■JJ0 
 
 
ainrncU IntoctattH ffiibrarg 
 
 attjaca, Siem lairk 
 
Cornell University Library 
 QP 481.S54 1920 
 
 Fovea! adaptation to, color. 
 
 3 1924 024 830 006 
 
Cornell University 
 Library 
 
 The original of this book is in 
 the Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924024830006 
 
FOVEAL ADAPTATION 
 TO COLOR 
 
 BY 
 
 HUBERT SHEPPARD 
 A THESIS 
 
 Presented to the Faculty of the Graduate School of 
 
 Cornell University in Partial Fulfillment of 
 
 the Requirements for the Degree of 
 
 DOCTOR OF PHILOSOPHY 
 
 Reprinted from The American Journai. of Pstcboloot 
 January 1920, Vol. XXXI, pp. 34-58 
 
FOVEAL ADAPTATION TO COLORE 
 
 By HuBBatT Sheppard 
 
 Contents 
 
 Page 
 
 I. Introduction: Historical 34 
 
 II. Experimental: Plan of the Experiment 36 
 
 III. Parti.: General Adaptation in Direct Simlight 37 
 
 Foveal Adaptation to Surface Colors (Preliminary 
 
 Series) 43 
 
 IV. Part II.: Foveal Adaptation to Surface Colors Under Con- 
 
 trolled Conditions 44 
 
 Chroma as a Determinant of Adaptation 49 
 
 V. Part III. : Foveal Adaptation to Spectral Colors 50 
 
 Ekiuation for Brightness 52 
 
 Chroma as a Determinant of Adaptation 52 
 
 Adaptation at High Intensities 55 
 
 Conclusions 58 
 
 Introduction 
 
 The primary aim of this study is to describe the course of 
 foveal adaptation to color. We have also obtained records 
 of the times of adaptation, and have attempted to show their 
 vairiation with variations of intensity and chroma; but our 
 main purpose throughout the ex{)eriment has been to secure 
 introspective reports of the attributive changes which occur 
 while the fovea or macula lutea is under stimulation by a color, 
 from the moment of exhibition until the color has disappeared 
 or until there is no further alteration of hue or tint or chroma. 
 The peripheral retina is discussed only in so far as its dis- 
 cussion bears upon adaptation at the fovea. 
 
 Wherever possible, all earlier experiments which seemed 
 to relate, directly or indirectly, to our problem were repeated, 
 and their results tested, before our own technique was finally 
 decided upon. Very little work has been done, however, that 
 bears directly upon foveal adaptation considered as a mode 
 of experience. Adaptation in general depends ( 1 ) upon wave- 
 length, wave-form, and wave-amplitude of light, and (2) 
 upon the time during which the light-waves affect the eye. 
 Hence the problem may be approached from the side of light- 
 dark adaptation, of color adaptation, or of negative after- 
 image. Previous investigators have studied the first and third 
 
 1 From the Psychological Laboratory of Cornell University. 
 
 34 
 
35 FOVEAL ADAPTATION TO COLOR 
 
 of these topics rather than the experienced course of color 
 adaptation. 
 
 The term adaptation, as applied to visual experiences, was first used 
 by Aubert, to denote " the accommodation of the eye to light-intensi- 
 ties."2 Nowhere in his writings does Aubert connect this term with 
 color-changes, for which he uses the term ' fatigue.' In his paper on 
 fatigue and retinal after-image,^ as elsewhere, he argues that there is 
 a close analogy of function between the fovea and the peripheral 
 retina. Colored objects, he says, appear colorless in both direct and 
 indirect vision, if the stimulus is sufficiently long continued. He is, 
 therefore, familiar with complete adaptation, but includes it under the 
 term ' fatigue.' 
 
 Exner, working under the direction of Helmholtz, made the first sys- 
 tematic study of adaptation (fatigue) to color-stimulation.* In an 
 experiment in which he describes the course of the after-image, he 
 finds that three parts of the spectrum are much less altered in ap- 
 pearance by ' fatigue ' than other parts, namely, R from the end of 
 the spectrum to a point between the lines C and D, G between E and 
 b, and B near G. He notes also that, if the after-image of a 
 color is projected upon another color of the same hue, this second 
 color appears much less saturated.'' 
 
 Von Kries investigated the problem of adaptation from both the 
 physical and the psychological points of view. He found that "with 
 long fixation, a color loses in brightness as well as in saturation."" 
 suffer no change in hue, while the other colors change in the direction 
 of F or 5 from G. 
 
 Hess made a study of the tone-changes of spectral colors when the 
 retina was fatigued with homogeneous light.'' He observed that all 
 colors possess either a 7 or a B valence. That is, all colors, from 
 near a pure i? to a pure Y, or near a pure G to a pure Y, will change 
 under adaptation towards Y; and all colors from near a pure i? to a 
 pure B, or near a pure G to a pure B, will change towards B. Further, 
 during the process of adaptation the stimulus-color always becomes 
 less and less saturated, until it disappears ; the complement of the 
 stimulus-color always remains unchanged in hue ; and the appearance 
 of a spectral color after adaptation corresponds in general to a mix- 
 ture of that color with its complement. 
 
 Burch was able to produce temporary color-blindness by means of 
 general adaptation at any intensity.* In an article entitled " On the 
 Production of Artificial Colour-blindness to Successive Contrast " he 
 describes the process of adaptation as complete for R, G, B and V. 
 In another experiment he was able to get complete adaptation in 3 
 min. when the intensity of the stimulating light was bright moonlight. 
 His method for very high intensities was as follows : The eye of an 
 
 2 H. Aubert, Physiol, d. Netehaut, 1865, 39 ff- 
 
 3 Moleschott's Untersuch., IV, 1858, 220 ff. 
 
 ■'J. A{tken,Roy. Scot. Sac. of Arts, Proc. 1871-2, and E. Hunt, Colour 
 Vision, Glasgow, 1892, made similar studies of adaptation. A detailed 
 account of their method and procedure is given later by Burch. 
 
 ^ S. Exner, 'Ueber einige neue subjective Gesichtserscheinungen,' 
 Arch. f. Physiol, I, 1868, 37S-394. 
 
 " Nagel's Handb. d. Physiol, d. Mensch., Ill, 1905, 213-220. 
 Further, K of a wave-length of 560 up., G of S0Ojti|U> and B of 460 mm 
 
 ' C. Hess, Arch. f. Ophth., 1890, XXXVI, Abtl. i, i ff. 
 
 8 G. J. Burch, Phil. Trans., 191 B, 1899, i-i3S- Roy. Soc. Proc, 66, 
 208 ff, 216-219. 
 
SHEPPARD 36 
 
 observer was exposed to bright sunlight, which had been focussed by 
 a burning-glass through a colored screen. After the eye had been 
 subjected to this intense light until all sensation of the particular color 
 was lost, the observer proved to be temporarily color-blind to the color 
 used in stimulation. Further, four parts of the spectrum, R from A 
 to B, G in the neighborhood of E, B about half-way between F and G, 
 and V at and beyond H, showed marked fatigue-effects, differing in 
 degree but not in kind. AH direct sensation of the stimulus-color was 
 temporarily lost: R for lo min., G for a still longer period, while the 
 recovery from F-blindness might require a couple of hours. When- 
 ever the eye is completely fatigued to any color, that color is absent 
 from the spectrum, whatever the fatiguing intensity may have been. 
 The observer is, in fact, color-blind for the particular color. After 
 fatiguing the eye with three of the four colors, Burch found that only 
 one color remained." 
 
 Troland has unpublished data from which he is able to determine a 
 theoretical course for the process of adaptation in all the adapting 
 sense departments. Retinal adaptation follows the same general course ' 
 as that of the other adapting organs.^" The rate of the process is 
 greater in the beginning, with a decrease rapid at first, then more 
 gradual, until it finally reaches a level at which it is constant.i^ Tro- 
 land finds also " that, at least for stimuli of fairly high intensity, 
 adaptation does not proceed to a point at which the resultant sensation 
 reduces to a neutral gray."i2 
 
 In sum, then, Exner, von Kries, and Hess come to almost exact 
 agreement with regard to the gross results of adaptation to color, (i) 
 The 'primary' colors, V SSO-S74mmi G 495-520 ma*, B 435-471 mm, do not 
 change in hue under the process of adaptation. Hess finds a psycho- 
 logical R outside of the solar spectrum which also does not change. 
 All other colors change either towards V or towards B. (2) A color, 
 after adaptation, corresponds in general to a mixture of this color 
 with its complement. 
 
 As regards adaptation to color at high intensity, there is difference 
 of opinion. Burch gets complete general adaptation with any intensity 
 which produces temporary color-blindness to the stimulus-color. Tro- 
 land, on the other hand, finds that color-adaptation at high intensity 
 does not reduce the color to a neutral gray. 
 
 Experimental: Plan of the Experiment 
 The present experimental work upon foveal adaptation to 
 color was begun in October, 1918, and ended in September, 
 1919. The observers were Dr. C. W. Bock (B), Mr. H. Shep- 
 pard (Sh) and Mr. M. J. Zigler (Z), instructors in psy- 
 chology, and Mr. H. Liddell (L), Miss A. K Sullivan (S), 
 and Miss M. Finck (F), graduate students in the department. 
 L, S and Z were given several weeks' training before any 
 data were taken for our records ; F was relatively unpractised. 
 
 A. W. Porter and F W. Edridge-Green, Proc. Roy. Soc, 85 B, 1912, 
 435 ff, find somewhat different results for B and G; but Burch, Proc. 
 Roy. Soc, 86 B, 1912-13, 117-18, argues that their work was not care- 
 fully done, and that their results are not strictly reliable. 
 
 10 L. T. Troland, Am. J. Psych., 25, 521 ff. 
 
 " Ibid.. 508. 
 
 12 Ibid., S20. 
 
37 FOVEAL ADAPTATION TO COLOR 
 
 Our object, as has been stated above, was to describe the 
 course of foveal adaptation to color under as many different 
 conditions as possible. We do not attempt a theory, nor do 
 we draw any theoretical conclusion whatever.^' We aim 
 only at a qualitative study of the experienced course of adap- 
 tation, with secondary reference to the times of adaptation 
 for the different colors used. 
 
 The experiment as a whole comprises three parts. In Part 
 I. we examine general adaptation to a surface color in direct 
 sunlight. Here the experienced course of adaptation is de- 
 scribed in its various stages. In this preliminary experiment 
 we also take account of foveal adaptation in a general and 
 tentative way. In Part II. we work with surface-colors, and 
 the process of adaptation takes place under certain controlled 
 conditions. These conditions greatly reduced eye movement, 
 and kept the illumination constant. Here we describe the 
 experienced course of foveal adaptation to 9 different colors, 
 R, O, Y, YG, G, BG, B, V, and C. In addition we consider 
 the effect of chroma upon adaptation. In Part III. pure spec- 
 tral colors are used with various intensities of illumination. 
 Here we describe the experienced course of foveal adaptation, 
 and note in what respect this process differs from the course 
 of adaptation when a surface-color is used. Difference of 
 chroma is also considered. Two colors, differing in hue, are 
 matched for chroma, and the process and time of adaptation 
 are recorded in each case. We also give the results of some 
 experiments made with high intensities of stimulus. 
 
 Part I. : Preliminary Experiment 
 
 GENERAL ADAPTATION IN DIRECT SUNLIGHT 
 
 We are here concerned to determine the course of general 
 adaptation when a surface-color, illuminated by direct sun- 
 light, stimulates the entire retina. In former experiments 
 peripheral adaptation to color has been found to be more rapid 
 than foveal.^* Our main problem throughout the preliminary 
 experiment was the study of the course of adaptation, by 
 introspective observation, as the process travelled over the 
 retina from periphery to fovea. If there is a rapid fading-out 
 of a color under peripheral stimulation, as is indicated in the 
 literature, it ought to be possible to observe this change in its 
 progress towards the central retina. 
 
 ^' Papers on ' An Experimental Treatment and Discussion of the 
 Theory of Adaptation and Color Vision ' as well as ' The Early 
 Development of Color Vision ' are now under consideration and will 
 follow shortly. 
 
 1* J. W. Baird, Color Sensithjity of the Peripheral Retina, ipos; 
 53-74- 
 
SHEPPARD 
 
 38 
 
 Procedure. — A drawing-board of 28 x 36 in. was fastened 
 in a window on the west side of the optics room. The board 
 served as a table upon which to place the stimulus-color. The 
 following papers were selected for stimuli in the preliminary 
 tests : I ( 
 
 Color 
 
 Brightness 
 Value 
 
 Match to 
 A Gray 
 
 Brightness Value 
 OF Match 
 
 
 Black White 
 
 
 Black White 
 
 R 
 BG 
 
 288° 72° 
 206 154 
 
 211° 
 149 
 
 88° 272° 
 
 Y 
 B 
 
 144 216 
 263 97 
 
 170 
 190 
 
 154 106 
 
 GY 
 
 V 
 
 162 198 
 285 75 
 
 125 
 235 
 
 112 248 
 
 C 
 G 
 B 
 
 259 101 
 250 110 
 
 208 
 132 
 
 97 263 
 
 
 G 
 
 B 
 
 216 144 
 
 92 
 120 
 148 
 
 112 248 
 
 A sheet of colored paper, 19 x 24 in., was spread upon 
 the table. In the center of the sheet was a very small light 
 dot for a fixation-point. O sat facing the stimulus with one 
 elbow resting upon the table and the chin in the palm of the 
 hand. This position brought the eye within about 25 cm. of 
 the fixation point, and was found to be convenient for fixation 
 in monocular vision. All the experimental work for this part 
 of the experiment was done during the months of July and 
 August, from 2 to 4:30 o'clock in the afternoon. We were 
 careful to select days when the sky was free from clouds or 
 haze." 
 
 The instruction to was to report all change in color sen- 
 sitivity, from the time the eye first fixated the stimulus until 
 the color either faded out to a gray, showing no trace of 
 color, or reached a stage where no further change in hue, tint, 
 
 IS It is very difficult to get complete adaptation to R, BG, B, C and 
 F in a reasonable length of time unless the sun is shining brightly. 
 The sky must be free from clouds or haze, or the times of adaptation 
 will vary considerably. 
 
39 FOVEAL ADAPTATION TO COLOR 
 
 or chroma was perceptible. was asked to give a running 
 report of all sensory change, and in as much detail as possible. 
 Results. — The results are given in summary for the different 
 observers and stimuli. 
 
 Observer B. Stimulus Y. — A gray film soon appeared in front of 
 the stimulus-color. This film came in from the periphery. There was 
 a slow change in chroma until the color faded out to a white or very 
 light gray. A small area surrounding the fixation-point persisted 
 longer than the other parts of the field. The change was slow at first, 
 then more rapid towards the last. 
 
 Stimulus BG. — The outer edge began to pale rapidly at first. This 
 paling extended inward towards the center. The center remained 
 richer in chroma than the outer edge for some time. Finally the 
 surface evened up into a gray field. 
 
 Stimulus B. — There was no change in hue. The haze appeared over 
 the surface from the outside, although very slowly as compared with 
 some of the other colors. The color persisted for a considerable time 
 around the fixation point. In the process of fading out there were 
 many bluish spots scattered about among white filmy spots. These 
 spots would even up into a white or light gray surface. 
 
 Stimulus R. — While the color was fading out from the periphery, 
 certain portions did not appear uniformly colored; little filmy red 
 dots were scattered about over the surface. Finally the red spots 
 paled out to gray, and then evened up over the whole surface. The 
 color around the fixation-point remained for some time as a dirty 
 brown color. The final gray surface was very dark. 
 
 Stimulus V. — There was a rough cloudy surface over the color for 
 a considerable time. The color began to pale out from the outer edge^ 
 and the change traveled towards the center. The roughness always 
 smoothed out into a gray film of a thick texture. It is very difficult 
 to tell just when adaptation is complete. 
 
 Stimulus 0. — It faded into a beautiful light yellow, beginning on 
 the outer edge and gradually decreasing towards the fixation-point. 
 From the Y it went into a gray which was somewhat darker than the 
 O. The steps were perceptible though gradual. All changes were from 
 the periphery towards the center. 
 
 Stimulus C. — Almost immediately the surface began to grow white 
 in the outer portion of the field. This white gradually approached the 
 center. The final stage was a lustrous silvery gray. 
 
 Stimulus G. — There was a slow change in hue from G through Y. 
 The Y finally changed into a very light gray, very much lighter than 
 the stimulus-color. 
 
 Stimulus GY. — The green faded out of the periphery very quickly. 
 This process always took place before the Y began to go. The Y then 
 changed into a very light gray. The surface of the stimulus-field at 
 times had a rough, uneven appearance. This gradually changed into 
 an even surface without any color. 
 
 Observer Sh. Stimulus Y. — Almost immediately the color began to 
 grow lighter and the chroma poorer. There was no change in hue. 
 As soon as the color began to fade a filmy surface appeared over the 
 surface of the color. The fading out process always came from with- 
 out towards the fixation-point. The periphery became white some time 
 before the fixation-point. The final stage was an almost white surface. 
 
 Stimulus BG. — There was a gradual, though rapid decrease in 
 saturation from the beginning. Also there appeared a very rough, 
 uneven surface about the fixation-point, which was much more satu- 
 
SHEPPARD 40 
 
 rated than the rest of the field. This area gradually evened up into 
 a gray. The hue changed slightly to B. 
 
 Stimulus B. — The blue faded out very slowly from the periphery 
 to the center. At all times the hue appeared to shine through a white 
 filmy surface which formed over the color. This effect persisted to 
 the end.18 The complete stage of adaptation was a very pretty whitish 
 gray, like an after-image of a velvet-black surface. There was no 
 hue-change whatever that could be observed. The entire process was 
 a slow fading away of the color into a filmy surface. Two cases of 
 staring blindness occurred during this series of observations. 
 
 Stimulus R. — The red is unlike any other color in adapting out to 
 a gray. It passes through different stages of hue-change from the first. 
 Almost immediately it passes over into a red-orange, then to an orange 
 of poor chroma, and finally into a dirty, dingy brown, which is very 
 persistent. This brown finally goes into a poor gray, then to a good 
 gray surface. 
 
 Stimulus V. — Violet is the most stubborn of all the pigment-colors 
 to a complete adaptation in sunlight. There is a feeling of uncer- 
 tainty as to the judgment, because the violet is almost of the same 
 brightness as the gray surface which persists when adaptation is 
 complete. (The two are so near alike that it is often difficult for O 
 to decide when the final stage is reached.) There is a rough cloudy 
 surface at first, which gradually smooths down with time. Two 
 changes in hue always take place. At first there is a tendency for 
 the red to persist; this presently adapts out, and leaves a bluish-violet 
 surface of very poor chroma. 
 
 Stimulus O. — Orange adapts out very much like red from the second 
 stage of the red, except that the orange is richer in chroma than the 
 red at this stage. When the color reaches the brown stage, it is 
 almost exactly like the same stage in the red, except that the orange 
 is brighter than the red. One effect of staring-blindness was observed, 
 which lasted for three or four seconds. 
 
 Stimulus C. — Almost, immediately the surface began to take on a 
 bright lustrous appearance. This effect gradually thickened, until the 
 whole surface was of a beautiful silver lustre. The hue changed some- 
 what to blue, thus causing the surface to become darker. 
 
 Stimulus G. — For the first few seconds the color did not appear to 
 change. Suddenly there is a change, the yellow appears and the green 
 seems to weaken. From this point there is a rapid fading 
 away. The yellow appears to go out first; the green, now 
 much less saturated, fades out into a beautiful gray surface 
 which is very smooth in texture. Two cases of staring blind- 
 ness were observed. 
 
 Stimulus GY. — The color is stubborn at first; the green suddenly 
 goes out; and then there is a gradual, though rapid, fading away of 
 the yellow to a beautiful gray surface. After the green passes away, 
 the process of adaptation is very much like that of the yellow. The 
 final stage, the gray surface, is similar to the same stage in the yellow 
 adaptation. 
 
 Observer Z. Stimulus Y. — The color began to fade in the per- 
 iphery almost immediately; this caused the center to appear much 
 lighter, and richer in chroma. The small center persisted for some time, 
 although slowly fading all the time. Finally the whole surface of the 
 stimulus-color became a uniform gray 
 
 i« To get the effect of the soft white surface, when adaptation is 
 complete, a Hering blue must be used. A paper that has a glossy 
 surface is very difficult to observe for this length of time. 
 
41 FOVEAL ADAPTATION TO COLOR 
 
 Stimulus BG. — There was no apparent change in hue. A thin filmy 
 mist came in from the edge of the paper, which caused all color to 
 disappear from the periphery. This colorless effect gradually went 
 towards the center, until the whole color disappeared. 
 
 Stimulus B. — No hue-change. The light or white foggy mist ap- 
 proached the center from the periphery until the whole surface was a 
 very poor B, almost gray. This state persisted for some seconds. 
 Finally all color faded away, leaving only a gray field. 
 
 Stimulus R. — There was always a decided hue-change. It went 
 first to a very poor yellowish O, then to a dirty yellow grray. This per- 
 sisted for some seconds. The final stage is a peculiar gray. It appears 
 to be a little darker than the stimulus-color. 
 
 Stimulus V. — There was a slight hue-change to B. The decrease 
 in chroma was marked at first However, the final state is difficult to 
 judge, since the final gray and the V, very poor in chroma, are much 
 alike. 
 
 Stimulus 0. — There was a change in hue to Y, poor in chroma. 
 This change began in the periphery, and traveled towards the center. 
 Finally the center went out, exactly like the Y. 
 
 Stimulus C. — This is a very stubborn color. It appears to be uneven 
 in surface during the adaptation-process. It changes from periphery 
 to center, but not so markedly as the other colors. The rough un- 
 even surface seems to smooth down into the dark gray film as it 
 thickens over the surface. There is a hue-change, but it is difficult to 
 observe in its course. It would appear to change to B, although this 
 can not be observed clearly. 
 
 Stimulus G. — Very slow change to Y. This change only comes 
 about when the color has lost nearly all chroma. The film thickens 
 from the periphery towards the center. This gradually conceals all 
 color. The resulting gray surface appears to be a little lighter than 
 the stimulus-color. 
 
 Stimulus GY. — The periphery began to fade very quickly. For 
 several seconds this part of the field looked like wool, while in the 
 center of the field there was a YG of very good chroma. Finally the 
 center began to fade out, rather suddenly, to a good gray surface. The 
 hue changed to Y, although the final stage came when the color was 
 very low in chroma. 
 
 Observer S. Stimulus Y. — The film came over the surface from the 
 periphery in 20 seconds. There was no apparent change in hue. The 
 final stage was a colorless paper. 
 
 Stimulus BG. — No change in hue. The surface began to fade out 
 from the periphery, all the time going towards the center. 
 
 Stimulus B. — Same process as the BG. The final gray was some- 
 what lighter than the stimulus-color. 
 
 Stimulus i?.— There was decided hue-change as the color began to 
 grow less saturated by the gray film which came in from the periphery 
 to the center. The color went through a poor O to a very weak 
 muddy Y. This last stage persisted for some seconds. 
 
 Stimulus V. — This is a very stubborn color. It is difficult to decide 
 when all the color has faded out. The film began to appear on the 
 periphery very quickly, moving to the center. The final stage appears 
 to be of about the same brightness as the stimulus-color. There is a 
 hue-change to B, although it is very difficult to judge. 
 
 Stimulus O. — The film began in the periphery. For a little, this 
 seemed to hang on the outer part of the color. Finally it moved over 
 
SHEPPARD 
 
 42 
 
 the center. At this time the hue changed to a very poor Y. Then the 
 process was the same as Y in the last stage. 
 
 Stimulus C. — The color made some change in hue, but it was so 
 slow and gradual that it was difficult to judge. There was a peculiar 
 uneven surface during the course of adaptation. The cloud appeared 
 from the periphery, moving to the center. This gradually weakened 
 the chroma. 
 
 Stimulus G. — A hue-change to yellow in the final stage. The chroma 
 gradually grew less, the center remaining longer. 
 
 Stmulus GY. — Same as G, except that the change to Y was more 
 marked. 
 
 The first part of the accompanying table shows the adapta- 
 tion-times under the conditions described. The second part 
 shows times obtained, still in direct sunlight, when a disc, 
 composed as indicated in the table, was viewed against a large 
 grey background. Observation in these circumstances was 
 difficult, owing to the lack of a fixation-point. 
 
 ADAPTATION TIMES, IN SECONDS. 
 
 AVERAGE OF FIVE OBSERVATIONS 
 
 6bs. 
 
 Sh 
 
 B 
 
 Z 
 
 S 
 
 
 Color 
 
 Av. 
 
 M.V. 
 
 Av. 
 
 M.V. 
 
 Av. 
 
 M.V. 
 
 Av. 
 
 M.V. 
 
 Av. of 
 Av. 
 
 R 
 
 
 
 Y 
 
 YG 
 
 G 
 
 BG 
 
 B 
 
 V 
 
 c 
 
 198.2 
 
 89.6 
 
 45.6 
 
 49.2 
 
 52.0 
 
 126.6 
 
 247.6 
 
 200.0 
 
 102.0 
 
 6.5 
 3.9 
 3.9 
 
 7.1 
 6.8 
 5.6 
 4.7 
 8.4 
 8.4 
 
 192.6 
 
 89.4 
 
 39.2 
 
 53.2 
 
 54.6 
 
 109.4 
 
 220.2 
 
 201.4 
 
 100.0 
 
 10.5 
 6.9 
 3.8 
 5.8 
 9.5 
 
 11.9 
 
 12.9 
 4.7 
 
 10.0 
 
 190.8 
 
 91.8 
 
 43.6 
 
 50.4 
 
 49.6 
 
 115.2 
 
 201.8 
 
 201.8 
 
 108.0 
 
 9.4 
 5.4 
 6.3 
 6.5 
 2.9 
 9.3 
 8.6 
 1.7 
 1.6 
 
 194.8 
 
 93.8 
 
 55.0 
 
 52.4 
 
 56.6 
 
 128.0 
 
 211.4 
 
 234.0 
 
 111.0 
 
 5.0 
 1.4 
 1.2 
 7.4 
 3.5 
 4.0 
 4.7 
 10.8 
 4.8 
 
 194.1 
 
 91.1 
 
 45.9 
 
 51.3 
 
 50.6 
 
 119.8 
 
 220.0 
 
 209.3 
 
 105.2 
 
 180° 
 
 OF Color + 180° of Black and White (Brightness- 
 
 
 
 
 Value) 
 
 
 
 R 
 
 93.8 
 
 7.0 
 
 
 104.0 
 
 8.6 
 
 96.8 
 
 12.6 
 
 98.2 
 
 O 
 
 42.0 
 
 5.2 
 
 
 45.5 
 
 3.2 
 
 45.2 
 
 4.2 
 
 44.2 
 
 Y 
 
 21.0 
 
 2.0 
 
 
 27.0 
 
 5.4 
 
 22.2 
 
 1.8 
 
 23.4 
 
 YG 
 
 25.4 
 
 4.7 
 
 
 29.0 
 
 2.1 
 
 26.0 
 
 4.8 
 
 26.8 
 
 G 
 
 26.0 
 
 7.8 
 
 
 23.5 
 
 6.6 
 
 21.6 
 
 4.3 
 
 23.7 
 
 BG 
 
 57.0 
 
 4.8 
 
 
 50.3 
 
 3.4 
 
 56.0 
 
 6.4 
 
 54.4 
 
 B 
 
 118.0 
 
 13.6 
 
 
 122.0 
 
 2.6 
 
 123.8 
 
 7.0 
 
 121.3 
 
 V 
 
 108.0 
 
 6.4 
 
 
 98.6 
 
 4.8 
 
 112.8 
 
 7.1 
 
 106.5 
 
 c 
 
 53.4 
 
 7.1 
 
 
 44.3 
 
 2.4 
 
 52.0 
 
 6.0 
 
 49.9 
 
43 
 
 FOVEAL ADAPTATION TO COLOR 
 
 FOVEAL ADAPTATION AT MEDIUM INTENSITY 
 PRELIMINARY SERIES 
 
 We next sought to determine' the experienced course of 
 adaptation to color, at medium intensity, when only the fovea 
 was stimulated by the color. 
 
 For stimuli, a circular disc, 100 cm. in diameter, was cut 
 from each one of the 9 colors shown in the preliminary experi- 
 ment. Every disc was mounted upon a gray background of 
 24 X 30 in. The background was free from any trace of color, 
 and had a brightness-value of 140° white and 220° black. A 
 frame on a table held the stimulus-card upright. The stimulus 
 was placed at a distance of 225 cm. from the eye of the obser- 
 ver; the image of the stimulus-color fell within the macula. 
 
 The experimental work was done in a large light optics- 
 room. Light window-shades were drawn over the windows, 
 so that the room was mildly illuminated with a soft white 
 light. The instructions to O were the same as in the prelimi- 
 nary experiment. 
 
 The quantitative results (times in sec.) are set forth in the 
 following table. The averages are of 5 observations. 
 
 Obs. 
 
 Sh. 
 
 S. 
 
 Color 
 
 Av. 
 
 M.V. 
 
 Av. 
 
 M.V. 
 
 R 
 
 137 
 
 7.5 
 
 178 
 
 8.1 
 
 O 
 
 140 
 
 6.5 
 
 140 
 
 8.4 
 
 Y 
 
 100 
 
 4.5 
 
 105 
 
 4.2 
 
 YG 
 
 88 
 
 4.5 
 
 95 
 
 7.0 
 
 G 
 
 80 
 
 5,5 
 
 110 
 
 6.2 
 
 BG 
 
 100 
 
 3.2 
 
 98 
 
 3.1 
 
 B 
 
 118 
 
 4.5 
 
 109 
 
 3.2 
 
 V 
 
 201 
 
 10.5 
 
 199 
 
 7.0 
 
 c 
 
 175 
 
 7.2 
 
 178 
 
 8.0 
 
 The reports of the observers may be summarized as follows : 
 
 R. The color began to lose in chroma very quickly. The outer 
 edge of the circle paled out first, the paling gradually approaching 
 the center. There was a little hue-change; the R faded slightly 
 towards a very poor O. The resultant gray was a little lighter than 
 the stimulus-color. 
 
 O. A thin gray film gradually thickened over the surface of the 
 color. This caused the chroma to become poorer with the progress 
 of adaptation. The hue changed slightly towards a Y, very poor in 
 chroma. 
 
 y. The hue faded away under the gray film until no color was per- 
 ceptible. There was no hue-change. 
 
 YG. There was a hue-change to Y, very poor in chroma. The 
 final stage was somewhat darker than the color. 
 
 G. The hue changed to Y, very poor in chroma. The gray film 
 
SHEPPARD 44 
 
 very quickly obscured the surface of the color, only a greenish gray 
 remained for some time at the last stage of adaptation. Slightly darker 
 than the stimulus-color. 
 
 BG. There was no perceptible hue-change, only a fading out to a 
 very pretty gray. The film began to come in from the edge of the 
 disc, gradually agptoaching the center of the color. The gray field 
 was a little darker than the stimulus-color. 
 
 V. The hue changed to R and then to B. The resultant gray was 
 lighter than the stimulus-color. 
 
 B. There was no perceptible hue-change in this color. In every 
 respect it faded out like the BG. 
 
 C. Gradually faded out to a B, very poor in chroma. This state 
 persisted for some time. Then the field was almost of the same 
 brightness as the stimulus-color. 
 
 It soon became evident that this experimental arrangement 
 was unsatisfactory. The observers complained of eye-move- 
 ment in almost every observation; and the difference in illu- 
 mination from day to day influenced the adaptation-time con- 
 siderably. Further, there is a marked difference between the 
 adaptation-times of the two observers for certain colors, 
 especially R and G. It was therefore necessary to secure 
 better conditions of observation. 
 
 Part II. : Foveal Adaptation at Medium Intensity 
 Controlled Series 
 
 Procedure.— The experimental work was done in a dark 
 room. The apparatus and stimuli were those described in 
 our account of the preliminary series. The illumination was 
 produced by three 75-watt, 110-volt, Edison-Mazda daylight 
 bulbs; it was equal to 212.4 c. p. at a distance of 2.5 m. from 
 the stimulus-color and a distance of 1 m. above and a little 
 in front of O. The current was supplied by a D. C. motor in 
 the university service. 
 
 In order to avoid eye-movement, as far as possible, the fol- 
 lowing precautions were taken. (1) A time-recording appar- 
 atus was constructed, to register accurately the time of every 
 change in visual sensation. This made it needless for O to 
 speak during an observation. The recording apparatus con- 
 sisted, first, of a kymograph, set to revolve once in 3 min. A 
 double electrical contact-marker worked over the smoked 
 drum. To one marker was attached in circuit a large seconds- 
 pendulum, while the other marker was in circuit with a tele- 
 graph-key operated by in reporting the sensory changes. (2) 
 A comfortable head-rest was fastened to the table in front of 
 the stimulus-frame. With this arrangement, could hold 
 the head in a constant position during an observation. 
 
 adjusted the head-rest, covered one eye with a black vel- 
 vet eye-shield, and made ready to fixate the stimulus when the 
 
45 FOVEAL ADAPTATION TO COLOR 
 
 light was turned on. The instructions were as follows: 
 " When the light is turned on, fixate the center of the colored 
 disc, and hold it if possible without moving the eye. If the 
 eye does shift from its fixation, bring it back to position as 
 quickly as possible. When the light is turned on and you have 
 fixated, press the telegraph-key five times in rapid succession. 
 Every time the eye moves in any way during fixation, press 
 the key once. For every color-change of any kind, press 
 the key twice. When the color fades out, or when it reaches 
 a point where it shows no further change, press the key again 
 five times. You will be asked to give an introspective report 
 of these changes at the end of every observation." 
 
 Results. — The results are given in summary for the differ- 
 ent observers and stimuli. 
 
 Observer S. Stimulus Y. — An after-image came very quickly. The 
 mist appeared from the outside, causing the color to become less 
 saturated and much lighter. As the mist grew thicker, there was a 
 tendency for it to fuse in the border before the center became solid; 
 thus the center was the last part to disappear. The cloud was much 
 lighter than the background. 
 
 Stimulus BG. — The after-image was noticed very quickly. Soon the 
 cloud gathered over the surface of the color like a fog or mist. This 
 caused the color to be much less saturated. For a little while I could 
 see the color through the fog as if it were behind the mist. The cloud 
 which covered the color was darker than the stimulus-color. 
 
 Stimulus B. — A soft white film began to form over the surface; at 
 about the same time I observed the after-image; in fact, the after- 
 image and the fog could be clearly recognized as one. The after- 
 image was merely superimposed over the color, and appeared to stand 
 out from the color. Where this haze fell on the outside edge of the 
 stimulus-color and on the bordering background, it took on the 
 nature of the after-image; where it was on the stimulus-color it was 
 a beautiful white fog or mist. As this mist or fog gradually grew 
 thicker, the color slowly lost in richness of chroma without any hue- 
 change. The resulting gray was of the same shade as the back- 
 ground. 
 
 Stimulus R. The gray film was peculiar in this color. It came in 
 from the outside. The color appeared very dark through it. Th« 
 mist was not smooth, but rough, very much like hazy clouds on a 
 summer afternoon. It never had a smooth texture, like the blue, 
 yellow or orange. At times I noticed shooting phenomena from the 
 outside, like smaller clouds suddenly moving across the surface of 
 the disc. 
 
 Stimulus V. — The cloud suddenly came in from the outside, at 
 about the same time as the after-image was noticed. It was very hazy 
 at first, but slowly thickened to a very dense mist. The color did not 
 seem to change in hue in any way, but the haze seemed to thicken 
 and conceal it from view. The cloudy gray was somewhat darker than 
 the background. 
 
 Stimulus 0. — The after-image was seen very quickly, bordering the 
 color almost all the wj.y around ; with this came a very peculiar cloud, 
 which seemed to be a part of the after-image, extending over the 
 color, hiding the color. For a while the view resembled the sun 
 behind a luminous fog. 
 
SHEPPARD 46 
 
 Stimulus C. — The color became almost a bright red behind the thin 
 misty cloud, when the cloud would move a little to one side. The 
 after-image and the cloud could be recognized as the same surface, 
 floating over the surface as the eye moved. The cloud was smooth 
 and not rough like the red. There were shooting phenomena as with 
 the red. 
 
 Stimulus G. — The cloud appeared about the time the after-image 
 was observed. When the haze became very thick, there was noticed 
 a little yellow mixed with the misty surface. This was probably a 
 hue-change of the stimulus-color. The cloud in the beginning seemed 
 to move in from the outside, growing thicker towards the center. 
 
 Stimulus GY. — The cloud appeared as if in front of the stimulus- 
 color in the form of a light hazy mist. It became thicker and 
 thicker until the color was obscured. The color could be seen for a 
 considerable time through the fog; as the mist grew thicker, the 
 color appeared to lose in richness of chroma, until only a gray of about 
 the same brightness as the background remained. 
 
 Observer Sh. Stimulus Y. — An after-image came very quickly, 
 which extended all the way around the stimulus-disc. The after- 
 image was at first very poor in chroma. Almost at the same time a 
 whitish film began to form over the surface of the color. As it 
 thickened, it appeared to be of the same brightness as the back- 
 ground. This gray surface began as a very thin sheet moving over 
 the surface. At times it appeared as threads or filaments, stretching 
 across the surface, as if they were in motion; these thickened into a 
 solid uniform surface. 
 
 Stimulus BG. — There was a short interval of time before the after- 
 image was seen on the border around the stimulus-color. The misty 
 haze about the same time was observed to move inward over the 
 surface of the color. The film soon thickened in texture so as to 
 obscure the color. 
 
 Stimulus B. — The after-image began to show very quickly around 
 the edge of the stimulus. The haze began to form almost as quickly, 
 and thickened into a very white misty film. This stood out away 
 from the color, and at no time did it appear to be a part of the stim- 
 ulus-color. The color gradually lost in richness of chroma as the film 
 thickened. When the cloud had taken on a filmy or sheet-like ap- 
 pearance, it was very easy to see that it was an extension of the after- 
 image over the stimulus-color, as a misty white luminous fog. It 
 looked very much like a thin sheet of tissue-paper spread over the 
 color; the film at all times was in front of the color and not a part 
 of the color. The whole adaptation-process for the blue was a fading 
 out or disappearing of the color behind the fog. 
 
 Stimulus R. — The after-image appeared very quickly as a bluish- 
 green border on one side of the disc. At first it was pale, or washed 
 out. but it gradually increased in chroma until it was very rich. At 
 the same time a haze began to form over the stimulus-color, which 
 gradually grew thicker. The haze appeared to come from the out- 
 side of the disc and to settle in front of the color. This haze for the 
 red was a rough, uneven mist or fog; not smooth in texture like that 
 of the yellow or orange. The hue showed a remarkable change during 
 the course of adaptation. It changed almost as soon as the haze 
 began to form over the color into a red-orange, then to an orange of 
 very poor chroma, and then to a very dirty, dingy brown. This 
 brown was persistent for some time; finally it changed into a gray 
 surface which was somewhat darker than the background. 
 
 Stimulus V. — A very pale, faded after-image soon appeared from 
 
47 FOVEAL ADAPTATION TO COLOR 
 
 behind the stimulus-color, scarcely visible on the gray surface sur- 
 rounding the stimulus. The cloud or fog seemed to float in from the 
 side of the color towards the center; not so pronounced as with the 
 red. The film at first was thin and more solid than with the red, but 
 not so smooth in texture as with the blue or orange. (In three 
 observations there was a slight hue-change towards blue, the red 
 going out first. In two there was only a gradual fading out of the 
 color into the fog). 
 
 Stimulus O. — The pale blue of the after-image, which was seen 
 almost from the first, soon increased in chroma until it was a rich, 
 well saturated blue. At the same time a beautiful hazy mist was 
 gradually thickening over the color. This film appeared very much like 
 the sun behind a thin fog. An interesting phenomenon occurred with 
 a shift of the eye. There could always be observed two distinct discs 
 in the visual field; the one was the stimulus-color, and the other was 
 the film or fog in front of the color. These two discs, when not 
 superimposed, the cloud upon the stimulus-color, always gave three 
 parts of a circle. The part of the luminous fog which was upon the 
 background showed a very rich after-jmage. The segment of the 
 stimulus-color which was uncovered by the fog, caused by a shift 
 of the eye, was a rich bright color, of the same hue as the stimulus- 
 color ; its brightness always depended upon the brightness-value of 
 the background with reference to the brightness of the stimulus-color. 
 The third portion was that part of the stimulus-color upon which a 
 part of the cloud was superimposed. This was a beautiful gray, in 
 which no color could be perceived in the final stage of adaptation. 
 
 Stimulus C. — The after-image was slow in appearing. The cloud 
 effect was generally noticed as soon as the after-image. A change 
 in brightness is always noticed as the filmy surface begins to form 
 over the color. At this point the observer can always tell when adap- 
 tation is really progressing rapidly, because the color begins to pale 
 out very rapidly. The luminous fog was always smooth, gradually 
 thickening from a filmy surface to a veil very much like a thin sheet 
 of paper spread over the surface of the color. This finally becajne 
 so thick in texture that the color was obscured. 
 
 Stimulus G. — The cloudy mist began to form over the surface of 
 the color by floating in from the outside towards the center of the 
 disc. This movement and the after-image came at the same time. 
 This haze grew thicker, which caused the color to become less and 
 less saturated until only a gray surface was left. Generally there was 
 a slight change in hue towards yellow.^' The yellow would appear 
 as if it were in the foggy veil which floated before the stimulus-color. 
 
 Stimulus GY. — The after-image began as a very pale, washed-out 
 carmine. It extended all the way around the stimulus disc. With an 
 increase in the richness of the chroma of the after-image also came 
 a thickening of the hazy film which formed with the after-image. 
 The cloud was apparently of the same brightness as the background. 
 In nearly every case the green was observed to adapt out first, leaving 
 a dirty, muddy surface which remained as a stubborn color for a 
 considerable time. 
 
 Observer L. Stimulus Y. — Soon a film formed over the surface 
 of the color, which did not appear uniform at first, but thicker in 
 spots. The color began to lose in saturation from the first, changing 
 its hue to a tan or very light unsaturated yellow. The haze finally 
 grew so thick that the color was obscured from view. 
 
 IT This was no doubt due to the fact that the Milton-Bradley G 
 contains a slight amount of Y. 
 
SHEPPARD 
 
 48 
 
 Stimulus SC— The after-image and the cloud-effect appeared very 
 quickly and at almost the same instant. The fog was very light and 
 fluffy. It stood out in front of the color. The hue of the color 
 gradually faded away behind this white mist as the fog became 
 thicker. The final stage of adaptation gave a gray surface equal in 
 brightness to the background. 
 
 Stimulus B. — The after-image was quickly observed; at about the 
 same time the misty cloud spread quickly over the surface of the 
 color. The fog was almost pure white and appeared to fit the stitn- 
 ulus-color. It seemed to stand out away from the color. This mist 
 gradually thickened into a thick cloud which hid the color of the 
 disc. 
 
 Stimulus R. — The cloud floated in over the surface of the color 
 from the outer edge of the disc. It was not smooth in texture, but 
 somewhat rough and uneven. The red disc changed at first into a 
 kind of orange, then to a yellowish color of very poor chroma, and 
 finally faded out to a dark gray. 
 
 Stimulus V. — iThe after-image appeared very bright from the 
 beginning. The misty haze is somewhat darker than the background. 
 It gradually thickened over the color. It is difficult to tell the final 
 gray, when adaptation is complete, from the violet when it is very 
 poorly saturated. 
 
 Stimulus O. — The haze over the color and the after-image around 
 the stimulus appeared at the same time. The cloud thickened and 
 obscured the color from view. For some time the color could be seen 
 as if shining through the fog. The final gray was of the same bright- 
 ness as the background. 
 
 Stimulus C. — The after-image was at first dim and slow in ap- 
 pearing. The cloud was very thin at first and somewhat darker than 
 the background. The color gradually paled away in the fog. 
 
 Stimulus G. — A very weak after-image appeared at first, which 
 gradually increased in chroma. A misty cloud came in from the out- 
 side of the color towards the center. This haze increased in thick- 
 ness until it obscured the stimulus-color. It is of the same brightness 
 as the background. 
 
 Stimulus GY. — Cloud-effect appeared very quickly. This caused 
 the color to become very much less saturated, as if a thin sheet of 
 white paper had been suddenly spread over the surface. The mist 
 thickened so as completely to hide the color. The final gray is the 
 same as the background. 
 
 ADAPTATION TIMES IN SECONDS 
 
 AVERAGE OF FIVE OBSERVATIONS 
 
 Obs. 
 
 ^ 
 
 Sh 
 
 L 
 
 
 Color 
 
 Av. 
 
 M.V. 
 
 Av. M.V. 
 
 Av. 
 
 M.V. 
 
 Av. of 
 
 
 
 
 
 
 
 
 Av. 
 
 R 
 
 84.8 
 
 7.0 
 
 91.4 
 
 7.9 
 
 98.0 
 
 6.4 
 
 91.4 
 
 O 
 
 88.2 
 
 7.4 
 
 92.2 
 
 5.4 
 
 92.4 
 
 11.1 
 
 90.1 
 
 Y 
 
 97.6 
 
 5.9 
 
 90.8 
 
 3.8 
 
 98.8 
 
 7.4 
 
 95.7 
 
 YG 
 
 62.0 
 
 6.4 
 
 63.2 
 
 3.4 
 
 58.4 
 
 7.5 
 
 61.2 
 
 G 
 
 55.2 
 
 1.3 
 
 58.4 
 
 3.5 
 
 55.0 
 
 3.2 
 
 56.2 
 
 BG 
 
 42.0 
 
 5.2 
 
 39.8 
 
 4.2 
 
 39.2 
 
 3,0 
 
 40.3 
 
 B 
 
 125.0 
 
 7.2 
 
 107.2 
 
 12.6 
 
 128.4 
 
 4.9 
 
 120.2 
 
 V 
 
 82.4 
 
 2.4 
 
 84.0 
 
 3.2 
 
 78.0 
 
 5.6 
 
 81.3 
 
 c 
 
 73.2 
 
 6.2 
 
 73.4 
 
 7.3 
 
 75.4 
 
 6.9 
 
 74.0 
 
49 
 
 FOVEAL ADAPTATION TO COLOR 
 
 The curve for the times of adaptation in sunlight differs 
 considerably from the curve for the same colors at medium 
 intensity. R is relatively shorter at medium intensity, while 
 B is slightly longer than B in sunlight. On the other hand, 
 the Y and G regions of the two curves take almost opposite 
 positions. There is not only a rapid rise in the Y region for 
 medium intensity, but there is a shift of the shortest time to 
 the BG region. This difference might evidently be due to a 
 difference in the chroma of the Y and G regions, caused by 
 the two sources of illumination.^* We therefore determined 
 the chromatic limen of observer Sh for every color under both 
 sources of illumination. 
 
 Procedure. — A color-mixer was placed in the window, on 
 the table used in the preliminary experiment, so that the discs 
 should revolve in a horizontal position. The disc was made up 
 of a black-velvet paper, a good white, and a colored paper. The 
 relative proportion of the brightness-value of the color was kept 
 constant by varying the black and white discs as the color com- 
 ponent was increased or decreased. The method of limits was 
 employed (five series). O did not see the disc until it was in 
 motion, and then only long enough to make a judgment. The 
 experiment was repeated in the dark-room with the illumina- 
 tion used for medium intensity, and also in diffused daylight. 
 
 The results are shown in the following table : 
 
 CHROMATIC LIMENS 
 
 COLOR 
 
 SUNLIGHT 
 
 DARK ROOM ILLUMINA- 
 TION 
 
 DIFFUSED DAYLIGHT 
 
 
 Av. 
 
 M.V. 
 
 Av. 
 
 M.V. 
 
 Av. 
 
 M.V. 
 
 R 
 
 18.0 
 
 3.6 
 
 22.0 
 
 6.4 
 
 19.3 
 
 1.6 
 
 
 
 34.0 
 
 3.8 
 
 20.0 
 
 6.2 
 
 8.0 
 
 1.3 
 
 Y 
 
 55.0 
 
 6.1 
 
 18.0 
 
 3.8 
 
 19.7 
 
 2.5 
 
 YG 
 
 46.0 
 
 4.3 
 
 28.0 
 
 5.2 
 
 23.0 
 
 3.6 
 
 G 
 
 37.0 
 
 4.3 
 
 32.0 
 
 4.3 
 
 22.2 
 
 2.3 
 
 BG 
 
 30.0 
 
 2.2 
 
 40.0 
 
 5.4 
 
 19.0 
 
 1.6 
 
 B 
 
 12.0 
 
 3.2 
 
 12.0 
 
 1.1 
 
 11.0 
 
 1.6 
 
 V 
 
 22.0 
 
 4.6 
 
 30.0 
 
 3.6 
 
 33.0 
 
 6.3 
 
 C 
 
 26.0 
 
 6.3 
 
 25.0 
 
 4.1 
 
 13.3 
 
 3.1 
 
 *' The gray background upon which the stimulus-color was mounted, 
 and which tested free from color under the daylight electric-bulb 
 illumination, gave the following composition when placed in direct 
 sunlight illumination: Black 215°, White 125°, B 12°, and G 8°. 
 
SHEPPARD 50 
 
 There can be no doubt as to the effect of chroma on the time 
 of adaptation in direct sunlight, and in illumination used in 
 the dark-room experiment. The proportion, through the 
 whole series of nine colors, runs in the right direction. 
 Wherever the adaptation-time is long, the chromatic Umen 
 is small, and conversely. For diffused daylight, there is a 
 slight variation; R and V are out of proportion. This ex- 
 ception may be due to the fact that R and V are the dark 
 colors; since the illumination was rather weak, the limen 
 would tend to be high. For all the lighter colors there is 
 a close approximation to the law stated above, that where the 
 time is high the limen is low. We shall later find that the 
 same thing holds for spectral colors; chroma is the principal 
 factor which influences the time of adaptation, with a possi- 
 bility of intensity as a minor factor. 
 
 Part III.: Pure Spectral Colors 
 We proceed to describe the course of foveal adaptation to 
 pure spectral colors, and to compare the behavior of these 
 colors with that of surface-colors. We were careful to present 
 the spectral color to O under controlled conditions. The wave- 
 length of the spectral band producing the stimulus was de- 
 termined before every observation; the intensity of the light 
 producing the spectral color was always known; and the 
 amount of light entering the eye was kept constant by the use 
 of an artificial pupil. 
 
 Procedure. — The apparatus consisted of a remodeled spec- 
 trophotometer.^* In addition to the original equipment of the 
 apparatus, we inserted a field in front of the lens, so that any 
 desired width of spectral band could be obtained in the form 
 of a circular disc of color. An artificial pupil, 2.5 mm. in di- 
 ameter, was placed 14 cm. in front of the stimulus field, and 
 2 mm. from the eye. A small projection-lantern, burning a 
 200 w. 115 V. Mazda nitrogen-filled daylight bulb, illuminated 
 the spectroscope to produce the spectral colors. Before the 
 lantern was placed a plate of thin and very finely ground glass 
 to cut out the image of the filament in the light-bulb. The 
 amount of light that entered the collimator of the spectro- 
 scope was 61.4 c. p. 
 
 The observations were made in the dark-room. All the 
 light for illuminating the spectroscope was thoroughly con- 
 cealed, so that only the color through the artificial pupil could 
 be observed. A head-rest was set in front of the spectro- 
 
 1" Manufactured by the Cambridge Scientific Instrument Co. 
 
51 
 
 FOVEAL ADAPTATION TO COLOR 
 
 scope.'" All changes in color-sensation were recorded in sec- 
 onds by means of the apparatus described in Part II. was 
 given 15 min. of dark-adaptation before beginning the experi- 
 mental work, and a period of 5 min. was allowed for rest be- 
 tween observations. Since monocular vision was employed, 
 and the eyes were alternated from observation to observation, 
 there was a period of at least 10 min. before an eye was re- 
 stimulated. 
 
 For stimuli, circular fields were cut from the spectral band 
 having the following wave-lengths ; R 740-770 fi/i, Y 545-575 
 /li/j,, BG 495-525 tifi, B 440-470 /xfi, V 405-435 /i^it. The field was 
 6 mm. in diameter, and was just large enough to fall well 
 within the macula lutea. The constancy of the light could be 
 tested at any moment by means of a switch which turned the 
 current through voltmeter and ammeter. 
 
 The results appear in the following table, as averages of five observations. 
 
 Adaptation Times in Sec 
 
 Obs. 
 
 Sh. 
 
 S 
 
 Z 
 
 F 
 
 L 
 
 Av. 
 of Av. 
 
 Color 
 
 R21 
 
 Y 
 
 BG 
 
 B 
 
 V 
 
 Av. Mv. 
 
 102 4 
 166 10 
 119 9 
 
 103 6 
 69 8 
 
 Av. Mv. 
 
 103 6 
 164 10 
 126 10 
 
 104 2 
 69 9 
 
 Av. Mv. 
 
 110 9 
 180 21 
 132 6 
 120 4 
 78 4 
 
 Av. Mv. 
 
 98 7 
 
 Failed 
 
 164 16 
 
 121 10 
 
 90 6 
 
 Av. Mv. 
 
 109 13 
 
 166 7 
 
 117 3 
 
 90 6 
 
 77 6 
 
 104.4 
 169.0 
 131.6 
 107.6 
 76.6 
 
 Observer F failed to get complete adaptation for Y at this intensity. 
 In explanation we can only state that she had had less practice than 
 the other O's when she began the observations. All the others re- 
 ported complete adaptation for all stimuli used. 
 
 The curve of the adaptation-times rises from R, reaches its 
 maximum at Y, and gradually descends to V. This is almost an 
 exact figure of the brightness-curve for spectral colors. *' 
 
 20 It was found at the beginning that any movement, even that of 
 winking the eye, influenced the time of adaptation. The head-rest 
 enabled the observer to hold the head in one position, leaving only 
 steady fixation to be attended to. After a little practice with the 
 artificial pupil, a good observer has no trouble in holding an observa- 
 tion period, often without even winking the eye. 
 
 21 A psychological R was made up from gelatins. It was composed 
 of i? and B gelatins so combined that no hue-change was observed 
 during the process of adaptation. The time of adaptation was : Sh 
 100, Mv. 3.2; S 104.2, Mv. 2.1; Z 104.9, Mv. 6.4; L 98.9, Mv. 11.2. 
 F did not observe in this part of the series. 
 
 22 W. de W. Abney, Researches in Color Vision, 78-1 11. 
 
SHEPPARD 
 
 52 
 
 Hence we may have recourse to two factors, to explain the 
 difference in the curves for surface-colors and for pure spec- 
 tral colors : brightness and chroma. To test the first factor, we 
 equated all the colors for brightness-value and then repeated 
 the series, with the following result: 
 
 Adaptation Times in Sec 
 
 Color 
 
 Sh. 
 
 S 
 
 L 
 
 
 Av. 
 
 Mv. 
 
 Av. 
 
 Mv. 
 
 Av. 
 
 Mv. 
 
 R 
 
 96 
 
 5.4 
 
 82 
 
 4.5 
 
 105 
 
 3.1 
 
 Y 
 
 115 
 
 6.6 
 
 120 
 
 3.0 
 
 130 
 
 6.9 
 
 BG 
 
 97 
 
 5.5 
 
 100 
 
 6.7 
 
 96 
 
 4.3 
 
 B 
 
 59 
 
 3.1 
 
 68 
 
 6.5 
 
 65 
 
 1.1 
 
 V 
 
 68 
 
 3.1 
 
 52 
 
 5.0 
 
 68 
 
 2.3 
 
 It is clear that an equation for brightness does not altogether 
 remove our difficulty. To test the second factor, namely, the 
 possible effect of chroma, we arranged a series of experiments 
 as follows. Three color-fields were used: 4.5 mm., 3 mm., 
 and 2 mm. in diameter. Since V has the lowest adaptation- 
 time, a stimulus of this wave-length and brightness was set 
 up on a second spectroscope. White light was added to an- 
 other color in the spectrophotometer until the observer judged 
 the two colors equal in chroma. When this point was reached, 
 the observer was given 15 min. of dark-adaptation, and then 
 observations were made with the three fields; the same colors 
 were employed as in the first part of this experiment. The fol- 
 lowing are the averages for 10 observations by every ob- 
 server on every field : 
 
 
 
 3mm 
 
 Field 
 
 
 
 Obs. 
 
 Sh. 
 
 S 
 
 Z 
 
 F 
 
 L 
 
 Color 
 
 Av. M.V. 
 
 Av. M.V. 
 
 Av. M.V. 
 
 Av. M.V. 
 
 Av. M.V. 
 
 R 
 Y 
 BG 
 B 
 
 V 
 
 47.2 5.4 
 47.2 3.4 
 47.2 2.6 
 41.8 1.8 
 44.4 2.4 
 
 63.4 3.3 
 66.4 2.3 
 65.2 7.4 
 57.8 5.4 
 63.0 4.4 
 
 48.0 6.0 
 48.2 5.1 
 46.0 4.4 
 45.6 3.4 
 46.2 4.2 
 
 62.4 4.8 
 
 62.8 9.6 
 
 60.5 3.1 
 
 61.9 3.1 
 
 62.6 8.2 
 
 67.0 9.2 
 63.8 5.4 
 74.2 4.8 
 70.0 2.8 
 67.8 3.4 
 
53 
 
 FOVEAL ADAPTATION TO COLOR 
 4.5mm. Field 
 
 R 
 
 64.4 6.7 
 
 57.8 3.5 
 
 59.2 4.6 
 
 64.4 5.3 
 
 55.0 9.6 
 
 Y 
 
 46.0 4.4 
 
 75.8 5.0 
 
 46.0 4.0 
 
 50.2 1.9 
 
 74.2 6.5 
 
 BG 
 
 47.4 6.8 
 
 73.4 4.5 
 
 43.6 3.7 
 
 89.0 1.2 
 
 80.8 3.8 
 
 B 
 
 54.6 6.8 
 
 63.4 8.8 
 
 46.6 7.1 
 
 71.6 7.9 
 
 67.0 6.4 
 
 V 
 
 46.0 3.2 
 
 63.8 2.6 
 
 44.8 1.8 
 
 106.0 8.8 
 
 64.4 8.6 
 
 2mm. Field 
 
 R 
 
 50.6 5.7 
 
 62.0 4.8 
 
 46.2 3.8 
 
 55.1 6.3 
 
 64.2 10.6 
 
 Y 
 
 62.6 1.9 
 
 59.4 8.3 
 
 53.8 3.8 
 
 61.2 6.8 
 
 64.8 7.4 
 
 BG 
 
 47.8 7.0 
 
 56.6 10.0 
 
 50.6 5.9 
 
 70.0 10.2 
 
 65.0 5.2 
 
 B 
 
 52.8 9.8 
 
 69.2 3.0 
 
 55.0 4.4 
 
 51.1 7.1 
 
 63.8 5.6 
 
 V 
 
 57.0 2.0 
 
 58.6 8.5 
 
 53.4 4.7 
 
 52.4 6.4 
 
 67.2 14.1 
 
 The introspective reports may be summarized as follows : 
 
 Observer Z. Y. — After about 30 sec. I observed a thin filmy cloud 
 which began to form on the rim of the circle. The cloud was darker 
 than the Y, and contained some color. This film became thicker and 
 darker until there was no color left. The Y under the film always 
 became brighter, paler, and gradually took on a washed-out appear- 
 ance. 
 
 R. — The red began to get lighter. The hue began soon to change 
 to a very good orange. This began about the time a thin film was 
 forming over the color. The filmy surface gradually grew darker and 
 lost color, until a good gray surface covered the color. At the same 
 time the hue changed from O to Y, and then finally to a gray. 
 
 G. — The green began to fade out to a F as soon as it began to 
 lose chroma. The loss of chroma began along the border of the 
 colored field. I could recognize the after-image, when the eye moved, 
 as being a portion of the filmy surface over the color. The film was 
 a soft white cloud. The Y then faded out to a very good gray. 
 
 B. — The whole surface began to brighten up together with a re- 
 duction in chroma. There was no change in hue. The chroma grad- 
 ually faded out to a good light gray. This process was like a cloud 
 gradually thickening so as to obscure a color beneath it. 
 
 y. — The V began to lose chroma very quickly ; at the same time 
 there was a change in hue; first the red faded very quickly to a 
 certain degree, then the blue, and finally both faded out together. 
 This loss of chroma was again in the form of a white film collecting 
 over the surface of the color. Began on the border of the field, and 
 then spread to the center, all the time growing thicker until the color 
 was obscured. 
 
 Observer S. Y. — There was no change in hue. In the beginning 
 of observation there appeared a light ring around the edge of the 
 stimulus-color due to contrast. The color gradually paled out to a 
 good light gray surface. 
 
 R. — The red went immediately to a very pretty orange, and from 
 this to a yellow. A dark gray cloud appeared in the center of the 
 disc. This gradually spread out over the surface of the color, and 
 covered the whole disc. 
 
SHEPPARD 54 
 
 G. — There was no hue-change ; only paled out to a good gray 
 surface. The final stage was almost a white. The cloud first ap- 
 peared in the center of the disc and spread over the whole surface of 
 the stimulus-color. 
 
 B. — The blue did not change in hue, but lost in chroma, until a 
 very pretty gray surface remained. The color gradually grew lighter 
 as it lost in chroma. There appeared a gray cloud in the center of 
 the disc which spread over the entire surface. 
 
 V. — The hue of the violet changed, first to blue, losing the red; 
 then the bluish-violet lost in chroma, until a very dark gray surface 
 remained. The cloud appeared in the center of the disc and spread 
 over the surface of the color. 
 
 Observer Sh. Y. — A whitish film began immediately to collect over 
 the surface of the stimulus-color. It appeared to be a little brighter 
 than the color. The film gradually grew thicker until it obscured all 
 the color, leaving only a light gray surface. There was no percep- 
 tible hue change. 
 
 R. — The red went through a color-change, first to orange, then 
 through O to yellow, and finally to a dark gray. A film was observed 
 forming over the surface of the color at the same time it began to 
 change in color. The film began to form in the center of the stim- 
 ulus-disc, and gradually spread outward, covering the whole disc. 
 
 G. — There was a change in brightness at the beginning. A bright 
 ring was noticed around the edge of the field. This extended inward 
 towards the center very quickly. Then a light film began to form over 
 the surface of the color. This gradually grew thicker until all color 
 was obscured. There was no hue-change. 
 
 B — There was a decided change in brightness. A light film formed 
 immediately over the color. This grew thicker until all color was 
 obscured. The resulting gray is almost a white. 
 
 V. — There was a change in hue towards the blue. The red faded 
 out first; at the same time there was a loss in chroma, by a thin 
 cloud forming over the surface. A very dark gray finally formed over 
 the color. 
 
 Observer L. Y. — The color became a little lighter, and lost in 
 chroma until it became a light gray. There was no hue-change. 
 
 R. — The R changed to R-0, 0, Y, and finally to a dark gray. 
 
 G. — There was no hue-change. A cloudy mist began to form in 
 the center of the circular field, and spread over the entire surface, 
 which decreased the color in chroma until it was completely faded 
 out. 
 
 B. — There was a gradual loss in chroma; no hue-change. A cloud- 
 mist formed immediately over the surface of the color, which grad- 
 ually thickened until no color was noticed. The final stage was almost 
 white. 
 
 V. — There was a change in hue to a violet-blue color. The red 
 almost faded out first; then the color was poor in chroma. The 
 final stage was a very dark gray. 
 
 Observer F. Y. — The color grew paler by losing in chroma. There 
 was no hue-change. 
 
 R. — There was decided hue-change in this color. It went to O, 
 then to a very poor YO, which reminded me of a dirty Y. Finally 
 lost in chroma, until a gray resulted which was darker than the color. 
 
 G. — The G lost in chroma by paling away. No hue-change. 
 
 B. — Same as the green. 
 
 V. — The V changed to R and then to B, all the time losing in 
 chroma. 
 
55 FOVEAL ADAPTATION TO COLOR 
 
 Since a chroma-match between two colors that differ in hue 
 is only a relative matter, there will always be some variation. 
 So far, therefore, as time of adaptation is concerned, we have 
 every reason to be satisfied with our results. The averages 
 for the 3 mm. field are, it is true, more uniform than those for 
 the other two; the 4.5 mm. and the 2 mm. fields show slight 
 fluctuations. But no variation is as great as with the bright- 
 ness-equation. 
 
 From the introspective reports we find that there is no change 
 in hue except for R and V. All the other colors which we used 
 faded away steadily to a gray field. The adaptation-process 
 makes its appearance in the form of a film or mist over the 
 color. This causes the color to become less and less saturated, 
 exactly as was the case with the surface colors. When there 
 is a shift of the eye from fixation, O always reports three con- 
 tinuous portions of the field : a crescent of after-image ; on the 
 opposite side to this, a crescent of stimulus-color, very rich in 
 chroma; and between the two crescents a white or gray sur- 
 face over the stimulus-color. 
 
 ADAPTATION AT HIGH INTENSITY 
 
 It was found, in the first series with spectral colors, that 
 foveal adaptation would not proceed to a neutral gray when 
 the light that entered the collimator of the spectroscope was 
 raised to more than 60 c. p. At the same time three important 
 facts were observed in regard to foveal adaptation at high 
 intensity. (1) All our records show that there is a tendency 
 for eye-movement to increase very rapidly at the end of a 
 fixation-period, in some cases as much as 4 or 5 times the 
 amount at the beginning of a period. (2) Our O's report that 
 adaptation is always very rapid at the beginning of a period, 
 no matter how high the intensity of illumination may be, pro- 
 vided it does not produce a blinding glare.^* (3) There is al- 
 ways a good deal of pain in fixating a bright colored light, 
 even with foveal fixation, when the intensity passes much be- 
 yond 100 c. p. This pain may be noticed even as long as two 
 or three days after the observation. It appears, then, that . 
 when a certain intensity is reached the eye moves so fre-f 
 quently, during the latter part of the period, that the process 
 of adaptation can not be completed. The pain is probably 
 caused by the excessive contraction of the pupil in response 
 to the reflex set up by the very intense beam of light striking 
 the fovea through the artificial pupil. In the following ex- 
 periment we tried to overcome these difficulties, and to study 
 the process of adaptation at high intensities. 
 
 22 See P. G. Nutting, Trans. Ilium. Eng. Soc, xi, 1916, 943. 
 
SHEPPARD 56 
 
 Procedure. — The exposure-apparatus was a projection-lan- 
 tern which burned a carbon 13 mm. in diameter, carrying a 
 30 amp. (125 v.) D. C. current. The lantern was mounted 
 with a projection-lens of 2 in. diameter. In front of the pro- 
 jection-lens was placed a short-focus lens to bring all the 
 rays of light to a pencil-focus. The actual amount of light 
 focussed by the lens was about 6,200 c. p. The head of the 
 observer was held in a steady position by means of a head-rest. 
 The focus of light was adjusted to fall within the eye. Col- 
 ored gelatin-filters (Eastman Kodak Co.) of known wave- 
 length and transmission-power were used for stimuli; a filter 
 of any desired color could be immediately fitted into 
 one side of a pair of goggles, made to hold the filter securely. 
 In the other side of the goggles was fitted a dark piece of card- 
 board, to exclude all light from the unstimulated eye. A 
 foveal comparison-light was set up in such manner that any 
 spectral color, from the intensity afforded by the lantern to 
 the lowest possible intensity, could be obtained immediately 
 by shifting the lever of a rheostat. We measured the actual 
 intensity of the light that entered the eye through the filters 
 by the following method. A 500 watt Mazda lamp was 
 screened to match daylight (acetylene flame and No. 79 filter) 
 and was set up at one end of the photometer bench This 
 procedure was adopted because the photometry of a carbon 
 arc for this purpose is not as precise as the photometry of an 
 incandescent lamp. A flicker photometer and a 20 c. p. com- 
 parison lamp were set up at the other end. When necessary, 
 neutral-tint filters were used to reduce intensities. 
 
 For determining the transmission, a setting was made with 
 the filter screening the 500 watt lamp, and then a setting was 
 made without the filter; the ratio of these two readings on 
 the candle-power scale gave the transmission. The following 
 is an average of several trials : 
 
 Color Filter No. Transmission x C.P. of Lan- 
 
 tern=About 
 R 70 7.6% 471.2 c.p. 
 
 Y 73 35.9% 2225.8 
 
 BG 75 18.6% 1153.2 " 
 
 B 48 19.5% 1209.0 
 
 V 76 1.6% 99.2 
 
 All work was done in the dark-room. adjusted the filter to 
 the eye, placed his head in the head-rest, and awaited the 
 stimulus-light. The instructions were : " Fixate the bright 
 circle of light until all color disappears. When you no longer 
 see any color, remove the goggles, fixate the spectroscope, and 
 describe what you observe." The following are the average 
 adaptation-times in sec. for four observers : 
 
57 FOVEAL ADAPTATION TO COLOR 
 
 The hue quickly changed to 0, then to Y, 
 
 Sh 92 very poor in chroma. The Y persisted for 
 
 Z 109 some time. Finally all hue disappeared. 
 
 R S 120 
 
 F 100 The foveal comparison-color was pure 
 
 — gray, mudi lighter than the stimulus-color. 
 
 Av. 105 . 3 All intensities were gray. 
 
 Sh 165 There was no change in hue. The color 
 
 Z 177 gradually faded away. 
 
 S 150 
 
 F 175 
 
 Av. 166.8 
 
 F 175 The comparison light at all intensities was 
 
 almost white; no trace of color. 
 
 Sh 120 No change in hue. 
 
 BG Z 112 
 
 S 98 Very light gray. No color at any inten- 
 
 F 130 sity. 
 
 Av. 120 
 
 Sh 240 No change in hue. 
 
 Z 182 
 
 B S 155 Very light gray for all intensities. 
 
 F 148 
 
 Av. 181.3 
 
 Sh 200 There was a change in hue; the R api^ar- 
 
 V Z 183 ently weakened first; then there was a notice- 
 
 S 185 able decrease in chroma for the B. The R 
 
 F 220 disappeared, leaving only a verj; weak un- 
 
 — saturated B, which slowly faded into a dark 
 
 Av. 197.0 gray. 
 
 The foveal comparison light showed no 
 trace of color for any intensity. 
 
 We were able to get complete foveal adaptation for all O's 
 with these intensities, as for all lower intensities. After a 
 long fixation, beyond the point where adaptation was complete, 
 we were able even to raise the comparison-light somewhat 
 higher than the standard, and still had complete adaptation in 
 the fovea. However, long fixation with a bright light is very 
 painful; the pain even persists for two or three days as an 
 ache or soreness within the eye. 
 
 It might be objected that the blinding glare of the positive 
 after-image here influenced the judgment. We were, how- 
 ever, careful to test this factor before finally deciding upon 
 the technique for the series. If the arc was fixated without the 
 filter, there always resulted a blinding glare. When the foveal 
 
SHEPPARD 58 
 
 comparison-light was fixated in this condition, there was only 
 a slight change in brightness of the color, which immediately 
 passed away. This was not true when a stimulus-color was 
 fixated. The foveal comparison-light was always colorless, 
 and so remained. It must, however, be small enough to fall 
 within the foveal cup, and must not extend to the periphery 
 of the macula.^* Otherwise a more pronounced effect will be 
 noted from glare. 
 
 Adaptation at high intensity is difficult to describe; but so 
 far as we were able to observe the "same general process takes 
 place that has been observed throughout the earlier experi- 
 ments of this study. 
 
 Conclusions 
 
 In general adaptation to direct sunlight, as well as under 
 other conditions of adaptation, both to surface and to spectral 
 colors, the fovea has a longer adaptation-time for color than 
 any other part of the retina. This law aparently holds for 
 high intensities as well as for medium and low intensities. 
 
 Chroma is the principal determinant of the rate of adaptation. 
 Intensity is a second factor. The effect of both chroma and 
 intensity has been observed with surface colors as well as with 
 spectral colors. 
 
 The rate of adaptation is always faster at the beginning of a 
 fixation-period, gradually decreasing, and becoming very slow 
 at the end. The rate at the beginning also depends upon the 
 chroma; the richer the chroma, the faster is the initial rate of 
 adaptation. It gradually becomes slower later, and the whole 
 process lasts longer than for a color poorer in chroma. 
 
 There is no difference in the experienced course of adaptation 
 between a surface color and a pure spectral color. A color, 
 under the process of adaptation, very quickly begins to decrease 
 in chroma. This decrease is observed in the form of a light film 
 which spreads over the color. The film continually thickens, 
 until the color is obscured under the fog. The appearance of a 
 color after adaptation corresponds in general to a mixture of 
 that color with its complement. This relation was observed 
 in nearly every observation. If the eye moved slightly from 
 fixation, there were always reported three parts in the field: 
 a crescent of the after-image, an opposed crescent of the 
 stimulus-color, and an intermediate neutral gray surface. 
 
 Under suitable conditions, adaptation in direct vision can 
 be carried to completeness at any physiological intensity. 
 
 2* J. von Kries, Z. f. Psych, u. Phys. d. Sinnesorg., xv, 1897, 327. 
 
't^'k '* 
 
 
 >*r?. 
 
 ^W 
 
 « '»"iflp*- 
 
 •«»• 
 
 -^ 
 
 •h^^ 
 
 iiK 
 
 
 ^m