simple "o4 ,illu/ration3 ,fig4 #cc ,"h1 = example1 ! sev]al attractions1 correspond+ & equal1 lie well )9 ! field ( vision4 ,! me?od foll[$ to balance !m is t :i* ,i h j describ$4 ,! c5t] ( equilibrium 0 f\nd & !n indicat$ by a symmetrical fram+4 ,move ! frame up or d[n1 "r or left1 & ! c5t] ( ! frame & ! c5t] ( ! attractions )9 x w no long] co9cide1 & ! balance w be lo/4 ,we might say ( ?is arrangem5t t x is a ,h>mony ( ,positions due to ! co9cid5ce ( two c5t]s1 ! c5t] ( ! attractions & ! c5t] ( ! fram+4 #cg4 ,x w be obs]v$ t ! =ce ( ! symmetrical inclosure %\ld be suffici5t to ov]p[] any sugge/ion ( movem5t :i* may lie 9 ! attractions inclos$ by x4 ,illu/ration3 ,fig4 #cd ,9 ?is case ! dots & ! inclosure are ab\t equally attractive4 ,illu/ration3 ,fig4 #ce ,9 ?is case ! =ce ( attractions 9 ! symmetrical outl9e is suffici5t to ov]p[] any sugge/ion ( in/ability & movem5t :i* may lie 9 ! attractions inclos$ by x4 ,!re is ano!r =m ( ,balance1 ! ,balance ( ,incl9ations1 b ,i w def] its consid]ation until ,i c illu/rate ! idea by l9es4 ,,! ,,ord] ,,( ,,rhy?m ,,9 ,,positions3 ,,directions1 ,,di/ances1 ,,int]vals #ch4 ,9 any unsymmetrical relation ( positions 7directions1 di/ances1 int]vals71 9 :i* ! balance-c5t] is n cle>ly & suffici5tly indicat$1 !re is a sugge/ion ( movem5t4 ,! eye1 n be+ held by any balance1 readily foll[s ?is sugge/ion4 ,illu/ration3 ,fig4 #cf ,9 ?is case we feel t ! gr\p ( dots is unbalanc$ 9 "* & un/able 9 its position or attitude4 ,x is easy1 inevitable inde$1 to imag9e ! gr\p fall+ away to ! "r4 ,?is is due1 no d\bt1 to ! visual habit ( imag9+ a base-l9e :5 x is n drawn4 ,our judgm5ts are con/antly made ) ref]5ce to ! imag9$ /&>ds ( v]ticality & horizontality4 ,we seem to be provid$ ) a plumb-l9e & a level )\t be+ consci\s ( ! fact4 ,illu/ration3 ,fig4 #cg ,9 ?is case !re is a sugge/ion ( fall+ d[n to ! left due to ! feel+ ( in/ability4 ,a symmetrical fram+ hold+ ! eye at ! c5t] ( equilibrium w\ld prev5t ! feel+ ( movem5t1 provid$ ! fram+ 7 suffici5tly /rong 9 its attractions4 ,9 ! examples ,i h giv5 7,fig4 #cf & ,fig4 #cg7 we h movem5t1 b no ,rhy?m4 #ci4 ,!re is ano!r type ( movem5t :i* we mu/ consid]1--! type ( movem5t :i* is caus$ by a gradual cr[d+ toge!r ( attractions4 ,illu/ration3 ,fig4 #ch ,!re is no?+ 9 ?is s]ies ( dots b ! h>mony ( correspond+ attractions & int]vals repeat$ 9 a h>mony ( direction4 ,if1 in/ead ( ! repetition ( equal int]vals1 we _h a regul> progression ( int]vals1 ei!r ari?metical or geometrical1 we %\ld feel a movem5t 9 ! direction ( dim9i%+ int]vals4 ,illu/ration3 ,fig4 #ci ,9 ! above example ! *anges ( int]val are ^? ( an ari?metical progression4 ,illu/ration3 ,fig4 #dj ,9 ,fig4 #dj ! *anges ( int]val are ^? ( a geometrical progression4 ,! movem5t to ! left ?r\gh ^! sequ5ces is1 no d\bt1 some:at *eck$ or prev5t$ by ! habit ( read+ to ! "r4 ,illu/ration3 ,fig4 #da ,! angle ,,fab is ! angle ( vision )9 :i* ! sequ5ce is obs]v$4 ,at ! end ,f ( ! sequ5ce !re is a great] numb] ( attractions 9 a giv5 angle ( vision ?an at ! end ,b1 s ! eye is drawn t[>ds ! left4 ,! pull on ! eye is great] at ! end ,f because ( ! great] numb] & ! cr[d+ toge!r ( attractions4 ,9 ! examples j giv5 7,figs4 #ci1 #dj71 we h n only movem5ts 9 c]ta9 directions1 b movem5ts 9 regul> & m>k$ measures4 ,! movem5ts are1 !re=e1 rhy?mical1 accord+ to ! def9ition ,i h giv5 ( ,rhy?m4 #dj4 ,x is evid5t t any relation ( positions1 balanc$ or unbalanc$1 may be sub/itut$ = ! s+le dots or po9ts 9 ! figures j giv54 ,su* sub/itutions h ! foll[+ possibilities4 #da4 ,fir/4 ,:5 ! po9ts lie 9 a s]ies1 at equal int]vals1 ! sub/itution ( a symmetrical gr\p ( positions at ea* po9t gives no ,rhy?m1 only ,h>mony4 ,illu/ration3 ,fig4 #db ,!re is no movem5t 9 ?is s]ies ( repetitions4 ,!re is consequ5tly no ,rhy?m4 ,disreg>d+ ! habit ( read+ to ! "r1 :i* induces ! eye to move 9 t direction1 x is z easy to move t[>d ! left z t[>d ! "r4 ,x requires m ?an repetitions at equal int]vals to produce ! feel+ ( ,rhy?m4 ,!re mu/ be movem5t1 & ! movem5t mu/ h a def9ite direction4 #db4 ,second4 ,! sub/itution at ea* po9t ( a symmetrical gr\p at equal int]vals1 z be=e1 b ) a progressive *ange ( scale1 w give u ,rhy?m4 ,! movem5t w be due to ! gradual cr[d+ toge!r ( attractions at "o end ( ! s]ies4 ,illu/ration3 ,fig4 #dc ,9 ?is case we h ! repetition ( a symmetrical relation ( positions at equal int]vals ) a gradation ( scale 9 ! repetitions4 ,! result is a ,rhy?m1 9 :i* ! movem5t is f left to "r1 ow+ to ! great] cr[d+ toge!r ( attractions at ! "r end ( ! s]ies4 ,! feel+ ( ,rhy?m is no d\bt some:at enhanc$ by our habit ( read+ to ! "r1 :i* facilitates ! movem5t ( ! eye 9 t direction4 #dc4 ,?ird4 ,! sub/itution ( an un/able gr\p at ea* po9t ( ! sequ5ce1 ! repetitions be+ at equal int]vals1 gives u a ,rhy?m1 due simply to ! movem5t ( ! gr\p itself1 :i* is un/able4 ,illu/ration3 ,fig4 #dd ,tak+ ! relation ( positions giv5 9 ,fig4 #cf & repeat+ x at equal int]vals1 x w be obs]v$ t ! fall+-to-!-"r movem5t1 :i* is ! result ( in/ability1 is convey$ to ! :ole s]ies ( repetitions4 ,to make x p]fectly cle> t ! movem5t ( ?is ,rhy?m is due to ! sugge/ion ( movem5t 9 ! relation ( positions :i* is repeat$1 ,i w ask ! read] to comp>e x ) ! repetition ( a symmetrical gr\p 9 ,fig4 #db4 ,!re is no movem5t 9 t case1 !re=e no ,rhy?m4 #dd4 ,f\r?4 ,! movem5t 9 ,fig4 #dd may be increas$ by a dim9ution ( scale & consequ5t cr[d+ toge!r ( ! dots1 provid$ ! movem5t ( ! gr\ps & ! cr[d+ toge!r h ! same direction4 ,illu/ration3 ,fig4 #de ,9 ?is case1 z ,i h said1 ! movem5t ( ,fig4 #dd is en=c$ by ! pres5ce ( ano!r elem5t ( movem5t1 t ( a gradation ( scale & consequ5t cr[d+ toge!r 9 ! gr\ps4 ,! two movem5ts h ! same direction4 ,! movem5t ( ! cr[d+ is n s /rong z t :i* is caus$ by ! in/ability ( ! gr\p itself4 #de4 ,fif?4 ,a symmetrical relation ( positions1 be+ repeat$ 9 a s]ies ) gradually dim9i%+ int]vals betwe5 ! repeats1 w give u a feel+ ( rhy?mic movem5t4 ,x w be due to a gradual increase 9 ! numb] ( attractions z ! eye passes f "o angle ( vision to ano!r4 ,see ,fig4 #da4 ,! ,rhy?m w1 no d\bt1 be some:at ret>d$ by ! s5se ( successive axes ( symmetry4 ,illu/ration3 ,fig4 #df ,9 ?is case a symmetrical gr\p is repeat$ 9 a progression ( measures4 ,! movem5t is t[>d ! great] numb] ( attractions at ! "r end ( ! s]ies4 ,?is increase 9 ! numb] ( attractions is due simply to dim9i%+ int]vals 9 t direction4 ,! eye moves ?r\gh a s]ies ( angles t[>d ! angle :i* conta9s ! greate/ numb] ( attractions4 ,! read] c h>dly fail to feel ! successive axes ( symmetry z a ret>d+ elem5t 9 ?is ,rhy?m4 #df4 ,six?4 ,symmetrical relations ( position may be repeat$ 9 progressions ( scale & ( int]vals4 ,9 t case we get two movem5ts1 "o caus$ by a gradual increase 9 ! numb] ( attractions 9 successive angles ( vision1 ! o!r be+ due to a gradual cr[d+ toge!r & conv]g5ce ( attractions 9 ! same s]ies ( angles4 ,illu/ration3 ,fig4 #dg ,comp>+ ?is ,rhy?m ) ! ,rhy?m ( ,fig4 #dc1 ! read] w appreciate ! =ce ( a dim9ution ( scale 9 connection ) a dim9ution ( int]vals4 #dg4 ,sev5?4 ,un/able gr\ps may be repeat$ 9 progressions ( int]vals1 9 :i* case ! movem5t 9 ! gr\p is convey$ to ! :ole s]ies1 9 :i* !re w be1 also1 ! movem5t ( a gradual increase ( attractions f "o angle ( vision to ano!r4 ,9 all su* cases contr>y motion %\ld be avoid$ if ! object is ,rhy?m4 ,! sev]al movem5ts %\ld h a h>mony ( direction4 ,illu/ration3 ,fig4 #dh ,9 ?is case ! movem5t 9 ! gr\p is felt ?r\gh\t ! s]ies1 & ! =ce ( ! movem5t is enhanc$ by ! =ce ( a gradual increase ( attractions f "o visual angle to ano!r1 9 ! same direction1 to ! "r4 ,by rev]s+ ! direction ( increas+ attractions & s gett+ ! two movem5ts into contr>y motion1 ! feel+ ( rhy?m w\ld be mu* dim9i%$4 ,su* contr>y motions are unsatisfactory unless ,balance c be a*iev$4 ,9 t case all s5se ( movem5t & ( rhy?m disappe>s4 #dh4 ,eigh?4 ,un/able gr\ps may be repeat$1 n only 9 a gradation ( int]vals1 b 9 a gradation ( scale1 9 :i* case we h a comb9ation ( ?ree causes ( movem5t3 lack ( /ability 9 ! gr\p repeat$1 a gradual increase 9 ! numb] ( attractions 9 ! sequ5ce ( visual angles1 & a cr[d+ or conv]g5ce ( ! attractions4 ,rhy?ms ( ?is type w n be satisfactory unless ! ?ree movem5ts h ! same direction4 ,illu/ration3 ,fig4 #di ,"h we h ! repetition ( an un/able gr\p ( attractions 9 a progression ( scale & also ( int]vals4 ,! arrangem5t gives u ?ree elem5ts ( movem5t1 all 9 ! same direction4 #di4 ,two or ev5 m ( su* rhy?ms z ,i h describ$ may be comb9$ 9 "o comp\nd rhy?m1 9 :i* ! eye w foll[ two or m di/9ct movem5ts at ! same "t4 ,x is important 9 all comp\nd rhy?ms t !re %\ld be no opposition or conflict ( movem5ts1 unless ( c\rse ! object is to a*ieve a balance ( contr>y movem5ts4 ,correspond+ rhy?ms 9 contr>y motion balance "o ano!r4 ,if "o ( ! movem5ts is to ! "r1 ! o!r to ! left1 ! balance w be symmetrical4 ,,attitudes ,,relations ,,( ,,position ,,9 ,,diff]5t ,,attitudes #ej4 ,giv5 any relation ( positions 7directions1 di/ances1 int]vals71 x may be turn$ ^u a c5t] & s made to take an indef9ite numb] & v>iety ( attitudes4 ,x may be inv]t$ & ! inv]sion may be turn$ ^u a c5t]1 produc+ ano!r s]ies ( attitudes4 ,except 9 cases ( axial balance1 ! attitudes ( ! second s]ies w be diff]5t f ^? ( ! fir/4 ,illu/ration3 ,fig4 #ej ,9 ?is case ! relation ( positions be+ turn$ ^u a c5t] *anges its attitude1 :ile ! positions )9 ! gr\p rema9 relatively un*ang$4 ,!re is no *ange ( %ape4 ,illu/ration3 ,fig4 #ea ,9 ?is case ! same gr\p has be5 inv]t$1 & a second s]ies ( attitudes is %[n1 diff]+ f ! fir/ s]ies4 ,illu/ration3 ,fig4 #eb ,9 ?is case1 h[ev]1 :i* is a case ( axial balance1 ! inv]sion ( ! gr\p & ! turn+ ( ! inv]sion on a c5t] gives no additional attitudes4 #ea4 ,among all possible attitudes !re are f\r :i* are pr9cipal or fundam5tal1 :i* we may di/+ui% z foll[s3-- ,illu/ration3 ,fig4 #ec ,^! pr9cipal attitudes are3 ,fir/1 ,i1 ! orig9al attitude1 :atev] x is2 second1 ,,ii1 ! s+le inv]sion ( t attitude1 to ! "r on a v]tical axis2 ?ird1 ,,iii1 ! d\ble inv]sion ( ! orig9al attitude1 fir/ to ! "r !n d[n2 &1 f\r?1 ,,iv1 ! s+le inv]sion ( ! orig9al position1 d[n across ! horizontal axis4 ,,! ,,ord] ,,( ,,h>mony ,,9 ,,attitudes #eb4 ,! repetition ( any relation ( positions )\t *ange ( attitude gives u ,h>mony ( ,attitudes4 ,illu/ration3 ,fig4 #ed ,9 ?is case we h n only a ,h>mony 9 ! repetition ( a c]ta9 relation ( positions & ( int]vals1 b a ,h>mony ( ,attitudes4 ,we h1 9 ! relation ( positions repeat$1 a c]ta9 %ape4 ,9 ! repetition ( ! %ape we h ,%ape-,h>mony4 ,9 ! repetition ( ! %ape 9 a c]ta9 attitude we h a ,h>mony ( ,attitudes4 ,illu/ration3 ,fig4 #ee ,9 ?is case we h lo/ ! ,h>mony ( ,attitudes :i* we _h 9 ,fig4 #ed1 b n ! ,h>mony ( a c]ta9 %ape repeat$4 #ec4 ,! possibilities ( ,h>mony 9 ! repetition ( any relation ( positions 9 ! same attitude has be5 discuss$4 ,a ,h>mony ( ,attitudes w occur1 also1 9 ! repetition ( any relation ( attitudes4 ,illu/ration3 ,fig4 #ef ,"h we h ,h>mony 9 ! repetition ( a relation ( two attitudes ( a c]ta9 gr\p ( positions4 ,! comb9ation ( ! two attitudes gives u ano!r gr\p ( positions & ! ,h>mony lies 9 ! repetition ( ?is gr\p4 ,,! ,,ord] ,,( ,,balance ,,9 ,,attitudes #ed4 ,x is to be obs]v$ t s+le inv]sions 9 any direction1 = example ! relation ( attitudes ,i & ,,ii1 ,,ii & ,,iii1 ,,iii & ,,iv1 ,,iv & ,i1 9 ,fig4 #ec1 %[s an opposition & ,balance ( ,attitudes ^u ! axis ( inv]sion4 ,! relation ( positions ,i & ,,ii & ,,iii & ,,iv1 ! relation ( ! two gr\ps on ! left to ! two gr\ps on ! "r1 illu/rates ! idea ( ,symmetry ( ,attitudes1 ! axis ( balance be+ v]tical4 ,by ,symmetry ,i mean1 9 all cases1 "r & left balance on a v]tical axis4 ,all d\ble inv]sions1 ! relation ( positions ,i & ,,iii1 & ,,ii & ,,iv1 9 ,fig4 #ec1 are ,attitude-,balances1 n on axes1 b on c5t]s4 ,! balance ( ^! d\ble inv]sions is n symmetrical 9 ! s5se 9 :i* ,i use ! ^w symmetry1 nor is x axial4 ,x is c5tral4 ,,! ,,ord] ,,( ,,rhy?m ,,9 ,,attitudes #ee4 ,:5 movem5t is sugge/$ by any s]ies ( attitudes & ! movem5t is regulat$ by equal or regul>ly progressive int]vals1 we h a ,rhy?m ( ,attitudes4 ,illu/ration3 ,fig4 #eg ,9 ?is case ! *anges ( attitude sugge/ a fall+ movem5t to ! "r & d[n4 ,9 ! regul> progression ( ?is movem5t ?r\gh m>k$ int]vals we h ! effect ( ,rhy?m1 9 spite ( ! fact t ! relation ( positions repeat$ has axial balance4 ,! int]vals 9 ?is case correspond1 produc+ ,int]val-,h>mony4 ,! =ce ( ?is ,rhy?m might be increas$ if ! relation ( positions repeat$ sugge/$ a movem5t 9 ! same direction4 ,we %\ld h ,rhy?m1 ( c\rse1 9 ! repetition ( any su* un/able attitude-rhy?ms at equal or lawfully v>y+ int]vals4 ,,l9es ,,def9ition ,,( ,,l9es #ef4 ,tak+ any dot & draw+ x out 9 any direction1 or 9 a s]ies or sequ5ce ( directions1 x becomes a l9e4 ,! l9e may be drawn 9 any tone1 9 any value1 color or color-int5sity4 ,9 ord] t ! l9e may be se51 ! tone ( x mu/ diff] f ! gr\nd-tone ^u :i* x is drawn4 ,! l9e be+ di/9ctly visible1 ! "q ( tone ne$ n be rais$ at ?is po9t ( our discussion4 ,we w /udy ! l9e1 fir/1 z a l9e1 n z an effect ( light4 ,! l9e may be drawn long or %ort1 broad or n>r[4 ,z ! l9e increases 9 bread?1 h[ev]1 x becomes an area4 ,we w disreg>d = ! pres5t all consid]ation ( wid?-measures 9 ! l9e & conf9e our att5tion to ! possible *anges ( direction 9 x1 & to possible *anges 9 its l5g?4 ,we c draw ! l9e 9 "o direction f beg9n+ to end1 9 :i* case x w be /raight4 ,if1 9 draw+ ! l9e1 we *ange its direction1 we c d ?is abruptly1 9 :i* case ! l9e becomes angul>1 or we c d x gradually1 9 :i* case x becomes curv$4 ,l9es may be /raight1 angul>1 or curv$4 ,!y may h two ( ^! *>act]i/ics or all ?ree ( !m4 ,! %apes ( l9es are ( inf9ite v>iety4 ,,*anges ,,( ,,direction ,,9 ,,l9es ,angles #eg4 ,reg>d+ ! l9e :i* is drawn z a way or pa? ^u :i* we move & proce$1 we mu/ decide1 if we *ange our direction1 :e!r we w turn to ! "r or to ! left1 & :e!r we w turn abruptly or gradually4 ,if we *ange our direction abruptly we mu/ decide h[ mu* ( a *ange ( direction we w make4 ,is x to be a turn ( #cj or #fj or #ij or #ace8 ,h[ mu* ( a turn %all x be8 ,illu/ration3 ,fig4 #eh ,! above illu/rations are easy to und]/& & require no explanation4 ,an abrupt *ange ( #ahj means1 ( c\rse1 return+ ^u ! l9e j drawn4 ,curves #eh4 ,9 turn+1 n abruptly b gradually1 *ang+ ! direction at e po9t1 t is to say 9 mak+ a curve1 ! "q is1 h[ mu* ( a turn to make 9 a giv5 di/ance1 ?r\gh h[ _m degrees ( ! circle to turn 9 "o in* 7#a71 9 half an in* 771 9 two in*es 7#b74 ,9 e/imat+ ! relation ( arcs1 z di/ances1 to angles ( curvature1 ! angles ( ! arcs1 ! read] w f9d x conv5i5t to ref] to :at ,i may call an ,arc-,met]4 ,! pr9ciple ( ?is met] is %[n 9 ! foll[+ diagram3-- ,illu/ration3 ,fig4 #ei ,if we wi% to turn #cj 9 1 we take ! angle ( #cj & look )9 x = an arc ( 4 ,! arc ( ! "r l5g? & ! "r angle be+ f\nd1 x c be drawn free-h& or me*anically1 by trac+ or by ! divid]s4 ,us+ ?is met]1 we are able to draw any curve or comb9ation ( curves1 approximately2 & we are able to describe & def9e a l9e1 9 its curvatures1 s accurately t x c be produc$ accord+ to ! def9ition4 ,ow+1 h[ev]1 to ! difficulty ( measur+ ! l5g? ( circul> arcs accurately1 we may f9d x simpl] to def9e ! circul> arc by ! l5g? ( its radius & ! angle ?r\gh :i* ! radius passes :5 ! arc is drawn4 ,illu/ration3 ,fig4 #fj ,"h1 = example1 is a c]ta9 circul> arc4 ,x is p]haps be/ def9$ & describ$ z ! arc ( a half in* radius & an angle ( n9ety degrees1 or 9 writ+1 m briefly1 rad4 #ij4 ,reg>d+ e curv$ l9e ei!r z a circul> arc or made up ( a s]ies ( circul> arcs1 ! curve may be def9$ & describ$ by nam+ ! arc or arcs ( :i* x is compos$1 9 ! ord] 9 :i* !y are to be drawn1 & ! attitude ( ! curve may be det]m9$ by />t+ f a c]ta9 tang5t drawn 9 a c]ta9 direction4 ,! direction ( ! tang5t be+ giv51 ! fir/ arc takes ! direction ( ! tang5t1 turn+ to ! "r ( x or to ! left4 ,illu/ration3 ,fig4 #fa ,"h is a curve :i* is compos$ ( f\r circul> arcs to be drawn 9 ! foll[+ ord]3-- ,tang5t up-"r #de1 arc "r radius #a #fj1 arc left radius #ij1 arc "r radius #ahj4 ,two arcs w (t5 come toge!r at an angle4 ,! def9ition ( ! angle mu/ be giv5 9 t case4 ,x is1 ( c\rse1 ! angle made by tang5ts ( ! arcs4 ,describ+ ! fir/ arc & ! direction 7"r or left s _m degrees7 :i* ! tang5t ( ! second arc takes f ! tang5t ( ! fir/ arc2 !n describ+ ! second arc & /at+ :e!r x turns f its tang5t to ! "r or to ! left1 we %all be able to describe1 n only our curves1 b any angles :i* may occur 9 !m4 ,illu/ration3 ,fig4 #fb ,"h is a curve :i*1 s f> z ! arcs are conc]n$1 ( :i* x is compos$1 resembles ! curve ( ,fig4 #fa2 b 9 ?is case ! ?ird arc makes an angle ) ! second4 ,t angle has to be def9$4 ,draw+ ! tang5ts1 x appe>s to be a "r angle4 ,! def9ition ( ! l9e giv5 9 ,fig4 #fb w read z foll[s3-- ,tang5t d[n "r #de1 arc left radius #a #fj1 arc "r radius #ij1 tang5t left #ij1 arc left #ahj4 #ei4 ,9 ?is way1 reg>d+ all curves z circul> arcs or compos$ ( circul> arcs1 we %all be able to def9e any l9e we see1 or any l9e :i* we wi% to produce1 s f> z *anges ( direction are conc]n$4 ,= ! purposes ( ?is discussion1 ,i %all consid] all curves z compos$ ( circul> arcs4 ,!re are _m curves1 ( c\rse1 :i* are n circul> 9 "*1 nor compos$1 /rictly speak+1 ( circul> arcs4 ,! ,spirals are 9 no "p circul>4 ,elliptical curves are 9 no "p circul>4 ,all curves may1 nev]!less1 be approximately drawn z compositions ( circul> arcs4 ,! approximation to curves :i* are n circul> may be easily c>ri$ beyond any p[] ( discrim9ation :i* we h 9 ! s5se ( vision4 ,! me?od ( curve-def9ition1 :i* ,i h describ$1 ?\gh x may n be /rictly ma!matical1 w be f\nd satisfactory = all purposes ( ,pure ,design4 ,x is v important t we %\ld be able to analyze our l9es ^u a s+le g5]al pr9ciple2 to discov] :e!r !y are illu/rations ( ,ord]4 ,we mu/ "k :e!r any giv5 l9e1 be+ ord]ly1 is ord]ly 9 ! s5se ( ,h>mony1 ,balance1 or ,rhy?m4 ,x is equally important1 if we wi% to produce an ord]ly z di/+ui%$ f a disord]ly l9e1 t we %\ld h "s g5]al pr9ciple to foll[ 9 do+ x1 t we %\ld be able to produce =ms ( ,h>mony or ,balance or ,rhy?m 9 a l9e1 if we wi% to d s4 ,,diff]5ces ,,( ,,scale ,,9 ,,l9es #fj4 ,hav+ drawn a l9e ( a c]ta9 %ape1 ei!r /raight or angul> or curv$1 or p>tly angul>1 p>tly curv$1 we may *ange ! measure ( ! l9e1 9 its l5g?1 )\t *ang+ its %ape4 ,t is to say1 we may draw ! l9e long] or %ort]1 keep+ all *anges ( direction1 su* z !y are1 9 ! same positions1 relatively4 ,9 t way ! same %ape may be drawn l>g] or small]4 ,t is :at we mean :5 we speak ( a *ange ( scale or ( measure :i* is n a *ange ( %ape4 ,,diff]5ces ,,( ,,attraction ,,9 ,,l9es #fa4 ,a l9e attracts att5tion 9 ! measure ( ! tone-contra/ :i* x makes ) ! gr\nd-tone ^u :i* x is drawn4 ,x attracts att5tion1 also1 accord+ to its l5g?1 :i* is an ext5sion ( ! tone-contra/4 ,x attracts m att5tion ! long] x is1 provid$ x lies1 all ( x1 well )9 ! field ( vision4 ,x attracts att5tion also 9 ! measure ( its conc5tration4 ,illu/ration3 ,fig4 #fc ,l9e 8a0 w\ld attract less att5tion ?an x does if ! tone-contra/1 black on a gr\nd ( :ite pap]1 7 dim9i%$1 if ! l9e 7 gray1 n black4 ,9 l9e 8b0 !re is twice ! ext5sion ( tone-contra/ !re is 9 8a40 ,= t reason 8b0 is m attractive4 ,if1 h[ev]1 8a0 7 black & 8b0 7 gray1 8a0 might be m attractive ?an 8b10 because ( ! great] tone-contra/4 ,illu/ration3 ,fig4 #fd ,9 ?is illu/ration ! curv$ l9e is m attractive ?an ! /raight l9e because x is m conc5trat$1 !re=e m def9ite4 ,! ext5t ( tone-contra/ is ! same1 ! l9es be+ ( ! same l5g?4 ,illu/ration3 ,fig4 #fe ,9 ?is l9e !re is no d\bt z to ! great] attraction ( ! twi/$ end1 on acc\nt ( ! great] conc5tration x exhibits4 ,! ext5t ( tone-contra/ is ! same at bo? ends4 ,! =ce ( attraction 9 ! twi/$ end ( ! l9e w\ld be dim9i%$ if ! twi/$ end 7 made gray in/ead ( black4 ,! pull ( conc5tration at "o end might1 conceivably1 be p]fectly neutraliz$ by ! pull ( a great] tone-contra/ at ! o!r4 ,illu/ration3 ,fig4 #ff ,9 8b0 we h a great] ext5sion ( tone-contra/ 9 a giv5 space4 ,! space becomes m attractive 9 consequ5ce4 ,?is might n be ! case1 h[ev]1 if ! great] ext5sion ( tone-contra/ 9 "o case 7 neutraliz$ by an increase ( tone-contra/ 9 ! o!r4 ,,! ,,ord] ,,( ,,h>mony ,,9 ,,l9es #fb4 ,h>mony ( ,direction means no *ange ( direction4 ,illu/ration3 ,fig4 #fg ,9 ?is case we h a ,h>mony ( ,direction 9 ! l9e1 because x does n *ange its direction4 #fc4 ,h>mony ( ,angles4 ,we may h ,h>mony 9 ! repetition ( a c]ta9 relation ( directions1 z 9 an angle4 ,illu/ration3 ,fig4 #fh ,! angle up #de & d[n #de is "h repeat$ sev5 times4 ,illu/ration3 ,fig4 #fi ,9 ?is case we h a great _m angles 9 ! l9e1 b !y are all "r angles1 s we h a ,h>mony ( ,angles4 ,illu/ration3 ,fig4 #gj ,9 ?is case we h ,h>mony 9 ! repetition ( a c]ta9 relation ( angles1 t is to say1 9 ! repetition ( a c]ta9 =m ( angul>ity4 #fd4 ,equality ( l5g?s or measures betwe5 ! angles ( a l9e means a ,h>mony ( ,measures4 ,illu/ration3 ,fig4 #ga ,9 ?is case1 = example1 we h no ,h>mony ( ,angles1 b a ,h>mony ( ,measures 9 ! legs ( ! angles1 z !y are call$4 #fe4 ,we h a ,h>mony ( ,curvature 9 a l9e :5 x is compos$ :olly ( arcs ( ! same radius & ! same angle4 ,illu/ration3 ,fig4 #gb ,?is is a case ( ,h>mony ( ,curvature4 ,!re is no *ange ( direction "h1 9 ! sequ5ce ( correspond+ arcs4 ,illu/ration3 ,fig4 #gc ,"h1 aga91 we h a ,h>mony ( ,curvature4 ,9 ?is case1 h[ev]1 !re is a regul> alt]nation ( directions 9 ! sequ5ce ( correspond+ arcs4 ,9 ?is regul> alt]nation1 :i* is ! repetition ( a c]ta9 relation ( directions1 !re is a ,h>mony ( ,directions4 ,illu/ration3 ,fig4 #gd ,9 ?is case ! *anges ( direction are abrupt 7angul>7 z well z gradual4 ,!re is no regul> alt]nation1 b ! h>mony ( correspond+ arcs repeat$ w be appreciat$1 nev]!less4 #ff4 ,arcs produc$ by ! same radius are 9 h>mony to t ext5t1 hav+ ! radius 9 common4 ,illu/ration3 ,fig4 #ge ,?is is an example ( a h>mony ( arcs produc$ by radii ( ! same l5g?4 ,! arcs v>y 9 l5g?4 #fg4 ,arcs ( ! same angle-measure produc$ by diff]5t radii are 9 ,h>mony to ! ext5t t !y h an angle-measure 9 common4 ,illu/ration3 ,fig4 #gf ,?is is an example4 ,arcs hav+ ! same l5g? b v>y+ 9 bo? radius & angle may be felt to be 9 ,measure-,h>mony4 ,x is d\btful1 h[ev]1 :e!r l9es ( ! same l5g? b ( v diff]5t curvatures w be felt to correspond4 ,if ! correspond5ce ( l5g?s is n felt1 visually1 x has no int]e/ or value f ! po9t ( view ( ,pure ,design4 #fh4 ,any l9e may be cont9u$ 9 a repetition or repetitions ( its %ape1 :atev] ! %ape is1 produc+ :at ,i call a ,l9e> ,progression4 ,9 ! repetitions we h ,%ape-,h>mony4 ,illu/ration3 ,fig4 #gg ,?is is an example ( ,l9e> ,progression4 ,! "* ( ! progression is det]m9$ by ! %ape-motive :i* is repeat$ 9 x4 #fi4 ,! repetition ( a c]ta9 %ape-motive 9 a l9e is n1 necess>ily1 a repetition 9 ! same measure or scale4 ,a repetition ( ! same %ape 9 ! same measure means ,measure & ,%ape-,h>mony 9 ! progression4 ,a repetition ( ! same %ape 9 diff]5t measures means ,%ape-,h>mony )\t ,measure-,h>mony4 ,illu/ration3 ,fig4 #gh ,"h we h ! repetition ( a c]ta9 %ape 9 a l9e1 9 a progression ( measures4 ,t gives u ,%ape-,h>mony & a ,h>mony ( ,proportions1 )\t ,measure-,h>mony4 #gj4 ,9 ! repetition ( a c]ta9 %ape-motive 9 ! l9e1 ! l9e may *ange its direction abruptly or gradually1 cont9u\sly or alt]nately1 produc+ a ,l9e> ,progression ) *anges ( direction4 ,illu/ration3 ,fig4 #gi ,9 ,fig4 #gi !re is a c]ta9 *ange ( direction z we pass f "o repetition to ! next4 ,9 ! repetition ( ! same *ange ( direction1 ( ! same angle ( div]g5ce1 we h ,h>mony4 ,if ! angles ( div]g5ce v>i$ we %\ld h no su* ,h>mony1 ?\gh we might h ,h>mony 9 ! repetition ( a c]ta9 relation ( div]g5ces4 ,any repetition ( a c]ta9 *ange or *anges ( direction 9 a l9e> progression gives a ,h>mony ( ,directions 9 ! progression4 ,illu/ration3 ,fig4 #hj ,9 ?is case !re is a regul> alt]nation ( directions 9 ! repeats4 ,! repeats are drawn fir/ to ! "r1 !n up1 & ! relation ( ^! two directions is !n repeat$4 #ga4 ,by inv]t+ ! motive ( any progression1 9 s+le or 9 d\ble inv]sion1 & repeat+ ! motive toge!r ) its inv]sion1 we are able to v>y ! "* ( ! progression indef9itely4 ,illu/ration3 ,fig4 #ha ,9 ?is case we h a s+le inv]sion ( ! motive & a repetition ( ! motive ) its inv]sion4 ,comp>e ?is progression ) ! "o 9 ,fig4 #gg1 :]e ! same motive is repeat$ )\t inv]sion4 ,illu/ration3 ,fig4 #hb ,"h we h ! same motive ) a d\ble inv]sion1 ! motive ) its d\ble inv]sion be+ repeat$4 ,! inv]sion gives u ,%ape-,h>mony )\t ,h>mony ( ,attitudes4 ,we h ,h>mony1 h[ev]1 9 a repetition ( ! relation ( two attitudes4 ,^! d\ble inv]sions are m int]e/+ f ! po9t ( view ( ,balance ?an ( ,h>mony4 ,,! ,,ord] ,,( ,,balance ,,9 ,,l9es #gb4 ,we h ,balance 9 a l9e :5 "o half ( x is ! s+le or d\ble inv]sion ( ! o!r half2 t is1 :5 !re is an equal opposition & consequ5t equilibrium ( attractions 9 ! l9e4 ,:5 ! axis ( ! inv]sion is v]tical ! balance is symmetrical4 ,illu/ration3 ,fig4 #hc ,!re is ,balance 9 ?is l9e because half ( x is ! s+le inv]sion ( ! o!r half4 ,! balance is symmetrical because ! axis is v]tical4 ,! balance1 al?\gh symmetrical1 is n likely to be appreciat$1 h[ev]1 because ! eye is sure to move along a l9e ^u :i* !re is no bett] reason = n mov+ ?an is f\nd 9 slight t]m9al contra/s4 ,! eye is n held at ! c5t] :5 !re is no?+ to hold ! eye on ! c5t]4 ,m>k ! c5t] 9 any way & ! eye w g to x at once4 ,a m>k or acc5t may be put at ! c5t]1 or acc5ts1 correspond+ & equal1 may be put at equal di/ances f ! c5t] 9 opposite directions4 ,! eye w !n be held at ! c5t] by ! =ce ( equal & opposite attractions4 ,illu/ration3 ,fig4 #hd ,9 ?is case ! eye is held at ! balance-c5t] ( ! l9e by a *ange ( "* at t po9t4 ,illu/ration3 ,fig4 #he ,9 ?is case ! *anges ( "* are at equal di/ances1 9 opposite directions1 f ! c5t]4 ,! c5t] is m>k$ by a break4 ,! axis be+ v]tical1 ! balance is a symmetrical "o4 #gc4 ,! appreciation ( ,balance 9 a l9e dep5ds v mu* ^u ! attitude 9 :i* x is drawn4 ,illu/ration3 ,fig4 #hf ,9 ?is case ! balance 9 ! l9e itself is j z good z x is 9 ,fig4 #he2 b ! axis ( ! balance be+ diagonal1 ! balance is less di/9ctly felt4 ,! balance is unsatisfactory because ! attitude ( ! l9e is "o :i* sugge/s a fall+ d[n to ! left4 ,x is ! in/ability ( ! l9e :i* is felt1 m ?an ! balance 9 x4 ,illu/ration3 ,fig4 #hg ,9 ?is case ( d\ble inv]sion1 also1 we h balance4 ,! balance is m di/9ctly felt ?an x 0 9 ,fig4 #hf4 ,! attitude is "o ( /ability4 ,?is balance is nei!r axial nor symmetrical1 b c5tral4 #gd4 ,a l9e balances1 9 a s5se1 :5 its incl9ations are balanc$4 ,illu/ration3 ,fig4 #hh ,?is l9e may be said to be 9 balance1 z x has no incl9ations1 ei!r to ! "r or to ! left1 to sugge/ in/ability4 ,! v]ticals & ! horizontals1 be+ /able1 look aft] !mselves p]fectly well4 ,illu/ration3 ,fig4 #hi ,?is l9e has two unbalanc$ incl9ations to ! left4 ,x is1 !re=e1 less satisfactory ?an ! l9e 9 ,fig4 #hh1 f ! po9t ( view ( ,balance4 ,illu/ration3 ,fig4 #ij ,! two incl9ations 9 ?is l9e c\nt]act "o ano!r4 ,"o incl9ation t[>d ! left is balanc$ by a correspond+ incl9ation t[>d ! "r4 ,illu/ration3 ,fig4 #ia ,9 ?is case1 also1 !re is no incl9ation t[>d ! left :i* is n balanc$ by a correspond+ incl9ation t[>d ! "r4 ,illu/ration3 ,fig4 #ib ,9 ?is l9e1 :i* is compos$ :olly ( incl9ations to ! "r or left1 e incl9ation is balanc$1 & ! l9e is1 !re=e1 ord]ly 9 ! s5se ( ,balance2 m s1 c]ta9ly1 ?an x w\ld be if ! incl9ations 7 n c\nt]act$4 ,?is is ! problem ( balanc+ ! directions or incl9ations ( a l9e4 #ge4 ,a l9e hav+ no balance or symmetry 9 itself may become balanc$4 ,! l9e may be reg>d$ z if x 7 a s]ies ( dots close toge!r4 ,! l9e is !n a relation ( positions indicat$ by dots4 ,x is a composition ( attractions correspond+ & equal4 ,x is only necess>y1 !n1 to f9d :at ,i h call$ ! c5t] ( equilibrium1 ! balance-c5t] ( ! attractions1 & to indicate t c5t] by a symmetrical inclosure4 ,! l9e w !n become balanc$4 ,illu/ration3 ,fig4 #ic ,"h is a l9e4 ,to f9d ! c5t] ( its attractions x may be consid]$ z if x 7 a l9e ( dots1 l ?is3-- ,illu/ration3 ,fig4 #id ,! pr9ciple accord+ to :i* we f9d ! balance-c5t] is /at$ on page #bc4 ,! balance-c5t] be+ f\nd1 x mu/ be indicat$ unmi/akably4 ,?is may be done by means ( any symmetrical inclosure :i* w draw ! eye to ! c5t] & hold x !re4 ,illu/ration3 ,fig4 #ie ,9 ?is case ! balance-c5t] is indicat$ by a rectangul> inclosure4 ,?is rectangle is n1 h[ev]1 9 h>mony ( "* ) ! l9e inclos$ by x1 :i* is curv$4 ,illu/ration3 ,fig4 #if ,9 ?is case ! balance-c5t] is indicat$ by a circle1 :i*1 be+ a curve1 is 9 h>mony ( "* ) ! inclos$ l9e1 :i* is also a curve4 ,i %all call ?is ,occult ,balance to di/+ui% x f ! unmi/akable ,balance ( ,symmetry & o!r comp>atively obvi\s =ms ( ,balance1 includ+ ! balance ( d\ble inv]sions4 ,z ,i h said1 on page #bd1 ! symmetrical fram+ mu/ be suffici5tly attractive to hold ! eye /eadily at ! c5t]1 o!rwise x does n s]ve its purpose4 ,,! ,,ord] ,,( ,,rhy?m ,,9 ,,l9es #gf4 ,! eye1 n be+ held on a v]tical axis or on a balance-c5t]1 readily foll[s any sugge/ion ( movem5t4 ,illu/ration3 ,fig4 #ig ,9 ?is case !re is no intimation ( any v]tical axis or balance-c5t]4 ,! figure is consequ5tly un/able4 ,!re is a s5se ( movem5t to ! "r4 ,?is is due1 n only to ! incl9ations to ! "r1 b to ! conv]g5ces 9 t direction4 ,illu/ration3 ,fig4 #ih ,9 ?is case ! movem5t is unmi/akably to ! left4 ,9 su* cases we h movem5t1 b no ,rhy?m4 #gg4 ,rhy?m requires1 n only movem5t1 b ! ord] ( regul> & m>k$ int]vals4 ,illu/ration3 ,fig4 #ii ,9 ,fig4 #ii we h a l9e1 a l9e> progression1 :i* gives u ! feel+ ( movem5t1 unmi/akably4 ,! movem5t1 :i* 9 ! motive itself is n rhy?mical1 becomes rhy?mical 9 its repetition at regul>1 & 9 ?is case equal1 int]vals4 ,! int]vals are m>k$ by ! repetitions4 #gh4 ,x is a "q ( "s int]e/ to decide h[ _m repetitions are requir$ 9 a ,rhy?m4 ,9 answ] to ?is "q ,i %\ld say ?ree z a rule4 ,a s+le repetition %[s u only "o int]val1 & we d n "k :e!r ! succe$+ int]vals are to be equal or progressive1 ari?metically progressive or geometrically progressive4 ,! rhy?m is n def9$ until ?is "q is decid$1 z x w be by two m repetitions4 ,! measures ( ! rhy?m might take ! =m ( a repeat$ relation ( measures2 a repetition1 = example1 ( ! measures two1 sev51 f\r4 ,9 t case ! relation ( ! ?ree measures w\ld h to be repeat$ at lea/ ?ree times be=e ! "* ( ! rhy?m c\ld be appreciat$4 #gi4 ,any contr>iety ( movem5t 9 ! motive is ext5d$1 ( c\rse1 to its repetitions4 ,illu/ration3 ,fig4 #ajj ,9 ?is case1 = example1 !re are conv]g5ces &1 consequ5tly1 movem5ts bo? up & d[n4 ,?is contr>iety ( movem5ts is felt ?r\gh ! :ole s]ies ( repetitions4 ,o!r ?+s be+ equal1 ,i believe ! eye moves up m readily ?an d[n1 s t conv]g5ces d[nw>d h less effect ^u u ?an correspond+ conv]g5ces upw>d4 ,illu/ration3 ,fig4 #aja ,9 ?is case1 by omitt+ ! long v]tical l9e ,i h dim9i%$ ! am\nt ( conv]g5ce d[nw>d4 ,9 t way ,i h giv5 pr$om9ance to ! upw>d movem5t4 ,in/ead ( altoge!r omitt+ ! long v]tical l9e1 ,i might h *ang$ its tone f black to gray4 ,t w\ld cause an approximate in/ead ( complete disappe>ance4 ,x %\ld be rememb]$ t 9 all ^! cases ! habit ( read+ comes 9 to facilitate ! movem5ts to ! "r4 ,x is easi] = ! eye to move to ! "r ?an 9 any o!r direction1 o!r ?+s be+ equal4 ,! movem5t back to ! beg9n+ ( ano!r l9e1 :i* is ( c\rse inevitable 9 read+1 produces comp>atively little impression ^u u1 no m ?an ! turn+ ( ! page4 ,! habit ( read+ to ! "r happ5s to be our habit4 ,! habit is n univ]sal4 #hj4 ,read+ repetitions & alt]nations to ! "r1 always1 ,i1 = a long "t1 reg>d$ su* repetitions & alt]nations z rhy?mical1 until ,pr(essor ,m[ll rais$ ! "q :e!r x is necess>y to read all alt]nations to ! "r :5 !re is no?+ 9 ! alt]nations !mselves to sugge/ a movem5t 9 "o direction r ?an ano!r4 ,:y n read !m to ! left z well z to ! "r8 ,we at once decid$ t ! movem5t 9 a ,rhy?m mu/ be det]m9$ by ! "* ( ! ,rhy?m itself1 n by any habit ( read+1 or any o!r habit1 on our "p4 ,x 0 9 t way t we came to reg>d repetitions & alt]nations z illu/rations ( ,h>mony r ?an ( ,rhy?m4 ,rhy?m comes into ! ,h>mony ( a ,repeat$ ,relation :5 ! relation is "o :i* causes ! eye to move 9 "o direction r ?an ano!r1 & :5 ! movem5t is c>ri$ on f repetition to repetition1 f measure to measure4 #ha4 ,! repetition ( a motive at equal int]vals1 :5 !re is no movem5t 9 ! motive1 gives u no feel+ ( ,rhy?m4 ,illu/ration3 ,fig4 #ajb ,9 ?is case1 = example1 we h a repetition 9 ! l9e ( a c]ta9 symmetrical %ape4 ,z !re is no movem5t 9 ! %ape repeat$1 !re is no ,rhy?m 9 ! repetition4 ,!re is no?+ to draw ! eye 9 "o direction r ?an ano!r4 ,! attractions at "o end ( ! l9e correspond ) ! attractions at ! o!r4 #hb4 ,! feel+ ( ,rhy?m may be induc$ by a regul> dim9ution ( measure or scale 9 ! repetitions ( ! motive & 9 ! int]vals 9 :i* ! repetitions take place4 ,illu/ration3 ,fig4 #ajc ,9 ?is case ! %ape repeat$ is /ill symmetrical1 b x is repeat$ ) a gradual dim9ution ( scale & ( int]vals1 by :i* we are giv5 a feel+ ( rhy?mic movem5t4 ,! *ange ( scale & ( int]vals1 to induce a s5se ( rhy?mic motion1 mu/ be regul>4 ,to be regul> ! *ange mu/ be 9 ! t]ms ( "o or ! o!r ( ! regul> progressions2 ! ari?metical progression1 :i* proce$s by a c]ta9 addition1 or ! geometrical1 :i* proce$s by a c]ta9 multiplication4 ,! "q may arise 9 ?is case 7,fig4 #ajc7 :e!r ! movem5t ( ! ,rhy?m is to ! "r or to ! left4 ,i feel1 myself1 t ! movem5t is to ! "r4 ,9 dim9i%+ ! scale ( ! motive & ( ! int]vals we h1 h>dly at all1 dim9i%$ ! ext5t ( ! tone-contra/ 9 a giv5 angle ( vision4 ,see ,fig4 #da1 p4 #bg1 %[+ ! increase ( attractions f "o visual angle to ano!r4 ,at ! same "t we come at ! "r end ( ! progression to two or m repetitions 9 ! space ( "o4 ,we h1 !re=e1 e/abli%$ ! attraction ( a cr[d+ toge!r at ! "r end ( ! s]ies4 ,see ! passage 7p4 #dc7 on ! attractiv5ess ( a l9e4 ,! =ce ( ! cr[d+ toge!r ( attractions is1 ,i feel1 suffici5t to cause a movem5t to ! "r4 ,x mu/ be rememb]$1 h[ev]1 t ! great] facility ( read+ to ! "r is add$ "h to ! pull ( a great] cr[d+ toge!r ( attractions 9 ! same direction1 s ! movem5t ( ! ,rhy?m 9 t direction may n be v /rong aft] all4 ,if ! direction ( any ,rhy?m is d\btful1 ! ,rhy?m itself is d\btful4 #hc4 ,! feel+ ( ,rhy?m may be induc$1 z ,i h said1 by a gradual increase ( ! numb] ( attractions f measure to measure1 an increase ( ! ext5t ( tone-contra/4 ,illu/ration3 ,fig4 #ajd ,increas+ ! ext5t ( tone-contra/ & ! numb] ( attractions 9 ! measures ( ! ,rhy?m 9 ,fig4 #ajc1 we are able to =ce ! eye to foll[ ! s]ies 9 ! direction contr>y to ! habit ( read+1 t is to say f "r to left4 ,a decrease 9 ! =ces ( attraction 9 connection ) a decrease ( scale is famili> to u all 9 ! ph5om5a ( p]spective4 ,! gradual disappe>ance ( objects 9 a]ial p]spective does away ) ! attraction ( a great] cr[d+ toge!r ( objects 9 ! di/ance4 ,illu/ration3 ,fig4 #aje ,9 ?is case ! dim9ution ( scale has be5 giv5 up & !re is no long] any cr[d+ toge!r4 ,!re is no *ance ( ?is rhy?m be+ read f left to "r except by an ef=t ( ! w4 ,! increase ( attractions t[>d ! left is mu* m ?an suffici5t to c\nt]act ! habit ( read+4 #hd4 ,! =ce ( a gradual com+ toge!r ( attractions1 induc+ movem5t 9 ! direction ( su* com+ toge!r1 is noticeable 9 spiral %apes4 ,illu/ration3 ,fig4 #ajf ,9 ?is case we h a s]ies ( /raight l9es ) a con/ant & equal *ange ( direction to ! "r1 comb9$ ) a regul> dim9ution ( measures 9 ! l5g? ( ! l9es1 ?is 9 ! t]ms ( an ari?metical progression4 ,! movem5t is 9 ! direction ( conc5tration & x is di/9ctly m>k$ 9 its measures4 ,! movem5t is !re=e rhy?mical4 ,illu/ration3 ,fig4 #ajg ,9 ?is case we h a s]ies ( /raight l9es ) a con/ant *ange ( direction to ! "r2 b 9 ?is case ! *anges ( measure 9 ! l9es are 9 ! t]ms ( a geometrical progression4 ,! direction is ! same1 ! pull ( conc5tration p]haps /rong]4 ,illu/ration3 ,fig4 #ajh ,9 ?is ,rhy?m !re is an ari?metical gradation ( measures 9 ! *anges ( direction1 bo? 9 ! l5g? ( ! legs & 9 ! measure ( ! angles4 ,! pull ( conc5tration is1 9 ?is case1 v mu* increas$4 ,x is evid5t t ! legs may v>y ari?metically & ! angles geometrically2 or ! angles ari?metically & ! legs geometrically4 #he4 ,if1 9 ! place ( ! /raight l9es1 :i* =m ! legs1 9 any ( ! examples giv51 are sub/itut$ l9es :i* 9 !mselves induce movem5t1 ! feel+ ( ,rhy?m may be /ill fur!r increas$1 provid$ ! directions ( movem5t are consi/5t4 ,illu/ration3 ,fig4 #aji ,9 ?is case ! movem5t is 9 ! direction ( increas+ conc5tration & 9 ! direction ( ! conv]g5ces4 ,if ! movem5t ( ! conv]g5ces be contr>y to ! movem5t ( conc5tration1 !re w be 9 ! figure a contr>y motion :i* may dim9i% or ev5 entirely prev5t ! feel+ ( ,rhy?m4 ,if ! movem5t 9 "o direction or ! o!r pr$om9ates1 we may /ill get ! feel+ ( ,rhy?m1 9 spite ( ! drawback ( ! o!r & contr>y movem5t4 ,illu/ration3 ,fig4 #aaj ,9 ?is case ! l9e> conv]g5ces sub/itut$ = ! /raight l9es are contr>y to ! direction ( increas+ conc5tration4 ,! movem5t is d\btful4 #hf4 ,correspond+ rhy?ms1 set 9 contr>y motion1 give u ! feel+ ( ,balance r ?an ( ,rhy?m4 ,! balance 9 su* cases is a balance ( movem5ts4 ,illu/ration3 ,fig4 #aaa ,?is is an example ( correspond+ & oppos$ rhy?ms produc+ ! feel+1 n ( ,rhy?m1 b ( ,balance4 ,,attitudes ,,l9es ,,9 ,,diff]5t ,,attitudes #hg4 ,any l9e or l9e> progression may be turn$ ^u a c5t]1 & s made to take an indef9ite numb] & v>iety ( attitudes4 ,x may be inv]t$ ^u an axis1 & ! inv]sion may be turn$ ^u a c5t] produc+ ano!r s]ies ( attitudes :i*1 except 9 ! case ( axial symmetry 9 ! l9e1 w be diff]5t f ^? ( ! fir/ s]ies4 ,illu/ration3 ,fig4 #aab ,9 ?is case ! l9e *anges its attitude4 ,illu/ration3 ,fig4 #aac ,9 ?is case ,i h inv]t$ ! l9e1 & turn+ ! inv]sion ^u a c5t] ,i get a diff]5t set ( attitudes4 ,illu/ration3 ,fig4 #aad ,9 ?is case1 :i* is a case ( axial symmetry 9 ! l9e1 ! inv]sion gives u no additional attitudes4 ,,! ,,ord] ,,( ,,h>mony ,,9 ,,! ,,attitudes ,,( ,,l9es #hh4 ,:5 any l9e or l9e> progression is repeat$1 )\t *ange ( attitude1 we h a ,h>mony ( ,attitudes4 ,illu/ration3 ,fig4 #aae ,?is is an illu/ration ( ,h>mony ( ,attitudes4 ,x is also an illu/ration ( ,int]val-,h>mony4 #hi4 ,we h a ,h>mony ( ,attitudes1 also1 9 ! repetition ( any relation ( two or m attitudes1 ! relation ( attitudes be+ repeat$ )\t *ange ( attitude4 ,illu/ration3 ,fig4 #aaf ,we h "h a ,h>mony ( ,attitudes due to ! repetition ( a c]ta9 relation ( attitudes1 )\t *ange ( attitude4 ,,! ,,ord] ,,( ,,balance ,,9 ,,! ,,attitudes ,,( ,,l9es #ij4 ,:5 a l9e or l9e> progression is inv]t$ ^u any axis or c5t]1 & we see ! orig9al l9e & its inv]sion side by side1 we h a ,balance ( ,attitudes4 ,illu/ration3 ,fig4 #aag ,! relation ( attitudes ,i1 ,,ii1 ( ,,iii1 ,,iv1 & ( ,i1 ,,ii1 ,,iii1 ,,iv1 is t ( ,symmetrical ,balance on a v]tical axis4 ,! relation ( attitudes ,i1 ,,iv1 & ( ,,ii1 ,,iii1 is a relation ( ,balance b n ( ,symmetrical ,balance4 ,?is is true1 also1 ( ! relation ( ,i1 ,,iii & ( ,,ii1 ,,iv4 ,d\ble inv]sions are nev] symmetrical1 b !y are illu/rations ( ,balance4 ,! balance ( d\ble inv]sions is c5tral1 n axial4 ,^! /atem5ts are all repetitions ( /atem5ts previ\sly made ab\t positions4 ,,! ,,ord] ,,( ,,rhy?m ,,9 ,,! ,,attitudes ,,( ,,l9es #ia4 ,x (t5 happ5s t a l9e repeat$ 9 diff]5t attitudes gives u ! s5se ( movem5t4 ,x does ?is :5 ! gr\p+ ( ! repetitions sugge/s in/ability4 ,! movem5t is rhy?mical :5 x exhibits a regul>ity ( *anges 9 ! attitudes & 9 ! int]vals ( ! *anges4 ,illu/ration3 ,fig4 #aah ,9 ?is case we h a movem5t to ! "r1 b no ,rhy?m1 ! int]vals be+ irregul>4 ,illu/ration3 ,fig4 #aai ,9 ?is case ! *anges ( attitude & ! int]vals ( ! *anges be+ regul>1 ! movem5t becomes rhy?mical4 ,! direction ( ! rhy?m is cle>ly d[n-to-!-"r4 #ib4 ,9 ! repetition ( any l9e we h a ,h>mony1 due to ! repetition4 ,if ! l9e is repeat$ 9 ! same attitude1 we h a ,h>mony ( ,attitudes4 ,if x is repeat$ 9 ! same int]vals1 we h a ,h>mony ( ,int]vals4 ,we h ,h>mony1 also1 9 ! repetition ( any relation ( attitudes or ( int]vals4 ,we h n yet consid]$ ! arrangem5t or composition ( two or m l9es ( diff]5t measures & ( diff]5t %apes4 ,,! ,,composition ,,( ,,l9es #ic4 ,by ! ,composition ( ,l9es ,i mean putt+ two or m l9es toge!r1 9 juxtaposition1 9 contact or int]lac+4 ,our object 9 ! composition ( l9es1 s f> z ,pure ,design is conc]n$1 is to a*ieve ,ord]1 if possible ,beauty1 9 ! sev]al modes ( ,h>mony1 ,balance1 & ,rhy?m4 ,,h>mony ,,9 ,,! ,,composition ,,( ,,l9es #id4 ,we h ,h>mony 9 l9e-compositions :5 ! l9es :i* are put toge!r correspond 9 all respects or 9 "s respects1 :5 !y correspond 9 attitudes1 & :5 !re is a correspond5ce ( di/ances or int]vals4 ,illu/ration3 ,fig4 #abj ,9 ?is case ! l9es ( ! composition correspond 9 tone1 measure1 & %ape1 b n 9 attitude2 & !re is no correspond5ce 9 di/ances or int]vals4 ,illu/ration3 ,fig4 #aba ,9 ?is case ! attitudes correspond1 z !y did n 9 ,fig4 #abj4 ,!re is /ill no correspond5ce ( int]vals4 ,illu/ration3 ,fig4 #abb ,"h we h ! correspond5ce ( int]vals :i* we did n h ei!r 9 ,fig4 #abj or 9 ,fig4 #aba4 ,!re is n only a ,h>mony ( ,attitudes & ( ,int]vals1 9 ?is case1 b ! ,h>mony ( a repetition 9 "o direction1 ,direction-,h>mony4 ,9 all ^! cases we h ! repetition ( a c]ta9 angle1 a "r angle1 & ( a c]ta9 measure-relation betwe5 ! legs ( ! angle1 giv+ ,measure & ,%ape-,h>mony4 #ie4 ,! repetition 9 any composition ( a c]ta9 relation ( measures1 or ( a c]ta9 proportion ( measures1 gives ,measure-,h>mony to ! composition4 ,! repetition ( ! relation "o to ?ree 9 ! legs ( ! angle1 9 ! illu/rations j giv51 gives to ! compositions ! ,h>mony ( a ,recurr+ ,ratio4 ,by a proportion ,i mean an equality betwe5 ratios1 :5 !y are num]ically diff]5t4 ,! relation ( "o to ?ree is a ratio4 ,! relation ( "o to ?ree & ?ree to n9e is a proportion4 ,we may h 9 any composition ! ,h>mony ( a ,repeat$ ,ratio1 z 9 ,figs4 #abj1 #aba1 #abb1 or we may h a ,h>mony ( ,proportions1 z 9 ! composition :i* foll[s4 ,illu/ration3 ,fig4 #abc #if4 ,to be 9 ,h>mony l9es are n necess>ily simil> 9 all respects4 ,z ,i h j %[n1 l9es may be 9 ,%ape-,h>mony1 )\t be+ 9 any ,measure-,h>mony4 ,l9es are approximately 9 h>mony :5 !y correspond 9 c]ta9 p>ticul>s1 ?\gh !y diff] 9 o!rs4 ,! m po9ts ( resemblance betwe5 !m1 ! great] ! h>mony4 ,:5 !y correspond 9 all respects we h1 ( c\rse1 a p]fect h>mony4 ,illu/ration3 ,fig4 #abd ,?is is a case ( ,%ape-,h>mony )\t ,measure-,h>mony & )\t ,h>mony ( ,attitudes4 ,illu/ration3 ,fig4 #abe ,9 ?is case we h a ,h>mony ( ,%apes & ( ,attitudes1 )\t ,measure-,h>mony or ,h>mony ( ,int]vals4 ,?is is a good illu/ration ( a ,h>mony ( ,proportions4 ,/raight l9es are 9 ,h>mony ( ,/raightness because !y are all /raight1 h[ev] mu* !y diff] 9 tone or measure4 ,!y are 9 ,h>mony ( ,measure :5 !y h ! same measure ( l5g?4 ,! measures ( wid?1 also1 may agree or disagree4 ,9 e agreem5t we h ,h>mony4 ,angul> l9es are 9 ,h>mony :5 !y h "o or m angles 9 common4 ,! recurr5ce ( a c]ta9 angle 9 diff]5t p>ts ( a composition br+s ,h>mony into ! composition4 ,design]s are v apt to use diff]5t angles :5 !re is no good reason = do+ s1 :5 ! repetition ( "o w\ld be m ord]ly4 ,illu/ration3 ,fig4 #abf ,! f\r l9es 9 ?is composition h "r angles 9 common4 ,to t ext5t ! l9es are 9 ,h>mony4 ,!re is also a ,h>mony 9 ! correspond5ce ( tones & ( wid?-measures 9 ! l9es4 ,consid]able ,h>mony ( ,attitudes occurs 9 ! =m ( p>allelisms4 ,illu/ration3 ,fig4 #abg ,^! two l9es h simply "o angle 9 common1 a "r angle1 & ! angle has ! same attitude 9 bo? cases4 ,!y diff] 9 o!r respects4 ,illu/ration3 ,fig4 #abh ,9 ^! ?ree l9es ! only elem5t mak+ = ,h>mony1 except ! same tone & ! same wid?1 is f\nd 9 ! pres5ce 9 ea* l9e ( a c]ta9 small arc ( a circle4 ,/raightness occurs 9 two ( ! l9es b n 9 ! ?ird4 ,!re is a ,h>mony1 !re=e1 betwe5 two ( ! l9es f :i* ! ?ird is exclud$4 ,!re is1 also1 a ,h>mony ( ,attitude 9 ^! two l9es1 9 c]ta9 p>allelisms4 ,,balance ,,9 ,,! ,,composition ,,( ,,l9es #ig4 ,l9es balance :5 9 opposite attitudes4 ,we get ,balance 9 all inv]sions1 :e!r s+le or d\ble4 ,illu/ration3 ,fig4 #abi ,"h simil> l9es are drawn 9 opposite attitudes & we get ,measure & ,%ape-,balance4 ,9 ! above case ! axis ( balance is v]tical4 ,! balance is1 !re=e1 symmetrical4 ,symmetrical ,balance is obta9$ by ! s+le inv]sion ( any l9e or l9es on a v]tical axis4 ,d\ble inv]sion gives a ,balance ( ,measures & ,%apes on a c5t]4 ,we h no ,symmetry 9 d\ble inv]sions4 ,all ?is has be5 expla9$4 ,illu/ration3 ,fig4 #acj ,we h ,measure & ,%ape-,balance on a c5t] 9 ?is case4 ,x is a case ( d\ble inv]sion4 ,x is int]e/+ to turn ^! d\ble inv]sions on _! c5t]s1 & to obs]ve ! v diff]5t effects !y produce 9 diff]5t attitudes4 #ih4 ,%apes 9 ord] to balance satisfactorily mu/ be drawn 9 ! same measure1 z 9 ,fig4 #aca :i* foll[s4 ,illu/ration3 ,fig4 #aca ,illu/ration3 ,fig4 #acb ,"h1 9 ,fig4 #acb1 we h ,%ape-,h>mony )\t ,measure-,h>mony4 ,x might be argu$ t we h 9 ?is case an illu/ration ( ,%ape-,balance )\t ,measure-,balance4 ,!oretically t is s1 b ,%ape-,balance )\t ,measure-,balance is nev] satisfactory4 ,if we want ! l9es 9 ,fig4 #acb to balance we mu/ f9d ! balance-c5t] betwe5 !m1 & !n indicate t c5t] by a symmetrical inclosure4 ,we %all !n h a ,measure-,balance 7occult7 )\t ,%ape-,balance4 #ii4 ,:5 measures correspond b %apes diff] ! balance-c5t] may be sugge/$ by a symmetrical inclosure or fram+4 ,:5 t is done ! measures become balanc$4 ,illu/ration3 ,fig4 #acc ,"h we h ,measure-,h>mony & a ,measure-,balance )\t ,%ape-,h>mony or ,%ape-,balance4 ,! two l9es h diff]5t %apes b ! same measures1 l5g?s & wid?s correspond+4 ,! balance-c5t] is f\nd = ea* l9e4 ,see pp4 #ed1 #ee4 ,betwe5 ! two c5t]s is f\nd ! c5t]1 ^u :i* ! two l9es w balance4 ,?is c5t] is !n sugge/$ by a symmetrical inclosure4 ,! balanc+ measures 9 su* cases may1 ( c\rse1 be turn$ ^u _! c5t]s1 & ! axis connect+ _! c5t]s may be turn$ 9 any direction or attitude1 ) no loss ( equilibrium1 s f> z ! measures are conc]n$4 ,illu/ration3 ,fig4 #acd ,! ,balance ( ,measures "h is j z good z x is 9 ,fig4 #acc4 ,! attitudes are *ang$ b n ! relation ( ! ?ree balance-c5t]s4 ,! *ange ( %ape 9 ! inclosure makes no diff]5ce4 #ajj4 ,measure-,balance )\t ,%ape-,h>mony or ,%ape-,balance is satisfactory only :5 ! balance-c5t] is unmi/akably indicat$ or sugge/$1 z 9 ! examples :i* ,i h giv54 #aja4 ,!re is ano!r =m ( ,balance :i* is to be inf]r$ f :at ,i h said1 on page #ah1 ( ! ,balance ( ,directions1 b x ne$s to be p>ticul>ly consid]$ & m fully illu/rat$4 ,i mean a ,balance 9 :i* directions or incl9ations to ! "r are c\nt]act$ by correspond+ or equival5t directions or incl9ations to ! left4 ,! idea 9 its simple/ & mo/ obvi\s =m is illu/rat$ 9 ,fig4 #bb1 on page #ah4 ,9 t case ! l9es ( incl9ation correspond4 ,!y d n necess>ily correspond except 9 ! ext5t ( contra/1 :i* may be di/ribut$ 9 v>i\s ways4 ,illu/ration3 ,fig4 #ace ,! balance ( incl9ations 9 ?is case is j z good z ! balance 9 ,fig4 #bb4 ,!re is no symmetry z 9 ,fig4 #bb4 ,?ree l9es balance aga9/ "o4 ,! ?ree l9es1 h[ev]1 %[ ! same ext5t ( contra/ z ! "o4 ,s f> z ! incl9ations are conc]n$ !y w balance 9 any arrangem5t :i* lies well )9 ! field ( vision4 ,! eye mu/ be able to appreciate ! fact t a disposition to fall to ! "r is c\nt]act$ by a correspond+ or equival5t disposition to fall to ! left4 ,illu/ration3 ,fig4 #acf ,?is arrangem5t ( ! incl9+ l9es is j z good z ! arrangem5t 9 ,fig4 #ace4 ,! incl9ations may be di/ribut$ 9 any way1 provid$ !y c\nt]act "o ano!r prop]ly4 ,illu/ration3 ,fig4 #acg ,9 ?is case ,i h aga9 *ang$ ! composition1 & hav+ sugge/$ ! balance-c5t] ( ! l9es1 z attractions1 by a symmetrical inclosure1 ,i h add$ ,measure-,balance 7occult7 to ,incl9ation-,balance4 ,! ,ord] 9 ,fig4 #acg is great] ?an ! ,ord] 9 ,figs4 #ace & #acf4 ,9 ,fig4 #acg two =ms ( ,balance are illu/rat$1 9 ! o!r cases only "o4 ,! value ( any composition lies 9 ! numb] ( ord]ly connections :i* x %[s4 ,illu/ration3 ,fig4 #ach ,9 ?is case ,i h tak5 a long angul> l9e & add$ a suffici5t numb] & ext5t ( opposite incl9ations to make a balance ( incl9ations4 ,! horizontal "p ( ! long l9e is /able1 s x ne$s no c\nt]action1 b ! o!r p>ts incl9e 9 v>i\s degrees1 to ! left or to ! "r4 ,ea* incl9+ "p requires1 !re=e1 ei!r a correspond+ l9e 9 a balanc+ direction1 or two or m l9es ( equival5t ext5sion 9 t direction4 ,9 "o case ,i h set ?ree l9es to balance "o1 b !y equal ! "o 9 l5g?1 t is to say1 9 ! ext5t ( contra/4 ,we h 9 ,fig4 #ach an illu/ration ( occult ,measure-,balance & ! ,balance ( ,incl9ations4 ,i h illu/rat$ ! idea ( ,incl9ation-,balance by v simple examples4 ,i h n consid]$ ! incl9ations ( curves1 nor h ,i gone1 at all1 into ! m difficult problem ( balanc+ av]ages ( incl9ation1 :5 ! av]age ( two or m diff]5t incl9ations ( diff]5t ext5ts ( contra/ has to be c\nt]act$4 ,9 ,tone-,relations ! incl9ations are ( tone-contra/s1 & a %ort incl9ation ) a /rong contra/ may balance a long incl9ation ) a slight "o1 or sev]al incl9ations ( slight contra/s may s]ve to balance "o ( a /rong contra/4 ,! =ce ( any incl9+ l9e may be increas$ by increas+ ! tone-contra/ ) ! gr\nd-tone4 ,9 tone-relations ! problem becomes complicat$ & difficult4 ,! :ole subject ( ,incl9ation-,balance is "o ( great int]e/ & wor?y ( a sep>ate treatise4 ,,rhy?m ,,9 ,,! ,,composition ,,( ,,l9es #ajb4 ,we w fir/ consid] ! ,measure-,rhy?ms :i* result f a gradual increase ( scale1 an increase 9 ! ext5t ( ! contra/s4 ,! int]vals mu/1 9 su* ,rhy?ms1 be regul> & m>k$4 ,!y may be equal2 !y may alt]nate1 or !y may be regul>ly progressive4 ,illu/ration3 ,fig4 #aci ,9 ?is case ,i feel t ! direction ( ! ,rhy?m is up-to-!-"r ow+ to ! gradual increase ( l5g? & consequ5tly ( ! ext5t ( contra/ 9 ! l9es1 9 t direction4 ,illu/ration3 ,fig4 #adj ,9 ?is case ,i h1 by means ( regul>ly dim9i%+ int]vals1 add$ ! =ce ( a cr[d+ toge!r ( contra/+ edges to ! =ce ( a gradual ext5sion ( !m4 ,! movem5t is /ill m /rongly up-to-!-"r4 ,illu/ration3 ,fig4 #ada ,9 ?is case a great] ext5sion ( contra/s pulls "o way & a great] cr[d+ ( contra/s ! o!r4 ,i ?9k t cr[d+ has ! be/ ( x4 ,! movem5t1 ?\gh mu* ret>d$1 is1 ,i feel1 d[n-to-!-left r ?an up-to-!-"r1 9 spite ( ! fact t ! great] facility ( read+ to ! "r is add$ to ! =ce ( ext5d$ contra/s4 #ajc4 ,sub/itut+ un/able = /able attitudes 9 ! examples j giv51 we are able to add ! movem5t sugge/$ by in/ability ( attitude to ! movem5t caus$ by a gradual ext5sion ( contra/s4 ,illu/ration3 ,fig4 #adb ,! movem5t up-to-!-"r 9 ,fig4 #aci is "h connect$ ) an incl9ation ( all ! l9es d[n-to-!-"r4 ,illu/ration3 ,fig4 #adc ,"h ! fall+ ( ! l9es d[n-to-!-left c\nt]acts ! movem5t 9 ! opposite direction :i* is caus$ by ! ext5sion ( contra/+ edges 9 t direction4 ,a cr[d+ toge!r ( ! l9es1 due to ! dim9ution ( int]vals t[>d ! left1 adds =ce to ! movem5t 9 t direction4 ,illu/ration3 ,fig4 #add ,9 ?is case a movem5t up is caus$ by conv]g5ces1 a movem5t d[n by cr[d+4 ,! conv]g5ces are all up1 ! cr[d+ d[n4 ,i ?9k t ! conv]g5ces h x4 ,i ?9k ! movem5t is1 on ! :ole1 up4 ,! int]vals ( ! cr[d+ d[n dim9i% ari?metically4 ,illu/ration3 ,fig4 #ade ,! conv]g5ces & ! cr[d+ ( attractions are1 "h1 bo? up-to-!-"r4 ,! ,rhy?m is mu* /rong] ?an x 0 9 ,fig4 #add4 ,! int]vals are ^? ( an ari?metical progression4 ,illu/ration3 ,fig4 #adf ,! movem5t "h is up-to-!-"r1 because ( conv]g5ces 9 t direction & an ext5sion ( contra/s 9 t direction4 ,illu/ration3 ,fig4 #adg ,9 ?is case ! two movem5ts "p company4 ,"o leads ! eye up-to-!-left1 ! o!r leads x up-to-!-"r4 ,! movem5t z a :ole is approximately up4 ,z ! direction ( ! int]vals is horizontal1 n v]tical1 ?is is a case ( movem5t )\t ,rhy?m4 ,! movem5t w become rhy?mic only 9 a v]tical repetition4 ,t is to say1 ! direction or directions ( ! movem5t 9 any ,rhy?m & ! direction or directions ( its repetitions mu/ co9cide4 ,9 ,fig4 #aci1 ! movem5t is up-to-!-"r1 & ! int]vals may be tak5 9 ! same direction1 b 9 ,fig4 #adg ! movem5t is up4 ,! int]vals _c be tak5 9 t direction4 ,x is1 !re=e1 impossible to get any feel+ ( ,rhy?m f ! composition4 ,we %all get ! feel+ ( ,rhy?m only :5 we repeat ! movem5t 9 ! direction ( ! movem5t1 :i* is up4 ,illu/ration3 ,fig4 #adh ,"h we h a v]tical repetition ( ! composition giv5 9 ,fig4 #adg4 ,! result is an upw>d movem5t 9 regul> & m>k$ int]vals1 answ]+ to our und]/&+ ( ,rhy?m4 ,illu/ration3 ,fig4 #adi ,9 ?is case we h a curv$ movem5t4 ,! l9es be+ spac$ at regul> int]vals1 ! movem5t is 9 regul> & m>k$ measures4 ,its direction is due to an increase 9 ! numb] ( attractions1 to cr[d+1 & to conv]g5ces4 ,! movem5t is1 accord+ly1 rhy?mical4 ,illu/ration3 ,fig4 #aej ,! movem5t ( ,fig4 #adi is "h p>tly de/roy$ by an inv]sion & opposition ( attitudes & directions4 ,! movem5t is1 on ! :ole1 up1 b x c h>dly be describ$ z rhy?mical1 because x has no repetition upw>ds1 z x has 9 ! next illu/ration1 ,fig4 #aea4 ,be=e proce$+1 h[ev]1 to ! consid]ation ( ,fig4 #aea1 ,i want to call ! att5tion ( ! read] to ! fact t we h 9 ,fig4 #aej a type ( ,balance to :i* ,i h n p>ticul>ly ref]r$4 ,x is a case ( unsymmetrical balance on a v]tical axis4 ,! balanc+ %apes & movem5ts correspond4 ,!y incl9e 9 opposite directions4 ,!y div]ge equally f ! v]tical axis4 ,! incl9ations balance4 ,at ! same "t ! composition does n answ] to our und]/&+ ( ,symmetry4 ,x is n a case ( "r & left balance on ! v]tical axis4 ,! %apes & movem5ts are n "r & left & opposite4 ,"o ( ! %apes is set high] ?an ! o!r4 ,! balance is on ! v]tical4 ,x is obvi\s1 b x is n symmetrical4 ,x is a =m ( ,balance :i* has _m & v int]e/+ possibilities4 ,illu/ration3 ,fig4 #aea ,! repetition1 9 ?is case1 ( some:at contr>y movem5ts1 a repetition at equal int]vals on a v]tical axis1 gives u m ,balance ?an ,rhy?m4 ,we feel1 h[ev]1 a g5]al upw>d movem5t ?r\gh ! repetitions &1 z ?is movem5t is regul>1 x mu/ be describ$ z rhy?mical4 ,! feel+ ( upw>d movem5t 9 ,fig4 #aea is1 no d\bt1 p>tly due to ! sugge/ion ( upw>d gr[? 9 c]ta9 =ms ( vegetation4 ,! sugge/ion is inevitable4 ,s f> z ! movem5t is caus$ by ?is association ( ideas x is a matt]1 n ( s5sation1 b ( p]ception4 ,! consid]ation ( su* associations ( ideas does n belong1 prop]ly1 to ,pure ,design1 :]e we are deal+ ) s5se-impressions1 exclusively4 #ajd4 ,rhy?m is n inconsi/5t ) ,balance4 ,x is only necess>y to get movem5ts :i* h ! same or ne>ly ! same direction & :i* are rhy?mical 9 "* to balance on ! same axis & we h a reconciliation ( ! two pr9ciples4 ,illu/ration3 ,fig4 #aeb ,"h we h a ,rhy?m1 ( some:at contr>y movem5ts1 ) ,balance1--,balance on a diagonal axis4 ,! ,balance is n satisfactory4 ,! ,balance ( ,incl9ations is felt m ?an ! ,balance ( ,%apes4 ,illu/ration3 ,fig4 #aec ,9 ?is case we h ! comb9ation ( a ,rhy?m ( some:at contr>y1 b on ! :ole upw>d1 movem5ts ) ,symmetry4 ,if ! div]g+ movem5ts ( ,fig4 #aec %\ld be made /ill m div]g+1 s t !y become approximately contr>y & opposite1 ! feel+ ( a g5]al upw>d movem5t w disappe>4 ,! ?ree movem5ts to ! "r w balance ! ?ree movem5ts to ! left1 & we %all h an illu/ration ( ,symmetrical ,balance1 ) no ,rhy?m 9 ! composition z a :ole4 ,x is d\btful :e!r ! balance ( contr>y & opposite movem5ts is satisfactory4 ,our eyes are drawn 9 opposite directions1 away f ! axis ( balance1 in/ead ( be+ drawn t[>d x4 ,our appreciation ( ! balance mu/1 !re=e1 be dim9i%$4 ,contr>y & opposite movem5ts neutralize "o ano!r1 s we h nei!r re/ nor movem5t 9 ! balance ( contr>y motions4 ,by br++ ! div]g5ces ( movem5t toge!r1 gradually1 we %all be able to increase1 consid]ably1 ! upw>d movem5t %[n 9 ,fig4 #aec4 ,at ! same "t1 ! sugge/ion ( an upw>d gr[? ( vegetation becomes /rong]4 ,! increase ( movem5t w be p>tly expla9$ by ?is association ( ideas4 ,illu/ration3 ,fig4 #aed ,"h all ! movem5ts are pull$ toge!r into "o direction4 ,! ,rhy?m is easi] & m rapid4 ,! ,balance is j z good4 ,! movem5t 9 ?is case is no d\bt facilitat$ by ! sugge/ion ( upw>d gr[?4 ,x is impossible to e/imate ! =ce :i* is add$ by su* sugge/ions & associations4 ,illu/ration3 ,fig4 #aee ,"h ! movem5ts come toge!r 9 ano!r way4 ,! numb] & v>iety ( ^! illu/rations might1 ( c\rse1 be indef9itely increas$4 ,^? :i* ,i h giv5 w1 ,i ?9k1 s]ve to def9e ! pr9cipal modes ( l9e-composition1 :5 ! l9es are su* z we *oose to draw4 ,,! ,,composition ,,( ,,v>i\s ,,l9es #aje4 ,9 mo/ ( ! examples ,i h giv5 ,i h us$ repetitions ( ! same l9e or simil> l9es4 ,:5 ! l9es :i* are put toge!r are n 9 h>mony1 :5 !y are drawn1 z !y may be1 )\t any reg>d to ! exig5cies ( ord]ly composition1 ! problem becomes "o ( do+ ! be/ we c ) our t]ms4 ,we try = ! greate/ possible numb] ( ord]ly connections1 connections mak+ = ,h>mony1 ,balance1 & ,rhy?m4 ,we arrange ! l9es1 s f> z possible1 9 ! same directions1 giv+ !m simil> attitudes1 gett+1 9 details1 z mu* ,h>mony ( ,direction & ( ,attitudes z possible1 & e/abli%+ z mu* ,h>mony ( ,int]vals z possible betwe5 ! l9es4 ,by spac+ & plac+ we try to get diff]5ces ( "* z f> z possible into regul> alt]nations or gradations 9 :i* !re w be a sugge/ion ei!r ( ,h>mony or ( ,rhy?m4 ,a sugge/ion ( ,symmetry is sometimes possible4 ,occult ,balance is possible 9 all cases1 z x dep5ds1 n ^u ! t]ms balanc$1 b ^u ! indication ( a c5t] ( attractions by a symmetrical fram+ ( !m4 ,let u take sev5 l9es1 ) a v>iety ( %ape-"*1 ) z little ,%ape-,h>mony z possible1 & let u try to put ^! l9es toge!r 9 an ord]ly way4 ,illu/ration3 ,fig4 #aef ,) ^! l9es1 :i* %[ little or no h>mony ( "*1 :i* agree only 9 tone & 9 wid?-measure1 l9es :i* w\ld n be select$ c]ta9ly z suitable mat]ial = ord]ly compositions1 ,i w make ?ree compositions1 gett+ z mu* ,ord] into ea* "o z ,i c1 j to illu/rate :at ,i mean4 ,i %all n be able to a*ieve a great deal ( ,ord]1 b en\gh1 probably1 to satisfy ! read] t ! ef=t has be5 wor? :ile4 ,illu/ration3 ,fig4 #aeg ,9 ?is case ,i h a*iev$ ! sugge/ion ( a ,symmetrical ,balance on a v]tical axis ) "s ,h>mony ( ,directions & ( ,attitudes & "s ,int]val-,h>mony4 ,illu/ration3 ,fig4 #aeh ,9 ?is case1 also1 ,i h a*iev$ a sugge/ion ( ,ord]1 if n ,ord] itself4 ,consid] ! comp>ative disord] 9 ,fig4 #aef1 :]e no arrangem5t has be5 attempt$4 ,illu/ration3 ,fig4 #aei ,"h is ano!r arrangem5t ( ! same t]ms4 ,=tunately1 9 all ( ^! cases1 ! l9es agree 9 tone & 9 wid?-measure4 ,t means consid]able ord] to beg9 )4 ,?is problem ( tak+ any t]ms & mak+ ! be/ possible arrangem5t ( !m is a mo/ int]e/+ problem1 & ! ability to solve x has a practical value4 ,we h ! problem to solve 9 e-day life2 :5 we h to arrange1 z well z we c1 9 ! be/ possible ord]1 all ! useful & indisp5sable articles we h 9 our h\ses4 ,to a*ieve a consi/5cy & unity ( effect ) a great numb] & v>iety ( objects is nev] easy4 ,x is (t5 v difficult4 ,x is p>ticul>ly difficult :5 we h no two objects alike1 no correspond5ce1 no lik5ess1 to make ,h>mony4 ,) ! possibility ( repetitions & inv]sions ! problem becomes comp>atively easy4 ,) repetitions & inv]sions we h ! possibility1 n only ( ,h>mony1 b ( ,balance & ,rhy?m4 ,) inv]sions we h ! possibility1 n only ( ,balance1 b ( ,symmetrical ,balance1 & :5 we h t we are n at all likely to ?9k :e!r ! t]ms ( :i* ! symmetry is compos$ are 9 h>mony or n4 ,we feel ! ,ord] ( ,symmetry & we are satisfi$4 ,illu/ration3 ,fig4 #afj ,9 ?is design ,i repeat an inv]sion ( ! arrangem5t 9 ,fig4 #aeh4 ,! result is a symmetry1 & no "o is likely to ask :e!r ! elem5ts ( :i* x is compos$ are h>moni\s or n4 ,by inv]sions1 s+le & d\ble1 x is possible to a*ieve ! ,ord] ( ,balance1 9 all cases4 ,illu/ration3 ,fig4 #afa ,= ?is design ,i h made ano!r arrangem5t ( my sev5 l9es4 ,! arrangem5t sugge/s movem5t4 ,9 repeat+ ! arrangem5t at regul> & equal int]vals1 )\t *ange ( attitude1 ,i produce ! effect ( ,rhy?m4 ,)\t resort+ to inv]sion1 x is difficult to make ev5 an approximation to ,symmetry ) su* t]ms 7see ,fig4 #aeg71 b !re is little or no difficulty 9 mak+ a consi/5t or fairly consi/5t movem5t out ( !m1 :i*1 be+ repeat$ at regul> int]vals1 )\t *ange ( attitude1 or ) a gradual *ange ( attitude1 w produce ! effect ( ,rhy?m4 ,up to ?is po9t ,i h spok5 ( ! composition ( l9es 9 juxtaposition1 t is to say1 ! l9es h be5 plac$ ne> toge!r s z to be se5 toge!r4 ,i h n spok5 ( ! possibilities ( ,contact & ,int]lac+4 ,! l9es 9 any composition may t\* "o ano!r or cross "o ano!r4 ,! result w be a composition ( connect$ l9es4 ,9 c]ta9 cases ! l9es w become ! outl9es ( areas4 ,i w def] ! illu/ration ( contacts & int]lac+s until ,i come to consid] ! composition ( outl9es4 ,,outl9es ,,def9ition ,,( ,,outl9es #ajf4 ,outl9es are l9es :i*1 return+ to !mselves1 make inclosures & describe areas ( diff]5t measures & %apes4 ,:at has be5 said ( l9es may be said1 also1 ( outl9es4 ,x w be wor? :ile1 h[ev]1 to give a sep>ate consid]ation to outl9es1 z a p>ticul>ly int]e/+ & important class ( l9es4 ,z 9 ! case ( dots & l9es1 ,i %all disreg>d ! fact t ! outl9es may be drawn 9 diff]5t tones1 mak+ diff]5t contra/s ( value1 color1 or color-int5sity ) ! gr\nd-tone ^u :i* !y are drawn4 ,i %all1 also1 disreg>d possible diff]5ces ( wid? 9 ! l9es :i* make ! outl9es4 ,i %all conf9e my att5tion1 "h1 to ! measures & %apes ( ! outl9es & to ! possibilities ( ,h>mony1 ,balance1 & ,rhy?m 9 ^? measures & %apes4 ,,h>mony1 ,,balance1 ,,& ,,rhy?m ,,9 ,,outl9es #ajg4 ,:at is ,h>mony or ,balance or ,rhy?m 9 a l9e is ,h>mony1 ,balance1 or ,rhy?m 9 an ,outl9e4 ,illu/ration3 ,fig4 #afb ,9 ?is outl9e we h ,measure-,h>mony 9 ! angles1 ,measure-,h>mony ( l5g?s 9 ! legs ( ! angles1 ,measure & ,%ape-,balance on a c5t] & ,symmetry on ! v]tical axis4 ,! same /atem5t w be true ( all polygons :i* are bo? equiangul> & equilat]al1 :5 !y are balanc$ on a v]tical axis4 ,illu/ration3 ,fig4 #afc ,9 ?is case we h ,measure-,h>mony ( angles b no ,measure-,h>mony ( l5g?s 9 ! legs ( ! angles4 ,we h lo/ ,measure & ,%ape-,balance on a c5t] :i* we _h 9 ! previ\s example4 ,illu/ration3 ,fig4 #afd ,9 ?is case ! angles are n all 9 a ,h>mony ( ,measure2 b we h ,measure-,h>mony ( l5g?s 9 ! legs ( ! angles1 & we h ,measure & ,%ape-,balance on a c5t]4 ,!re is a c]ta9 ,h>mony 9 ! repetition ( a relation ( two angles4 ,illu/ration3 ,fig4 #afe ,9 ?is case we h ,measure-,h>mony 9 ! angles1 :i* are equal1 & a ,h>mony due to ! repetition ( a c]ta9 measure-relation 9 ! legs ( ! angles4 ,z 9 ,fig4 #afb1 we h "h a ,measure & ,%ape-,balance on a c5t] & ,symmetry on ! v]tical axis4 ,?is polygon is n equilat]al1 b its sides are symmetrically dispos$4 ,_m int]e/+ & beautiful figures may be drawn 9 ^! t]ms4 ,illu/ration3 ,fig4 #aff ,we h 9 ! circle ! mo/ h>moni\s ( all outl9es4 ,! ,h>mony ( ! circle is due to ! fact t all sections ( x h ! same radius & equal sections ( x h1 also1 ! same angle-measure4 ,! circle is1 ( c\rse1 a p]fect illu/ration ( ,measure & ,%ape-,balance on a c5t]4 ,! balance is also symmetrical4 ,we h a ,h>mony ( ,directions 9 ! repetition ( ! same *ange ( direction at e po9t ( ! outl9e1 & we h a ,h>mony ( ,di/ances 9 ! fact t all po9ts ( ! outl9e are equally di/ant f ! balance-c5t]1 :i* is unmi/akably felt4 ,illu/ration3 ,fig4 #afg ,! ,ellipse is ano!r example ( ,measure & ,%ape-,balance on a c5t]4 ,9 ?is attitude x is also an illu/ration ( ,symmetry4 ,illu/ration3 ,fig4 #afh ,9 ?is case we /ill h balance b no symmetry4 ,! attitude sugge/s movem5t4 ,we _c help feel+ t ! figure is fall+ d[n to ! left4 ,a repetition at equal int]vals w\ld give u ,rhy?m4 ,illu/ration3 ,fig4 #afi ,9 ?is case we h an outl9e produc$ by ! s+le inv]sion ( a l9e 9 :i* !re is ! repetition ( a c]ta9 motive 9 a gradation ( measures4 ,t gives ,%ape-,h>mony )\t ,measure-,h>mony4 ,?is is a case ( ,symmetrical ,balance4 ,x is also a case ( rhy?mic movem5t upw>d4 ,! movem5t is ma9ly due to conv]g5ces4 ,illu/ration3 ,fig4 #agj ,9 ?is case1 also1 ! %apes repeat$ on ! "r side & on ! left side ( ! outl9e %[ movem5ts :i* become 9 repetitions almo/ rhy?mical4 ,! movem5t is up 9 spite ( ! fact t ea* "p ( ! movem5t is1 9 its end+1 d[n4 ,we h 9 ^! examples symmetrical balance on a v]tical axis comb9$ ) rhy?m on ! same axis4 ,x may be desirable to f9d ! balance-c5t] ( an outl9e :5 only ! axis is indicat$ by ! "* ( ! outl9e4 ,! pr9ciple :i* we foll[ is ! "o already describ$4 ,9 ,fig4 #afi we h a symmetrical balance on a v]tical axis1 b !re is no?+ to indicate ! balance-c5t]4 ,x lies on ! axis some:]e1 b !re is no?+ to %[ u :]e x is4 ,reg>d+ ! outl9e z a l9e ( attractions1 ! eye is presumably held at _! balance-c5t]1 :]ev] x is4 ,exactly :]e x is is a matt] ( visual feel+4 ,! balance-c5t] be+ asc]ta9$1 x may be indicat$ by a symmetrical outl9e or inclosure1 ! c5t] ( :i* _c be d\btful4 ,illu/ration3 ,fig4 #aga ,! balance-c5t]1 z det]m9$ by visual feel+1 is "h cle>ly indicat$4 ,9 ?is case besides ! balance on a c5t] we h also ! ,symmetry :i* we _h 9 ,fig4 #afi4 ,illu/ration3 ,fig4 #agb ,! s5se ( ,balance is1 9 ?is case1 mu* dim9i%$ by ! *ange ( attitude 9 ! balanc$ outl9e4 ,we h our balance ^u a c5t]1 all ! same2 b ! balance on ! v]tical axis be+ lo/1 we h no long] any ,symmetry4 ,x w be obs]v$ t balance on a c5t] is n inconsi/5t ) movem5t4 ,if ?is figure 7 repeat$ at equal int]vals )\t *ange ( attitude1 or ) a gradual *ange1 we %\ld h ! ,rhy?m ( a repeat$ movem5t4 ,9 "s outl9es only c]ta9 p>ts ( ! outl9es are ord]ly1 :ile o!r p>ts are disord]ly4 ,illu/ration3 ,fig4 #agc ,9 ! above outl9e we h two sections correspond+ 9 measure & %ape-"* & 9 attitude4 ,we h1 !re=e1 c]ta9 elem5ts ( ! outl9e 9 h>mony4 ,we feel movem5t b n rhy?m 9 ! relation ( ! two curves4 ,!re is no balance ( any k9d4 ,we ought to be able to recognize elem5ts ( ord] z !y occur 9 any outl9e1 ev5 :5 ! outl9e1 z a :ole1 is disord]ly4 ,illu/ration3 ,fig4 #agd ,9 ord] to balance ! some:at irregul> outl9e giv5 9 ,fig4 #agc1 we foll[ ! proc$ure already describ$4 ,! effect1 h[ev]1 is unsatisfactory4 ,! composition lacks /ability4 ,illu/ration3 ,fig4 #age ,! attitude ( ! figure is "h made to con=m1 z f> z possible1 to ! %ape & attitude ( ! symmetrical fram+3 ?is = ! sake ( ,%ape & ,attitude-,h>mony4 ,! *ange ( attitude gives great] /ability4 ,,int]ior ,,dim5sions ,,( ,,an ,,outl9e #ajh4 ,a di/9ction mu/ be drawn betwe5 ! measures ( ! outl9e1 z an outl9e1 & ! measures ( ! space or area ly+ )9 ! outl9e3 :at may be call$ ! int]ior dim5sions ( ! outl9e4 ,illu/ration3 ,fig4 #agf ,9 ?is case we mu/ di/+ui% betwe5 ! measures ( ! outl9e & ! dim5sions ( ! space inclos$ )9 x4 ,:5 we consid] ! measures--n ( ! outl9e1 b ( ! space or area inside ( ! outl9e--we may look 9 ^! measures1 also1 = ,h>mony1 = ,balance1 or = ,rhy?m1 & = comb9ations ( ^! pr9ciples4 ,,h>mony ,,9 ,,! ,,int]ior ,,dim5sions ,,( ,,an ,,outl9e #aji4 ,we h ,h>mony 9 ! int]ior dim5sions ( an outl9e :5 ! dim5sions correspond or :5 a c]ta9 relation ( dim5sions is repeat$4 ,illu/ration3 ,fig4 #agg ,9 ?is case we h an outl9e :i* %[s a ,h>mony 9 ! correspond5ce ( two dim5sions4 ,illu/ration3 ,fig4 #agh ,9 ?is case we h ,h>mony 9 ! correspond5ce ( all v]tical dim5sions1 ,h>mony 9 ! correspond5ce ( all horizontal dim5sions1 b no relation ( ,h>mony betwe5 ! two4 ,x might be argu$1 f ! fact t ! int]val 9 "o direction is twice t 9 ! o!r1 t ! dim5sions h some?+ 9 common1 namely1 a common divisor4 ,x is v d\btful1 h[ev]1 :e!r ?is fact is appreciable 9 ! s5se ( vision4 ,! recurr5ce ( any relation ( two dim5sions w\ld1 no d\bt1 be appreciat$4 ,we %\ld h1 9 t case1 ,%ape-,h>mony4 ,illu/ration3 ,fig4 #agi ,9 ?is circle we h a ,measure-,h>mony ( diamet]s4 ,illu/ration3 ,fig4 #ahj ,9 ?is case we h a ,h>mony due to ! repetition ( a c]ta9 ratio ( v]tical int]vals3 #a3c1 #a3c1 #a3c4 #aaj4 ,any gradual dim9ution ( ! int]val betwe5 opposite sides 9 an outl9e gives u a conv]g5ce 9 :i* ! eye moves m or less rapidly t[>d an actual or possible contact4 ,! m rapid ! conv]g5ce ! m rapid ! movem5t4 ,illu/ration3 ,fig4 #aha ,9 ?is case we h n only symmetrical balance on a v]tical axis b movem5t1 9 ! upw>d & rapid conv]g5ce ( ! sides ,,ba & ,,ca t[>d ! angle ,a4 ,! "q may be rais$ :e!r ! movem5t1 9 ?is case1 is up f ! side ,,bc to ! angle ,a or d[n f ! angle ,a t[>d ! side ,,bc4 ,i ?9k t ! read] w agree t ! movem5t is f ! side ,,bc into ! angle ,a4 ,9 ?is direction ! eye is m def9itely guid$4 ,! opposite movem5t f ,a t[>d ,,bc is a movem5t 9 div]g+ directions :i* ! eye _c foll[ to any di/ance4 ,z ! di/ance f ,,bc t[>d ,a decreases1 ! conv]g5ce ( ! sides ,,ba & ,,ca is m & m helpful to ! eye & produces ! feel+ ( movem5t4 ,! eye f9ds itself 9 a small] & small] space1 ) a m & m def9ite impulse t[>d ,a4 ,x is a "q :e!r ! movem5t f ,,bc t[>d ,a is rhy?mical or n4 ,! movem5t is n connect$ ) any m>k$ regul>ity ( measures4 ,i am incl9$ to ?9k1 h[ev]1 t ! gradual & ev5 *ange ( measures produces ! feel+ ( equal *anges 9 equal measures4 ,if s1 ! movem5t is rhy?mical4 ,:5 ! movem5t ( ! eye 9 any conv]g5ce is a movem5t 9 regul> & m>k$ measures1 z 9 ! example :i* foll[s1 ! movem5t is rhy?mical1 )\t d\bt4 ,illu/ration3 ,fig4 #ahb ,! upw>d movem5t 9 ?is outl9e1 be+ regulat$ by measures :i* are m>k$ & equal1 ! movem5t is c]ta9ly rhy?mical1 accord+ to our und]/&+ & def9ition ( ,rhy?m4 ,comp>+ ,fig4 #aha ) ,fig4 #ahb1 ! "q arises1 :e!r ! movem5t 9 ,fig4 #ahb is felt to be any m rhy?mical ?an ! movem5t 9 ,fig4 #aha4 ,! measures ( ! movem5t 9 ,fig #aha are n m>k$1 b ,i _c p]suade myself t !y are n felt 9 ! ev5ness ( ! gradation4 ,! movem5t 9 ,fig4 #aha is easi] ?an x is 9 ,fig4 #ahb1 :5 ! m>k+ ( ! measures int]f]es ) ! movem5t4 ,illu/ration3 ,fig4 #ahc ,9 ?is case we h ano!r illu/ration l ,fig4 #ahb1 only ! measures ( ! rhy?m are diff]5tly m>k$4 ,! =ce ( ! conv]g5ce is greate/ 9 ,fig4 #aha4 ,x is some:at dim9i%$ by ! measure-m>ks 9 ,fig4 #ahb4 ,x is /ill fur!r dim9i%$1 9 ,fig4 #ahc1 by ! angles t break ! measures4 ,illu/ration3 ,fig4 #ahd ,9 ?is case ! movem5t is m rapid aga91 ! measures be+ measures ( an ari?metical progression4 ,!re is a cr[d+ toge!r ( attractions 9 ! direction ( ! conv]g5ce1 & ! movem5t is easi] ?an x is 9 ,fig4 #ahc1 9 spite ( ! fact t ! l9es ( conv]g5ce are m brok5 9 ,fig4 #ahd4 ,!re is an ari?metical dim9ution ( horizontal z well z ( v]tical l9es 9 ,fig4 #ahd4 ,illu/ration3 ,fig4 #ahe ,9 ?is case ! measures ( ! rhy?m are 9 ! t]ms ( a geometrical progression4 ,! cr[d+ toge!r ( attractions is /ill m rapid 9 ?is case & ! di/ance to be trav]s$ by ! eye is %ort]4 ,! conv]g5ce1 h[ev]1 is less compell+1 ! l9es ( ! conv]g5ce be+ s mu* brok5--unnecess>ily4 ,! movem5t w be v mu* ret>d$1 if n prev5t$1 by hav+ ! movem5t ( ! cr[d+ & ! movem5t ( ! conv]g5ce oppos$4 ,illu/ration3 ,fig4 #ahf ,!re is no d\bt t 9 ?is example1 :i* is to be comp>$ ) t ( ,fig4 #ahd1 ! upw>d movem5t is almo/ prev5t$4 ,!re are "h two oppos$ movem5ts3 t ( ! conv]g5ce upw>d & t ( a cr[d+ toge!r ( attractions d[nw>d4 ,! conv]g5ce is /rong]1 ,i ?9k1 ?\gh x mu/ be rememb]$ t x is probably easi] = ! eye to move up ?an d[n1 o!r ?+s be+ equal4 #aaa4 ,! movem5ts 9 all ( ^! cases may be enhanc$ by sub/itut+ = ! /raight l9es %apes :i* are 9 !mselves %apes ( movem5t4 ,illu/ration3 ,fig4 #ahg ,"h1 = example1 ! movem5t ( ,fig4 #ahd is facilitat$ & increas$ by a *ange ( %ape 9 ! l9es1 l9es ) movem5t be+ sub/itut$ = l9es :i* h no movem5t1 beyond ! movem5t ( ! conv]g5ce4 ,illu/ration3 ,fig4 #ahh ,9 ,fig4 #ahh all ! %apes h a d[nw>d movem5t :i* contradicts ! upw>d movem5t ( conv]g5ce4 ,! movem5t d[n almo/ prev5ts ! movem5t up4 #aab4 ,! movem5t ( any conv]g5ce may be /raight1 angul>1 or curv$4 ,illu/ration3 ,fig4 #ahi ,9 ?is case ! movem5t ( ! conv]g5ce is angul>4 ,x %\ld be obs]v$ t ! movem5t is di/ribut$ 9 ! measures ( an ari?metical progression1 s t we h1 n only movem5t1 b rhy?m4 ,illu/ration3 ,fig4 #aij ,9 ?is case ! movem5t ( conv]g5ce is 9 a curve4 ,! /ages ( ! movem5t1 n be+ m>k$1 ! movem5t is n rhy?mical1 unless we feel t equal *anges are tak+ place 9 equal measures4 ,i am incl9$ to ?9k t we d feel t4 ,! "q1 h[ev]1 is "o :i* ,i w\ld r ask ?an answ]1 def9itely4 ,illu/ration3 ,fig4 #aia ,9 ?is case ! movem5t is1 unque/ionably1 rhy?mical1 because ! measures are cle>ly m>k$4 ,! measures are 9 an ari?metical progression4 ,!y dim9i% gradually 9 ! direction ( ! conv]g5ce1 caus+ a gradual cr[d+ toge!r ( attractions 9 t direction4 ,sub/itut+1 9 ! measures1 %apes :i* h movem5t1 ! movem5t ( ! rhy?m may be consid]ably increas$1 z is %[n 9 ! example :i* foll[s4 ,illu/ration3 ,fig4 #aib ,?is is a case 9 :i* ! movem5t is1 no d\bt1 facilitat$ by an association ( ideas1 ! sugge/ion ( a gr[?4 #aac4 ,! m obvi\s ! sugge/ion ( gr[?1 ! m inevitable is ! movem5t 9 ! direction ( x1 :atev] t direction is4 ,x mu/ be und]/ood1 h[ev]1 t ! movem5t 9 su* cases is due to an association ( ideas1 n to ! pull ( attractions 9 ! s5se ( vision4 ,! pull ( an association ( ideas may or may n be 9 ! direction ( ! pull ( attractions4 ,illu/ration3 ,fig4 #aic ,9 ,fig4 #aic we h an illu/ration ( a rhy?mic movem5t upw>d4 ,! upw>d movem5t is due q z mu* to an association ( ideas1 ! ?\ght ( a gr[? ( vegetation1 z x is to m]e visual attractions4 ,x happ5s t ! figure is also an illu/ration ( ,symmetrical ,balance4 ,z we h ,h>mony 9 ! simil>ity ( ! opposite sides1 ! figure is an illu/ration ( comb9$ ,h>mony1 ,balance1 & ,rhy?m4 ,!re is ano!r po9t :i* is illu/rat$ 9 ,fig4 #aic4 ,x is ?is3 t we may h rhy?mic movem5t 9 an outl9e1 or1 inde$1 9 any composition ( l9es1 :i* %[s a gradual & regul> *ange f "o %ape to ano!r2 :i* %[s a gradual & regul> evolution or developm5t ( %ape-"*2 provid$ ! *anges are di/ribut$ 9 regul> & m>k$ measures & ! direction ( ! *anges1 ! evolution1 ! developm5t1 is unmi/akable2 z x is 9 ,fig4 #aic4 ,! *anges ( %ape 9 ! above outl9e are *anges :i* are gradual & regul> & sugge/ an upw>d movem5t unmi/akably4 ,! movem5t1 h[ev]1 involves a comp>ison ( %ape ) %ape1 s x is z mu* a matt] ( p]ception z ( s5sation4 ,evolutions & developm5ts 9 ,space1 9 ! field ( vision1 are z int]e/+ z evolutions & developm5ts 9 ! duration ( ,"t4 ,:5 ! *anges 9 su* movem5ts are regul>1 :5 !y take place 9 regul> & m>k$ measures1 :5 we mu/ take !m 9 a c]ta9 ord]1 ! movem5ts are rhy?mical1 :e!r 9 ,"t or 9 ,space4 ,,! ,,attitudes ,,( ,,outl9es #aad4 ,any outl9e1 no matt] :at dim5sions or %ape x has1 may be turn$ ^u a c5t] & 9 t way made to take a great numb] & v>iety ( attitudes4 ,n only may x be turn$ ^u a c5t] b inv]t$ ^u an axis4 ,be+ inv]t$1 ! inv]sion may be turn$ ^u a c5t] & made to take ano!r s]ies ( attitudes1 & ?is second s]ies ( attitudes w be diff]5t f ! fir/ s]ies1 except 9 cases ( axial symmetry 9 ! outl9e or area4 ,x mu/ be cle>ly und]/ood t a *ange ( attitude 9 any outl9e or area is n a *ange ( %ape4 #aae4 ,:at has be5 said ( ,h>mony1 ,balance1 & ,rhy?m 9 ! attitudes ( a l9e applies equally well to outl9es & to ! spaces def9$ by !m4 ,,! ,,arrangem5t ,,& ,,composition ,,( ,,outl9es #aaf4 ,by ! composition ( outl9es ,i mean putt+ two or m outl9es 9 juxtaposition1 9 contact or int]lac+4 ,9 all cases ( int]lac+1 ( c\rse1 ! %ape-"* ( ! int]lac+ outl9es is lo/4 ,! outl9es become ! outl9es ( o!r areas & ( a l>g] numb] ( !m4 ,our object 9 putt+ outl9es toge!r is1 9 ,pure ,design1 to illu/rate ! ord]s ( ,h>mony1 ,balance1 & ,rhy?m1 to a*ieve ,ord]1 z mu* z we c1 if possible ,beauty4 ,i w n[ give a s]ies ( examples ) a brief analysis or explanation ( ea* "o4 ,illu/ration3 ,fig4 #aid ,9 ?is case we h ,%ape-,h>mony 9 ! outl9es & also a ,h>mony ( ,attitudes4 ,illu/ration3 ,fig4 #aie ,"h we h ano!r illu/ration ( ! ,h>mony ( ,%apes & ( ,attitudes1 ) a ,h>mony ( ,int]vals1 :i* we did n h 9 ,fig4 #aid4 ,illu/ration3 ,fig4 #aif ,9 ?is case we h a ,h>mony ( ,attitudes & ( ,int]vals 7! ,h>mony ( a repeat$ ,relation ( ,int]vals7 9 :at may be call$ an ,all-,ov] ,repetition4 ,illu/ration3 ,fig4 #aig ,9 ?is case we h a ,h>mony ( ,attitudes 9 ! repetition ( a relation ( two opposite attitudes2 ?is ) ,%ape-,h>mony & ,int]val-,h>mony4 ,illu/ration3 ,fig4 #aih ,9 ?is case we h a ,symmetry ( ,attitudes1 ) ,%ape-,h>mony & ,int]val-,h>mony4 ,turn+ ! composition (f ! v]tical axis we %\ld h ,balance b no ,symmetry4 ,! balance-c5t] w be felt 9 all possible attitudes ( ?is composition4 ,illu/ration3 ,fig4 #aii ,9 ?is case ,i h repeat$ a c]ta9 outl9e1 :i* gives me ! ,h>mony ( a repetition1--?is 9 connection ) a progression 9 scale1 s t ! ,h>mony is ,%ape-,h>mony1 n ,measure-,h>mony4 ,we h 9 ! attitude ( ?is repetition a ,symmetrical ,balance4 ,! movem5t is rhy?mical & ! direction ( ! rhy?m is up4 ,! movem5t 9 ,fig4 #aii might be indef9itely increas$ by ! introduction into x ( a gradation ( attractions1 increas+ 9 numb]4 ,t means t ! ext5t ( contra/+ edges is increas$ f measure to measure4 ,illu/ration3 ,fig4 #bjj ,! addition ( details1 increas+ 9 numb] f measure to measure upw>d1 increases ! movem5t ( ! rhy?m 9 t direction4 ,illu/ration3 ,fig4 #bja ,tak+ ! arrangem5t ( ,fig4 #aii & repeat+ x six times at div]g+ angles ( sixty degrees1 we get :at may be call$ a radial balance ^u ! basis ( a hexagon4 ,outl9es may be drawn "o inside ( ! o!r or sev]al inside ( "o4 ,illu/ration3 ,fig4 #bjb ,?is is a case ( outl9es-)9-outl9es & ( ,%ape-,h>mony )\t ,measure-,h>mony4 ,!re is1 also1 a ,h>mony ( ,attitudes1 b no ,h>mony ( ,int]vals4 ,int]e/+ results may be produc$ by draw+ a s]ies ( outl9es simil> 9 %ape1 ! second inside ( ! fir/1 ! ?ird inside ( ! second1 & s on4 ,illu/ration3 ,fig4 #bjc ,9 ?is case1 = example1 we h ! outl9es drawn "o inside ( ! o!r4 ,! outl9es h all ! same %ape1 b diff]5t measures4 ,x is a case ( ,%ape-,h>mony & ,h>mony ( ,attitudes1 )\t ,measure-,h>mony1 & )\t any ,h>mony ( ,int]vals4 ,?is is a v int]e/+ & important =m ( ,design :i* has _m applications4 ,illu/ration3 ,fig4 #bjd ,9 ?is case1 also1 we h ,%ape-,h>mony )\t ,measure-,h>mony4 ,we h a ,h>mony ( ,attitudes & also ( ,int]vals1 ! spaces betwe5 ! outl9es correspond+4 ,illu/ration3 ,fig4 #bje ,"h we h ! ,h>mony ( an alt]nation ( ,attitudes repeat$1 ) ,%ape-,h>mony1 )\t ,measure-,h>mony4 ,9 all =ms ( design 9 :i* we h ! conc5tric repetition ( a c]ta9 outl9e we h1 9 connection ) ! feel+ ( a c5tral balance1 ! feel+ ( a movem5t or movem5ts t[>d ! c5t]4 ,^! movem5ts are due to conv]g5ces4 ,movem5ts c>ry+ ! eye away f ! c5t]1 9 opposite directions1 int]f]e ) ! feel+ ( balance4 ,! feel+ is enhanc$1 h[ev]1 :5 ! movem5ts conv]ge & come toge!r4 ,we may h n only an alt]nation ( attitudes 9 ^! cases1 b an alt]nation ( %ape-"*4 ,illu/ration3 ,fig4 #bjf ,! repetition ( outl9es-)9-outl9es may be conc5tric or ecc5tric4 ,! repetition is conc5tric 9 ,fig4 #bjd4 ,x is ecc5tric 9 ! example :i* foll[s4 ,illu/ration3 ,fig4 #bjg ,9 all ecc5tric repetitions l ?is we h a lack ( balance & ! sugge/ion ( movem5t4 ,! direction ( ! movem5t is det]m9$ by ! direction ( conv]g5ces & ( ! cr[d+ toge!r ( attractions4 ,! movem5t 9 ,fig4 #bjg is up-to-!-left1 unmi/akably4 ,repeat+ ! composition ( ,fig4 #bjg1 at regul> int]vals & )\t *ange ( attitude1 ! movem5t up-to-!-left w\ld be ext5d$ to ! repetitions & ! movem5t w\ld be rhy?mical4 ,! movem5t is rhy?mical 9 ! composition itself1 z %[n 9 ,fig4 #bjg1 because ! movem5t 9 ! composition is regul> 9 "*1 regul> 9 its measures1 & unmi/akable 9 direction4 ,illu/ration3 ,fig4 #bjh ,?is is ano!r example ( ecc5tric repetition 9 outl9es-)9-outl9es4 ,z 9 ,fig4 #bjg1 we h movem5t1 & ! movem5t is rhy?mical4 ,9 ! examples ,i h giv5 !re h be5 no contacts & no int]lac+s4 ,contacts & int]lac+s are possible4 ,illu/ration3 ,fig4 #bji ,"h1 = an example1 is an in/ance ( contact1 ) ,h>mony ( ,attitudes & a ,symmetrical ,balance on a v]tical axis4 ,illu/ration3 ,fig4 #baj ,9 ?is case we h contacts1 ) no ,h>mony ( ,attitudes4 ,! balance :i* is c5tral z well z axial is 9 ?is attitude ( ! figure symmetrical4 ,illu/ration3 ,fig4 #baa ,"h we h a simil> composition ) int]lac+s4 ,:5 ! outl9es h diff]5t %apes z well z diff]5t measures1 p>ticul>ly :5 ! outl9es are irregul> & ! %apes to be put toge!r are1 9 !mselves1 disord]ly1 ! problem ( composition becomes m difficult4 ,! be/ plan is to arrange ! outl9es 9 a gr\p1 mak+ z _m ord]ly connections z possible4 ,tak+ any composition ( outl9es & repeat+ x 9 ! diff]5t ways :i* ,i h describ$1 x is g5]ally possible to a*ieve ord]ly if n beautiful results4 ,illu/ration3 ,fig4 #bab ,"h are five outl9es1 v diff]5t 9 %ape-"*4 ,let u see :at c be done ) !m4 ,a lot ( exp]im5ts h to be tri$1 to f9d out :at connections1 :at arrangem5ts1 :at effects are possible4 ,! possibilities _c be pr$ict$4 ,us+ trac+-pap]1 a great _m exp]im5ts c be tri$ 9 a %ort "t1 ?\gh x may take a long "t to rea* ! be/ possible results4 ,illu/ration3 ,fig4 #bac ,9 ?is example ,i h tri$ to make a good composition ) my five outl9es4 ,! problem is difficult4 ,! outl9es to be comb9$ h s little ,h>mony4 ,! only ,h>mony we c a*ieve w be ! ,h>mony ( ! same arrangem5t ( %apes repeat$1 :i* am\nts to ,%ape-,h>mony4 ,inv]sions w give u ! satisfaction ( ,balance4 ,inv]sions on a v]tical axis w give u ! satisfaction ( ,symmetry4 ,9 ! design above giv5 ,i h a*iev$ simply ! ,h>mony ( a relation ( %apes repeat$1 ) ,rhy?m4 ,! ,rhy?m is due to ! repetition ( a decid$ly unbalanc$ gr\p ( elem5ts ) a pr$om9ance ( conv]g5ces 9 "o direction4 ,! movem5t is on ! :ole up1 9 spite ( c]ta9 d[nw>d conv]g5ces4 ,! upw>d conv]g5ces pr$om9ate4 ,!re are m incl9ations to ! "r ?an to ! left1 b ! composition :i* is repeat$ is un/able 9 its attitude & sugge/s a fall+ away to ! left4 ,! resultant ( ^! slight div]g5ces ( movem5t is a g5]al upw>d movem5t4 ,illu/ration3 ,fig4 #bad ,9 ?is case ,i h less difficulty ?an 9 ,fig4 #bac1 hav+ left out "o ( my five outl9es1 ! "o mo/ difficult to use ) ! o!rs4 ,!re is a great ga9 ( ,h>mony4 ,!re is a ,h>mony ( ,int]vals & a ,h>mony 9 ! repetition ( ! same gr\p+ ( outl9es4 ,9 ! outl9es !mselves we h a ,h>mony ( curv$ "*1 & ! curves fit "o ano!r v well1 ow+ to a correspond5ce ( measure & %ape-"* 9 c]ta9 p>ts4 ,9 su* cases we are able to get consid]able ,h>mony ( ,attitudes into ! composition4 ,!re is a ,h>mony ( ,attitudes 9 ! repeats1 z well z 9 c]ta9 details4 ,comp>+ ,fig4 #bad ) ,fig4 #bac1 ,i am sure ! read] w agree t we h 9 ,fig4 #bad ! l>g] measure ( ,h>mony4 ,illu/ration3 ,fig4 #bae ,9 ,fig4 #bae ,i h us$ inv]sions & repetitions ( ! r disord]ly outl9e :i* gave me s mu* difficulty :5 ,i tri$ to comb9e x ) ! o!r outl9es4 ,:atev] m]it ! composition has is due solely to ! art ( composition1 to ! pres5ce ( ,attitude-,h>mony1 ,int]val-,h>mony1 & to ! inv]sions & repetitions2 inv]sions giv+ ,balance1 repetitions giv+ ,h>mony4 ,:ile x is important to recognize ! limitation ( ! t]ms 9 ?is problem1 x is important to yield to any def9ite impulse :i* y may feel1 ?\gh x c>ries y beyond y\r t]ms4 ,! value ( a rule is (t5 f\nd 9 break+ x = a good & suffici5t reason2 & !re is no bett] reason ?an t :i* all[s y1 9 ,design1 to foll[ any impulse y may h1 provid$ t x is consi/5t ) ! pr9ciples ( ,ord]4 ,illu/ration3 ,fig4 #baf ,9 ?is case an ef=t has be5 made to modify ! t]ms already us$ s z to produce a m rapid & consi/5t movem5t4 ,advantage has be5 tak5 ( ! fact t ! eye is drawn into all conv]g5ces1 s all po9t+ d[n has be51 s f> z possible1 avoid$4 ,! movem5t is di/9ctly rhy?mical4 ,9 ! previ\s examples ,i h avoid$ contacts & int]lac+4 ,x 0 n necess>y to avoid !m4 ,illu/ration3 ,fig4 #bag #aag4 ,:at is done1 9 e case1 dep5ds ^u ! design] :o does x4 ,he foll[s ! sugge/ions ( 8 imag9ation1 n1 h[ev]1 ) p]fect lic5se4 ,! imag9ation acts )9 def9ite limitations1 limitations ( t]ms & ( pr9ciples1 limitations ( c]ta9 modes 9 :i* t]ms & pr9ciples are unit$4 ,9 spite ( ^! limitations1 h[ev]1 if we give ! same t]ms1 ! same pr9ciples1 & ! same modes to diff]5t p1 !y w produce v diff]5t results4 ,individuality expresses itself 9 spite ( ! limitation ( t]ms & modes1 & ! "w ( "o man w be v diff]5t f ! "w ( ano!r1 inevitably4 ,we may h ,ord]1 ,h>mony1 ,balance1 or ,rhy?m 9 all cases1 ,beauty only 9 "o case1 p]haps 9 no case4 ,x mu/ be rememb]$ h[1 9 ! practice ( ,pure ,design1 we aim at ,ord] & hope = ,beauty4 ,beauty is f\nd only 9 supreme in/ances ( ,ord]1 intuitively felt1 in/9ctively appreciat$4 ,! end ( ! practice ( ,pure ,design is f\nd 9 ! love ( ! ,beautiful1 r ?an 9 ! production ( beautiful ?+s4 ,beautiful ?+s are produc$1 n by ! practice ( ,pure ,design1 b out ( ! love ( ! ,beautiful :i* may be develop$ by ! practice4 ,,areas #aah4 ,i h already consid]$ ! measures ( areas1 9 discuss+ ! int]ior dim5sions ( outl9es1 & 9 discuss+ ! outl9es !mselves ,i h consid]$ ! %apes ( areas4 ,x rema9s = me to discuss ! tones 9 :i* ! areas may be drawn & ! tone-contra/s by :i* !y may be di/+ui%$ & def9$--9 _! positions1 measures1 & %apes4 ,,l9e> ,,areas #aai4 ,be=e proce$+1 h[ev]1 to ! subject ( tones & tone-relations1 ,i mu/ speak ( a peculi> type ( area :i* is produc$ by increas+ or dim9i%+ ! wid? ( a l9e4 ,i h po/pon$ ! discussion ( measures ( wid? 9 l9es until n[4 ,a l9e may *ange its wid? 9 c]ta9 p>ts or passages4 ,x may become wid] or n>r[] z ! case may be4 ,! wid] x is ! m x is l an area4 ,if x is suffici5tly wide1 ! l9e ceases to be a l9e1 & becomes an area4 ,! l9e may *ange its wid? abruptly or gradually4 ,! effect ( ! l9e is by ^! *anges indef9itely v>i$4 ,! l9e ( ,design is n ! l9e ( ,geometry4 #abj4 ,consid]able int]e/ may be giv5 to :at ,i h call$ ,l9e> ,progressions by *ang+ ! wid? ( ! l9e at c]ta9 po9ts1 9 c]ta9 passages1 & m or less abruptly4 ,! *anges w be l acc5ts 9 ! l9e1 giv+ v>iety &1 possibly1 an add$ int]e/4 ,illu/ration3 ,fig4 #bah ,let u take ?is l9e z ! motive ( a l9e> progression4 ,we c give x a diff]5t "*1 p]haps a m int]e/+ "*1 by wid5+ all ! v]tical passages1 z foll[s3-- ,illu/ration3 ,fig4 #bai ,?is is :at we get = a motive by wid5+ all ! v]tical passages4 ,illu/ration3 ,fig4 #bbj ,?is is :at we get = a motive by wid5+ all ! horizontal passages4 ,illu/ration3 ,fig4 #bba ,comp>e ?is ,progression1 9 :i* ,i h us$ ! motive ( ,fig4 #bai1 ) t ( ,fig4 #gg1 p4 #dg4 ,! acc5ts1 :i* 9 ,fig4 #bba occur 9 e repetition ( ! motive1 might occur only 9 c]ta9 repetitions1 at c]ta9 int]vals4 ,illu/ration3 ,fig4 #bbb ,x is n necess>y t ! *anges 9 ! wid? ( ! l9e be abrupt1 z 9 ! examples j giv54 ,! wid? ( ! l9e may increase or dim9i% gradually1 9 :i* case we may h1 n only acc5ts 9 ! l9e1 b movem5ts due to gradations ( dim5sion1 to conv]g5ces1 or to an increase or gradual cr[d+ toge!r ( attractions 9 a s]ies ( visual angles4 ,illu/ration3 ,fig4 #bbc ,9 ?is case we h a gradual increase foll[$ by a dim9ution ( ! wid? ( ! l9e 9 c]ta9 p>ts1 & ^! *anges occur at equal int]vals4 ,a c]ta9 am\nt ( rhy?mic movem5t is giv5 to ! progression by su* acc5ts1 provid$ ! direction ( movem5t is unmi/akable1 :i* x is n 9 ?is case4 ,x is n at all cle> :e!r ! movem5t is d[n-to-!-"r or up-to-!-left4 ,x seems to me ab\t z easy to move 9 "o direction z 9 ! o!r4 ,illu/ration3 ,fig4 #bbd ,9 ?is case !re is less d\bt ab\t ! movem5t4 ,x seems to be d[n-to-!-"r4 ,! eye is pull$ ?r\gh an increase ( wid?-measures t[>d a great] ext5sion & cr[d+ toge!r ( contra/+ edges4 ,illu/ration3 ,fig4 #bbe ,sub/itut+ outl9es = areas 9 ! previ\s illu/ration1 we are surpris$1 p]haps1 to f9d t ! movem5t is rev]s$4 ,we g up-to-!-left 9 ?is case1 n d[n-to-!-"r4 ,! pull ( a great] ext5sion ( tone-contra/ 9 a giv5 area 01 9 ,fig4 #bbd1 suffici5t to ov]come ! pull ( a less evid5t conv]g5ce 9 ! o!r direction4 ,by increas+ or dim9i%+ ! wid? ( l9es1 do+ x gradually or abruptly1 we are able to control ! movem5t ( ! eye to an indef9ite ext5t4 ,?is is "o ( ! important res\rces ( ! design]'s art4 ,its use is n limit$ to =ms ( ,l9e> ,progression1 b may be ext5d$ to all =ms ( ,design 9 :i* l9es are us$4 ,illu/ration3 ,fig4 #bbf ,9 ?is case1 = example1 ! eye foll[s ! direction ( conv]g5ces1 b we c easily =ce x to turn & move 9 ! opposite direction1 by wid5+ ! l9es 9 t direction1 ?us increas+ ! ext5t ( contra/+ edge until x m ?an outbalances ! conv]g5ces2 z 9 ! foll[+ illu/ration3-- ,illu/ration3 ,fig4 #bbg ,,! ,,arrangem5t ,,& ,,composition ,,( ,,areas #aba4 ,:at has be5 said ab\t ! composition ( ,l9es & ,outl9es applies equally well to ! composition ( ,areas1 s f> z !y are di/+ui%$ & def9$ by outl9es4 ,we w n[ proce$ to consid] areas z di/+ui%$ & def9$1 n by outl9es1 b by tone-contra/s4 ,! composition ( l9es & outl9es is "o ?+1 ! composition ( tones 9 diff]5t positions1 measures1 & %apes is ano!r4 ,9 putt+ l9es & outl9es toge!r we draw4 ,! po9t ( view is t ( draw+4 ,9 putt+ tones 9 diff]5t positions1 measures1 & %apes we pa9t4 ,! po9t ( view is t ( ! pa9t+4 ,,tones ,,& ,,tone-,,relations #abb4 ,up to ?is po9t ,i h avoid$ ! consid]ation ( ,tones & ,tone-,relations4 ,i h spok5 ( possible *anges ( tone 9 dots & 9 l9es2 *anges ( value1 ( color1 ( color-int5sity2 b x is n 9 dots nor 9 l9es t ^! *anges call = p>ticul> att5tion4 ,our int]e/ has be5 9 ! positions1 measures1 %apes1 & attitudes ( dots & l9es1 & 9 ! possibilities ( arrangem5t & composition4 ,:5 x comes to ! consid]ation ( areas & area-sy/ems1 h[ev]1 ! subject ( tone-relations becomes "o ( ! greate/ int]e/1 because areas are def9$ & di/+ui%$1 n only by _! outl9es1 b q z mu* by diff]5ces ( tone2 t is to say1 by tone-contra/s4 ,,! ,,process ,,( ,,pa9t+ ,,z ,,di/+ui%$ ,,f ,,draw+ #abc4 ,! fir/ ?+ to consid] is ! tone ( ! surface ^u :i* y are go+ to pa9t4 ,y !n take a tone diff]+ f ! gr\nd-tone1 9 value1 9 color1 or 9 color-int5sity1 y put x 9 a c]ta9 position1 & y spread x ov] a c]ta9 ext5t ( space4 ,9 s do+ y give to ! space a c]ta9 %ape4 ,?is is ! process ( ,pa9t+1 z di/+ui%$ f ! process ( ,draw+4 ,9 ,draw+ we ?9k ( l9es & outl9es fir/4 ,9 ,pa9t+ we ?9k ( ,tones fir/1 ( positions1 measures1 & %apes aft]w>ds4 ,,def9ition ,,( ,,! ,,^w ,,tone #abd4 ,9 produc+ tones we use1 necess>ily1 c]ta9 pigm5t-mat]ials & mixtures ( ^! mat]ials4 ,! effect ( light produc$ by any p>ticul> mat]ial or mixture we call its tone4 ,?\gh ,i h be5 us+ ! ^w ,tone ,i h n yet def9$ its mean+4 ,i w n[ d t4 ,,tone-,,analysis1--,,value1 ,,color1 ,,int5sity1 ,,neutrality #abe4 ,9 e tone we h to di/+ui% two elem5ts1 ! quantity ( light 9 x--:at we call its value--& ! quality ( ! light 9 x--its color2 & ! color1 :atev] x is1--,r$1 ,orange1 ,yell[1 ,gre51 ,blue1 or ,violet1--may be int5se or neutral4 ,by int5sity ,i mean ! quality ( a color 9 its highe/ or 9 a v high degree4 ,by ! int5sity ( ,r$ ,i mean ,r$ :5 x is z r$ z possible4 ,! mixture ( ,v]milion & ,rose ,madd]1 = example1 gives u a ,r$ ( great int5sity4 ,t is ab\t ! /ronge/ ,r$ :i* we are able to produce ) ! pigm5t-mat]ials :i* we use4 ,int5sity mu/ n be conf\nd$ ) value nor value ) int5sity4 ,by value ,i mean m or less light4 ,by int5sity ,i mean a great purity & brilliancy ( color4 ,int5sity /&s 9 opposition to neutrality1 9 :i* no color c be di/+ui%$4 ,! m color we h 9 any tone ! m int5sity we h4 ,! less ! int5sity ! less color1 & ! abs5ce ( color means neutrality or grayness4 ,neutrality or grayness1 ?\gh x is ! negation ( color1 ! z]o ( color1 s to speak1 mu/ be class$ z a color because ^u analysis x proves to be a result ( color comb9ation or mixture4 ,:5 ,i speak1 z ,i %all f "t to "t1 ( ! neutral z a color1 x w be und]/ood t ,i am speak+ ( a comb9ation or mixture ( colors 9 :i* no p>ticul> color c be di/+ui%$4 ,i speak ( ! neutral z a color j z ,i speak ( z]o z a numb]4 ,we use z]o z a numb] ?\gh x is no numb]1 & c\nts = no?+4 ,,/udy ,,( ,,tones ,,& ,,tone-,,relations #abf4 ,! /udy ( tones & tone-relations means ! /udy ( pigm5t-mat]ials & _! effects1 to f9d out :at quantities ( light we c produce1 :at qualities ( color1 :at int5sities ( color1 :at neutralizations4 ,t is ! problem ( tones & tone-relations4 ,we _c "k mu* ab\t tones & tone-relations until we h _h exp]i5ce 9 ! use ( pigm5t-mat]ials4 ,we mu/ be able to di/+ui% tones1 h[ev] slight ! diff]5ces ( value or ( color or ( color-int5sity1 & we mu/ be able to produce tones accord+ to our discrim9ations3 ?is ) exact precision4 ,9 ord] to ?9k 9 tone-relations we mu/ h def9ite ideas ( tone & ( tone-relations1 9 ! =m ( visual images4 ,9 ord] to express our ideas we mu/ be able to pa9t4 ,we mu/ h practice 9 pa9t+ & a great deal ( x4 ,i propose to describe ?is practice 9 tones & tone-relations3 :at x ought to be1 :at =ms x %\ld take4 ,,pigm5t-,,mat]ials #abg4 ,( pigm5ts ,i use ^!3 ,blue ,black1 ,madd] ,lake 7,deep71 ,rose ,madd]1 ,indian ,r$1 ,v5etian ,r$1 ,v]milion1 ,burnt ,si5na1 ,cadmium ,orange1 ,yell[ ,o*re1 ,pale ,cadmium1 ,aureol91 ,cremnitz ,:ite1 8,em]aude0 ,gre5 7,gre5 ,oxide ( ,*romium1 transp>5t71 ,cobalt ,blue1 ,fr5* ,ultram>9e ,blue4 ,^! are ! pigm5ts :i* ,i sugge/ = oil-pa9t+4 ,9 wat]-color pa9t+ ,i %\ld sub/itute ,*>coal ,gray = ,blue ,black4 8,em]aude ,gre50 is (t5 call$ ,viridian 9 ! =m 9 :i* x is us$ 9 wat]-color4 ,= ,cremnitz ,:ite ,i %\ld sub/itute1 9 wat]-color pa9t+1 ,*9ese ,:ite4 ,^! are ! pigm5t-mat]ials :i* ,i use myself & recomm5d to o!rs4 ,!re are1 ( c\rse1 _m o!r pigm5ts :i* may be us$1 b ^! w1 ,i ?9k1 be f\nd suffici5t = all purposes4 ,provid$ ) ^! pigm5ts1 ) a palette ^u :i* to put !m1 ) bru%es & o!r mat]ials necess>y = pa9t+1 we are prep>$ to take up ! /udy ( tones & tone-relations4 ,,! ,,scale ,,( ,,values #abh4 ,x is evid5t t we h 9 black pa9t ! lea/ quantity ( light :i* we c produce4 ,black is ! l[e/ ( all values4 ,x is equally evid5t t 9 :ite pa9t we h ! greate/ possible quantity ( light4 ,:ite is ! highe/ ( all values4 ,mix+ ,black & ,:ite 9 diff]5t proportions we c produce an indef9ite numb] ( int]m$iates4 ,we d n want1 h[ev]1 to be indef9ite 9 our t]ms2 on ! contr>y we want to be z def9ite z possible4 ,let u1 !re=e1 e/abli%1 betwe5 ,black & ,:ite1 a ,middle ,value 7,m72 betwe5 ,black & ,middle ,value an int]m$iate ,d>k 7,d72 betwe5 ,middle ,value & ,:ite an int]m$iate ,light 7,lt71 & betwe5 ^! five values ! int]m$iates1 ,l[ ,d>k 7,,ld71 ,high ,d>k 7,,hd71 ,l[ ,light 7,l,lt71 & ,high ,light 7,h,lt74 ,fur!r int]m$iates 7eight7 may be e/abli%$1 b to ^! we ne$ n give any p>ticul> names4 ,if we h occasion to ref] to any "o ( !m we c say t x lies betwe5 c]ta9 quantities or values ( light = :i* we h names4 ,we c speak1 = example1 ( ! int]m$iate betwe5 ,middle & ,high ,d>k1 & x may be describ$ 9 writ+ by ! =mula ,m-,,hd4 ,) ?is t]m9ology we %all be able to describe ! pr9cipal quantities or values ( light bo? 9 spee* & 9 writ+4 ,9 ord] to /udy ! pr9cipal quantities or values ( light & ! possibilities ( contra/ :i* !y af=d x is wise to avoid all diff]5ces ( color & color-int5sity4 ,to d t we produce our ,scale ( ,values 9 t]ms ( p]fect neutrality1 9 :i* no color c be di/+ui%$4 ,:5 we use ! names ( diff]5t values x is und]/ood t !y are values ( ,neutrality4 ,! t]m ,m1 = example1 /&s = ,neutral ,middle1 ,d = ,neutral ,d>k1 ,lt = ,neutral ,light4 ,,contra/s ,,( ,,! ,,scale ,,( ,,values #abi4 ,hav+ produc$ a scale ( n9e neutral values1 includ+ ,:ite & ,black1 ! "q arises z to ! numb] ( contra/s :i* x af=ds1 & x is easy to see t ! numb] is ?irty-six4 ,! v]tical l9es 9 ! foll[+ diagram indicate ! possible contra/s ( value 9 ! ,scale ( ,values4 ,c\nt+ ! l9es1 we see t ! numb] ( contra/s is ?irty-six4 ,produc+ ^! contra/s1 we %all see :at ea* "o am\nts to4 ,illu/ration3 ,diagram #a ,,def9ition ,,( ,,value-,,relations #acj4 ,! be/ me?od ( describ+ & di/+ui%+ ^! value-contra/s w be to use ! value-names 9 a =m ( fractions4 ,= example1 ,lt_/,d w\ld mean a contra/ ( ,d>k on ,light1 ,d_/,lt w\ld mean a contra/ ( ,light on ,d>k1 ,wt_/,blk w\ld mean a contra/ ( ,black on ,:ite4 ,t is to say1 ,:ite is subdivid$ or cross$ by ,black4 ,:5 we wi% to describe sev]al contra/s 9 comb9ation1 we set ! value ( ! gr\nd-tone above ! l9e1 always1 ! value ( ! tone or tones put ^u x bel[1 ?us3-- ,lt ----------- ,wt ,blk ,?is =mula means1 spots ( ,:ite & ,black on a gr\nd ( ,light4 ,lt -------------- ,wt ,blk --- ,m ,?is =mula means spots ( ,:ite & ,black on a gr\nd-tone ( ,light1 ) a spot ( ,middle on ! ,:ite1 ! ,middle be+ altoge!r sep>at$ f ! ,light by ! ,:ite4 ,!re is no def9ite ?9k+ except 9 def9ite t]ms1 & )\t "s su* t]m9ology z ,i h devis$ & describ$1 x w be impossible to ent] ^u an exp]im5tal practice 9 value-relations ) ! hope ( def9ite results4 ,) def9ite t]ms1 h[ev]1 we c take up ! practice 9 value-relations ) a good *ance ( le>n+1 9 ! c\rse ( "t1 all t !re is to be le>n$4 ,,scales ,,( ,,colors ,,9 ,,diff]5t ,,values #aca4 ,we mu/ n[ proce$ to ! consid]ation ( ! qualities ( light beyond ! ,scale ( ,neutral ,values1 9 ! region ( colors & color-int5sities1--a region ( tones :i* we h n yet explor$4 ,x is evid5t t no color c exi/ ei!r 9 ! value ( ,black or 9 ! value ( ,:ite1 b 9 e o!r value we h ! possibility ( all colors4 ,t is to say1 we may h ,r$ 7,r7 or ,orange 7,o7 or ,yell[ 7,y7 or ,gre5 7,g7 or ,blue 7,b7 or ,violet 7,v7 or any ( ! colors ly+ int]m$iate betwe5 !m1--,r$ ,orange 7,,ro71 ,orange ,yell[ 7,,oy71 ,yell[ ,gre5 7,,yg71 ,gre5 ,blue 7,,gb71 ,blue ,violet 7,,bv71 or ,violet ,r$ 7,,vr73 all ^!1 9 any value ( ! ,scale ( ,values1 except 9 ! value ( ,black & 9 ! value ( ,:ite4 ,! possibilities ( value & color1 9 tones1 are exhibit$ 9 ! foll[+ diagram3-- ,,diagram ,,( ,,values ,,& ,,colors ,wt ,wt ,h,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,h,lt ,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,lt ,l,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,l,lt ,m ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,m ,,hd ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,hd ,d ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,d ,,ld ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,ld ,blk ,blk ,diagram #b ,,def9ition ,,( ,,! ,,color-,,t]ms #acb4 ,x is important t ! words :i* we use = ! diff]5t colors %\ld be well und]/ood1 t 9 us+ !m we use !m ) ! same mean+s4 ,by ,r$ ,i mean ! only positive color :i* %[s no elem5t ei!r ( ,yell[ or ( ,blue4 ,x is ! color :i* we (t5 describe by ! ^w crimson1 & we produce x by ! mixture ( ,rose ,madd] & ,v]milion4 ,by ,yell[ ,i mean ! only positive color :i* %[s no elem5t ei!r ( ,r$ or ,blue4 ,x is ! color ( ! primrose :i* may be produc$ by ! pigm5t ,aureol94 ,by ,blue ,i mean ! only positive color :i* %[s no elem5t ei!r ( ,yell[ or ( ,r$4 ,blue is se5 9 a cle> sky aft] ra9 & 9 ! pigm5t ,cobalt4 ,by ,orange ,i mean a positive color %[+ equal elem5ts ( ,r$ & ( ,yell[4 ,by ,gre5 ,i mean a positive color %[+ equal elem5ts ( ,yell[ & ( ,blue4 ,by ,violet ,i mean a positive color %[+ equal elem5ts ( ,blue & ,r$4 ,! "* ( ! int]m$iates is cle>ly indicat$ by _! sev]al names4 ,9 ea* "o we see ! adjac5ts 9 equal measures4 ,?is def9ition ( ! colors is only approximate4 ,x does n pret5d to be sci5tific1 b x may help to br+ u to a common und]/&+4 ,to c>ry ^! def9itions f>!r1 ,i %\ld h to produce examples4 ,?is ,i c d 9 my class-room1 produc+ ea* color accord+ to my idea1 exactly4 ,i might rea* ! same result approximately by color-pr9t+1 b ! result w\ld n1 probably1 be p]man5t4 ,! samples produc$ by h&1 = use 9 ! class-room1 c be reproduc$ f "t to "t :5 !y no long] answ] to ! ideas :i* !y are int5d$ to express4 ,9 ?is treatise ,i %all use a t]m9ology in/ead ( color$ illu/rations :i* w\ld n be satisfactory1 or1 if satisfactory1 n s p]man5tly4 ,,color-,,int5sities ,,9 ,,diff]5t ,,values #acc4 ,if we proce$ to c>ry out ! idea ( ,diagram #b1 produc+ all ! twelve colors 9 all ( ! sev5 values int]m$iate betwe5 ! extremes ( ,black & ,:ite1 mak+ ! colors1 9 e case1 z /rong1 z int5se1 z is possible ) ! pigm5t-mat]ials we h *os5 to use1 we %all discov] t ! twelve colors rea* _! greate/ int5sities 9 diff]5t values2 t is to say1 9 diff]5t quantities ( light4 ,r$ rea*es its greate/ int5sity 9 ! value ,high ,d>k1 ,orange 9 ,l[ ,light1 ,yell[ 9 ,high ,light1 ,gre5 9 ,l[ ,light1 ,blue 9 ,high ,d>k1 ,violet 9 ,l[ ,d>k1 approximately2 & ! int]m$iate colors rea* _! greate/ int5sities 9 ! int]m$iate values1 approximately4 ,9 ord] to indicate ?is fact 9 our diagram1 we w m>k ! positions ( greate/ int5sity by putt+ ! color signs 9 l>g] type4 ,,diagram ,,( ,,values1 ,,colors1 ,,& ,,color-,,int5sities ,wt ,wt ,h,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,hl ,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,lt ,l,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,l,lt ,m ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,m ,,hd ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,hd ,d ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,d ,,ld ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,ld ,blk ,blk ,diagram #c ,,tones ,,( ,,! ,,spectrum ,,& ,,( ,,pigm5ts #acd4 ,x is probable t we h 9 ! ,spectrum an indication ( ! natural value-relations ( ! diff]5t colors :5 9 _! highe/ int5sities4 ,ow+ to ! limitations ( pigm5t-mat]ial1 h[ev]1 x is impossible to reproduce ! int5sities ( ! ,spectrum satisfactorily4 ,an approximation is all t we c a*ieve 9 pa9t+4 ,,! ,,spectrum ,,sequ5ce ,,& ,,! ,,circuit ,,( ,,! ,,colors #ace4 ,hav+ produc$ ! scale ( twelve colors 9 ! values ( _! greate/ int5sities1 & z int5se z possible1 we get an approximation to ! ,spectrum ) ?is diff]5ce1 t ! color ,violet-,r$ 7,purple7 :i* we get 9 pigm5ts & mixtures ( pigm5ts does n occur 9 ! ,spectrum &1 s f> z we "k1 does n belong 9 ! ,spectrum4 ,we h 9 ! ,spectrum a sequ5ce :i* beg9s ) ,r$ & ends ) ,violet4 ,x is a sequ5ce1 n a circuit4 ,9 pigm5t-mixtures1 h[ev]1 we h a circuit1 cle>ly en\gh1 & ,violet-,r$ is a connect+ l9k betwe5 ,violet & ,r$4 ,,! ,,complem5t>ies #acf4 ,consid]+ ! circuit ( ! colors :i* we are able to produce ) our pigm5t-mat]ials1 ! "q arises1 ,:at contra/s ( color are ! /ronge/8 :at int]val 9 ! ,scale ( ,colors gives u ! /ronge/ possible color-contra/8 ,produc+ ! twelve colors 9 ! values ( _! greate/ int5sities1 & z int5se z possible1 & sett+ ! tones 9 a circuit & 9 _! natural & inevitable ord]1 y w obs]ve t ! greate/ color-contra/ is ! contra/ betwe5 colors at ! int]val ( ! sev5?3 = example1 ! contra/s ( ,r$ & ,gre51 or ,orange & ,blue1 or ,yell[ & ,violet4 ,! colors at ! int]val ( ! six? are less /rong 9 contra/4 ,! contra/ dim9i%es gradually z we pass f ! int]val ( ! sev5? to ! int]val ( ! second4 ,! contra/ ( colors at ! int]val ( ! sev5?1 ! greate/ possible contra/1 is call$ ! contra/ ( ! complem5t>ies4 ,9 e/imat+ int]vals we c\nt ! colors betwe5 :i* ! int]vals occur4 ,a ,,g5]al ,,classification ,,( ,,tones #acg4 ,tak+ ea* color 9 ! value ( its greate/ int5sity 7z %[n 9 ! ,spectrum71 & z int5se z possible1 ! color may be neutraliz$ 9 ! direction ( ,black 7neutral d>kness7 or ,:ite 7neutral light7 or 9 ! direction ( any value ( neutrality int]m$iate betwe5 ,black & ,:ite1 includ+ ! value ( ! color 9 its greate/ int5sity4 ,if we ?9k ( five degrees ( neutralization1 includ+ ! extremes ( ,int5sity & ,neutrality1 we %all get z def9ite a t]m9ology = color-int5sities & color-neutralizations z we h = colors & = values4 ,! *oice ( five degrees is arbitr>y4 ,x is a "q h[ f> ! classification %all g1 :at x %all include4 ,we are deal+ ) inf9ity1 & our limitations are necess>ily arbitr>y4 ,9 ,diagram #c we h a g5]al classification ( tones z to value1 color1 color-int5sity1 & color-neutralization4 ,( values we h n9e4 ,( colors we h twelve4 ,( degrees ( int5sity & ( neutralization we h five4 ,,color-,,int5sities ,,& ,,color-,,neutralizations #ach4 ,x is important to di/+ui% betwe5 degrees ( int5sity & degrees ( neutralization4 ,! degrees ( color-int5sity & ( color-neutralization1 9 any value1 are describ$ by fractions4 ,! =mula ,d-,r means1 value ,d>k1 color ,r$1 int5sity ?ree qu>t]s4 ,! =mula ,d-,r1 ,n means1 value ,d>k1 color ,r$1 ?ree qu>t]s neutraliz$4 ,! =mula ,m-,o means1 value ,middle1 color ,orange1 int5sity "o half4 ,! =mula ,m-,o1 ,n means1 value ,middle1 color ,orange1 half neutraliz$4 ,m-,o1 ,n is a tone some:at less int5se 9 color ?an ,m-,o1 z may be se5 on ! diagram4 ,! degree ( neutralization has ref]5ce1 9 all cases1 to ! maximum int5sity = ! giv5 value4 ,:at t is1 !oretically1 may be se5 by ref]r+ to ! triangle ( ! color1 9 :i* ! possibilities ( int5sity1 9 diff]5t values1 are cle>ly indicat$4 ,,! ,,def9ition ,,( ,,p>ticul> ,,tones #aci4 ,to def9e any tone1 9 ?is classification1 we mu/ "n its value1 its color1 & ! degree ( color-int5sity or neutralization4 ,,! ,,classification ,,( ,,tones ,,necess>ily ,,!oretical #adj4 ,! g5]al classification ( tones 9 :i* is %[n all ! possibilities ( value1 color1 color-int5sity1 & color-neutralization1 9 reflect+ pigm5ts1 is necess>ily !oretical1 or r ideal1 because ! degrees ( int5sity obta9able 9 any value dep5d ^u ! pigm5t-mat]ials we h to use1 or *oose to use4 ,no v great int5sity ( ,yell[1 ev5 9 ! value ( ,high ,light1 c be obta9$ if we *oose to use a mixture ( ,yell[ ,o*re ) ,ultram>9e ,blue & ,:ite to produce x4 ,x is only :5 we use ! mo/ brilliant pigm5ts--! ,madd]s1 ,v]milion1 ! ,cadmiums1 ,aureol91 & ,cobalt ,blue--t we c approximate t[>d ! highe/ int5sities1 z indicat$ 9 our diagram & exhibit$ 9 ! ,spectrum4 ,,! ,,def9ition ,,( ,,p>ticul> ,,tone-,,relations #ada4 ,! numb] ( tone-contra/s--contra/s ( value1 ( color1 & ( color-int5sity or neutralization--is1 evid5tly1 beyond calculation4 ,! me?od ( describ+ any p>ticul> contra/ or contra/s is easy to und]/&4 ,we h only to def9e ! tones & to indicate h[ !y cross "o ano!r4 ,,ro1 ,n --------- ,,vr ,?is =mula means t a spot ( ,violet-,r$ 7,d>k1 full int5sity7 is put on a gr\nd-tone ( ,middle ,r$-,orange1 half neutraliz$4 ,,ro ---------- ,,vr ,wt -- ,,yg ,?is =mula means t spots ( ,l[ ,d>k ,violet-,r$ 7full int5sity7 & ,:ite are put on a gr\nd-tone ( ,middle ,r$-,orange1 half int5sity1 & t on ! spot ( ,l[ ,d>k ,violet-,r$ 7full int5sity71 z a gr\nd-tone1 is put a spot ( ,light ,yell[-,gre5 7full int5sity74 ,x is n necess>y to "n ! value :5 ! color occurs 9 ! value ( its greate/ int5sity1 & x is n necess>y to describe ! int5sity1 9 any value1 :5 ! greate/ int5sity possible to t value is meant4 ,9 ! fir/ case ! value is und]/ood1 9 ! second case ! int5sity--! greate/ = ! value--is und]/ood4 ,,sequ5ces ,,( ,,values ,,& ,,colors #adb4 ,:51 9 view ( all possible tones1 z indicat$ 9 ! g5]al classification ( tones1 accord+ to value1 color1 & color-int5sity1 or color-neutralization 7,diagram ( ! ,triangles71 we try to ?9k :at tones we %all use1 :at contra/s ( tone we %all produce1 we are sure to be v mu* 8at sea10 because ( ! great numb] & v>iety ( possibilities4 ,ev5 :5 we disreg>d diff]5ces ( int5sity & consid] simply ! possibilities ( value & ( color1 z %[n 9 ! g5]al classification ( tones accord+ to value & color 7,diagram ( ,values & ,colors1 p4 #acg71 we h /ill too _m possibilities to consid]1 & our *oice ( tones is det]m9$ by accid5t or habit r ?an by cle> vision or delib]ate pref]5ce4 ,we %all f9d x wor? :ile to limit our range 9 ea* exp]im5t to "s p>ticul> sequ5ce ( values & colors1 or to "s p>ticul> comb9ation ( sequ5ces4 ,in/ead ( try+ to ?9k 9 ! range ( all values1 all colors1 we ought to limit our ?9k+1 9 ea* case1 to ! range ( a few values & a few colors1--a few def9ite tones ) :i* we c become p]fectly famili> & ( :i* we c h def9ite visual images4 ,x is only :5 we c imag9e tones vividly t we c ?9k satisfactorily 9 tone-relations4 ,we %all a*ieve ?is p[] ( ?9k+ 9 tones & tone-relations be/ ?r\gh self-impos$ limitations4 #adc4 ,we ought to beg9 our /udy ( ,tones & ,tone-,relations ) ! ,scale ( ,neutral ,values 7see p4 #ace74 ,we ought to "w ) ! n9e tones ( ?is scale or sequ5ce until we "k !m well1 until we c visualize !m cle>ly1 & until we c produce !m accurately2 until we c readily produce any s+le tone ( ! scale & any ( ! ?irty-six possible contra/s :i* ! scale af=ds4 ,besides ! ,scale ( ,neutral ,values !re are ?ree types ( ,value & ,color ,sequ5ce :i* we may use4 #add4 ,fir/4 ,we h ! sequ5ces :i* may be describ$ z ^? ( ! ,v]tical2 sequ5ces :i* may be indicat$ by v]tical l9es drawn across ! ,diagram ( ,values & ,colors4 ,9 ea* ( ^! sequ5ces1 twelve 9 numb]1 we h "o color 9 all ! values ( ! ,scale ( ,values1 except ,black & ,:ite4 ,^! sequ5ces ( ! ,v]tical1 z ,i %all call !m1 are ( v little use 9 ,pure ,design4 ,!y give u value-contra/s & contra/s ( color-int5sity 7int5sities ( "o color71 b no color-contra/s1 no diff]5ces ( color4 ,! tones 9 ^! sequ5ces are monoton\s 9 color4 #ade4 ,second4 ,we h ! sequ5ces :i* may be describ$ z ^? ( ! ,horizontal2 sequ5ces :i* may be indicat$ by horizontal l9es drawn across ! ,diagram ( ,values & ,colors4 ,9 ^! sequ5ces we h diff]5ces ( color & color-int5sity1 b all 9 "o value4 ,^! sequ5ces give u color-contra/s 7diff]5t colors 9 diff]5t degrees ( int5sity71 b no value-contra/s4 ,! tones 9 ^! sequ5ces are monoton\s 9 value4 ,! sequ5ces ( "o horizontal are ( v little use4 #adf4 ,?ird4 ,we h ! sequ5ces :i* may be describ$ z ^? ( ! ,diagonal2 sequ5ces :i* may be indicat$ by l9es drawn diagonally across ! ,diagram ( ,values & ,colors4 ,9 draw+ ^! sequ5ces ! read] mu/ n =get t ! ,scale ( ,colors is a circuit1 s :5 he rea*es ! end ( ! diagram he returns & cont9ues f ! o!r end4 ,! diagram might1 = conv5i5ce 9 draw+ ^! sequ5ces1 be ext5d$ to sev]al repetitions ( ! ,scale ( ,colors4 ,9 ! sequ5ces ( ! ,diagonal we h contra/s bo? ( value & ( color4 ,! color 9 ^! sequ5ces *anges f value to value ?r\gh ! ,scale ( ,values4 ,ea* sequ5ce gives u c]ta9 colors 9 c]ta9 values1 & 9 no case h we two colors 9 ! same value4 ,to ^! sequ5ces ( ! ,diagonal we mu/ give our p>ticul> att5tion4 ,!y are ! sequ5ces :i* we %all use con/antly1 9 ,repres5tation z well z 9 ! practice ( ,pure ,design4 #adg4 ,! sequ5ces ( ! ,diagonal fall into two divisions4 ,fir/1 !re are ! sequ5ces :i* we draw ?r\gh ! ,diagram ( ,values & ,colors f ,black up-to-!-"r to ,:ite4 ,i %all call ^! ! ,sequ5ces ( ! ,"r ,mode 7,sign 74 ,second1 !re are ! sequ5ces :i* we draw f ,black up-to-!-left to ,:ite4 ,i %all call ^! ! ,sequ5ces ( ! ,left ,mode 7,sign 74 ,tak+ ! l[e/ color 9 ! sequ5ce z ! keynote1 we h = ! ,"r ,mode1 9 ! ,scale ( ,twelve ,colors1 twelve di/9ct sequ5ces ( :i* ?is :i* foll[s is an example4 ,seq4 ,,ld-,,bv1 #b;ds ,wt ,h,lt - ,,oy ,lt - ,o ,l,lt - ,,ro ,m - ,r ,,hd - ,,vr ,d - ,v ,,ld - ,,bv ,blk ,9 ?is sequ5ce ! colors are tak5 at ! int]val ( ! second4 ,t is :at is meant by ! abbreviation #b;ds4 ,tak+ ! l[e/ color ( ! sequ5ce z its keynote1 z be=e1 we h = ! ,left ,mode twelve di/9ct sequ5ces1 ( :i* t :i* foll[s is an example4 ,seq4 ,,ld-,,oy1 #b;ds ,wt ,h,lt - ,,bv ,lt - ,v ,l,lt - ,,vr ,m - ,r ,,hd - ,,ro ,d - ,o ,,ld - ,,oy ,blk ,9 ?is sequ5ce1 z 9 ! "o previ\sly giv51 ! colors are tak5 at ! int]val ( ! second4 #adh4 ,! colors 9 ^! diagonal sequ5ces may be tak5 n only at int]vals ( ! second1 b at int]vals ( ! ?ird1 ! f\r?1 ! fif?1 ! six?1 & ! sev5?4 ,tak+ ! colors at ^! diff]5t int]vals we h1 = ea* int]val1 tw5ty-f\r di/9ct sequ5ces2 twelve = ! ,"r ,mode1 twelve = ! ,left ,mode2 9 all "o hundr$ & =ty-f\r diff]5t sequ5ces4 #adi4 ,among ! sequ5ces ( ! ,diagonal ^? 9 :i* ! colors are tak5 at ! int]val ( ! fif? are p>ticul>ly int]e/+4 ,! colors tak5 at ! int]val ( ! fif? fall into f\r triads1--! fir/1 ,r-,y-,b1 ! second1 ,,ro-,,yg-,,bv1 ! ?ird1 ,o-,g-,v1 ! f\r?1 ,,oy-,,gb-,,vr4 ,tak+ ! colors 9 any ( ^! triads 9 ! two modes1 ! ,"r & ! ,left1 we get six sequ5ces ( diff]5t colors 9 diff]5t values = ea* triad4 ,( ^! ,triad-,sequ5ces ,i w give "o z an example4 ,seq4 ,,ld-,r1 #e?s ,wt ,h,lt - ,r ,lt - ,y ,l,lt - ,b ,m - ,r ,,hd - ,y ,d - ,b ,,ld - ,r ,blk ,! ,triad-,scales1 :e!r 9 ! ,"r ,mode or 9 ! ,left ,mode1 are ( great int]e/ bo? 9 ,pure ,design & 9 ,repres5tation4 ,9 ,repres5tation1 h[ev]1 ! numb] ( tones betwe5 ! limits ( ,black & ,:ite w\ld1 z a rule1 be increas$1 z 9 ! ext5d$ diagram giv5 f>!r on4 #aej4 ,in/ead ( tak+ ! colors at a c]ta9 int]val 9 "o mode or ! o!r1 x is possible to take ! colors 9 a c]ta9 relation ( int]vals repeat$2 ?is 9 ei!r mode4 ,! relation ( a ?ird to a fif?1 = in/ance1 be+ repeat$1 9 "o mode or ! o!r1 gives u "s v int]e/+ sequ5ces4 ,! "o :i* foll[s is an example4 ,,ld-,v1 #e?-#c;d ,wt ,h,lt - ,y #c;d ,lt - ,o #e? ,l,lt - ,v #c;d ,m - ,b #e? ,,hd - ,y #c;d ,d - ,o #e? ,,ld - ,v ,blk ,! relation ( a sev5? foll[$ by two fif?s1 :5 repeat$1 9 ei!r mode1 gives a l>ge numb] ( sequ5ces ( v great int]e/1 p>ticul>ly = ,repres5tation4 #aea4 ,any two ( ! sequ5ces :i* ,i h describ$ z ^? ( ! ,v]tical1 or m ?an two1 may be comb9$ & us$ toge!r4 ,9 t case we h two or m colors to a value4 ,! monotony :i* is inevitable 9 any s+le v]tical sequ5ce is avoid$ 9 ! comb9ation ( two or m su* sequ5ces4 ,seq4 ,r & ,seq4 ,y ,wt ,r ,h,lt ,y ,r ,lt ,y ,r ,l,lt ,y ,r ,m ,y ,r ,,hd ,y ,r ,d ,y ,r ,,ld ,y ,blk ,?is is an example ( ! comb9ation ( two v]tical sequ5ces--! sequ5ce ( ,r$ & ! sequ5ce ( ,yell[4 ,i h n f\nd ! sequ5ces ( ?is type v int]e/+4 ,9 us+ !m 9 ,repres5tation ,i h f\nd x desirable to h ! int5sities increase gradually t[>d :ite1 or1 :at am\nts to ! same ?+1 to h ea* color neutraliz$ z x loses light4 ,t happ5s1 con/antly1 9 ,nature4 #aeb4 ,any two ( ! sequ5ces :i* ,i h describ$ z ( ! ,horizontal1 or ev5 m ?an two1 may be comb9$ & us$ toge!r4 ,seq4 ,lt & ,d1 #c;ds ,lt ,r ,o ,y ,g ,b ,v ,d ,r ,o ,y ,g ,b ,v ,?is scale gives u a v>iety ( color-contra/s ) "o value-contra/4 ,! colors are tak5 at ! int]val ( ! ?ird4 ,!y might be tak5 at any int]val up to t ( ! sev5?1 9 :i* case we %\ld h a contra/ ( complem5t>y colors 9 two values1 ea* color occurr+ 9 ea* value4 ,! monotony ( value :i* is inevitable 9 any s+le horizontal sequ5ce is 9 ! comb9ation ( two or m su* sequ5ces avoid$4 ,i h us$ ! ,r$-,yell[-,blue triad 9 ?ree & 9 five values ) satisfaction4 ,ea* value repres5ts a plane ( light 9 :i* c]ta9 diff]5ces ( color are obs]v$4 #aec4 ,any two ( ! sequ5ces :i* ,i h describ$ z ( ! ,diagonal may be comb9$1 9 two ways4 ,fir/1 two sequ5ces ( ! same mode may be comb9$4 ,second1 two sequ5ces ( diff]5t modes1 "o ( ! ,"r ,mode & "o ( ! ,left ,mode1 may be comb9$4 ,,ld-,,gb #c;ds ) ,,ld-,,ro #c;ds ,wt ,,gb ,h,lt ,,ro ,,bv ,lt ,,oy ,,vr ,l,lt ,,yg ,,ro ,m ,,gb ,,oy ,,hd ,,bv ,,yg ,d ,,vr ,,gb ,,ld ,,ro ,blk ,9 ?is case we h a comb9ation ( two diagonal sequ5ces ( ! ,left ,mode 9 :i* ! colors are tak5 at ! int]val ( ! ?ird4 ,*ang+ ! mode ( ^! two sequ5ces we get !m inv]t$1 ?us3-- ,,ld-,,gb ,,ld-,,ro #c;ds ,wt ,,gb ,h,lt ,,ro ,,yg ,lt ,,vr ,,oy ,l,lt ,,bv ,,ro ,m ,,gb ,,vr ,,hd ,,yg ,,bv ,d ,,oy ,,gb ,,ld ,,ro ,blk ,"h ! mode is *ang$ & ! comb9$ sequ5ces inv]t$4 ,! comb9$ sequ5ces may be bo? 9 ! same mode or 9 diff]5t modes4 ,:5 ! modes are diff]5t ! sequ5ces come into contact1 & 9 "s cases cross "o ano!r4 ,,ld-,v #b;ds ) ,,ld-,v #b;ds ,wt ,y ,,oy ,,yg ,o ,g ,,ro ,,gb ,r ,b ,,vr ,,bv ,v ,blk ,9 ?is case we h a comb9ation ( two diagonal sequ5ces4 ,"o ( ! sequ5ces is 9 ! ,"r1 ! o!r is 9 ! ,left ,mode4 ,! colors are 9 ! values ( _! greate/ int5sities4 ,seq4 ,,ld-,,gb #c;ds ) ,,ld-,,gb #c;ds ,wt ,,gb ,,yg ,,bv ,,oy ,,vr ,,ro ,,vr ,,oy ,,bv ,,yg ,,gb ,blk ,9 ?is case ! comb9$ sequ5ces cross "o ano!r 9 ! tone ( ,m-,,ro4 ,! comb9$ sequ5ces h ?ree tones 9 common4 ,x may happ5 t ! sequ5ces comb9$ w h no tones 9 common4 ,?is is %[n 9 ! sequ5ce :i* foll[s3-- ,,ld-,o #e?s ) ,,ld-,b #e?s ,wt ,o ,h,lt ,b ,g ,lt ,y ,v ,l,lt ,r ,o ,m ,b ,g ,,hd ,y ,v ,d ,r ,o ,,ld ,b ,blk #aed4 ,in/ead ( hav+ two colors to a value 9 ! comb9ation ( two v]tical sequ5ces1 we may h an alt]nation ( colors 9 ! values1 giv+ "o color to a value1 ?us3-- ,wt ,r ,h,lt ,lt ,y ,r ,l,lt ,m ,y ,r ,,hd ,d ,y ,r ,,ld ,blk ,x has seem$ to me t ! sequ5ces 9 :i* we h "o color to a value give bett] results ?an ^? 9 :i* we h two or m colors to a value4 #aee4 ,in/ead ( hav+ ea* color 9 two values 9 ! comb9ation ( two horizontal sequ5ces1 we may h ! colors1 tak5 at equal int]vals1 occurr+ alt]nately fir/ 9 "o value & !n 9 ! o!r4 ,lt ,r 4 ,y 4 ,b 4 ,d 4 ,o 4 ,g 4 ,v #aef4 ,^! alt]nat+ sequ5ces may proce$1 n only v]tically & horizontally1 b diagonally across ! diagram4 ,9 t case ! alt]nations w be betwe5 diff]5t value-int]vals 9 a s]ies ( equal color-int]vals or betwe5 diff]5t color-int]vals 9 a s]ies ( equal value-int]vals4 ,wt ,h,lt ,,vr ,lt ,,bv ,l,lt ,,gb ,v ,m ,,yg ,b ,,hd ,,oy ,g ,d ,y ,,ld ,o ,blk ,9 ?is case ! alt]nation is betwe5 diff]5t value-int]vals ?r\gh ! ,scale ( ,colors4 ,! movem5t be+1 z a :ole1 up-to-!-"r1 is 9 ! ,"r ,mode4 ,i h n us$ any ( ! sequ5ces1 ( ?is type1 9 :i* ! value-int]vals alt]nate1 fir/ 9 "o mode !n 9 ! o!r1 ) a con/ant color-int]val1 b ,i h us$1 frequ5tly1 ! alt]nation ( two diff]5t color-int]vals 9 a s]ies ( equal value-int]vals4 ,! sequ5ces produc$ 9 ?is way are among ! mo/ int]e/+ ( all ! _m ,i h us$4 ,i w give sev]al examples4 ,wt ,h,lt ,y #g? ,lt ,v #e? ,l,lt ,o #g? ,m ,b #e? ,,hd ,r #g? ,d ,g #e? ,,ld ,v ,blk ,9 ?is case ! alt]nation is f ! keynote1 ,l[ ,d>k ,violet1 up fir/ 9 ! ,left ,mode a fif?1 !n up 9 ! ,"r ,mode a sev5?1 !n 9 ! ,left ,mode a fif?1 & s on up to ,:ite4 ,?is p>ticul> alt]nation might be describ$ z ! relation ( a fif? & a sev5? repeat$1 9 ! ,left ,mode4 ,wt ,h,lt ,y #g? ,lt ,v #e? ,l,lt ,g #g? ,m ,r #e? ,,hd ,b #g? ,d ,o #e? ,,ld ,v ,blk ,9 ! sequ5ce j giv5 ! alt]nation is1 f ! keynote ,l[ ,d>k ,violet1 fir/ 9 ! ,"r ,mode a fif?1 !n 9 ! ,left ,mode a sev5?3 ?is ?r\gh ! ,scale ( ,values up to ,:ite4 ,! ord] ( ! previ\s sequ5ce is inv]t$4 ,?is p>ticul> alt]nation might be describ$ z ! relation ( a fif? & sev5? repeat$ 9 ! ,"r ,mode4 ,! alt]nation ( int]vals ( ! fif? ) int]vals ( ! ?ird gives "s int]e/+ sequ5ces1 9 :i* ! alt]nation ( int]vals is1 necess>ily1 an alt]nation ( modes4 ,wt ,h,lt ,y #c;d ,lt ,g #e? ,l,lt ,o #c;d ,m ,y #e? ,,hd ,r #c;d ,d ,o #e? ,,ld ,v ,blk #aeg4 ,i h by no means exhau/$ ! possibilities ( value & color comb9ation1 b ,i h indicat$ a suffici5t numb] to s]ve ! purposes ( exp]im5tal practice 9 tone-relations1 = a long "t to come4 ,! sequ5ces :i* ,i h f\nd mo/ int]e/+1 9 my own exp]im5ts1 h be5 ! diagonal sequ5ces ( ! two modes1 us+ int]vals ( ! fif?1 & ! diagonal sequ5ces 9 :i* ) equal value-int]vals !re is an alt]nation ( c]ta9 color-int]vals1--! sev5? & ! fif?1 & ! sev5? & two fif?s4 ,x may v well be t ^! p>ticul> sequ5ces int]e/ me because ,i h us$ !m m ?an o!rs & consequ5tly ?9k 9 !m m easily4 #aeh4 ,= ! purposes ( ,pure ,design ! ,scale ( ,n9e ,values1 includ+ ,black & ,:ite1 w be f\nd suffici5t2 b :5 x comes to ! comb9ation ( ,design ) ,repres5tation1 & p>ticul>ly to ,repres5tation 9 ,full ,relief1 x w be necess>y to introduce int]m$iates into ! ,scale ( ,values4 ,) ?is purpose 9 view ,i give "o m diagram 9 :i* int]m$iates ( value h be5 introduc$4 ,= conv5i5ce 9 draw+ out ! diff]5t sequ5ces ^u ?is diagram ,i h repeat$ ! ,scale ( ,colors %[+ ! connection ( ,violet-,r$ ) ,r$4 ,?is diagram 7#e7 is simply an ext5sion ( ! ,diagram ( ,values & ,colors giv5 on p4 #acg4 #aei4 ,we may use ! v>i\s sequ5ces ,i h describ$ )\t mix+ ! tones1 us+ ! tones "o at a "t z !y may be requir$2 b if we *oose we may mix adjac5ts or ?irds or ev5 ?rees4 ,9 t way ! tone-possibilities ( ea* sequ5ce may be v mu* ext5d$4 ,x may be well to %[ :at ! ext5sion am\nts to by giv+ "o ( ! sequ5ces ) an indication ( ! result ( mixtures )9 ! limits describ$4 ,seq4 ,,ld-,r #e?s ,wt ,h,lt ,r ,lt ,y ,l,lt ,b ,m ,r ,,hd ,y ,d ,b ,,ld ,r ,blk ,?is is ! sequ5ce 9 :i* we decide to mix adjac5ts1 ?irds1 & ?rees4 ,a ,,diagram ,,( ,,values ,,& ,,colors ,wt ,wt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,h,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,h,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,l,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,l,lt ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,m ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,m ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,hd ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,hd ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,d ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,d ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,ld ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,,ld ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,r ,,ro ,o ,,oy ,y ,,yg ,g ,,gb ,b ,,bv ,v ,,vr ,blk ,blk ,diagram #e #b;ds #c;ds #cs ,wt ,h,lt ,r ,o ,lt ,y ,v ,n ,g ,l,lt ,b ,o ,n ,v ,m ,r ,g ,n ,o ,,hd ,y ,v ,n ,g ,d ,b ,o ,n ,v ,,ld ,r ,blk ,?is diagram %[s ! results ( mix+ seconds1 ?irds1 & ?rees4 ,x is evid5t t 9 mix+ ! tones ( any sequ5ce 9 ?is way we g beyond ! /rict limitations ( ! sequ5ce1 p>ticul>ly 9 mix+ ?irds & ?rees4 ,! results obta9$ are fairly def9ite1 h[ev]1 & ! tones obta9able are /ill )9 ! range ( def9ite ?9k+4 ,if we %\ld g f>!r1 to ! mixture ( tones beyond ! int]val ( ! ?ird1 we %\ld get into ! region ( indef9ite possibilities4 #afj4 ,x mu/ be cle>ly und]/ood t our object 9 us+ ^! sequ5ces & m or less re/rict$ mixtures is to limit our ?9k+ s t x may ga9 9 def9it5ess :at x loses 9 ext5t4 ,:5 we limit our ?9k+ 9 any case to a few tones1--c]ta9 colors 9 c]ta9 values1--we come to "k ^? tones s well t we c imag9e any "o ( !m vividly1 )\t see+ x4 ,x is only :5 we h 9 m9d def9ite tone-images t we beg9 to ?9k 9 tone-relations & rise to ! possibilities ( imag9ative composition 9 tones4 ,9 us+ ! sequ5ces :i* ,i h describ$ ! tones mu/ be c>efully mix$ & prep>$ ^u ! palette & set !re 9 ! ord] ( ! sequ5ce1 :atev] x is4 ,t gives ! pa9t] c]ta9 colors 9 c]ta9 values4 ,t is to say1 he has a c]ta9 numb] ( tones to consid] & to use4 ,he takes "o ( ! tones into 8 bru% & gives x a position1 a measure1 & a %ape4 ,t done1 he takes ano!r tone & gives to t a position1 measure1 & %ape4 ,proce$+ 9 ?is way he creates a c]ta9 relation%ip ( tones1 positions1 measures1 & %apes1 ! t]ms ( :i* relation%ip are p]fectly def9ite4 ,he repeats :at he f9ds satisfactory4 ,he avoids :at he f9ds unsatisfactory4 ,exp]im5t+ 9 ?is way1 9 su* def9ite t]ms1 he ought to make a sure & /eady progress t[>d ! discov]y ( :at is ord]ly & beautiful4 ,! use ( any p>ticul> sequ5ce ( values & colors is l ! use1 by ! musician1 ( a well-tun$ in/rum5t4 ,x is at once a def9ition ( t]ms & a s\rce ( sugge/ion & ( inspiration4 ,!re is no?+ occult or sacr$ ab\t ^! sequ5ces & comb9ations ( sequ5ces4 ,9 us+ !m we are 9 no way safegu>d$ aga9/ error4 ,us+ ^! sequ5ces1 we c produce bad effects ( light & ( color z well z good ones4 ,:e!r ! results ( us+ ^! sequ5ces are good or bad dep5ds ^u ! us]--:at 8 ?9k+ am\nts to4 ,x w be a grave mi/ake to reg>d any ( ^! sequ5ces z recipes = righte\sness1 :5 !y are simply modes ( ?\ght4 ,!y are no?+ m ?an ! sections or divisions ( a g5]al classification ( tones4 ,9 us+ any p>ticul> sequ5ce we obs]ve t ! same value & color-relations recur repeat$ly4 ,t is always desirable f ! po9t ( view ( ,design4 ,x means ,h>mony4 #afa4 ,! beauty ( any s*eme ( values & colors dep5ds1 n only ^u ! pigm5t-mat]ials us$1 ^u ! sequ5ce ( values & colors *os5 & ^u ! p>ticul> tones produc$1 b q z mu* ^u ! relative positions & juxtapositions giv5 to ! tones1 ! quantities or measures 9 :i* !y are us$1 &1 la/ly1 ! way 9 :i* ! pa9t is h&l$4 ,to f9d out :at tones to use z gr\nd-tones1 :at tones to put ^u ^! gr\nd-tones1 & 9 :at quantities or measures1 is a matt] ( exp]im5tal practice & ( visual & imag9ative discrim9ation4 ,hav+ def9$ ! ^w tone & its elem5ts1 value1 color1 & color-int5sity1 & hav+ e/abli%$ a g5]al classification ( tones to %[ ! possibilities ( tone1 ,i mu/ g on to describe :at w be ord]ly 9 tone-relations4 ,ord] & ,beauty 9 tone-relations w be f\nd 9 ,tone-,h>monies1 ,tone-,balances1 & ,tone-,rhy?ms4 ,,tone-,,h>mony #afb4 ,by ,tone-,h>mony ,i mean a relation ( lik5ess 9 tones4 ,tones are 9 ,h>mony :5 !y resemble "o ano!r 9 all or 9 c]ta9 respects4 ,to be 9 ,h>mony two or m tones mu/ h at lea/ some?+ 9 common1 ei!r value or color4 ,if !y h ! same color !y may be 9 ! same degree ( int5sity1 giv+ a ,h>mony ( ,int5sities4 ,tone-,h>mony resolves itself into ,value-,h>mony1 ,color-,h>mony1 & ! ,h>mony ( ,int5sities4 ,! ,h>mony ( ,int5sities lies betwe5 tones ( ! same color1 :5 !y are equally neutraliz$ or neutraliz$ 9 approximately ! same degree4 ,:5 diff]5t colors are neutraliz$ we h ! ,h>mony ( a common neutrality or grayness ( color4 #afc4 ,tones may be h>moniz$ on ! palette be=e !y are us$1 t is to say1 be=e any positions1 measures1 & %apes are giv5 to !m on pap] or canvas1 or !y may be h>moniz$ aft] positions1 measures1 & %apes h be5 giv5 to !m4 ,to h>monize tones on ! palette1 z to value1 we mu/ br+ !m approximately to ! same value1 ) z little *ange ( color z possible4 ,to h>monize tones on ! palette z to color we mu/ br+ !m approximately to ! same color1 ) z little *ange ( value z possible4 ,if two or m tones h ! same color !y may be int5sifi$ or neutraliz$ until !y are br\ght approximately to ! same degree ( int5sity4 ,! tones ( ,r$ %[+ a ,h>mony ( ,int5sities w\ld lie on a v]tical l9e drawn ?r\gh ! triangle ( ,r$ 9 ! ,diagram ( ! ,triangles4 ,z ! tone-effect :i* we produce dep5ds v l>gely ^u ! positions1 measures1 & %apes :i* we give to our tones1 we may n be satisfi$ ) an effect :i* has be5 produc$ ) previ\sly prep>$ & h>moniz$ tones4 ,we may wi% to *ange ! effect1 to a*ieve a /ill great] ,h>mony4 ,giv5 a c]ta9 arrangem5t or composition ( tones1 c]ta9 tones 9 c]ta9 positions1 measures1 & %apes1 & giv5 ! problem to h>monize ^? tones1 :at d we d8 #afd4 ,suppose x is ,value-,h>mony :i* we want2 :at is our proc$ure8 ,wt ,h,lt - ,b ,lt - ,v ,l,lt - ,y ,m - ,,ro ,,hd - ,g ,d - ,o ,,ld - ,,vr ,blk ,"h1 let u say1 are ! tones ( a design1 c]ta9 colors 9 c]ta9 values4 ,:at %all we d ) ^! tones to br+ !m into ,value-,h>mony8 ,illu/ration3 ,wt ,h,lt - ,b ,lt - ,v ,l,lt - ,y ,m - ,,ro ,,hd - ,g ,d - ,o ,,ld - ,,vr ,blk ,diagram #f ,= ,value-,h>mony we mu/ dim9i% ! range ( values t[>d "o value4 ,9 ! above diagram ,i h %[n a dim9ution ( ! range ( values t[>d ,high ,light3 ?is 9 sev]al degrees4 ,illu/ration3 ,wt ,h,lt - ,b ,lt - ,v ,l,lt - ,y ,m - ,,ro ,,hd - ,g ,d - ,o ,,ld - ,,vr ,blk ,diagram #g ,foll[+ ! indications ( ?is diagram1 we pull ! colors toge!r t[>d ,light 9 "o case 7,b71 t[>d ,middle 9 ano!r 7,a71 t[>d ,d>k 9 a ?ird 7,c74 ,z we d ?is we increase ! ,value-,h>mony4 ,9 reproduc+ ! tones 9 a dim9i%$ range ( values1 rais+ ! colors 9 value or l[]+ !m1 we are n oblig$ to *ange ! colors except 9 cases :]e !y become1 possibly1 conf\nd$ ) ,black or ) ,:ite4 ,x w (t5 happ51 h[ev]1 t ! int5sity ( a color has to be dim9i%$ :5 ! value is *ang$4 ,= example1 if ,r$ ,orange1 9 ! illu/ration giv51 is 9 its greate/ int5sity1 ! color may rema9 un*ang$ 9 ,sy/em 8,a10 b its int5sity w be1 necess>ily1 dim9i%$ 9 ,sy/em 8,b10 or ,sy/em 8,c40 ,see ,diagram ( ! ,triangles4 ,= ! sake ( ,value-,h>mony we dim9i% ! range ( values1 mak+ z little *anges ( color z possible1 & only ^? *anges ( color-int5sity :i* are inevitable4 ,a complete & p]fect ,value-,h>mony is1 z a rule1 undesirable because x means t all ! colors are r$uc$ to "o value :i* gives a monotony ( value4 ,approximate ,h>mony ( ,values is g5]ally suffici5t4 ,! range ( values is n>r[$1 ! contra/s are dim9i%$1 & an ev5 tonality is secur$4 ,t is all we require1 9 mo/ cases1 an approximation to "o value4 #afe4 ,suppose x is ,color-,h>mony :i* we want to a*ieve3 :at proc$ure %all we foll[8 ,wt ,b ,v ,y ,r ,g ,o ,blk ,"h are c]ta9 tones1 c]ta9 colors 9 c]ta9 values4 ,:at %all we d ) ^! tones to get ,color-,h>mony8 ,we mu/ dim9i% ! range ( color-contra/s by giv+ pr$om9ance to "o color1 ei!r to "o ( ! colors to be h>moniz$ or to "s o!r4 ,t may be done by mix+ "o color into all our tones4 ,wt ,wt ,b ,v ,v ,,vr ,y ,giv+ pr$om9ance to ,r$1 we get3 ,o ,r ,r ,g ,n ,o ,,ro ,blk ,blk ,diagram #h ,! range ( color-contra/ is 9 ?is way dim9i%$ to ! int]vals betwe5 ,violet1 ,orange1 & ,neutral4 ,! process1 s f> z color is conc]n$1 disreg>d+ value-relations1 is fully expla9$ 9 ! foll[+ diagram3-- ,illu/ration3 ,diagram #i ,suppose1 in/ead ( giv+ pr$om9ance to ,r$1 z 9 ! example above1 we give pr$om9ance to ,blue1 tak+ ! same range ( colors4 ,wt ,wt ,b ,b ,v ,,bv ,y ,giv+ pr$om9ance to ,blue we get3 ,g ,r ,v ,g ,,gb ,o ,n ,blk ,blk ,diagram #aj ,! range ( color-contra/ is 9 ?is way dim9i%$ to ! int]vals betwe5 ,gre51 ,violet1 & ,neutral4 ,! process1 s f> z color is conc]n$1 is fully expla9$ 9 ! foll[+ diagram3-- ,illu/ration3 ,diagram #aa ,9 ! diagrams :i* ,i h giv5 ! pr$om9ance is 9 ! measure ( "o half4 ,t is to say1 ! mixtures are half & half1 !oretically speak+4 ,! !oretical result is a range ( int]m$iate colors4 ,! pr$om9ance is n necess>ily 9 ! measure ( "o half4 ,x may be 9 any measure4 ,! pres5ce ( ,r$ or ,blue 9 all ! tones may be h>dly noticeable or x may am\nt to a g5]al r$ness or blu5ess 9 :i* o!r colors are di/+ui%$ ) m or less difficulty4 #aff4 ,suppose x is ! h>mony ( grayness1 a ,h>mony ( ,neutralization1 :i* we want4 ,:at is ! proc$ure to foll[8 ,illu/ration3 ,diagram #ab ,! proc$ure is %[n 9 ?is diagram4 ,we see "h :at is meant by a ,h>mony ( ,neutralization1 )\t *anges ( value4 ,! neutralization is 9 ! measure ( "o half 9 ea* case4 ,r$ ,orange & ,gre5 are ! only colors :i* exi/ 9 _! maximum int5sities4 ,_! int5sities are dim9i%$ to ! half-po9t1 )\t *ange ( value1--f ,,ro to ,,ro 9 "o case1 & f ,g to ,g 9 ! o!r4 ,! o!r colors are r$uc$ 9 _! int5sities proportionally4 ,! value 9 ea* case rema9s un*ang$4 #afg4 ,hav+ consid]$ ! me?ods ( gett+ ,value-,h>mony & ,color-,h>mony sep>ately1 ,i mu/ n[ describe ! me?od ( gett+ ! comb9ation ( ,value-,h>mony ) ! ,h>mony ( ,neutralization4 ,to d ?is we mu/ set ! colors 9 positions reg>d+ ! ,scale ( ,neutral ,values1 :i* w indicate _! sev]al values1 & 9 ea* case ! degree ( int5sity4 ,we mu/ !n decide :e!r to neutralize ! sev]al tones t[>d ,black or ,:ite1 or t[>d "s neutral value betwe5 ^! extremes4 ,illu/ration3 ,diagram #ac ,?is illu/rates ! me?od ( a neutralization t[>d ,black 9 ! measures ( "o qu>t]1 "o half1 & ?ree qu>t]s4 ,illu/ration3 ,diagram #ad ,?is illu/rates ! me?od ( a neutralization t[>d ,:ite 9 ! measures ( "o ?ird & two ?irds4 ,illu/ration3 ,diagram #ae ,?is illu/rates ! me?od ( a neutralization t[>d ! ,middle ,neutral1 betwe5 ,black & ,:ite1 a neutralization 9 ! measures ( "o qu>t]1 "o half1 & ?ree qu>t]s4 ,9 br++ tones into h>mony1 by "o or ano!r or all ( ^! v>i\s me?ods1 we mu/ rememb] t :5 we h dim9i%$ ! contra/ ( value & ( color beyond a c]ta9 po9t ! result is monotony1 a monotony :i* may be undesirable4 ,x is easy to get into a /ate ( m9d 9 :i* we dislike all contra/s4 ,9 ?is /ate ( m9d we f9d no /hetic satisfaction except 9 monotony4 ,su* a /ate ( m9d %\ld be avoid$4 ,monotony is ! ,nirvana ( /heticism4 #afh4 ,we may h a ,h>mony 9 ! repetition & recurr5ce ( two or m contra/+1 ev5 /rongly contra/+1 tones4 ,we may h ,h>mony 9 ! repetition & recurr5ce ( a contra/ 9 :i* !re is p]haps no ,h>mony4 ,= example3 ,i may repeat ! contra/ ,orange-,blue any numb] ( times 9 a c]ta9 composition4 ,!re is no ,h>mony ( ,value or ( ,color 9 ! contra/1 b 9 repeat+ ! contra/ ,i h ! ,h>mony ( a ,repetition1 j z ,i h a ,h>mony 9 ! repetition ( a c]ta9 l9e or outl9e 9 :i* !re is no ord] ( any k9d4 ,! ,h>mony lies solely 9 ! repetition or recurr5ce4 ,9 ?is way ,i may repeat1 at equal int]vals all ov] a c]ta9 space1 ! v>i\s contra/s indicat$ by ! foll[+ diagram3-- ,illu/ration3 ,diagram #af ,!re is no ,h>mony 9 ! relation ( tones "h indicat$1 b we %all get ,h>mony 9 ! repetition ( ?is relation4 ,illu/ration3 ,diagram #ag ,! ,h>mony "h indicat$ w lie 9 ! repetition ( c]ta9 contra/s 9 :i* !re is no ,h>mony4 ,! ,h>mony ( a repeat$ contra/1 or contra/s1 is a v important =m ( ,tone-,h>mony4 ,x means t a c]ta9 effect ( light due to ! juxtaposition or association ( c]ta9 tones recurs repeat$ly1 let u say at equal or approximately equal int]vals1 all ov] a c]ta9 space4 ,! result is sure to be h>moni\s1 no matt] h[ /rong ! tone-contra/s are 9 ! gr\p or s]ies1 provid$ t ! repetitions are well )9 ! range ( vision s t !y may be comp>$1 & ! recurr5ce ( ! same effect ( light appreciat$4 ,we mu/ n be too ne> to ! arrangem5t1 = 9 t case ! contra/s ( ! repeat$ gr\p w be m noticeable ?an ! ev5 tonality ( ! all-ov] repetition4 ,e ev51 all-ov] effect ( light1 no matt] :at ! contra/s are :i* produce x1 gives u ! feel+ ( ,h>mony4 #afi4 ,9 su* compositions z ! "o indicat$ 9 ,diagram #ag pr$om9ance may be giv5 to "o tone by hav+ x recur 9 l>g] spots 9 ea* gr\p or 9 a great] numb] ( spots1 two or m 9 ea* gr\p4 ,9 ?is way1 9 a composition ( _m colors 9 diff]5t values1 pr$om9ance may be giv5 to ,middle ,blue or ,light ,orange or ,d>k ,blue-,violet1 or any o!r p>ticul> tone4 ,pr$om9ance may be giv5 to neutral gray ( a c]ta9 value1 by hav+ x recur 9 l>g] spots or 9 num]\s small spots4 #agj4 ,neutral gray may be made to pr$om9ate 9 ano!r way2 by s compos+ ! tones1 9 ! gr\p to be repeat$1 t !y neutralize "o ano!r at a c]ta9 di/ance1--! po9t ( view ( ! obs]v]4 ------------------ ,y ,g ,y ,g ,y ,g ------------------ ,r ,v ,r ,v ,r ,v ------------------ ,y ,g ,y ,g ,y ,g ------------------ ,r ,v ,r ,v ,r ,v ------------------ ,diagram #ah ,9 ?is case ,yell[ & ,violet w neutralize "o ano!r & ,r$ w neutralize ,gre54 ,! effect ( ! repetition ( ^! complem5t>y oppositions ought to be1 at a c]ta9 di/ance1 a v lively neutral4 ,x has be5 ! idea ( c]ta9 pa9t]s ( our "t to subject e tone-impression to analysis1 & to produce ! effect ( ! tone by an arrangem5t or composition ( its elem5ts4 ,_m int]e/+ & "s beautiful results h be5 produc$ 9 ?is way4 ,illu/ration3 ,diagram #ai ,9 ?is case we h a repetition ( ! triad ,r$-,yell[-,blue1 :i*1 at a c]ta9 di/ance1 ought to produce ! effect ( a middle neutral4 ,! pr9ciple ( ^! arrangem5ts is "o ( ! mo/ important 9 tone-composition4 #aga4 ,!re is ano!r consid]ation :i* ought to keep u f any morbid int]e/ 9 h>moni\s monotonies1 :i* ought to reconcile u to contra/s1 ev5 /rong contra/s1 & to a great v>iety 9 tones4 ,h>mony is only "o pr9ciple ( composition 9 ,design2 we h two o!rs :i* are equally important1--! pr9ciple ( ,balance & ! pr9ciple ( ,rhy?m4 ,! pr9ciples ( ,balance & ,rhy?m are consi/5t ) ! greate/ possible contra/s ( tone4 ,! tone-contra/s 9 =ms ( ,balance & ,rhy?m may be /rong1 ev5 h>%1 & ! appreciation & enjoym5t ( ! ,balance or ( ! ,rhy?m 9 no degree dim9i%$4 ,we w n[ proce$ to ! consid]ation ( ,tone-,balance & ,tone-,rhy?m4 ,,tone-,,balance #agb4 ,tones1 simply z tones1 disreg>d+ ! positions1 measures1 & %apes :i* may be giv5 to !m1 balance1 :5 ! contra/s :i* !y make ) ! gr\nd-tone ^u :i* !y are plac$ are equal4 ,we h an indication ( su* a balance ( tones1 simply z tones1 9 ! foll[+ =mula3-- ,,ld - ,v -------------------- ,h,lt - ,y ,h,lt - ,y ,two spots ( ,high ,light ,yell[ occur on a gr\nd-tone ( ,l[ ,d>k ,violet4 ,! two spots ( ,yell[ make equal contra/s ) ! gr\nd-tone1 & = t reason balance z tones1 no matt] :at positions1 measures1 & %apes are giv5 to !m4 ,! value-contra/ is t ( ! int]val ( ! sev5? 9 ! ,scale ( ,values2 ! color-contra/ is t ( ! int]val ( ! sev5? 9 ! ,scale ( ,colors4 ,we mu/ assume t ! int5sities are s adju/$ z n to di/urb ! balance4 ,m - ,v ----------------- ,lt - ,o ,d - ,o ,9 ?is case ! values mak+ ! contra/s diff]4 ,! contra/s are1 nev]!less1 equal because ! value-int]vals are equival5t int]vals4 ,! value diff]5ce betwe5 ,light & ,middle is equival5t to ! value diff]5ce betwe5 ,d>k & ,middle4 ,?\gh ! contra/+ elem5ts diff]1 ! contra/s are equal4 ,9 ?is case ! contra/+ colors are ! same & ! color-contra/s correspond4 ,we mu/ assume t ! int5sities are s adju/$ z n to di/urb ! balance4 ,,ld - ,v -------------------- ,l,lt - ,o ,l,lt - ,g ,9 ?is case ! contra/+ colors diff]1 b ! contra/s are equal because ! color-int]val betwe5 ,orange & ,violet is ! same z ! color-int]val betwe5 ,gre5 & ,violet4 ,9 ?is case ! value-contra/s correspond4 ,we mu/ assume "h1 z be=e1 t !re is no diff]5ce ( color-int5sity to di/urb ! balance4 ,d - ,r ------------------ ,,hd - ,o ,,ld - ,v ,9 ?is case ! two tones :i* balance on ! gr\nd-tone diff] bo? 9 value & 9 color4 ,!y balance1 nev]!less1 because bo? ! value & ! color-contra/s are ( ! int]val ( ! ?ird4 ,aga9 we mu/ assume t !re is no disp>ity ( int5sities to di/urb ! balance4 #agc4 ,! read] w f9d ! ,diagram ( ,values & ,colors 7,no4 #e7 v useful 9 mak+ calculations = tone-balances1 s f> z value-contra/s & color-contra/s are conc]n$1 leav+ out consid]ations ( color-int5sity4 ,tak+ any tone indicat$ on ! ,diagram z a gr\nd-tone1 any tones at equal di/ances 9 balanc+ directions w balance on t gr\nd-tone4 ,illu/ration3 ,diagram #bj ,! v>i\s types ( tone-balance are %[n 9 ! above diagram4 ,! tones :i* balance1 "o aga9/ ! o!r1 on ! gr\nd-tone ( ,blue-,violet1 are ! tones m>k$ by ! same numb]4 ,! value & color-balances be+ a*iev$1 ! int5sities may be adju/$1 increas$ or dim9i%$1 until ! balance is p]fect4 #agd4 ,z y increase ! color-int5sity 9 any tone x attracts m att5tion1 & unless y increase ! int5sity 9 ! opposite tones !re w be a disp>ity :i* w di/urb y\r balance4 ,:5 ! int5sity 9 any tone is too great1 y c increase ! color-contra/ or ! value-contra/ ( ! opposite tones until ! balance is a*iev$4 #age4 ,up to ?is po9t ,i h be5 speak+ ( ,tone-,balance 9 ! ab/ract1 ( ,tone-,balance z su*4 ,i h spok5 ( ,tone-,balance z some?+ ap>t f ,position1 ,measure1 & ,%ape-,balance1 z if tones c\ld balance )\t hav+ any positions1 measures1 or %apes assign$ to !m4 ,! fact is t a tone does n exi/ until y give x a position1 a measure1 & a %ape4 ,x foll[s t ,tone-,balance is1 9 all cases1 m or less complicat$ by consid]ations ( position1 measure1 & %ape4 #agf4 ,! pr9ciple ( balance be+ t equal attractions balance at equal di/ances & unequal attractions at di/ances inv]sely proportional to !m1 x foll[s1 t if ! attraction ( a tone is increas$ by quantity1 ! attraction ( quantity may be balanc$ aga9/ ! attraction ( contra/4 ,! calculation ( su* balances may be made on ! ,diagram ( ,values & ,colors4 ,illu/ration3 ,diagram #ba ,9 ?is case1 = example1 we h ! indication ( a possible balance ( two p>ts ( ,light ,r$ & "o "p ( ,d>k ,gre5 on a gr\nd-tone ( ,middle ,violet1 ! diff]5ce ( contra/ 9 "o case mak+ up = a diff]5ce ( quantity & ( contra/+ edge 9 ! o!r4 #agg4 ,s f> z ,tone-,balance dep5ds ^u positions1 measures1 & %apes1 ! problem is ! problem ( ,position1 ,measure1 & ,%ape-,balance1 :i* we h already consid]$4 ,giv5 c]ta9 tones 9 c]ta9 measures & %apes1 ! inv]sion ( ! measures & %apes involves an inv]sion ( ! tones1 s we h a ,tone-,balance z well z a ,measure & ,%ape-,balance4 ,! inv]sion 9 any case may be s+le or d\ble4 ,illu/ration3 ,diagram #bb ,9 ?is case we h an in/ance ( s+le inv]sion1 :i* gives u a ,symmetrical ,balance1 ( tones1 z well z ( measures & %apes4 ,illu/ration3 ,diagram #bc ,9 ?is case we h an in/ance ( d\ble inv]sion ( tones1 z well z ( measures & %apes4 #agh4 ,! tones & tone-contra/s on "o side ( a c5t] or axis are n necess>ily ! same z ^? on ! o!r side4 ,we may h a ,tone-,balance 9 :i* v diff]5t tones & tone-contra/s are oppos$ to "o ano!r4 ,?is br+s u to ! consid]ation ( ,occult ,balance 9 ,tones1 ,measures1 & ,%apes4 ,a balance ( any tones & ( any tone-contra/s1 9 any measures & 9 any %apes1 is obta9$ :5 ! c5t] ( tone-attractions is unmi/akably indicat$1 ei!r by ! symmetrical "* ( ! balance or by a symmetrical inclosure :i* w indicate ! c5t]4 ,giv5 any comb9ation ( tones1 measures1 & %apes1 & ! problem to f9d ! balance-c5t]1 h[ %all we solve ! problem8 ,x _c v well be done by reason+4 ,x mu/ be done by visual feel+4 ,! pr9ciple ( ,balance be+ cle>ly und]/ood1 f9d+ ! c5t] ( any tone-contra/s is a matt] ( exp]im5tal practice 9 :i* ^? p]sons succe$ be/ :o are mo/ s5sitive to diff]5ces ( tone1 & :o make ! greate/ ef=t to feel ! c5t]s & to indicate !m accurately4 ,exp]i5ce & practice are necess>y 9 all cases4 ,illu/ration3 ,fig4 #bbh ,"h1 )9 ?is circle1 are ! attractions to be balanc$4 ,! problem is to f9d ! balance-c5t]1 & to indicate t c5t] by a symmetrical inclosure :i* w br+ ! tones1 measures1 & %apes into a ,balance4 ,! c5t] is "h indicat$ by ! circle4 ,:e!r x is correctly indicat$ is a matt] ( judgm5t 9 :i* !re may be a diff]5ce ( op9ion4 ,!re is a c5t] some:]e ^u :i* ! attractions are balanc+4 ,! "q is1 :]e is x8 ,! illu/ration :i* ,i h giv5 is 9 ! t]ms ( ! ,scale ( ,neutral ,values4 ,diff]5ces ( color & color-int5sity w\ld complicate ! problem1 b w\ld n 9 any way affect ! pr9ciple involv$4 ,i "k ( no m int]e/+ problem or ex]cise ?an ?is3 to a*ieve ,tone-,balance :]e !re is no ,tone-,symmetry4 #agi4 ,x w sometimes happ51 t a gradation ( tones or measures w draw ! eye 9 a c]ta9 direction1 t[>d ! great] contra/1 :ile a l>g] mass or measure ( tone1 on ! o!r side1 w be hold+ x back4 ,9 su* a case we may h a mass balanc+ a motion4 ,illu/ration3 ,fig4 #bbi ,9 ?is case ! eye is drawn along1 by a gradation ( values1 to ! "r1 t[>d ! edge ( great] contra/1 away f a l>ge d>k mass ( tone 9 :i* !re is no movem5t4 ,! t5d5cy ( ! d>k mass is to hold ! eye at its c5t]4 ,! problem is to f9d ! balance-c5t] betwe5 ! motion & ! mass4 ,i h done ?is1 & ! balance-c5t] is indicat$ by ! symmetrical outl9e ( ! diagram4 #ahj4 ,"s %apes hold ! eye ) peculi> =ce1 & 9 su* cases ! attractions ( tone or measure or %ape on ! o!r side h to be increas$ if we are to h a balance4 ,symmetrical %apes h a t5d5cy to hold ! eye at c5t]s & on axes4 ,giv5 c]ta9 attractions on ! o!r side1 we mu/ be sure t !y are suffici5t to balance ! =ce ( ! symmetry 9 addition to ! =ce ( its tone-contra/s1 :atev] !y are4 ,illu/ration3 ,fig4 #bcj ,9 ?is case we h an approximate balance 9 :i* ! =ce ( a symmetry1 ) contra/+ edges1 on "o side1 is balanc$ by contra/s & c]ta9 movem5ts on ! o!r4 ,if ,i %\ld turn d[n ! upp] spot on ! "r1 we w\ld feel a loss ( balance due to ! turn+ ( two movem5ts1 :i* comb9e to make "o movem5t to ! "r1 into two movem5ts d[n to ! "r4 ,if ,i %\ld increase ! =ce ( ! symmetry1 by fill+ 9 ! c5t] ) black1 x w\ld be necess>y ei!r to move ! symmetry ne>] to ! c5t] or to move ! opposite attractions away f x4 ,an un/able attitude 9 ! symmetry w\ld h to be c\nt]act$1 9 "s way1 on ! o!r side4 ,intricate %apes f :i* ! eye _c easily or quickly escape (t5 hold ! eye ) a =ce :i* mu/ be add$ to t ( _! tone-contra/s4 ,illu/ration3 ,fig4 #bca ,9 ?is case ! %ape on ! "r requires a pretty /rong d>k spot to balance its contra/s & its intricacy4 ,! problem is fur!r complicat$ :5 !re are1 also1 incl9ations1 to ! "r or to ! left1 to be balanc$4 ,illu/ration3 ,fig4 #bcb ,9 ?is case ,i h tri$ to balance1 on ! c5t] ( a symmetrical inclosure1 v>i\s ext5sions & incl9ations ( tone-contra/1 ! movem5t ( a conv]g5ce1 & ! =ce ( a some:at intricate & un/able symmetry4 ,^! occult =ms ( ,balance are n yet well und]/ood1 & ,i feel consid]able hesitation 9 speak+ ( !m4 ,we h c]ta9ly a great deal to le>n ab\t !m4 ,!y are f> bett] und]/ood by ! ,*9ese & by ! ,japanese ?an by u4 #aha4 ,:5 any l9e or spot has a mean+1 :5 !re is any symbolism or repres5tation 9 x1 x may ga9 an indef9ite =ce ( attraction4 ,?is1 h[ev]1 is a =ce ( attraction = ! m9d r ?an = ! eye4 ,x affects diff]5t p]sons 9 diff]5t measures4 ,! consid]ation ( su* attractions1 sugge/ions1 mean+s1 or significations does n belong to ,pure ,design b to ,symbolism or to ,repres5tation4 ,,tone-,,rhy?m #ahb4 ,! idea ( ,tone-,rhy?m is express$ 9 e regul> & p]fect gradation ( ,tones2 ( values1 ( colors or ( color-int5sities1 provid$ ! eye is drawn ?r\gh ! gradation 9 "o direction or 9 a s]ies or sequ5ce ( directions4 ,?is happ5s :5 !re is a great] tone-contra/ at "o end ( ! gradation ?an at ! o!r4 ,:5 ! t]m9al contra/s are equal !re is no reason :y ! eye %\ld move ?r\gh ! gradation 9 any p>ticul> direction4 ,accord+ to our def9ition ( ,rhy?m1 ! gradation %\ld be m>k$ 9 its /ages or measures1 & ! /ages or measures %\ld be regul>4 ,t is c]ta9ly true1 b 9 all regul> & p]fect gradations ,i feel t correspond+ *anges are tak+ place 9 correspond+ measures1 & ,i get ! same feel+ f su* a gradation t ,i get f x :5 x is m>k$ (f 9 equal sections4 ,?\gh ! measures 9 regul> & p]fect gradations are n m>k$1 !y are1 x seems to me1 felt4 ,!y seem suffici5tly m>k$ by ! regul>ity & p]fection ( ! gradation1 any irregul>ity or imp]fection be+ appreciable z a break 9 ! measure4 ,i am incl9$1 !re=e1 to say ( any regul> & p]fect gradation t x is rhy?mical provid$ ! direction ( movem5t is unmi/akable4 ,! direction1 z ,i h said1 dep5ds ^u ! relation ( t]m9al contra/s4 ,! eye is drawn t[>d ! great] contra/1 :atev] t is & :]ev] x is4 ,a few examples w make ?is cle>4 ,m ----------------------------------- ,blk ,,ld ,d ,,hd ,m ,l,lt ,lt ,h,lt ,wt ,9 ?is case we h ! gradation ( ! ,scale ( ,values set on a gr\nd-tone ( ! middle value4 ,"h !re are two oppos$ gradations ) equal contra/s at ! opposite ends4 ,! result is ,balance1 n ,rhy?m4 ,wt ----------------------------------- ,blk ,,ld ,d ,,hd ,m ,l,lt ,lt ,h,lt ,wt ,9 ?is case we h a gradation ( values beg9n+ ) ,:ite on ,:ite1 no contra/ at all1 & rea*+ ultimately ! contra/ ( ,black & ,:ite4 ,! eye is drawn ?r\gh ! tones ( ?is gradation 9 ! direction ( ?is contra/1 t is to say1 f "r to left4 ,x is a cle> case ( ,rhy?m4 ,if1 in/ead ( :ite1 we _h black1 z a gr\nd-tone1 ! movem5t ( ! rhy?m w\ld be 9 ! opposite direction1--f left to "r4 ,wt ,h,lt ,lt ,l,lt ,m ,,hd ,d ,,ld ,blk ------------------------------------- ,blk ,,ld ,d ,,hd ,m ,l,lt ,lt ,h,lt ,wt ,9 ?is case1 z 9 ! fir/1 we h equally great contra/s at ! ends & no contra/ at ! middle4 ,! result is ,balance1 n ,rhy?m4 ,v ------------------- ,y ,,yg ,g ,,gb ,b ,,bv ,9 ?is case1 disreg>d+ possible diff]5ces ( value & color-int5sities1 !re w be a color-rhy?m proce$+ f "r to left4 ,! contra/ to :i* ! eye w be drawn is ! color-contra/ ( ,yell[ & ,violet4 ,,ld-,v ------------------------------------- ,d-,y ,,hd-,,yg ,m-,g ,l,lt-,,gb ,lt-,b ,h,lt-,,bv ,9 ?is case1 disreg>d+ possible diff]5ces ( int5sity1 !re w be a rhy?m ( color mov+ f "r to left & a rhy?m ( values mov+ f left to "r4 ,assum+ t we are equally attract$ by correspond+ value & color-contra/s1 ^! two rhy?ms1 :5 produc$1 w neutralize "o ano!r & we %all h an illu/ration ( ,tone-,balance r ?an ,tone-,rhy?m4 ,if correspond+ color & value-contra/s are n equally attractive we %all h an unequal tug-(-w> betwe5 ! two rhy?ms4 ,,ld-,v ------------------------------------- ,h,lt-,y ,lt-,,yg ,l,lt-,g ,m-,,gb ,,hd-,b ,d-,,bv ,9 ?is case we h two rhy?ms1 "o ( values & "o ( colors1 9 a ,h>mony ( ,direction4 ,! direction ( movem5t w be f "r to left4 ,h,lt-,y ---------------------------------------- --------------- ,h,lt-,y ,h,lt-,y ,h,lt-,y_/ ,h,lt-,y ,h,lt-,y_/ ,h,lt-,y ,9 ?is case we h no *ange ( color & no *ange ( value1 b a rhy?m ( ! int5sities ( "o color1 9 "o value4 ,! movem5t w be f "r to left4 ,! gr\nd-tone might be ,neutral ,high ,light1 ! z]o-int5sity ( ,yell[4 ,t w\ld n *ange ! direction ( ! movem5t4 ,,ld-,y ---------------------------------------- --------- ,h,lt-,y ,lt-,y ,l,lt-,y_/ ,m-,y ,,hd-,y_/ ,d-,y ,9 ?is case ,i h indicat$ a comb9$ movem5t ( values & color-int5sities4 ,! direction ( ! movem5t w be f "r to left4 ,! tone-rhy?ms :i* ,i h describ$ are bas$ ^u ! repetition at regul> int]vals ( a c]ta9 *ange ( value1 ( color or ( color-int5sity4 ,we h ,h>mony1 ( c\rse1 9 ! repetition ( equal *anges1 ?\gh ! *anges are n ! same *anges4 ,! *ange ( value f ,middle to ,l[ ,light is equal to ! *ange f ,l[ ,light to ,light1 ?\gh ^! *anges are n ! same *anges4 ,! ,h>mony is1 !re=e1 ! ,h>mony ( equival5t contra/s :i* are n ! same contra/s4 #ahc4 ,we h m or less movem5t 9 e composition ( tones :i* is unbalanc$1 9 :i* ! eye is n held betwe5 equival5t attractions1 ei!r ^u a v]tical axis or ^u a c5t]4 ,9 all su* cases1 ( tones unbalanc$1 ! movem5t is 9 ! direction ( ! greate/ contra/4 ,unless ! movem5t is regul> & m>k$ 9 its measures1 z ,i ?9k x is 9 all regul> & p]fect gradations1 ! movem5t is n rhy?mical4 ,we get ,rhy?m1 h[ev]1 9 ! repetition ( ! movem5t1 :atev] x is1 9 equal or lawfully v>y+ measures1 provid$ ! direction ( ! movem5t rema9s ! same or *anges regul>ly or gradually4 ,if ! l9e ( ! movem5t is up-to-!-"r =ty-five degrees we h rhy?m 9 ! repetition ( ! movem5t at equal or lawfully v>y+ int]vals1 )\t *anges ( direction2 b we %\ld h ,rhy?m1 also1 if ! direction ( ! movem5t1 9 its repetitions1 7 *ang$1 regul>ly or gradually2 if1 = example1 ! direction 7 *ang$ fir/ f up-"r =ty-five degrees to up-"r =ty degrees1 !n to up-"r ?irty-five degrees1 !n to up-"r ?irty degrees1 ?is at equal or at lawfully v>y+ int]vals4 ,9 ?is way ! movem5t ( ! composition repeat$ may be c>ri$ on & gradually develop$ 9 ! movem5t ( ! s]ies4 ,a ref]5ce to ,fig4 #afa1 p4 #id1 & to ,fig4 #aai1 p4 #fh1 w help ! read] to und]/& ^! /atem5ts4 #ahd4 ,:5 any unbalanc$ composition ( tones is s+ly inv]t$ ^u a v]tical axis & ! movem5t ( ! composition foll[s ! axis1 ei!r up or d[n1 & ?is movem5t1 up or d[n1 is repeat$1 up or d[n1 we get =ms ( ,tone-,rhy?m :i* are also =ms ( ,symmetrical ,balance4 ,9 ! inv]sions & repetitions ( ! tone-composition we h ,tone-,h>mony4 ,z ! tones 9 ! repeat$ composition h c]ta9 positions1 measures1 & %apes1 ! ,h>mony1 ! ,balance1 & ! ,rhy?m are ( ,positions1 ,measures1 & ,%apes z well z ( ,tones2 s we get ! comb9ation ( all ! t]ms ( ,design 9 all ! ?ree modes ( ,design4 ,,composition ,,?ree ,,g5]al ,,rules #ahe4 ,x is q impossible = me1 9 ?is discussion ( t]ms & pr9ciples1 to indicate1 9 any measure1 ! possibilities ( composition1 9 l9es & spots ( pa9t1 9 tones1 measures1 & %apes4 ,?is is 9 no s5se a ,book ( ,designs4 ,all ,i h und]tak5 to d is to give a few v simple examples & to indicate ! k9d ( reason+ to be foll[$1 recomm5d+ ! same k9d ( reason+ 9 all cases4 ,!re are ?ree g5]al rules1 h[ev]1 :i* ,i mu/ /ate4 ,fir/4 ,giv5 a c]ta9 outl9e & c]ta9 tones1 measures1 & %apes to be put into x1 x is ! ,problem ( ,pure ,design to d ! be/ we c1 gett+ z _m connections mak+ unity z possible4 ,! process is "o ( exp]im5t+1 obs]v+1 comp>+1 judg+1 arrang+ & re>rang+1 tak+ no end ( "t & pa9s to a*ieve ,ord]1 ! utmo/ possible ,ord]1 if possible ! ,beautiful4 ,second4 ,:5 only an outl9e is giv5 & we c put into x l9es & spots ( pa9t1--tones1 measures1 & %apes1--ad libitum1 we mu/ be sure t 9 ! addition & multiplication ( features we d n get less ,ord] ?an we _h 9 ! simple outl9e ) :i* we />t$1 :5 x _h no?+ 9 x4 ,z we proce$ to add features we mu/ be sure t we are n dim9i%+ ! ord] ( ! composition z a :ole4 ,if ! composition z a :ole is ord]ly1 we d n want to make x less s by cutt+ x up & introduc+ additional attractions :i* may be disord]ly & confus+4 ,x may be h>d] to a*ieve ,ord] ) a great] numb] & v>iety ( t]ms4 ,we may des]ve cr$it = ov]com+ ?is difficulty1 b x is a difficulty :i* confronts u only :5 ! t]ms are giv5 & we h to make ! be/ ( !m4 ,:5 no t]ms are giv51 only a p]fectly ord]ly outl9e1 we %\ld hesitate be=e we put any?+ into x4 ,if we add any?+ we mu/ be sure t x does n dim9i%1 9 ! slighte/ degree1 ! ord] we _h be=e1 :5 we _h no?+ b ! outl9e4 ,! ord] ( ! :ole mu/ nev] be dim9i%$4 ,?ird4 ,:5 we h an outl9e ) c]ta9 tones1 measures1 & %apes 9 x1 ! "q is3 :e!r we c increase ! ord] by add+ o!r tones1 o!r measures1 or o!r %apes4 ,illu/ration3 ,fig4 #bcc ,arrangem5t 8a0 is less ord]ly ?an arrangem5t 8b10 s ,i h act$ wisely 9 add+ ! o!r outl9es4 ,illu/ration3 ,fig4 #bcd ,9 ?is case1 h[ev]1 ,i h add$ features )\t a*iev+ any increase ( ,ord] 9 ! composition4 ,! ord] is less ?an x 0 be=e4 ,! additions h no int]e/ f ! po9t ( view ( ,pure ,design4 ,i may add features = ! sake ( v>iety or novelty1 to give a *ange ( feel+1 a new s5sation1 b su* motives are n ! motives ( ,pure ,design4 ,9 ,pure ,design our motive is1 always1 to a*ieve ,ord]1 9 ! hope t 9 s do+ we may a*ieve a supreme in/ance ( x :i* w be beautiful4 ,illu/ration3 ,fig4 #bce ,consid] ^! illu/rations4 ,arrangem5t 8b0 is m ord]ly ?an arrangem5t 8a10 s ,i am ju/ifi$ 9 mak+ ! additions4 ,! additions h br\ght occult balance into ! composition ) ,direction & ,int]val-,h>mony4 ,arrangem5t 8c0 is less ord]ly ?an 8b10 less ord]ly ?an 8a40 ,x has1 !re=e1 no value = u4 ,!re is no m]it 9 ! multiplication ( features :i* x exhibits4 ,! surface is 8enri*$0 at ! exp5se ( ,direction-,h>mony1 ,int]val-,h>mony1 & ,%ape-,h>mony4 ,!re may be an approximation to an occult balance 9 arrangem5t 8c10 b y _c feel x unmi/akably z y d 9 8b40 ,its value is1 !re=e1 less4 #ahf4 ,i object to ! ^w 8decoration10 z commonly us$ by design]s1 because x implies t additions are likely to be improvem5ts1 t to multiply features1 to enri* surfaces1 is wor? :ile or desirable4 ,! fact is1 t additions are1 z a rule1 to be avoid$4 ,!re is no m]it 9 ! m]e multiplication ( features4 ,x is a mi/ake4 ,! rule ( ,pure ,design1 & x is ! rule = all ,design1 is simplification r ?an complication4 ,z design]s we ought to avoid additions1 if possible4 ,we ought to make !m only :5 9 s do+ we are able to increase ! ord] ( ! :ole4 ,we make additions1 inde$1 to a*ieve ! great] simplicity ( ,ord]1 & = no o!r reason4 ,our object 9 all cases is to a*ieve ,ord]1 if possible a supreme in/ance ( ,ord] :i* w be beautiful4 ,we aim at ,ord] & hope = ,beauty4 ,,! ,,/udy ,,( ,,ord] ,,9 ,,nature ,,& ,,9 ,,works ,,( ,,art #ahg4 ,9 connection ) ! practice ( ,pure ,design1 z ,i h describ$ x1--! composition & arrangem5t ( l9es & spots ( pa9t2 ( tones1 measures1 & %apes3 ?is 9 ! modes ( ,h>mony1 ,balance1 & ,rhy?m1 = ! sake ( ,ord] & 9 ! hope ( ,beauty1--! /ud5t %\ld take up ! /udy ( ,ord] 9 its ?ree modes1 z reveal$ 9 ,nature & a*iev$ 9 ,works ( ,art4 #ahh4 ,! me?od ( /udy %\ld be a comb9ation ( analysis ) syn!tic reproduction4 ,tak+ any in/ance ( ,ord]1 :e!r 9 ,nature or 9 "s "w ( ,art1 ! fir/ ?+ to d is to consid] its t]ms1--its positions1 its l9es1 its areas1 its measure & space-relations1 its tones & tone-relations1--br++ e elem5t to sep>ate & exact def9ition4 ,! next ?+ to d is to note e occurr5ce ( ,h>mony1 ( ,balance1 ( ,rhy?m1--e connection mak+ = consi/5cy1 unity1 ,ord]4 ,9 t way we %all get an exact k ( ! case4 ,we %all "k all ! facts1 s f> z ! t]ms & ! pr9ciples ( ,design are conc]n$4 ,t is :at ,i mean by analysis4 ,by a syn!tic reproduction ,i mean a reproduction ( ! effect or design1 :atev] x is1 foll[+ ! images :i* we h 9 m9d z ! result ( our analysis4 ,! reproduction %\ld be made )\t ref]5ce to ! effect or design :i* has be5 analyz$4 ,!re %\ld be no direct imitation1 no copy+4 ,we mu/ n dep5d s mu* ^u ! memory z ^u ! imag9ation4 ,hav+ reproduc$ ! effect or design 9 ?is way1 foll[+ ! sugge/ions ( ! imag9ation1 ! reproduction %\ld be br\ght into comp>ison ) ! effect or design reproduc$ & ! diff]5ces not$4 ,diff]5ces %\ld be c>efully obs]v$ & ! previ\s analysis %\ld be review$ & reconsid]$4 ,:5 ?is is done ano!r attempt at reproduction %\ld be made4 ,?is process %\ld be repeat$ until ! effect or design is ?or\ghly und]/ood & imag9atively grasp$4 ,! evid5ce ( und]/&+ & compreh5sion w be se5 9 ! reproduction :i* is made1 :i* ought to h an ess5tial b n a lit]al correspond5ce ) ! orig9al4 ,analysis %\ld prec$e2 syn!sis %\ld foll[4 ,i hope1 9 ano!r book or books1 to be publi%$ lat]1 to give "s examples ( ,ord] 9 natural objects or effects1 also examples ( ,ord] 9 ,works ( ,art1 ) a c>eful analysis ( ea* "o1 %[+ h[ ! po9ts1 l9es1 & areas1 ! measure & ! space-relations1 ! tones & tone-relations come toge!r 9 ! =ms ( ,h>mony1 ,balance1 & ,rhy?m1 9 ! modes ( ,ord]1 9 in/ances ( ,beauty4 ,9 ! mean "t1 z ! me?ods ( analytic /udy & ( syn!tic practice are cle>ly indicat$ 9 ! prec$+ pages1 ! /ud5t :o has tak5 pa9s to und]/& :at he has read w f9d himself well prep>$ = ! "w4 ,he c take up ! /udy ( ,ord] 9 ,nature & ( ,design 9 ,works ( ,art )\t fur!r assi/ance4 ,,conclusion #ahi4 ,x does n foll[1 ev5 :5 our m9ds1 9 consequ5ce ( ! /udy & ! practice :i* ,i h describ$1 are ri*ly /or$ ) ! t]ms & ! motives ( ,design1 t we %all produce any?+ important or rem>kable4 ,important "w comes only f important p4 ,:at we accompli%1 at be/1 is m]ely ! measure & expression ( our own p]sonalities4 ,nev]!less1 ?\gh we may n be able to produce any?+ important1 x is some?+ to appreciate & enjoy :at is a*iev$ by o!rs4 ,if our /udies & our "w br+ u to ! po9t ( visual discrim9ation1 to /hetic appreciation & enjoym5t1 & no f>!r1 we are di/+ui%$ among m54 ,! r>e/ ?+ 9 ! _w is creative g5ius1 ! faculty :i* creates great works4 ,next to t comes ! faculty ( appreciation4 ,t1 too1 is r>e4 ,we mu/ n believe t appreciation is easy4 ,x is true t ! recognition ( ,ord] is in/9ctive & spontane\s1 b untra9$ p recognize x only 9 a few simple & obvi\s =ms4 ,ord] 9 its high] =ms--! ord] ( a great numb] & v>iety ( t]ms & ( diff]5t pr9ciples 9 comb9ation--lies altoge!r beyond ! appreciation ( untra9$ p4 ,x is only z we are tra9$1 ex]cis$1 & practic$ 9 ! use ( t]ms & 9 foll[+ pr9ciples t we rise to ! appreciation ( great a*ievem5ts4 ,! s5se ( ord]1 :i* we all h1 9 a measure1 ne$s to be ex]cis$ & develop$4 ,! spontaneity ( undevelop$ faculty does n c\nt = mu*4 ,x c>ries u only a little way4 ,let no "o believe t )\t /udy & practice 9 ,design he c recognize & appreciate :at is be/ 9 ,design4 ,appreciation & enjoym5t are ! rew>ds ( h>d ?9k+ ) h>d "w4 ,9 ord] to appreciate ! ma/]piece we mu/ h "s k ( ! t]ms :i* ! arti/ has us$ & ! pr9ciples :i* he has foll[$4 ,we "k ! t]ms only :5 we h ourselves us$ !m1 & ! pr9ciples :5 we h tri$ to foll[ !m4 ,! reason :y ! appreciation ( excell5ce 9 spee* & 9 writ+ is s widespread is due to ! fact t we all speak & write1 con/antly1 & try1 s _m ( u1 to speak & write well4 ,! reason :y !re is s little appreciation ( excell5ce 9 o!r =ms ( art is due to ! fact t ! t]ms are n 9 g5]al use & ! pr9ciples are n und]/ood1 z !y %\ld be1 9 ! light ( p]sonal exp]i5ce & ef=t4 ,x is = ?is reason t ,i am anxi\s to see ! tea*+ & practice ( ,design introduc$ into ! s*ools1 public & private1 ev]y:]e1 & into our colleges z well z our s*ools4 ,i h no idea t _m able design]s w be produc$1 b :at ,i expect1 z a result ( ?is tea*+1 is a m g5]al und]/&+ ( ,design1 m int]e/ 9 x1 & m appreciation & enjoym5t ( its a*ievem5ts4 ,among ! _m :o w appreciate & enjoy w be f\nd ! few :o w create & produce4 ,! purpose ( :at is call$ art-tea*+ %\ld be ! production1 n ( objects1 b ( faculties1--! faculties :i* be+ ex]cis$ w produce objects ( ,art1 naturally1 inevitably4 ,in/ead ( try+ to tea* p to produce ,art1 :i* is absurd & impossible1 we mu/ give !m a tra9+ :i* w induce visual s5sitiv5ess ) /hetic discrim9ation1 an int]e/ 9 ! tones1 measures1 & %apes ( ?+s1 ! p]ception & appreciation ( ,ord]1 ! s5se ( ,beauty4 ,9 ^! faculties we h ! causes ( ,art4 ,induc+ ! causes1 ,art w foll[ z a matt] ( c\rse4 ,9 ex]cis+ & develop+ ! faculties :i* ,i h nam$1 :i* naturally & inevitably produce ,art1 we are do+ all t c be done by tea*+4 ,!re is no bett] tra9+ = ! visual & /hetic faculties ?an is f\nd 9 ! practice ( ,pure ,design1 induc+1 z x does1 discrim9ation 9 tones1 measures1 & %apes1 & ! appreciation ( :at is ord]ly & beautiful4 ,! result ( ! practice w be a wide spread ( visual & /hetic faculty :i* w h1 z its natural & inevitable result1 ! appreciation & ! production ( ,works ( ,art4 ,our object1 !n1 9 ! /udy & practice ( ,pure ,design is1 n s mu* ! production ( ,works ( ,art1 z x is to induce 9 ourselves ! art-lov+ & art-produc+ faculties4 ,) ^! faculties we %all be able to discov] ,ord] & ,beauty ev]y:]e1 & life w be happi] & bett] wor? liv+1 :e!r we produce ,works ( ,art1 ourselves1 or n4 ,we %all h an impulse :i* w lead u to produce ,works ( ,art if we c4 ,at ! same "t we %all h ! judgm5t :i* w tell u :e!r :at we h done is or is n beautiful4 ,,p>agraph ,,index #a1 p4 #a4 ,! ,mean+ ( ,design4 #b1 p4 #a4 ,! ,ord] ( ,h>mony4 #c1 p4 #a4 ,! ,ord] ( ,balance4 #d1 p4 #b4 ,! ,ord] ( ,rhy?m4 #e1 p4 #b4 ,relations ( ,h>mony1 ,balance1 & ,rhy?m4 #f1 p4 #d4 ,beauty a supreme in/ance ( ,ord]4 #g1 p4 #d4 ,! ,arts z diff]5t modes ( ,expression4 #h1 p4 #e4 ,draw+ & ,pa9t+4 #i1 p4 #e4 ,two modes ( ,draw+ & ,pa9t+4 #aj1 p4 #e4 ,pure ,design4 #aa1 p4 #f4 ,appli$ ,design4 #ab1 p4 #g4 ,repres5tation4 #ac1 p4 #g4 ,repres5tation 9 ,=ms ( ,design4 #ad1 p4 #i4 ,! ,def9ition ( ,positions4 #ae1 p4 #i4 ,! ,relation ( ,directions & ,di/ances4 #af1 p4 #aj4 ,directions def9$4 #ag1 p4 #aa4 ,di/ances def9$4 #ah1 p4 #aa4 ,positions det]m9$ by ,triangulations4 #ai1 p4 #aa4 ,int]vals4 #bj1 p4 #ab4 ,scale 9 ,relations ( ,position4 #ba1 p4 #ab4 ,h>mony ( ,positions4 #bb1 p4 #ab4 ,h>mony ( ,directions4 #bc1 p4 #ac4 ,h>mony ( ,di/ances4 #bd1 p4 #ad4 ,h>mony ( ,int]vals4 #be1 p4 #af4 ,int]vals 9 any s]ies ( ,positions4 #bf1 p4 #ag4 ,positions & _! possibilities4 #bg1 p4 #ag4 ,balance ( opposite ,directions4 #bh1 p4 #ag4 ,balance ( ,di/ances 9 opposite ,directions4 #bi1 p4 #ah4 ,balance ( ,directions n opposite4 #cj1 p4 #ah4 ,balance ( ,di/ances 9 ,directions n opposite4 #ca1 p4 #ai4 ,positions 9 ,balance4 #cb1 p4 #ai4 ,/able equilibrium ( v]tical & horizontal directions4 #cc1 p4 #bj4 ,symmetry def9$4 #cd1 p4 #ba4 ,! c5tral axis %\ld pr$om9ate 9 symmetrical ,balances4 #ce1 p4 #bb4 ,balance 9 ,relations ( ,position1 :5 inv]t$4 #cf1 p4 #bb4 ,f9d+ ! c5t] ( equilibrium 9 unbalanc$ relations ( position4 ,indication ( c5t]s by symmetrical inclosures4 #cg1 p4 #bd4 ,t5d5cy ( symmetrical inclosures1 :5 suffici5tly attractive1 to prev5t movem5t4 #ch1 p4 #be4 ,h[ un/able equilibrium sugge/s movem5t4 #ci1 p4 #bf4 ,rhy?mic movem5t 9 a gradual increase 9 ! numb] ( attractions ?r\gh a s]ies ( visual angles4 #dj1 p4 #bg4 ,! possibilities ( rhy?mic movem5t 9 relations ( position4 #da1 p4 #bg4 ,balanc$ attractions at equal int]vals give no movem5t1 consequ5tly no ,rhy?m4 #db1 p4 #bh4 ,! gradual increase ( attractions 9 a s]ies ( visual angles1 z produc$ by gradual *anges ( scale1 causes rhy?mic movem5t4 #dc1 p4 #bh4 ,h[ unbalanc$ gr\ps ( positions be+ repeat$ at equal int]vals produce rhy?mic movem5t4 #dd1 p4 #bi4 ,rhy?mic movem5ts produc$ by ! repetition ( unbalanc$ relations ( position & by a gradual dim9ution ( scale4 #de1 p4 #cj4 ,rhy?mic movem5ts produc$ by ! repetition ( a balanc$ relation ( positions ) a gradual dim9ution ( int]vals1 caus+ a gradual increase ( attractions ?r\gh a s]ies ( visual angles4 #df1 p4 #cj4 ,rhy?mic movem5ts produc$ by ! repetition ( a balanc$ relation ( positions ) dim9ution ( int]vals & ( scale4 #dg1 p4 #ca4 ,rhy?mic movem5ts produc$ by ! repetition ( an unbalanc$ relation ( positions ) a cr[d+ due to gradual dim9ution ( int]vals4 #dh1 p4 #ca4 ,rhy?mic movem5ts produc$ by ! repetition ( an unbalanc$ relation ( positions ) a dim9ution ( measure 9 ! int]vals & ( scale 9 ! gr\ps4 ,! comb9ation ( two or m rhy?ms4 #di1 p4 #cb4 ,! comb9ation ( two or m rhy?ms4 #ej1 p4 #cb4 ,relations ( position 9 diff]5t attitudes4 #ea1 p4 #cc4 ,pr9cipal ,attitudes4 #eb1 p4 #cd4 ,h>mony 9 ,attitudes4 #ec1 p4 #ce4 ,h>mony 9 ! repetition ( any relation ( attitudes4 #ed1 p4 #ce4 ,balance 9 ,attitudes4 #ee1 p4 #cf4 ,rhy?m 9 ,attitudes4 #ef1 p4 #cg4 ,! ,l9e4 #eg1 p4 #cg4 ,*anges ( ,direction 9 a l9e4 ,angles4 #eh1 p4 #ch4 ,gradual *anges ( ,direction 9 a l9e4 ,curves4 #ei1 p4 #da4 ,curves reg>d$ z compositions ( circul> arcs4 #fj1 p4 #db4 ,diff]5ces ( scale 9 l9es4 #fa1 p4 #db4 ,diff]5ces ( attractive =ce 9 l9es4 #fb1 p4 #dd4 ,h>mony ( ,direction 9 l9es4 #fc1 p4 #dd4 ,h>mony ( ,angles 9 l9es4 #fd1 p4 #de4 ,h>mony 9 ,legs ( ,angles4 #fe1 p4 #de4 ,h>mony 9 ,curvatures4 #ff1 p4 #df4 ,h>mony 9 ,arcs :5 !y h ! same radius4 #fg1 p4 #df4 ,h>mony 9 ,arcs :5 !y h ! same angle4 #fh1 p4 #dg4 ,l9e> ,progressions4 #fi1 p4 #dg4 ,v>iations ( scale 9 ,l9e> ,progressions4 #gj1 p4 #dh4 ,*anges ( ,direction 9 ,l9e> ,progressions4 #ga1 p4 #di4 ,inv]sions 9 ,l9e> ,progressions4 #gb1 p4 #ej4 ,balance 9 a ,l9e4 #gc1 p4 #ea4 ,appreciation ( ,balance 9 a l9e dep5ds v mu* on its attitude4 #gd1 p4 #eb4 ,balance ( ,incl9ations 9 a l9e4 #ge1 p4 #ed4 ,f9d+ ! c5t] ( equilibrium ( a l9e & indicat+ t c5t] by a symmetrical inclosure4 #gf1 p4 #ef4 ,rhy?m 9 a ,l9e4 #gg1 p4 #ef4 ,rhy?m requires m ?an movem5t4 ,! movem5t mu/ be 9 regul> & m>k$ measures4 #gh1 p4 #eg4 ,! numb] ( repetitions requir$ 9 a ,rhy?m4 #gi1 p4 #eg4 ,contr>y movem5ts 9 ,rhy?ms4 #hj1 p4 #eh4 ,regul> alt]nations 9 space n necess>ily rhy?mical4 ,t dep5ds ^u ! "* ( ! motive4 #ha1 p4 #ei4 ,repetition & alt]nation )\t ,rhy?m4 #hb1 p4 #ei4 ,rhy?m due to gradation ( scale4 #hc1 p4 #fj4 ,rhy?m due to ! gradual increase 9 ! numb] ( attractions f measure to measure4 #hd1 p4 #fa4 ,rhy?m 9 ,spiral ,conc5trations4 #he1 p4 #fc4 ,direct & ,contr>y ,motion 9 ,spiral l9es4 #hf1 p4 #fd4 ,! ,balance ( correspond+ b oppos$ ,rhy?ms4 #hg1 p4 #fd4 ,l9es 9 diff]5t ,attitudes4 #hh1 p4 #fe4 ,h>mony 9 ,attitudes ( l9es4 #hi1 p4 #ff4 ,h>mony 9 ! repetition ( any relation ( attitudes4 #ij1 p4 #ff4 ,balance 9 ,attitudes ( ,l9es4 #ia1 p4 #fg4 ,rhy?m 9 ,attitudes ( ,l9es4 #ib1 p4 #fh4 ,recapitulation4 #ic1 p4 #fh4 ,! ,composition ( ,l9es4 #id1 p4 #fi4 ,h>mony 9 ! ,composition ( ,l9es4 #ie1 p4 #gj4 ,measure-,h>mony ( ratios & ( proportions4 #if1 p4 #gj4 ,elem5ts mak+ = ,h>mony 9 dissimil> l9es4 #ig1 p4 #gc4 ,balance 9 ! ,composition ( ,l9es4 #ih1 p4 #gd4 ,%ape-,h>mony )\t ,measure-,h>mony4 #ii1 p4 #gd4 ,measure-,balance )\t ,%ape-,balance4 #ajj1 p4 #gf4 ,! c5t]s ( equilibrium 9 m]e measure-balances %\ld be indicat$ by symmetrical inclosures4 #aja1 p4 #gf4 ,balance ( ,incl9ations4 #ajb1 p4 #gi4 ,measure-,rhy?m 9 ! ,composition ( ,l9es4 #ajc1 p4 #hj4 ,! comb9ation ( v>i\s types ( rhy?mic movem5t4 #ajd1 p4 #hf4 ,rhy?m n necess>ily inconsi/5t ) ,balance4 #aje1 p4 #hi4 ,! ,composition ( v>i\s l9es4 #ajf1 p4 #if4 ,outl9es4 #ajg1 p4 #if4 ,h>mony1 ,balance1 & ,rhy?m 9 ,outl9es4 #ajh1 p4 #ajb4 ,int]ior ,dim5sions ( an ,outl9e4 #aji1 p4 #ajb4 ,h>mony ( ,int]ior ,dim5sions4 #aaj1 p4 #ajd4 ,conv]g5ce z a cause ( movem5t4 #aaa1 p4 #ajh4 ,rhy?m ( ,conv]g5ce4 ,contr>y ,motion 9 ,conv]g5ces4 #aab1 p4 #aji4 ,*anges ( ,direction 9 ,conv]g5ces4 #aac1 p4 #aaj4 ,ideas ( association 9 rhy?mic movem5ts4 ,rhy?m 9 *anges ( %ape4 #aad1 p4 #aab4 ,outl9es 9 diff]5t ,attitudes4 #aae1 p4 #aab4 ,h>mony1 ,balance1 & ,rhy?m 9 ! ,attitudes ( ,outl9es4 #aaf1 p4 #aab4 ,! ,composition ( ,outl9es4 #aag1 p4 #abd4 ,! purpose ( design+ to induce ! s5se ( ,beauty :i* is ! cause ( all t is f9e 9 ,design4 #aah1 p4 #abe4 ,areas4 #aai1 p4 #abe4 ,l9e> ,areas4 #abj1 p4 #abe4 ,*anges ( wid?-measure 9 ,l9e> ,progressions4 #aba1 p4 #abi4 ,areas def9$ by outl9es1 & also by tone-contra/s4 #abb1 p4 #aca4 ,! ,composition ( ,areas z def9$ by tone-contra/s4 #abc1 p4 #aca4 ,diff]5ce betwe5 draw+ & pa9t+1 if !re is any4 #abd1 p4 #aca4 ,def9ition ( ! ^w ,tone4 #abe1 p4 #acb4 ,tone-,analysis4 #abf1 p4 #acb4 ,! /udy ( ,tones & ,tone-,relations4 #abg1 p4 #acc4 ,pigm5t-,mat]ials4 #abh1 p4 #acc4 ,! ,scale ( ,neutral ,values4 #abi1 p4 #acd4 ,contra/s ( ! ,scale ( ,values4 #acj1 p4 #ace4 ,def9ition ( ,value-,relations4 #aca1 p4 #acf4 ,scales ( ,colors 9 ,diff]5t ,values4 #acb1 p4 #acg4 ,def9ition ( ! t]ms us$ to describe diff]5t ,colors4 #acc1 p4 #ach4 ,color-,int5sities f\nd 9 diff]5t values4 #acd1 p4 #aci4 ,value-,relation ( diff]5t ,colors %[n 9 ! ,spectrum4 #ace1 p4 #aci4 ,! ,spectrum a sequ5ce n a circuit2 a circuit 9 pigm5ts only4 #acf1 p4 #adj4 ,! ,complem5t>ies4 #acg1 p4 #adj4 ,a ,g5]al ,classification ( ,tones z to ,value1 ,color1 ,color-,int5sity1 & ,color-,neutralization4 #ach1 p4 #ada4 ,! di/9ction betwe5 ,color-,int5sities & ,color-,neutralizations4 #aci1 p4 #ada4 ,def9ition ( p>ticul> tones4 #adj1 p4 #ada4 ,!oretical "* ( our classification ( tones4 #ada1 p4 #adb4 ,def9ition ( p>ticul> tone-relations4 #adb1 p4 #adc4 ,sequ5ces ( ,values & ,colors4 #adc1 p4 #adc4 ,! ,sequ5ce ( ,neutral ,values4 #add1 p4 #add4 ,v]tical ,sequ5ces4 #ade1 p4 #add4 ,horizontal ,sequ5ces4 #adf1 p4 #add4 ,diagonal ,sequ5ces4 #adg1 p4 #ade4 ,diagonal ,sequ5ces ( ! ,"r & ,left ,modes4 #adh1 p4 #adf4 ,diff]5t ,int]vals 9 ,diagonal ,sequ5ces4 #adi1 p4 #adf4 ,peculi> value ( ! ,diagonal ,sequ5ce ( ,colors at ! int]val ( ! ,fif?4 ,! f\r ,triads4 #aej1 p4 #adg4 ,sequ5ces 9 :i* a c]ta9 relation ( int]vals is repeat$4 #aea1 p4 #adg4 ,! comb9ation ( two or m ,v]tical ,sequ5ces4 #aeb1 p4 #adh4 ,! comb9ation ( two or m ,horizontal ,sequ5ces4 #aec1 p4 #adi4 ,! ,comb9ation ( ,diagonal ,sequ5ces ( ! same & diff]5t modes4 #aed1 p4 #aea4 ,alt]nations 9 ,v]tical ,sequ5ces4 #aee1 p4 #aea4 ,alt]nations 9 ,horizontal ,sequ5ces4 #aef1 p4 #aea4 ,alt]nations ( diff]5t value-int]vals 9 color-sequ5ces ( equal int]vals4 #aeg1 p4 #aec4 ,alt]nations ( diff]5t color-int]vals 9 value-sequ5ces ( equal int]vals4 ,p>ticul> ,sequ5ces recomm5d$4 #aeh1 p4 #aec4 ,possibility ( ext5d+ ! classification ( values & colors to a scale ( sev5te5 values1 includ+ ,black & ,:ite4 #aei1 p4 #aed4 ,! me?od ( us+ ! ,sequ5ces describ$4 ,possible ext5sion ( ! sequ5ce by mixtures4 #afj1 p4 #afe4 ,! value ( ! sequ5ces f\nd 9 ! m def9ite ?9k+ :i* !y make possible1 & 9 ! ,h>mony ( repetitions4 #afa1 p4 #aef4 ,consid]ations ( position1 measure1 & %ape 9 tone-relations4 #afb1 p4 #aeh4 ,tone-,h>mony4 #afc1 p4 #aeh4 ,tones h>moniz$ on ! palette or by *anges 9 ! design4 #afd1 p4 #aei4 ,value-,h>mony4 #afe1 p4 #afa4 ,color-,h>mony4 #aff1 p4 #afc4 ,h>mony ( proportional neutralizations4 #afg1 p4 #afd4 ,value-,h>mony & ! ,h>mony ( ,proportional ,neutralizations comb9$4 #afh1 p4 #afg4 ,h>mony 9 ! repetition ( a c]ta9 relation ( tones n 9 itself h>moni\s4 #afi1 p4 #agj4 ,h>mony ( a pr$om9ant tone 9 ! repetition ( a c]ta9 relation ( tones4 #agj1 p4 #agj4 ,! ,h>mony ( a grayness induc$ by ! opposition ( tones :i* neutralize "o ano!r 9 ! s5se ( vision4 #aga1 p4 #aga4 ,/rong contra/s1 inconsi/5t ) ,h>mony1 may be p]fectly consi/5t ) bo? ,balance & ,rhy?m4 #agb1 p4 #agb4 ,tone-,balance 9 ! ab/ract4 #agc1 p4 #agc4 ,use ( ! ,diagram ( ,values & ,colors = ! calculation ( tone-balances4 #agd1 p4 #agd4 ,elem5t ( ,color-,int5sity 9 tone-balances4 #age1 p4 #agd4 ,tone-,balances always connect$ ) ,measure & ,%ape-,balances4 #agf1 p4 #age4 ,tone & ,measure-,balance4 #agg1 p4 #age4 ,tone-,relations 9 ,s+le & 9 ,d\ble ,inv]sions4 #agh1 p4 #agf4 ,occult ,tone1 ,measure & ,%ape-,balances4 #agi1 p4 #agh4 ,fur!r consid]ations on ! same subject4 #ahj1 p4 #agh4 ,fur!r consid]ations on ! same subject4 #aha1 p4 #aha4 ,! effect ( ,repres5tations 9 ,tone-,balances4 #ahb1 p4 #ahb4 ,tone-,rhy?m4 #ahc1 p4 #ahd4 ,attitudes 9 ,tone-,rhy?ms4 #ahd1 p4 #ahe4 ,inv]sions 9 ,tone-,rhy?ms4 #ahe1 p4 #ahf4 ,composition ( tones1 measures1 & %apes4 ,?ree g5]al rules4 #ahf1 p4 #ahh4 ,design & 8,decoration40 #ahg1 p4 #aij4 ,! /udy ( ,ord] 9 ,nature & 9 ,works ( ,art4 #ahh1 p4 #aij4 ,me?od ( /udy by ,analysis ) ,syn!tic ,p]=mance4 #ahi1 p4 #aib4 ,conclusion4 ,! practice ( ,pure ,design4 ,its purpose & end4 ,! ,riv]side ,press ,,cambridge ,,massa*usetts ,u ,s ,a