CIHM Microfiche Series (Monographs) ICMH Collection de microfiches (monographies) Canadian Institute for Hls'.orlc.^l Micrct^productions / Institut canadJen de microreproductions historiques Technical and Bibliographic Notes / Notes techniques et bibliographiques The Institute has 3''empted to obtain the best original copy available for filrriing. Features of this copy which may be bibliographically unique, which may aher any of the images in the reproduction, or which may significantly change the usual method of filming are checked below. D n u Coloured covers / Couverture de cou'eur □ Covers damaged / Cou" erture endommagee I I Covers restored and/or laminated / Couverture restauree et/ou pelliculee I Cover title missing / Le titre de couverture manque I I Coloured maps / Cartes geographiques en couleur n Coloured ink (i.e. other than blue or black) / Encre de couleur (i.e. autre que bleue ou noire) I j Coloured plates and/or illustrations / Planches et/ou illustrations en couleur Bound with other material / Relie avec d'autres documents Only edition available / Seule edition disponibie Tight binding may cause shadows or distortion along interior margin / La rpliure sprree peut causer de Tombre ou de la distorsion le long de la n-iarge interieure. Blank leaves added during restorations may appear within the text. V^'henever possible, these have been omitted from filming / II se peut que certaines pages blanches ajoutees lors d'une restauraticn apparaissenl dans le texte, mais, lorsque cela etait possible, ces pages n ont pas ete filmses. Additional comments / Paqjn.iiion i- Commentaires supplementaires L'Institut a microfilme le meilleur exemplaire qu'il lui a ete possible de se procurer Les details de cet exem- plaire qui sont peut-etre uniques du point de vue bibli- ographique, qui peuvent modifier une image reproduite, ou qui peuvent exiger une modification dans la metho- de normale de filmage sont indiques ci-dessous. Coloured pages / Pages de couleur I I Pages damaged / Pages endommagees n n n n n a Pages restored and/or laminated / Pages restaurees el'ou pellicuiees □ Pages discoloured, stained or foxed / Pages decolorees, tachetees ou piquees Pages detached / Pages detachees Showthrough / Transparence Quality of pnnt vanes / Qualite inegale de I'lmpression Includes supplementary material / -J Comprend du matenel supplementaire Pages wholly or partially obscured by errata slips, tissues etc., have been refilmed to ensure the ber-t possible image / Les pages totalemient . j partiellement obscurcies par un feuiilet d'errata, une pelure, etc., ont ete filmees a nouveau de fagon a obtenir la meilleure image possible. Opposing pages with varying colouration or aiscolourations are dimed twice to ensure the best possible image / Les pages s'opposant ayant des colorations variables ou des decolorations sont filmees deux fois afm d'obtemr la meilleure image possible. .1;. follows: (). Ul.)-'**!. This Item is filmed at the reduction ratio checked below / Ce document est filme du taux de reduction indique ci-dejiscus lOx 14x 18x 22x 26x 30x 1 ; 1 1 , . , . 1 12x 16x 20x 24x 2ex 32x The copy filmed here has been reproduced thanks to the generosity of: Uouy Idb I ibrdry Oueen'b Univcrbity L'exemplaire filmd fut reproduit grace A la g^n^rositd de Douglas Library Queen's University The images appearing here are the best quality possible considering the condition and legibility of xt. original copy and in keeping with the filmirig contract specifications. Las images suivantes oni 6t6 reproduites avec le plus grand soin, compte tenu de I9 condition et de la nertet* de lexemplaire filmd et en conformity avec les conditions du contrat de filmage. Original copies in printed paper covers are filmed beginning with the front cover and ending on the last page with a printed or illustrated impres- sion, or the back cover when appropriate All other original copies are filmed beginning on the first page with a printed or illustrated impres- sion, and ending on the last page with a printed or illustrated impression Les exemplaires originaux dont la couverture en papier est imprim*e sont filmds en commencant par le premier plat et en terminant soit par la derniAre page gui compona una empremte d'impression ou d'illustration. soit par le second plat, selon le cas Tous les autres exemplaires originaux sont filmAs en commenpant par la premiere page qui comporte une empremte d'impression ou d'illustration et en terminant par la derniAre page qui comporte une telle ampreinte The last recorded frame on each microfiche shall contain the symbol » (meaning "CON- TINUED I, or the symbol V (meaning 'END' ), whichever applies Un des symboles suivants apparaitra sur la derni^re image de cheque microfiche selon le CJS le symbole ^^ signifie 'A SUIVRE , le symbole V signifie FIN ' Maps, plates, charts, etc . may be filmed at different reduction ratios Those too large to be entirely included m one exposure are filmed beginning in the upper left hand corner, left to right and top to bottom as many frames as required The following diagrams illu.itrate the metnod Las cartes, planches, tableaux etc peuvent etre film^s d des taux de reduction differents Lorsque le docui.ient est trop grand pour etre reproduit en un seul cliche, il est filme ^ partir de I angle supAneur gauche de gauche d drone et de haut en bas. en prenant le nombre d images necessaire Les diagrammas suivants illustrent la mdthode 1 2 3 1 2 3 1 4 5 6 MICPOCOPV RESOIUTION TEST CHART iNS' 'j''_t iVJ n'y lmak; %. .' 1.0 I.I 1.25 ;i!;2 8 #2^ '^ m III 1.8 1-4 i 1.6 I'lUt I EXPERIMENTS ON THE AESTHETIC OF LIGHT AND COLOUR SECOND ARTICLK m •:?• r .SI'KCiRAl.l.V Pl-RK COI.OfkS IN BINARY COMBINATIONS i-MMA s. haki-;r, b.a. ■^ J I'l.ATl-: A M SI'I-XTRALLV ITRIi COI.orKS IX HIXAKV COMBINATIONS In tilt fourth number of the INycholo^ic.il .Series of I'niver- sitv of Toronto Studies' was published an article containing the results of experiments on the cesthetic value of c<}lour com- binations. The colours u-^ed in those experiments were pijj- ment colours, and the surfaces were the same in sha]K" and of e()ual size. In view of the importance of the relative propor- tions of the surfaces, it was thouj.;ht desirable to conduct a second series of experinunts, in which it nii^ht be possible to vary the size of the surface. This phase of the problem was in- vestigated during the years 1899-1900 and 1900-01, in room 16 of the Psychologic.d Laborator\'-. Till, apparatus .lud method employed were entirely difTerent from those used in the former experinivnts. The api)aratns was that described by I)r. \V. H. Lane, in an article entitled " Space Threshold of Colours and its Dependence upon Contrast," pub- lished in No. I of the Psychological Series of Universitv of Toronto Studies. .\s Dr. Lane's article contained an illustra- tion and a detailed explanatitm of the apparatus, we shall con- fine ourselves here to a short description of it, particularizing the changes which our experiments necessitated, and illustrated by a schematic representation (Plate A) which gives a view of the apparatus from above. In the ])late, A is the central table ; upon the front edge is erected the upright with the micrometer diaphragm /?, of which Figure 2 in Dr. Lane's article gives an illu.stration. .At the other end of this table, in a somewhat slanting position, there is another upright //, through which is visible one of the coloured pigments arranged on a movable disc i behind it. The pigment is illuminated by the electric light ■'/ adjustable to various dis- tances in box H. The rays of this light have to pa.ss through one of the combinations of gelatines which are ar;anged in the form of a revolving wheel hnu,^ h^'ht ju^t a,, pure ,. if a certain region of the spectrum itself had heen 't,ken but u,.t limited in intensity and space extensu.n. This liu'ht is then seen thro'i^di the observation tube 1, ,„d thn.UKh the aperture of the diaphra^nn -7, whilst the opaque pirts of the latter are covere.l with another pi-nient paper ilUnninated in an analo^rous manner bv means <;f /-' ni box L. .\t the end ..f box C is inserted a ^^elatine preparation as re- ..nired bv the -roumi colour. Siace the ground colour remains constant'durin.r , whole series of exiK-rinients taere is no need f„r the ^-elatiue preparations to be arranged m the form of a cMsc Thev are identicallv the same as those for the other box h„t tliev have the form of squares of the same sr/e as the open. i„.. of the box. The observation tube T is .ittached to the niov- i„r. part uf the diaphragm . by a lever arran^^-ineut .in suc i a ,„;;nuer that Its end moves with half speed. Tins keeps the n.id.lle part of the diaphraujm aUvavs in the centre of ^l^'-' '"-^ «'■ •M the ocular end of this tube . a lar«e screen .shuts off the eve ..f the observer from all other parts nf the appar.atus. 1 he eve-piece A the ^-raduated .sIit,^ and the .creeii /^ render it pn>- .ihle to compare the li«ht .seen throuj,di the tube with real spectral li^ht generated by a lantern and spectral apparatus m ;;„ „ijoinin,^ room (the first annex nf Room ibV '1 hrou.di the oh..ri,n. tube can now be seen the two surfaces the .round colour on the .liaphrasm, illnmlnate.l by/., and throng i >- openin- -^f the diaphra>.nn the colour to be combined with the erotmd^-nlour. illuminated bv //. There were twelve colours ■;,, all which could be chau^^ed by a revolution of discs r and Set- also Iir. Line's article, pp. 23-24. 29 ■was con-^tant throuj^h a series uf trials, was j^ivfii, by ilie i)o>.i- ti(>n of /.'. tile intrii i!^ whii^h tlie saiiif colour had as seen throu^'h tiic uiaplir.i^iu. li i> i.-a.>il\ j)i)S.-,ibl^- by adjustiujc the lnin;i^ to arraiij,;e the intensities in such a niannur that the j^^roiinil colour an(l tlie tran>iiiitte a distance of about two feet between the two surfaces, the olisfr\er at the end of tlu' tilbt.- could not possibh' 'kli-ct the sliirliu-.t deviation from a uuifcjrmh' coloured surface, no m.ittL-r wli it was the aperture of the diaphra^'m revealini;^ the transmitted colour. .\nd, indeed, before each series of tri.ds this impossi- bil'.t)- of distiuL^uishiu:; the surfaces w.is madi- the criterion for the correctness of the intensity cjf the ^^round colour. \\V next come to the method of procedure. The room was d.irkiiied and tlu- obserwr placeil himself .it the end of tube T where he s.iw tir.st the Lrround cokair on!\-, illuminated as described. Then by turnin.i( the micrometer handle there wa.s made to ajipe-r in the centre one of the other tuche colours. This colour ajipeared first a.s a little s])ot. As it increased and ro>v abo\-L- the characteristic sjjace thresh<.ild- the o}iscr\-er stated whetlur the cond)ination was jdeasant. and whether it became more or less ple.i.sant with the chan<:;-e in th,- couijiarative proportions of the two components; for as the transmitted colour iucre.isi.-(l the ;^round colour decreased nV ' ilnalh- only a small border of it was kft. The obser\-er, who had now seen nil the possibU- r(dati(.ins in size of the two conii)onents, was asked to decide if the combination on the whole was \'erv [ileasant. pleasant, indifTerent, or unpleasant. This was not • Miss Baker l^IU -.Ik- iiu.ilitics otitained hv the alime 'luscribca tnetiioil •' sj«.ctr,illy pvire," ineatr.iis thereby that a Hiiiitucl reKi,'alion was undertaken, n.inielv th.it of tile inlhience of the relati\e size; the jn(l;,„Mnent was taken to ^ee if under entirelv different conditions the results of our former exjieriments would be corroborated. \'>y a conipari- son of (.■ur\es i-i.? with curves I-XXI\' of our former article to which .itt'.r.tiun will be more fully c.iUed fiirtluT on, it will be seen th.it th.e results of t'-e two series of experiments do ncjt esseiui.ill- liiiTer. .\> We ri\er>eil the i)roceedin'^', the tr.iusuiitted colour becomini; '.^rf.idn.dlv le^s and icss, the observer -tated NcluTe the conibinatii'U w.;s most jile.is.int fV, if uniile.is.iut, where it w. is coiiipar.iti\ el\ m'>-t ]iIiMs.int. I''..Kh full turn i if the microinelL r hiudle ch ill'.^cs tile inutu.ll s, ..^e relati'iiis lutweell tile di.iUleter.-. of the two e'lidur- bv tin.- twelltittll of ,1.. illcli. liidoubltul ca.si-s to sicure ,i definite decision the jimcedure w.i> npe.ited if nfce.s.sarv, sevir.i! tiim \r. tlii-- w.i\- all i1k elevm (.uhmrs were treated successively ■- i the same ,i;roiind colour. .\ whole Series of tri.ils w.is t.ikeii .it one sittin;;. The --.ime ])r;!iieiils of \er\ pleasant, pU.is.int, indifferent or uniiK.isant. Of these twetltv-tive obsi r\ .it lou^, ele\eli weie m.ide bv reL'Ul.ir observers who tonk the whole series, and in sesei.d c.ises rciieated the exi>erinietits. M.niy took about h.df the .series, and in e.ich ind.ividual series there were a few casual observers. It is ouh litliii'.^ lure tu .ickii' 'W ii-d^c our xie.it obli^.itions to all these 1. idles .md j,'eiit!eiiHii, wlm. ,.fteu .it .1 ;.;reat incon- venience to themselves, were so f.iitlifui in tlw ir .ippoiiitnients and m.inifested such .in iiiUte-l in tin pi.^blein l-or the pli.i.>e of the problem which refers i,, tlie size of the CMinpuiunts there were ^;i»>) complete Ir.msitiuiis from tile 31 Si'KiTRosci')PiCAi. Analysis mi- tiii: C'M.oius I'si-.n. Wilh N.irtuw silt. N;iTTirof Colour. Visiblr part of Krfciun of great-' \ *i»il>le icirt of K CK'iiri ijf (Wral SjtfcTrum e»t intensity SiH'ttrmn ot tiil^iiMly in ^fi '"H-h ^7 '•' Mi" 2.51'' - 5^" >n fj.fi Red 635—610 ^"=•7 5 >iiS oranve-Kiil ''JJ.5— ,';''•'. s 612.5—592.5 635— 5>w 62^,5-59^.5 1 lr;ui>;<'. '''■■S— .SSJ-S 5S5-562.5 6J2.5-547 5 6i>7-5 -.562.5 1 ir.iiiKf-YclliAV. 5^7-5— 5-t7.5 5'>2 5-557.5 '"75-537 5 6"2.5-S55 Vfll.i'.v. ,VS— 462.5 497 5-43" 475— 45S Vidltt-rtirple. 1 6.^7..s— 6hs ( 4\5— 4411 462.5—455 I 7.»j — 665 4H75-4,V' :7" 452.5 I'urplp 1 ;""' -4<'-i ( 6K.J— A35 497-5— 4V> 17 s )'« 32 l--.,r each '-1 tlic^c transition.- which the smallest to the lar,. ^t. ^^^^. ^^^ ^^.^ ^^.^ ^,^^ ^^'"''■"^^^■^^■';i:r";;;vcrTa,.,c. I. n. niandw were cnli values in our represent the tabulated rcMilts nf.dl the^e expe tahk riiueiil- from the lie coii-trucU-d in an ■' , r • ,.. Tlu-^i- t,i 1 e^ are con-Li ni.^v: ,„.er point '■f;-;^^,, ,„„„,expernn.nt>. Tables -'''«':;■-'■'•'"'""';:, L nit ■■conilnu.ti.ni. and III and •'"'^ " "''"7"' ';■ in -tnnibe „r,,.tten that we deal V ihr ■■ ven 1> ea-.int. It nin-i n ,-. . , ; .,,■ ^ L ur-t 'rticle with iH^-luv >atniate.l co,o,l-. ler.- a~ m -nr lir-t ..rt.. ,,„.,„,,,u -s almost without ' 1 „r ,,t hi'di saturation are m theni>cl\.s ai a ^"^"'": '' "^ .... 1 h,.n- ..„ubinatu,ns.if n^t .Associated former point < a I I h ,1, ,mf de.'ree as combinations m wlian 'K r, so? lesser s,aur,>tion. The pvescnt iiunurv ,s tuere- '"'"', „ iM the .iesiled duectioU. Th,-- ■,uvest„.>tloU ::cLz:^z^^^>^-- , other. o„,biuat.ou,,ne. :lturatcdonou:s,whK.h.illbe.irmuchnioieon aepr,.c.Ka „,,„ ,loes the subject of this discussion, is .o to.losv, ,„T.d,le!tlKinh,.d leuers at tlK. top ,n liori..nit.U ordvr ,^.,„.,..„,Uie .round colours, red, or,.n,e-rcd.e,.-. ^ -^ - :'T:':^r;:::;;:;:oi:;.nw,.:c, u three t,mes w,th vuih oran-e as ,, ple.is.uit com. 'P.ib'ie III is .imiotjon^ led w.is the ul ,„,,„.e-re3 TaIU.K I. — i'l.i:A> \\ !■ C' 'MIUN \i'I( iNS K, OK, (I. () Y. \-. \- c, C, c. 1!. n. \-. vr. P. ^ i^ H I -^ -'i JS "< .■ . . ; ; ;S2 t). R II V. U) s 1 !■' I" I. \ I'. !\ 16 1 i 'i ■- !(> -■; .'4 -'' i'* I*-. M 1" ifi-, ; '. :. ." -■; Ji 1^ M f ; 146 17 I' '^ iv :i 21 114 170 " ^ • 17 Ml • '. II 1! V iv 12 I'l 1*1 !■, I~ 23 r- I'- 17 ^,. itj :^ 17 16 I^ n 1 ' ■ t i : I ■; M 1 : SI I 7 ' \ I ^ U 7 M i\S ' I '45 ■ l-ii I'' li IV i» ,20 i3 j i.i I an j ij . 19 , ,. ,91 i6.; H2 if,2 I tS 1-^ 175 1169 '151 :i37 157 ,i 14 iHhj 'I'AIi!.!-; II. — I'l.i \-\\r C' iMi;i\ \Ti. /N>. R. O.K. O. O.Y. Y. Y.G. (, (V;; H v \ p v 34S ^ .■=; J,i js _i-> 17 ti 14 37 ,v, 21 ;J05 17 13 26 ,-,y 44 1.. ;; .2 ;, .>2 -<08 :; ,., 27 y, 41 j„ ,j ., ,, '^^^ '- -" V> 3.^ 37 3: 2S ;. ;M8 js .i^ "t 39 1" V» 317 li 1^ I : 323 'J 7 •!'^ •^^ 37 R. R ( 1 I) V V ;7 4- v.<; <;. 298 -, I', .4 3-^ '-..H. 272 M ri 29 H 302 I V 294 1 - V !■ 335 I' 3728 2 1864. 35 Ai;i.!-; III.— V i'U\ 'I KASANT Ci)MniNATroN> II R R. lO.R. O. O.Y.I Y. Y.G. O I6 IG.B. ' V. V.P. P. 6 J 7 94 6 6 4 82 II V V G. n.n. V !■ o 1 3 1 I 2 1 5 ; ' 5 1 « i 4 8 .S 1 9 10 r.s '2 1 13 SI 88 ; 6a _Zi I 68 I* I (2 54 n» 44 3^ 73 50 48 38 39 47 40 88 H- 41 ' 48 - 679 3^' WAS. !V.— Vi.K'. IM.KASANT CoMr.iN.vnoNS. , ,,v. V, \r.. r.. c.v. v.. v v.P. P. 6 IS K. OK < 139 , - „ ,. ,~ 2S » 15 " - 123 1^ A 14 17 -•^ '4 '^ '* '' 94 , ., -- 1-' i; I" '■» 8T '\ 161 - 1-' '7 1'* 2.' •■! ;^ 112 -, ) ) '' - iS 121 J , - i6 108 2-679. enlniir at tl'.c lu k! <>i the cciluniii and .A>o at the riL,'ht-haiiil t.-nii (if the horizontal ; for exaniplt-, _:?45 in Tabic II indicates the >iini total fur all the jud^nneiits for the colour red. In order that these re.-nlt- may he eoinpared with those of the former experiments, the\- liave likewise been expressed in L!'.r\ I.S With the series of colonrs in spectral order as the abscissa line and the freijnencv of their ^election a> ordinate^. These curves are desit^Miated b\ Arabic numerals, and the letters ii h ,, in addition to the name of the res])ecti\e colours to who-.f com- binations ihe onr\es respecti\eh- refer. The lower curve in e.icli ca-e represents the " \ er\ plea-ant " combination ; the upper, the "jileasant " inclndins;- the " \er\ ]i!e.isant." Cnr\e I ,?, '■', and ; i represents the resuhs i.f j'.] the com- binations in vUiich the colour red t^ok jiart. Cnr\e _'. those in which the colour oran.!.;e-red took i>,irt. Curve ;. oran-e, etc.; i in each case represents the results whci! the colour indicated helow was the ^ri>und colour, lud corresipouiis thus ;,, tin- vertical columui in Tables I .uid III, '> .,n the other h.ind st.inds foi the results when the colour indicated below, as trans- mitted color.r. w.is cond)in(.d with .dl tht. othii-, thus corres- ]«i!idiu<,' to the hori/ont.d clunu's of T.ddes 1 ,;n.l III, The hei;.:ht of the ..rdinates is taken from Tables 1 .iiid ill. Thus iu Curve i om. where red w.is ihe -roimd colour, it w.ischosin three times with or.iiiLre-red, seven times witil i)ran;..;e as .1 pleasant comlnnatioii , when red w.,- the tr.iusniitted colour ( /o it was chosen live times with orar.^^e-ieii .ind ei'^hteen times with oran^a- as a jilea.s.iiu combin.itiou, I';n,ill\ , is the cmbin.i- lion of th.- rw,,, ti,,,^ oorrcspondin^ to the results in Tiblts II .niid I\. In these fxpeiinicnts, as iti our former ones, eacli colour .i|>- peared twice ; in tills case once as the ;.iround colotir and once as the transmitted colour. When the surf.ices were of eipi.il .sliaiw ami size.md weri' pl.iced side by side.is in the former e\peri. nieiits the rt'sults could unhesU,itinj,dy be combined ; but in our prf.sentf.vpcritiK-nts, as can e.isily he seen, this could not be done, for when the colour apfK-ars as the L;rouiid colour it forms a bol- der, whereas when it .ippears ,is the tr.iusmittecrie. -t rnrvL-^ %vii; ~h-%v Uii^ ,lit7.Tcnr. . C-ii^.-iu.-.ulv the .-L-urvc^ arc ni re-l.ttivclv much 1- - si-niticmcc. Curve 1 :; :> I cnTuhiniliuu ..f all the curves. .1 reitre^er.ls that each ..f the clcur^ IkImnv ... >;r„uu,ime -ener.d ch,>racteri-tic> in -.jiite ot the .^roit dilTerence in the cond.itious. It nnist be uotice.l, however. That these ch.iracteristics cannot here be exiKCted t^ • be .is marked, as in the former ci-e, since we now li.Ae oniv h.di the „,„„ber of colours. In ,,11 th.o.e curves ♦'or or.in^e and. vellow we tiud the same bro.,,1, pl.iteaudike shape. We tllld. too, th.lt like those with which we ne Colup.iriu.Lt tllelll, tlus,- c-rves do notre.icht.. so .^r^Al ,. lurjht as those cxpressin- the other .inalilie-.. In maiiv ..f the other curves we can tr.ice .i strikm- simil.mtN with the curves..! the f.,rmer series, but it must not be for^.-lteii th.at It would never be p.-ssiblc L ■ trace all the J.r..- mmences in these, on .icc-MUt of the s„,,dler number .4 quali- ties, nor th.lt the chan-ed .s]Mce conditions al..ne w..u;.l prevent a coiniilete .lualo.^v. It IS intere-tiiii; to notice the dilTerence in the m.i.xiiunm when the colour is the i,'round col.,nr on and when it is the transmute.! coh.nr i 'o. In the curves f..r the - pb.Ms.iut ■' ;n no rase ,1.. these cMiicide e\cei>l in th.lt . .f j^Meeli, wliere the m.ixi- muin m both c.ises is .is red. In the ••m..sl ple.is.iut." however, there is ,1 comci.lence in the cases of led, -reeii, blue, .md vio- Ut-i>nrple. In this series of evperniuuts ,is m the f..rnie' it w,is not kn..wn until .ifter the experiments were completed which ..f the cuhmrs used. U .nn. wete complemeiil.irv . It will be retncm- bere.i th.lt, in i'Kler to h.ive e.ich colour as free as possible from -^ s YC I'l. \ ri. r. r S I 39 nrliiiixture with other cnl,,.,irtMne-.t!K-c..I..!ir.;;-;,-.l \vtiVnl.;.ii,K.l liv ilhiiiiiiialiii^r pi^nnent jvipci. with !i-ht whirli lii.l uiuicr-Miic a str-.nj,' sel<.'cti\<_- absorption. Tlii> iii.iJ.e it imp. ,>m1)U- t.>",^at the coiiiplc'iuntarv relation of these clonr^ in the o,,lin,,rN uCiv. /.'., l)v niakiiii; colonr ecpi ition- ,1^ wa> ,lone m t!ie former ex- periments. We have totliank Mr. .\. 11. .\l,l„,tt, fn^t ,i.-,:.t- ant in the I'svehoh.-ie.il L,ihorator\ , fertile sn.i^'-e-tion of t'n. folIo\vin>,r method whieh was adopted. The iip;i-!it M.iekened hoard, to which was attadied the niet.d. di;iphr,,-m, w i- replaced \n another in which an openin- ,,f tw.. and a h.nf inches wa- in. ide, and witl! which all other sjjace rel.itinn. were i.re.erved. Xow, l,v l<,okin); thron-h the ohservin- tnhe the tr.niMnitted colonr can he seen as t>ef.>re. To the luck of thi> npri-ht ho.ird w.r- attached a small electric motor i., ..prr.ite ,i rotati.,n app.n- atnsin fr..nt of the tward. Thi> l.itter make- it p:..-il,l,'. to rot.ite, a- a kind of e])i.scoti>tei, .sector. ,,f ,,nv colonr \\l„,-,e .in- sular v.dne could he v.iried hetween tlie !ini:t ..f lo,,,,,] ,r„, dei^'rees, llic measurement t.. he oht.nned from a i^nad, ition .it the i'eripherN. When rot.atin- we see the mixtnre of tile colonr of the rotatin- .-ectors ;„,d that tran>niitted ihron^h the .apertnre. iluth col.,nr> are in tli , illnmin.ition .i- in oi:r tri.d> ; .md if'-..v,,oi onr col.mrs we actn.illv comj.lenient.irv we shouM, h varvii',- the .sectors of Jie rotatin- jMrt (;;;onnd coI,,„ri till we reach colonrless li-hl :,> :, resnlt, be able to c~t,il,lisii ., e-mplete Colonr e.|nati,.n. ( 'nfortnnatel v it wa- found th.it tliere was n,,t a .sin.:^!e p.iir of c.miplen -itarie- anion.^r ,,nr twelve cohnir.s, and -ince with .,nr method i- w,,- n..t po,sil,K. t,, mix more th.ni tw,, colour-, we conld onlv ni.ike out tliat the coniplement,.rv of ,, cert.iin colour was between two otlu',-' .n„i ne.irer to one than the other. Tlie com]dementarv rel.iti,= M. " = s = • •:- — "" ^ - ? 3i i'r -'■■.^ 41 .\!:l.]'. \'I. — Maximal I'uints ok 1,1 \> ANT, ' 'r.iii:;f-re.l. ' ''aiigf, ' iraii)4e-ye!iinv VelLnv-^rccM (jree'i. (''reen-I'lue. Klue. \'i'.;et, \!i)lt-t-purple. Purple . Number ijf Jiuljjiticnls. iH. '■.,-R. V.-P. C. K. R, I). O. V. V.i:Y.-G. Y. J I ' I ". 21 O.- IV li -'4 V. 2', R. 22 R. 21 R. 21 (). IV V I'.^iR (". ,v.(; NMrnLei of N inil.rr of JU'lKiiiftus. i^ ' 1" '.klinnits :V(-. 47 2-i''. 44 2-, c.-n 41 2' C. ,\I1 .'^'^ -■1 V IV 4" 2 1 K 45 22 R. 47 22 () 42 J.VO. 40 '9 Y. 39 21 Y. 40 ' 2.^ Y-O. 40 TAHLH \-II.-Ma\imai. I>. I iIN'i'S I 'V M I ',M 1 1.I-; \s.\N T. ■ .f Ri-i. ' 'raiij»c-re.i. <)raiixe-vcl!o« YeU.)w,' Vrllo»-j;re.-li, (Vreeii. Creen-blue Hhic Violet. Violet-purr,;,;, Piirple. JuilnuivrHs. V.oli>ijr . I .VK. -K.\Y. &Y. -i:. Niimli*rof JU'Jgiiitnl-,1 Colour ■SO .S(", 10 vr, ,(; ■ .■i V ,v\- -I 12 (_i,V ISR. I4<>,.K,VV 10 Y. 12 Y, 12 Y r V. li.VV ifikv. QlV. '-. !V 12 \ ,V1' - I'.VK 1 1; - K 4(J.-R.i 1'. Y. u y I- \' Numt*r of Judgments 28 »3 14 22 IS aS 20 14 |8>Y. 21 .'2 42 In Ti'i'a- \"I !ii1 VII tlv.' rv.uU-r will tunl the iiiaxim.i Uhu- latL-d. I>\ oin-ulf.-.i- T.ibk- V llw oiiiiplciiu'iit irv rci;iti()n> cm \)v (.-.i^ily ;i>ci.TtaiiK(l. [v. ci'ii^iik'iiii.L,' thv rvl.itioii ir. wiiirh .uir rcsult< slaiul to the thfiir\- th.il the iiio-t a.!:,rj-t.cal)k- coinliinalinns are f.jund with cuin]i!emeiitar\- colours, we shall clas-.if\- them into j^roup-. as we dill with the re-'alts,,f ijir lurnRr uxpcri-.nent-. I-":r-t, we ("iiiil iinUaverv few cases wiiere we n.ijl.l -ay that the niaxi- nnnn ini;.,dit jMissihlv coincide with tlu- e.inii)lenientar>- colour. Tims, t'.ir iii.-tancc, fur urai!:j.L-red wIm-l' cmplL-nK-itarx is he- tweeii t^reeii-hlne and .tureen. b\it nearer tt> j,'reen-l)lne, we find the maxinnmi for Curve fi. " pleasant," at j,-reeii-l)l\ie, while for the same curve, " ni"-l p^.asant," it is divided between ^reen- bliie and bine ; whereas for the rest of the curves, the niaxiniuin is at j^recn. The >anie is \alid f"r the oranjje, where also several of the curves have their inaxiniinn at the }j;reen-bluc, which is the nearest to the eouipleuKntary. • »ran;;e-yell()w and yellow have also one iii.ixiinuni each, nearest to the coin]>iementary, but the latter (vellow I nil in undisputed niaxiinuin, .ind the former onlv in the combinatiiin Curve c for tile " ple.isant." I'or the rest of the cur\is, one can say decidedly that a colour somewhat approachiuii; the complementary has the pre- ference, but it is never the one which an\otii; our twelve qualities would Ix.- the nriinsf to the complementary. Thus for ^'recn we thid throu^jhont all the curves red in the maximum, while the complemeut.nv to jrreen is beyond the ]mrp!e towards violet-purple. In ^^reiii-blue we find red and oraujje diicfly the ma.ximum, whilst the complementary is at oranj^e-yellow. ( )nc innsl not be mi.sk;reeii-b!ne of the pure colours w.is by no means identical with the >^reen- bhie of th«. pijjmeiit colours of tjur former cxj^erinieiits 115), as the «Hfference in the comi>leinentaries shows. For bine we find in two of the six cirves aj;iiti an inclination towards a prefer- ence f< r thf complementary, but there .ire always other maxima as w'H, ati(' the colour in question, yellow, has no better claim than the next colour, yellow-yreen. For violet, violet-purple and pnrjile, yellow seeir •' •• ■• -'v ' — ■(■'• .- ' ,-<• 43 docs tlic in.ixiiniiiu c-incidc witli tliat msc ,,f ,„„ t,vcl\c .iu.ili- tio whit'h !•- iRMr(. th.il the trills •.vith >;[,cctr,.Iiv piiix- cnl^urs .-s^cii- tially snhstantialL- ilic UMilts .,I,t,iiiu alwavs.-ne sonirxehat related to the conipleinuiiarv, hut scarcely ever' the latter it>eif. Tims //trrr n :rry little jusHficatw,, I r t/„ ol,l dulum , e 8 o in o J5 >"*» o A i r. V 1 ' o CD u u ** •4 CI O 1 i 3 o o y. y. 'w"