\Sl 
 
 
 .-t¥^ 
 
 '^-^.^^r^-'^'i 
 
 
 
 '■^•■^n 
 
 
 4>*f*V 
 
afotnell IttiwerHitg ffithrarti 
 
 Jlljara, Sftm lark 
 
Cornell University Library 
 BF241 .M38 1922 
 
 Film surface and bulky colors and thei 
 
 olin 
 
 3 1924 029 086 838 
 
% Cornell University 
 M Library 
 
 The original of tiiis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924029086838 
 
FILM, SURFACE. AND BULKY 
 
 COLORS AND THEIR 
 
 INTERMEDIATES 
 
 By Mabel F. Martin 
 
 A thesis presented to the Faculty of the Graduate 
 School of Cornell University in partial fulfilment of the 
 requirements for the Degree of Doctor of Philosophy 
 
 Reprinted from the American Journal of Psychology 
 October, 1922, Vol. xzziii, pp. 451-480 
 
THE AMERICAN 
 
 JOURNAL OF PSYCHOLOGY 
 
 Founded in 1887 
 
 by 
 G. Stanley Hall 
 
 Vol. XXXIII OCTOBER 1922 No. 4 
 
 FILM, SURFACE, AND BULKY COLORS AND THEIR 
 INTERMEDIATES! 
 
 By Mabel F. Martin 
 
 TABLE OF CONTENTS 
 
 PAGE 
 
 I. Inthodtjction 451 
 
 II. Surface to Film 
 
 Series I. Effect of varying size of hole 455 
 
 Series II. Effect of localization 464 
 
 Series III. Observations of familiar objects 468 
 
 III. Bulk to Film 
 
 Series IV. Episcotister before surface with fixation 
 
 point 469 
 
 Series V. Episcotister before surface without fixation 
 
 point 472 
 
 Series VI. Control Series 475 
 
 IV. The Monocular Film 
 
 Series VII. Holes of various sizes 476 
 
 Series VIII. Monocular observations with episcotister. . 478 
 V. Conclusions 479 
 
 I. Introduction 
 
 Katz has made us familiar with a mass of detail concerning 
 the various modes of appearance of colors, which were first 
 hinted at by Hering.^ There is a general tendency among psy- 
 chologists to accept Katz' detail, but there seems to have been 
 no definite effort made to confirm it from the fovmdation. This 
 work of verification, desirable in itself, is all the more important 
 because Katz' interest lay not so much in phenomenology for 
 its own sake as in other and more special problems. For these 
 reasons we have attempted a further study of those modes of ap- 
 pearance which seem fundamental. 
 
 •From the Psychological Laboratory of Cornell University. 
 *D. Katz, Die Erschmnungsweisen der Farhen, 1911. 
 
452 MAKTIN 
 
 In the opening chapter of his book Katz describes and com- 
 pares eight modes of appearance: (i) fihn colors (Fldchenfarben),^ 
 (2) surface colors (Oberfldchenfarben), (3) transparent plane col- 
 ors {durchsichtige Flachen), (4) bulky colors (Raumfarben) , (5) 
 mirrored colors {gespiegelte Farben), (6) lustre (Glanz), (7) lum- 
 inosity (Leuchten), and (8) glow {Gluhen). Fihn colors, surface 
 colors, and bulky colors are the most important, because all 
 other modes of appearance may be regarded as forms or com- 
 binations of these.* 
 
 In comparing surfaces and films, Katz takes paper as a typical ex- 
 ample of surface, and the colors seen in a spectroscope as a typical example 
 of film, although many other examples are also given. The observable dif- 
 ferences between film colors and surface colors he groups xmder five head- 
 ings: localization, texture, orientation, configuration, and aesthetic effect.' 
 
 1. Localization. "The spectral color of the usual apparatus is not 
 localized with the same definiteness at a precisely determinable distance 
 from the observer as the color of the paper." In saying that the film color 
 is indefinitely localized, Katz does not mean that its distance varies from 
 moment to moment. He says emphatically that, whether the observation 
 is continuous or recurrent, with constant Einstellung of the eyes the dis- 
 tance of the color in no way changes. By indefinite localization he means 
 positively indeterminate. It is true that probable limi ts may be set; but 
 these limits vary with the perceptive conditions under which the film color 
 is seen. For example, the limits he much farther apart for the film of the 
 sky than for that of the color seen in the spectroscope. Within these ex- 
 traneous Umits, the locaUzation is positively indeterminate. 
 
 2. Texture. There is a baffling contradiction about the appearance 
 of the film color. It seems to suggest depth, but proves in fact to be im- 
 penetrable. "The paper has a surface in which its color Ues. The plane in 
 which the spectral color stretches through space before the observer does 
 not possess a surface in the same sense."* One almost feels able to pen- 
 etrate more or less deeply into the spectral color, whereas with the color 
 of paper the gaze is forced to stop at the surface. The openness, the soft- 
 ness, the insubstantiaUty which the spectral color betrays is, however, 
 not of such a kind that one can speak of a clearly bulky mode of appearance 
 of the spectral color, that is, of a visible filling of space in three dimensions, 
 or of any sort of colored transparency. Like the color of paper, the spectral 
 color is bidimensional and screens the space behind from view. 
 
 3. Orientation. There is a marked difference in the orientation of 
 the two types of color. The surface in which the color of an object, such as 
 paper, seems to lie can take any orientation to the line of regard. The 
 film color tends always to assume a frontal-parallel position. Indirect 
 vision of course constitutes an exception to this rule, and there are certain 
 other exceptions. For example, the apparent orientation of film colors is 
 greatly influenced by the onentation of neighboring objects of perception. 
 Katz describes an experimental arrangement whereby a bit of the sky, 
 viewed through a cardboard tipped at an angle, takes on an apparent 
 
 'We could find no entirely satisfactory term for the appearance of the 
 "completely reduced" color. Katz' word, Fldche, does not suggest an 
 English equivalent. After trial of several words and after consultation 
 witnour Ob we decided upon the term "film." Our "film," however, as is 
 shown in the text, is not absolutely identical with Katz' Fldche. 
 
 *Ibid., 6. 
 ^Ihid., 7. 
 'Ibid., 7. 
 
FILM, SURFACE, BULKY COLORS AND INTERMEDIATES 453 
 
 orientation about midway between that of the cardboard and the usual 
 frontal-parallel position.' In spite of such susceptibility to secondary in- 
 fluences, the film color possesses in and of itself a tendency to appear in 
 frontal-parallel orientation. 
 
 4. Configuration. The surface of a body may be either smooth or 
 ridged and curved in various ways. The fihn color on the contrary always 
 lies in a practically smooth flat plane.' 
 
 5. Aesthetic Effect. The spectral color has something delicate about 
 it, something more pleasing aesthetically, than the color of paper. — 
 
 This completes Katz' list of differences between film colors and 
 surface colors. In connection with the third difference, that of orientation, 
 Katz discusses, however, another difference, which he seems to regard as 
 more fundamental than any of these five. Surface colors not only conform 
 to the surfaces of objects, but they seem also actually to be stable properties 
 of the objects. We can shadow or illuminate one and the same surface. 
 A shadowed or illuminated fihn, however, is simply a new film color. Sur- 
 face colors are color-qualities of objects. Film colors are unreferred colors, 
 colors as such. 
 
 "So far as the psychological conditions are concerned, the conscious- 
 ness of having an object before one on which the colors arise is for the 
 
 perception of surface colors, I venture to say, of decisive significance 
 
 As compared with the great influence which the inner Einstellung exerts 
 on the occurrence of this or that mode of appearance of colors, one may 
 designate as almost irrelevant the physical source of the radiation setting 
 up the impression."' 
 
 The relations between film colors and surface colors can easily be 
 studied in the laboratory, because any sort of surface color can be reduced 
 to film color by the use of a screen with a small hole, or (stfll better) of a 
 double-screen.^" The exact form of double screen is not important pro- 
 vided the following essential conditions are satisfied, (i) The screen must 
 cover the object completely except for the part seen through the opening. 
 (2) At the same time it must not permit any structure or grain that may 
 possibly be present to be recognized. (3) It must not permit any non- 
 frontal orientation of the portion of surface seen to be recognized. (4) The 
 hole must not be too large. 
 
 Between surface colors and film colors, all possible phenomenological 
 transitions occur.'' Many Os experience such a transition when monocular 
 observation is substituted for the usual binocular vision. An intermediate 
 which approaches more nearly to the true film can be secured by placing 
 before the eye a lens too strong to be overcome by accommodation. — 
 
 The third of the essential modes of appearance of colors is that of the 
 bulky colors. These differ from film and surface alike in that they are 
 tridimensional and partially transparent. They seem to fill a definite space 
 in its three dimensions. "According to my observation," says Katz, "they 
 show this property in a distinct way only when they are at the same time 
 in a true sense transparent," that is, so long as objects are actually seen 
 through them." When no object can be distinguished through the color, 
 Katz would call it a film color. 
 
 Bulky colors, like surface colors, can be reduced to film colors. For 
 this reason, Katz considers the film as the original mode of appearance of 
 colors, and regards aU other modes as derived from the film. 
 
 Ubid., 75. 
 'Ibid., 12. 
 'Ibid., 9. 
 "lUd., 9 f. 
 "7M., 9. 
 "IMd., 17. 
 
454 MARTIN 
 
 The upshot of Katz' whole discussion iSj evidently, that the film repre- 
 sents the ultimate stuff of vision. Color or light in the psychological sense 
 is filmy. If we assume with Eatz that the film is preperceptive, it follows 
 that localization, orientation, and configuration cannot be attributed to 
 it as such, and appear under experimental conditions only because and in 
 so far as secondary perceptive conditions are introduced by the exjjeri- 
 mental arrangement. Going beyond Katz' empirical statements, we might 
 lay down the following laws, (i) The film is not locaUzable. If it is local- 
 ized, that is merely because, e. g., we are famihar with the instrument 
 through which we are observing. (2) The film has no orientation. If it 
 appears always in a frontal-parallel relation to us that is simply because, 
 in making it an object of observation, we give it the easiest and most ob- 
 vious orientation; hang it, so to speak, on or across the line of regard. (3) 
 The film has no configuration. As sheer sensory quality, it must show as a 
 plane of homogeneous color. 
 
 Before the psychological status of the film is fully established, these 
 statements must be confirmed experimentally by other Os. Psychologists 
 have perhaps tended to take Katz' statements to be systematic; they are, 
 in fact, merely empirical. The reader of the introductory chapter can 
 scarcely help noticing many apparent contradictions. If, e. g., Katz is 
 correct in assuming that the fundamental difference between surface and 
 film colors is that surface colors are ascribed to objects and film colors are 
 not, in what sense can he assert the existence of transitions between film 
 and surface? Surely the source ascribed to the color either is or is not an 
 object. 
 
 Our experimental series were directed toward answering the following 
 questions: (l) Are there intermediates between film and surface? What 
 is their phenomenology, and in what sense may they be regarded as in- 
 termediates? (2) Can it then be shown experimentally that the object- 
 consciousness is sufficient of itself to ensure the perception of surface color? 
 (3) If the film color really is the ultimate stuff of vision, what happens to 
 it that changes it to surface color in the one case and to bulk in the other? 
 
 The Os were Dr. L. B. Hoisington (H), assistant professor of psy- 
 chology; Dr. H. G. Bishop (Bi), instructor in psyohologjr; Miss Cf. C. 
 Braddock (Br) and Miss G. Adams (A), graduate students in psychology; 
 and the writer (M). All had had training as Os in previous psychological 
 investigations. 
 
 The apparatus used for reducing surface to film was a double screen, 
 similar in essential details to that described by Katz. It consisted of two 
 entirely separate screens, constructed as follows. To a wooden frame, 71 
 cm. X 56.5 cm., was tacked a sheet of heavy grey cardboard which ap- 
 proximately matched in brightness no. 10 of the Hering series of grey 
 papers. In the cardboard was a circular hole, 25 cm. in diam., whose center 
 was approximately 31.5 cm. from the bottom of the screen, 25 cm. from the 
 top, and 35.5 cm. from each side. To the back of each screen was fastened 
 a strip of wood so cut that it formed a slot into which could be dropped 
 smaller sheets of the grey cardboard. In the center of these smaller screens 
 was a hole. The holes varied from 2 cm. to 20 cm. by steps of 2 cm., and 
 were numbered from o to 9. There was a pair of every size, one to fit each 
 of the two members of the double screen. The back of each hole was 
 ground down with emery paper to present a sharp, knife-like edge. When 
 one of the small screens was fitted into the slot, the small hole was con- 
 centric with the large hole. 
 
 The experiments were performed in a dark room, approximately 6 x 
 2.5 m. It was Ughted from above by three 'dayUght' lamps, arranged in a 
 single row along the middle of the cefling. The table to which the apparatus 
 was fastened was set up at the side of the room, almost exactly midway be- 
 tween two of the overhead lamps, so that the screens and the stimulus be- 
 hind them received an almost equal amount of light from the two sides. 
 
FILM, SURFACE, BULKY COLORS AND INTERMEDIATES 455 
 
 There was practically no brightness-contrast between the fore and back 
 screens in these experiments. The screens, being entirely separate, could 
 be set at any desired distance from each other, but in all regular series they 
 were 25 cm. apart. Other distances which might be varied were the dis- 
 tance of the stimulus from the back screen and the distance of the from 
 the fore screen. 
 
 From a few preUminary trials it seemed probable that reduction is 
 favored by increased distance of from the fore screen and by increased 
 distance of the stimulus from the back screen, just as it is favored by a re- 
 latively wide distance between the two screens themselves. In all regular 
 series, the sat at a distance of i m. from the fore screen. This position 
 was kept constant by the use of a fixed headrest. 
 
 II. Surface to Film 
 Series I. Effect of Varying Size of Hole 
 
 Series I was originally planned to answer our first question : 
 Are there intermediates between film and surface? 
 
 The colored stimuli were cloths stretched very smooth on 
 wooden frames, and Milton Bradley papers pasted on sheets of 
 stiff cardboard. These could be fastened by thimib-tacks to a 
 wooden and cardboard frame, so that a quick and easy shift from 
 one color to another was possible. 
 
 When viewed without the double screen, these colored 
 stimuli were, of course, ordinary surfaces. When they were 
 viewed through the very small holes, their appearance approx- 
 imated in most respects to the film as described by Katz. There 
 were certain differences, to be discussed in detail later, but these 
 need not concern us here. It seemed plausible to suppose that 
 by varying the size of the holes, we should find, somewhere be- 
 tween the largest holes, which gave surface, and the smallest, 
 which gave an approximation to film, some of those intermed- 
 iates to which Katz refers. 
 
 The surfaces, like the films to which they were to be reduced, 
 were flat and possessed a frontal-parallel orientation. The cri- 
 teria of the true film were assumed to be indefinite {i.e., only 
 secondarily motived) localization; bidimensionahty; loose, 
 soft, insubstantial texture; and complete lack of objective refer- 
 ence. We purposely omitted aesthetic effect, because we could 
 not agree with Katz in regarding this as an observable difference 
 on the same level as the others. All Os, however, occasionally 
 exclaimed at the beauty of the film colors, in spite of the fact 
 that aesthetic effect was not mentioned in the instructions. 
 
 It will be remembered that Katz' Os fixated the edge of the 
 hole in the back screen," because this fixation was most natural 
 and convenient and was moreover adequate to the purposes of 
 his experiments. Since, however, we were to ask for reports of 
 localization and dimensionality, it seemed best to instruct the 
 Os to fixate the color under observation. 
 
 ^Hbid., 38, 73, etc. 
 
456 MARTIN 
 
 The original instructions for H and M read as follows. "After the 
 usual 'Ready, Now' signal, a grey screen and a brown disc will be exposed for 
 3 sec. Fixate the center of the disc. Immediately after the exposure, 
 characterize the disc as either (i) surface, or (2) film, or (3) intermediate. 
 Then describe it in terms of texture, apparent localization, dimensionality, 
 and objectivity, and any other striking characteristics. In subsequent 
 exposures, it is unnecessary to repeat in detail items that remain the same 
 as in the previous exposure. Report fully whatever changes." The in- 
 structions for the other Os were the same, except that they were not asked 
 to report on more than two descriptive characters at a time. In addition 
 to the typewritten instructions, the Os were given typewritten slips with 
 the following definitions. " 'Intermediate' means like surface in some 
 respects and like film in others. 'Objectivity' means attachment or refer- 
 ence to a particular object, such as a board, a piece of cloth, a piece of 
 colored glass, etc. 'Localization' means position, either (i) before the 
 screen, or (2) at the screen, or (3) behind the screen, with a rough estimate 
 of distance in the first and third cases. 'Texture' means material nature 
 or substantial character, as transparent, translucent, opaque, hard, soft, 
 soUd, etc. 'Dimensionality' means extension solely in two or definitely 
 in three dimensions." The term "mixed" was soon substituted for "in- 
 termediate," because under the conditions of experimentation no true 
 (psychological) intermediates between film and surface were found. Later 
 the term "mixed" was also dropped. For greater clearness and freedom 
 from suggestion the definitions of t«rms were also revised, to read as follows. 
 " 'Objectivity' means the conscious reference of the observed color at the 
 time of observation to some concrete thing. 'Localization' means a given 
 position, whether constant or temporary, of the observed color in the third 
 dimension of space. 'Texture' means material nature or substantial 
 character of the observed color. 'DimensionaUty' means extension of the 
 observed color solely in two or definitely in three dimensions." 
 
 We at first attempted to tabulate the results as fast as they 
 were secured, but soon found that their complexity made a tabu- 
 lar mode of arrangement practically impossible. We shall in- 
 stead treat of the observed phenomena in groups. It will be 
 noticed that there is considerable overlapping between the 
 groups. 
 
 Group I. Surface Colors 
 
 The phenomenology of surface colors presented no real dif- 
 ficulties. The descriptions were simple, brief, and consistent. 
 
 Localization. Bi (9)" "About a foot and a half behind the screen." 
 A (7) "It looked about two or three feet from the screen." Br (8) "Local- 
 ized at a slight distance behind the back screen." H (9) "Surface behind 
 the screen at some distance, a meter or so." H (9) "About .5 m. behind 
 back screen." M (7) "Surface, quite definitely localized behind the back 
 screen, though I can't give the exact distance in feet and inches." 
 
 Objectivity. Bi (8) "I could make out the inequality of the cloth, the 
 weave, the fabric." A (9) "Rough green cloth." Br (8) "More uniform 
 than cloth, more like paper." H (7) "Fairly definite object, cloth, or paper, 
 or almost anything with a grain to it." M (9) "Opaque soft cloth." 
 
 Texture. Bi (9) "Weave of cloth." A (8) "Rough. I can see ridges, 
 very close together, blurred." Br (7) "Smooth shiny cloth with darkish 
 
 "Numbers in parentheses refer to the number of the hole used. 
 
FILM, StTRFACE, BtJLKY COLORS AND INTERMEDIATES 457 
 
 grain." H (9) "Coarse woven cloth, slightly fleeced, so that texture is just 
 a little obscured. Solid. Opaque." M (7) "Solid, opaque, soft cloth with 
 ridged weave." 
 
 Dimensionality. Surfaces were always reported as bidimensional. 
 
 Group II. Film Colors 
 
 The Os were unanimous in describing the fihn color as in- 
 definitely localized; but as regards the stability of the localiza- 
 tion of film colors we obtained results that at first seemed op- 
 posed to those of Katz. It will be remembered that he combats 
 the contention of HiUebrand that indefinite localization involves 
 shift of position," and asserts that the film color does not 
 change its distance, although it is positively indefinite in local- 
 ization. Hence we were surprised to get occasional results like 
 the following. 
 
 H (3) "One can do stunts to those things, so far as localization is 
 concerned. They are iastable. One can read all kinds of meanings into 
 them." H (2) "Localization is shifty. At first it seemed to fill all the space 
 behind the screens; but I took a localization attitude later, and it was 
 easy to throw the thing farther back." H (i) "Slightly shifting local- 
 ization. I don't mean that at any time I could say just where it was, only 
 that sometimes it seemed nearer and sometimes farther." H (4) "Local- 
 ization tends to shift from at the screen to far behind." Br (4) "It lay be- 
 hind the back screen, came forward a bit, and then seemed to shift up to 
 the fore screen." Bi (o) "I'm not sure whether it was always in the same 
 place or not. I think there was a little play back and front. Not much." 
 
 The opposition between these results and thoSe of Katz was, 
 however, apparent rather than real. Katz says that the localiza- 
 tion of the film color is stable with constant Einstellung of the 
 eyes. With a little more practice our Os soon discovered that 
 the apparent shifts in localization were due to shifts in fixation. 
 It will be remembered that the original instructions read "Fixate 
 the center of the disc." These instructions in a sense demanded 
 the impossible, because there was nothing there to fixate. 
 
 H (i) "When you try to fixate, you have just that meaning of looking 
 for something you can't see. You tend to stare. The eyes are held at an 
 immense strain." Bi (6) "Almost impossible to fixate the center of the 
 disc. My accommodation is for two or three different places !" 
 
 In hunting as it were for a place to anchor, the fixation natur- 
 ally- shifted about, and the film color, having no locaUzation of 
 its own, followed the changes of convergence. It therefore 
 seemed advisable to alter the instructions with respect to fixa- 
 tion to read : "Regard the white spot at the lefthand edge of the 
 screen opening as a fixation point and begin every observation 
 with your eyes directed to that. You need not, however, feel 
 obliged to maintain this fixation throughout the period of ex- 
 posure." Under these instructions the Os made more accurate 
 reports of their fixation, and it was found that apparent shifts 
 in the localization of the color were always correlated with shifts 
 of fixation. 
 
 ^Ibid., II. 
 
458 MAKTIN 
 
 Bi (7) "I can't tell where it is, unless I hunt for it by changing fixation. 
 In that case, the film seems to be anywhere I look for it, so long as I'm 
 looking behind the second screen." H (i ) "If I look at the back screen, then 
 the color is bwond it. There is no color between me and the fixated screen. 
 If, however, I try to fixate the color, then it is absolutely not localized." 
 Br (i) "It jerked forward two or three times and then back again. I think 
 these were eye-movements." A (3) "Every time my fixation shifts, the 
 color shifts farther away or closer. I don't know how far away it was. 
 All I could tell was that it was shifting." H (3) "It doesn't change a bit 
 so long as one holds fixation." M (i) "The whole thing is perfectly stable 
 so long as fixation is held, but tends to shift with fixation." M (i) "Color 
 might be either at the back screen or slightly in front of it, or almost any 
 distance behind; not that it ever seemed to be precisely in any of these 
 positions. It remained absolutely stable so long as fixation was held, but 
 shifted with shifting fixation, in this sense, that when I looked for it in one 
 place it always seemed to be somewhere else. I never saw it actually shift. 
 I never saw it in motion." 
 
 In regard to the tendency of the unlocalised fihn color to link 
 itself with perceptive experiences in the visual field, our results 
 also agree with those of Katz. 
 
 H (i) "It's just out there, apparently localized only in connection 
 with some other experience." H (i) "Meaning of forcing the gaze at 
 nothing in particular. In that sense you may say that the thing is not 
 locaUzed at all. But even so, one must force the gaze beyond the second 
 screen and that seems to carry the meaning that the color is beyond that. 
 One can't focus the eyes at a point and say the color is there. In that sense 
 the locaUzation is absolutely mdefinite." Bi (4) "When I fixate the edge of 
 the disc, I can see that the brown is behind that edge somewhere. No 
 front or back limit to the brown." 
 
 In certain prehminary experiments, in which we tried to 
 determine what length of exposure was most convenient for 
 observation, we found that this tendency of the film color to link 
 itself to the nearest perceptive object, the screen, is enhanced by 
 very brief exposure times. 
 
 H "With the shorter exposures, the tendency for the color to be linked 
 up with the back screen is a little more pronoxmced. With the i sec. ex- 
 posure, it became almost a part of the screen, only it had no definite 
 surface, nor any objective reference: but it belonged to the same depth 
 level." 
 
 Thus our results with regard to the localization of the film 
 colors are in full agreement with those of Katz. This was not 
 the case with dimensionality. The colors observed were not 
 unequivocally bidimensional. 
 
 H (i) "Comes right up to the screen, but extends back indefinitely." 
 H (2) "Depth, but not penetrable." Br (i B)'^ "I couldn't look through 
 it. It wasn't gauzy, but it seemed to have a bulkiness about it. I shouldn't 
 say definitely tridimensional. Indefinitely so." H (i) "Tridimensional? 
 
 "Letters following the numbers in parentheses refer to the colored 
 stimuli. When no letter appears, a dark brown rep was used as a stimulus. 
 Other backgrounds used in experiments from which these introspections 
 are quoted were bright green cheesecloth (G), red-orange cheesecloth (O), 
 purple lawn (P), and light brown gingham (B). 
 
FILM, SURFACE, BTJLKT COLORS AND INTERMEDIATES 459 
 
 Hard to say. It might have been an unlocalized bidimensional something. 
 The fact of its not having a beginning is the reason for calling it tridi- 
 mensional. It may be due precisely to the fact that it was not locaUzed." 
 H (i) "I fixated the white spot practically all through the exposure. Under 
 those conditions, I should call the color bidimensional. At least I didn't 
 see any tridimensionaUty." M (i) "Not definitely tridimensional nor 
 definitely bidimensional. I should have been inclined to call it tridimen- 
 sional, but could not do so, because it has neither front, back, nor middle. 
 My reason for wanting to call it tridimensional is that it is not locaUzed 
 in a single plane; but calling it tridimensional would not reaUy help be- 
 cause it is not locaUzed in any bulk of space either." M (i) "Tridimensional 
 in the sense that very dense fog and thick, non-transparent dust clouds 
 are tridimensional. Certainly not tridimensional in the sense that fluids 
 in bottles or masses of jelly or blocks of colored glass are tridimensional. 
 The brown color here had no beginning and no ending in the third dimen- 
 sion." A (iP) "I think I'd caU it three dimensions, but I'd have to stretch 
 what I mean by that. It certainly is more than two dimensions." A (iP) 
 "I wouldn't say it was definitely in three dimensions. It was very in- 
 definite. It's different from what I'm in the habit of caUing dimensions. 
 Dimensions are usually either soUd or they've got some boundary that's 
 flat or that blocks the thing off in some way. This sticks out in front, but 
 it sticks out in such a vague way. It's not soUd, and it hasn't any bound- 
 ary." Bi (oB) "Not sure about the dimensionaUty, beyond the fact that 
 it was spread out." A (oG) "I can't locaUze it, and I can't teU about its 
 dimensions either." H (i) "I don't know. It certainly both is and isn't 
 tridimensional, if one can make any sense out of that. Bidimensional, I 
 beUeve." H (i) "One is left with the meaning that you don't know whether 
 the thing is tridimensional or not." H (i) "How in the world can you say 
 anything about the dimensionaUty of a thing if you don't know where it is?" 
 H (i) "Bidimensional. I beUeve that the difficulty with the dimensionaUty 
 is one of locaUzation primarily." M (iB) "Bidimensional in the sense 
 of having no definite extent in the third dimension, but tridimensional 
 in the sense of not being locaUzed precisely in one plane." 
 
 From these and similar observations it would appear that 
 the phenomena with which we were dealing were neither bidi- 
 mensional nor tridimensional. They were merely extended, 
 without dimensions in the geometrical sense. We might almost 
 have anticipated this result. For, after all, dimensionahty in 
 that sense is already a perceptive character. If a color is local- 
 ized in one plane we call it bidimensional; if it is local- 
 ized in more than one plane, we call it tridimensional; but if it 
 is not localized at all, on what basis can we call it either? In 
 describing the color as bidimensional, the Os had reference to 
 its smoothness and impenetrability; in describing it as tridi- 
 mensional, they had reference to its softness, looseness, and "in- 
 vitation to penetrate." In either case, what the Os were describ- 
 ing was not dimensionality proper, but the predimensional nature 
 of visual quality. So in the light of our results it seems evident 
 that the product of the "reduction" of surf aceff cannot be called 
 bidimensional in precisely the same sense that a surface color is 
 bidimensional. The film is bidimensional in the negative sense, 
 in that it is not yet positively tridimensional. It is also tridimen- 
 sional in the negative sense, in that it is not yet positively bidi- 
 mensional. 
 
460 MARTIN 
 
 The question of objective reference gave no difficulty. All 
 the Os reported that the pure film colors had absolutely no ob- 
 jective reference. 
 
 The film colors had very characteristic textures, but the Os 
 found considerable difficulty in describing these except by the 
 use of perceptive analogies. 
 
 A (i) "It's hazy, but not haze. In the woods where there's been a fire 
 and smoke lies in a thin layer on the ground you'd have something like 
 this." H (i) "Like something viscous or gelatinous, if you could see such 
 a thing without a definite front face." H (3) "Haze, soUd, translucent, tri- 
 dimensional, — like colored glass, — and yet, no! No analogy under the sun 
 is exactly correct. The light is what bothers me. I don't see Ught through 
 it nor Ught reflected from surface, nor is the thing especially luminous." 
 Bi (oP) "It had a curious soft penetrableness." A (oG) "You wouldn't 
 say it's luminous, but there's light in it in some way. It wasn't reflected 
 from it." H (general observation) "The darker ones are always more like 
 looking into a colored hole, whereas the brighter ones seem glowing." M 
 (i) "Opaque in the sense of shutting off space behind, but not opaque in 
 the way that a solid is opaque^ for the shutting off did not occur at any 
 particular place. I could look mto the color, but not through it." M (i) 
 "It was not opaque as solids are opaque. Yet it could not in any sense 
 be seen through. The meaning of penetrabiUty seems to be associated with 
 the effort to gauge localization." Bi (5 O) "Self-radiant, somehow. I'm 
 sure that if I didn't force it down by careful fixation it would be a tre- 
 mendously glowing color. It's something like the glare we get at night 
 when we approach automobile lights. It has the same beaming, dense 
 character." Bi (3) "Haze points too much toward fogginess. This is 
 
 beautifully clear. This is too dense for thin smoke It's Uke 
 
 looking into a window from a lighted room into the dark. The dark 
 comes clear up to the window, but you don't know where it leaves off. 
 Same kind of thing here exactly." Bi (2) "No texture in the usual 
 sense. Foglike, but not so dense as fog. Localization is uncertain. It 
 seemed to stand behind the hole in space. There is a penetrabiUty about 
 a fog that this does not have." H (i O) "I don't know whether it's trans- 
 lucent or not. It's something like translucent. Rather like dense gas, 
 though not exactly Uke any I've ever seen." H (oP) "It's just a patch of 
 color that isn't objective, and yet one tries to apply textural terms to it. 
 If you have a texture, it must be a texture of somethmg." 
 
 While in general there is no harm in saying that the tissue 
 of the film color is soft and insubstantial, our reports indicate 
 very clearly that the film is pretextural in the same sense in which 
 it is predimensional. Texture is a perceptive character, and 
 therefore cannot in its ordinary sense be attributed to a sensory 
 quality. Texture, like dimensionaUty, seems thus to be bound 
 up in some way with fixation. 
 
 H (3) "It's easy for that to be two different things: (i) a transparent 
 (I don't like the word glassy) thing that becomes graduaUy less translucent, 
 and then gets so dense that you can't see any farther into it. The thing 
 shifts from that to (2) something at or almost on the screen, lustrous, al- 
 most luminous. I have never seen anything like it. I'm certain these two 
 things depend on a shift of fixation." 
 
 Intermediates 
 No true psychological intermediates were found, under the 
 experimental conditions of this series. With the holes inter- 
 
riLM, SURFACE, BULKY COLOK8 AND INTEKMEDIATES 46I 
 
 mediate in size between those giving ordinary surface and those 
 giving pure film, we obtained appearances that involved both 
 simultaneously. These were usually described as a surface, seen 
 more or less obscurely, and more or less definitely locab'zed be- 
 hind a semi-transparent medium of the same color. Similar ap- 
 pearances were frequently obtained with holes that at other 
 times under precisely the same external conditions gave ordinary 
 surface on the one hand, or pure film on the other. 
 
 The phenomenological intermediates found with different 
 colored stimuli and various sizes of holes differed in the locaUza- 
 tion and texture of the surfaces and in the depth and density of 
 the obscuring media. Strictly speaking, they did not fall into 
 sharply defined classes. For convenience, however, we have 
 classified the results more or less arbitrarily into four fairly 
 distinct groups. Starting with surface as given, we have the 
 following. 
 
 Group A. Soft Surface 
 
 Br (5) "More like surface than anything else, a surface that is a little 
 thick; like a wooUy surface, loose-woven so that you can look into it; like 
 thick, blankety cloth." M (9B) "Cloth. Bidimensional. Definitely lo- 
 calized behind screen at uncertain distance. Very soft, velvety surface." 
 H (7) "Solid without being highly resistant, like cut Uver. Velvety comes 
 nearest, but — ! Something that by vision shows it is soft to touch. Bi- 
 dimensional with a hint of tridimensionaJity." A (7) "Like a woolen 
 sweater or a chinchilla coat or soft snow." Bi (9) "Nearly surface. The 
 color is one foot or more behind the screen. Surface is irregular mottling 
 of color." 
 
 Group B. Veiled Surface 
 
 Br (8) "Surface with a sUght fihn over it." Bi (7) "Surface, i ft. be- 
 hind screen. Out in front there was a little bit of brown haze." Br (8) 
 "I don't know where the front or back of the film lies, but I do know 
 where the surface component is." A (6) "As if you had a surface color and 
 a haze in front of it." M (7) "Plainly cloth, not quite definitely localized. 
 Suggestion of very thin, indefinitely localized film of brown haze some- 
 where in front, perhaps contiguous with the surface, perhaps detached 
 from it. The cloth itself appears velvety, i. e., the color is not localized in 
 one plane, and still less is there any distinction of localization of separate 
 threads to form a clean-cut cloth tejcture." H (7) "Very thin cloud or 
 gauze before a solid, opaque, slightly irregular surface." 
 
 Group C. Fogged Surface 
 
 M (6) "Smooth, stony surface. It might be cardboard, stone, wall- 
 paper, or cloth. The brightness pattern suggests cloth, but the surface 
 appears too smooth; or rather^ it lacks the definite distinction of local- 
 ization of separate threads, which is characteristic of cloth. The surface 
 as a whole is not quite certainly locaUzed, and appears to Ue behind a 
 medium which is brown but nearly transparent, and is of indefinite depth 
 and uncertain localization." Br (6) "Sort of brownish haze, looking back 
 onto surface. Surface hard to distinguish." Bi (6) "I felt as if I could make 
 out a Uttle of the fleckiness of the cloth. Hazy foregroimd of color, as if 
 at the back limit of it there was a surface; rather defimitely localized. 
 Localization of the haze is uncertain. All this is somewhere behind the 
 back screen." H (5) "A surface, adhering to an object, and then something 
 
462 MARTIN 
 
 else in front, just as one might see through a vaporous film. The tridi- 
 mensional seemed set off from the bidimensional objective. The surface 
 was locahzed at about 25 cm. behind the screen." A (6) "It looks like a 
 surface behind, and thick haze in front. The surface looks like soft cloth." 
 
 Group D. Immersed Surface 
 
 M (4) "A surface, or at least something opaque, behind a medium of 
 decreasing transparency. The medium began at or close behind the 
 screen, and extended to the unidentified object, which was rather doubt- 
 fully localized, somewhere between 10 and 20 cm. behind the back screen." 
 M (2) "Very indefinitely objective. Close behind the screen, not be- 
 ginning in any definite plane, was a vaporous brown color, very dense and 
 very still, almost so dense as to suggest a viscous Uquid. This vapor was 
 nearly transparent where it began, but became progressively less so farther 
 back. Behind it at an indefinite distance, not very far, however, was a 
 gelatinous or perhaps solid something whose tejrture was completely 
 obscured by the medium in front." A (6) "Something with a haze in front 
 of it. Too rough to call it surface, i. e., not smooth enough across." Br 
 (2) "More haze than surface. It stops somewhere indefinite." Bi (5) "Haze 
 seemed to end in a surface, but I didn't see the surface, either." H (4) 
 "SoUd and opaque behind, but with a gradual transition through cloudy 
 to a thin, feathery gauze, and to a vaporous almost-transparency. Tri- 
 dimensional with a hint of surface at the back. The haze obscures too 
 much to allow objective reference." 
 
 It is clear that what we have phenomenologically before us, in 
 this series of observations, is the gradual break-down of a visual 
 object. We begin with the colored surface of a determinate 
 thing, and we end with a sheer quality of color that is not the 
 color of anything. Phenomenologically, we agree with Katz, the 
 transition is continuous. The object of which the color is the 
 surface becomes more and more vague, indefinite, unrecogniz- 
 able, until only a vestige of objective surface, a hint of the sur- 
 face of an indeterminate something, is left; and finally this last 
 vestige of objectivity itself disappears." 
 
 Psychologically, however, the picture is very different. At 
 first we have the fairly simple perception of the surface of a 
 material object;'* this perception grows more complex, as the 
 surface recedes and the haze before it becomes prominent; in 
 the stage which we have distinguished as Group D the percep- 
 
 "It was probably one of these intermediates, and not the film proper, 
 that was perceived by the patients of Gelb and Goldstein. See K. Gold- 
 stein und A. Gelb, Psychologische Analysen hirnpathologischer Falle auf 
 Grund von Untersuchungen Hirnverletzter: IV. A. Gelb, Ueber den Weg- 
 fall der Wahrnehmung von Oberflachenfarben, Zts. f. Psych., 84, 1920, 
 193-257- 
 
 "'Surface' is such a visual space-phenomenon as it has been worth 
 while to name, and therefore to make into a formal, text-book 'perception'. 
 If any of the other phenomena here noticed corresponded regularly with 
 our modes of apprehension of the outside world; if, that is to say, veiled, 
 or fogged, or immersed surfaces were part of our every^day experience; 
 then they, too, would have been listed as definite 'perceptions'. They are, 
 however, so uncommon that they have not been identified or stabiUzed 
 by name. 
 
riLM, SURFACE, BULKY COLORS AND INTERMEDIATES 463 
 
 tion is at the height of its complexity: we have the vestige of a 
 surface, covered by a haze which is denser in its remote portion 
 and becomes clearer as it extends nearer to us; and in the next 
 following stage, that of the film proper, we have (with the disap- 
 pearance of the last remnant of surface-color) no perceptive ob- 
 ject at all, but a sensory datum, ultimate and unanalysable. If 
 we persist in taking up toward this datum the perceptive atti- 
 tude which we have maintained throughout the series, we are 
 obUged to characterize the 'object' by negatives: it is not now 
 localizable, it is not dimensional, it is not objective. There is, 
 then a continuous increase of psychological complexity up to 
 (and including) Group D, and then there is a sudden break, from 
 'perception' to 'sensation,' from highly complex to absolutely 
 simple. 
 
 If we may trust the reports of the Os, the determining factor 
 throughout this experiment is localization, and the phenomeno- 
 logical continuity of the series depends upon the steadily in- 
 creasing difficulty of localization as cleancut surface is left and 
 film is approached. The following reports are typical. 
 
 Bi (2 G) "When I get my fixation on the cloth, it flattens right out. 
 Localization is definite. It's hard to keep it that way. It wiU cloud right 
 up and be a disembodied green." Bi (3 G) "The color fogged up as I ran 
 my eyes back, but as soon as I got definite localization there was nothing 
 but surface, bidimensional and definitely locahzed." Bi (o G) "The filmi- 
 ness and localization vary with fixation." Bi (3) "That film is determined 
 as much by fixation as by the screens. I've got more or less at the same 
 exposure from the same screen." H (o P) "That thing comes so near to 
 bein^ nowhere that it's almost nothing." A (o P) "I believe if I could 
 definitely localize how far away it was, I could get it as a surface, and the 
 thing works both ways." Br (9) "I can get a surface or a film either one, 
 depending on where I focus." Br (3) "Film at first. Then surface. These 
 are changes in convergence, I'm sure." H (i G) "I can fixate that so that 
 it becomes just a cloth surface. My eyes water and ache. When I do 
 fixate it, it is localized fairly definitely behind the second screen, 15 or 20 
 cm." Bi (5 G) "Fixation would sUp and fiilm would form. Then I would 
 get fixation and film would disappear." 
 
 Summary: Effect of Size of Hole 
 
 In reporting the results of this Series I, we have given the size of the 
 hole in connection with every introspection quoted. There was much 
 variation (both from to 0, and with the same at different times) so 
 far as the exact size of hole is concerned at which the various modes of 
 appearance were reported; there was, on the other hand, great constancy 
 in the descriptions. 
 
 FUm^ were reported most frequently by all Os with holes o and i 
 (2 and 4 cm. diam.). H and Br occasionally reported film with holes of 
 other sizes, up to 14 cm. diam. The fiJm-reports with the larger holes 
 came in the early stages of practice. 
 
 Surfaces were most frequently reported by all Os for the largest open- 
 ing (20 cm. diam.). With the cloths of very open weave, especially the 
 green and orange cheesecloths, surface was often reported with smaller 
 oles. 
 
4^4 MARTIN 
 
 The intermediates showed a good deal of overlapping. The medians 
 (all Os) were: soft surface, 1 8 cm.; veiled surface, 14 cm.; fogged surface, 
 10 cm.; immersed surface, 8 cm. 
 
 Series II. Effect of Localization 
 
 From the results of Series I it seemed probable that with our 
 experimental arrangement the essential condition for the ap- 
 pearance of surface on the one hand or of film on the other is the 
 locaHzation or non-locahzation of the color quahty. If this in- 
 ference proves true, we have aheady answered half of our third 
 question. We have found what changes the sheer color quality 
 into the color of something. We should now be able, therefore, to 
 change film colors to surface colors by giving the stimuli a defi- 
 nite localization. In Series II we attempted to test this hypoth- 
 esis by bringing a strong localization-motive to bear upon what 
 had been reported as a film color. 
 
 The colored stimuh used were Milton Bradley colored papers 
 and the dark brown rep, because in the preceding series these 
 had given pure film colors with larger holes than had any of the 
 others. The most natural and obvious cue to localization was a 
 fixation point in the center of the color-field viewed through the 
 double screen. 
 
 We used both black and colored fixation points. They were cut with a 
 beveled circular punch and were slightly more than i cm. in diam. The 
 usual e^osure time in this series was 5 sec. Longer and shorter times were 
 tried without affecting the results, except when prolonged fixation was 
 accompanied by staring. The following instruction was given. "After 
 the usual 'Ready, Now' signal, a colored disc and a differently colored 
 fixation spot will be exposed. Be sure to maintain fixation at the fixation 
 spot. Report on the character of the spot and of the colored field in terms 
 of localization, dimensionality, texture, and objectivity." 
 
 At first the Os had great difficulty iu fixating the point, be- 
 cause the edge of the back screen offered rival fixation points. 
 When they had learned to maintain fixation upon the spot, the 
 color was almost always seen as surface. This procedure was 
 possible for all except the 2 cm. hole, which did not permit of 
 binocular fixation of the fixation spot. 
 
 A (2 BL, 0)'° "Behind the back screen several inches. Small round piece 
 of orange paper placed on top of a piece of blue paper." M (i D, bk) 
 "Cloth with a black spot on it. Very definitely locaujed. Texture and 
 weave veiy plain." H (3 D, bk) "If I attempt to fixate the color beside 
 the spot, it is possible to make the color a surface in a plane with the spot." 
 H (2 BL, o) "When I first looked at that, I was sure it was simply one bit 
 of paper on another. Definitely surface behind the screen, opaque and 
 
 "The large letters after the figure in parentheses indicate the back- 
 ground; the smaUletters, the color of the fixation point. The backgrounds 
 used in experiments from which these introspections are quoted were 
 dark brown rep (D), and the following Milton Bradley colored papers: 
 blue green (BG), dark blue (BL), and red (R). The fixation points were 
 orange (o), yellow (y) and black (bk). Many other combinations were of 
 course tried. The colors used did not affect the results at all. 
 
riLM, SURFACE, BULKY COLORS AND INTERMEDIATES 465 
 
 solid." Br (2 R, bk) "A bit of red paper with the black stuck on it." Bi 
 (i R, bk) "I saw the spot pasted on in front. Both the purple and thespot 
 were surface colors." 
 
 The spot and the color nearly always appeared as surface at 
 first. Sometimes under steady fixation, even with the 5 sec. 
 exposures, they took on a filmy appearance. 
 
 Br (2 R, y) "Yellow circle on a red background, which seemed at first 
 to be surface; but as I looked, that seemed to be red air or space, not filmy 
 enough for air. Momentarily it looked as if the yellow were in front of it, 
 floating." H (2 BL, o) "Orange surface, bidimensional, definitely 15 cm. 
 behind the screen. Definitely paper. The blue was simply some kmd of 
 blue surface. It might have been anything opaque and soUd. As I stared 
 at the orange spot, the blue became indefinitely locaUzed, except that it 
 dropped behind the orange." 
 
 In spite of these complications, we had shown conclusively 
 that a visual cue to fixation was sufficient to change to surface 
 an appearance which would otherwise have been filmy. This 
 result in isolation would have been equivocal, however, because 
 a bicolored fihn might perhaps be a visual impossibihty. We 
 therefore decided to see whether the cue to localization might 
 not just as well come from some other sense department. The 
 sense departments chosen for experiment were the tactual-audi- 
 tory and the auditory. 
 
 In the lower part of each member of the double screen and 
 also of the cardboard to which was pasted the colored stimulus 
 were cut small holes 2.5 cm. square. A long rod passed freely 
 through these holes. A piece of cardboard could be set up either 
 against the color screen or in front of it in any desired position. 
 was Allowed to feel through the screens with the rod until the 
 tip of it touched the cardboard and then to tap, while looking at 
 the colored field. Before the stimulus color was actually ex- 
 posed, E put the pointer through the double screen in position 
 for the tapping. Then the color was exposed for s sec, during 
 which time tapped the cardboard and looked at the color. 
 
 At first the Os found it impossible to connect the surface 
 felt with the color seen. 
 
 A (2 BL, poking) "I don't think I'm realizing very well that I'm 
 poking the identical thing. That's a hard thing to do. I'm looking at this 
 thing up there and poking this thing down here, and there's no connection 
 between them." Br (i BL, poking color screen) "I can't think of those 
 two as being the same. It was surface I was touching back there, but it 
 didn't look and feel the same." M (2 BG, poking color screen) "The object 
 poked seems to have no relation to the color." H (2 BL, poking in front 
 of color screen) "Curious feeling of the unrelatedness 01 the two things. 
 The things simply don't belong together. I certainly was not poking at the 
 color;forthemost part behind it. The color is not at the screen. Itisunre- 
 latedto thescreen, but doesn'tseem very far beyond it. Onewould be per- 
 fectly willing to grant that it might be twenty feet beyond. For the most 
 part I was punching behind where the color lay." Bi (2 BL, poking) "I 
 don't see any point to sticking this pointer down there. I see color up here 
 in this position and away off in some other universe I stick this thing 
 through. I see the color and the color is nearer than the pointer." 
 
466 MARTIN 
 
 After about six hours' practice, however, the color, which had 
 appeared filmy, could be transformed to surface by the tapping, 
 though it was rarely possible to hold the surface throughout the 
 period of exposure. 
 
 Bi (2 BL, poking) "Sometimes while I was localizing the tip of the 
 pointer, I seemed to see a surface." Bi (2 BL, poking) "If I turn my atten- 
 tion away from the color and think where that pointer is, I can see the 
 color as blue paper." Br (i BL, poking color screen) "At first they seemed 
 quite incompatible, and then the blue receded backwards while I poked 
 and became for a moment surface lying some distance behind the hole, 
 but I couldn't continue seeing it that way." Br (i BG, poking) "At one 
 of the pokes, the color receded and became a flat surface and the tip of the 
 pointer was touching the same surface." H (2 BI, poking color screen) "If 
 
 1 attend very closely to the sensations (both tactual and auditory) that 
 come from the punching, then the thing does change. I'm. not sure just 
 what happens. The color does get thrown back there and fairly definitely 
 localized, pretty definitely bidimensional. It certainly becomes more like 
 an object, though just what I couldn't say; more like a smooth, painted 
 surface, fairly opaoue." H (i BI, poking in front of color screen) "If I 
 definitely focus at some distant pomt, although I can't see the point, — 
 but yet I have this tactual cue that helps me to establish and hold it, — 
 then the color seems to retreat and become very suggestive of a surface." 
 H (i BL, poking in front of color screen) "I then attended to the punching 
 and tried to fixate a plane perpendicular to the end of the stick. At first 
 the color was indefinite in localization but, as fixation steadied, there was 
 a sudden shift of that color. It seemed to start to retreat, and then became 
 very suggestive of a surface perfectly stable in localization for a second or 
 two." A (2 BL, poking and told to concentrate attention on the poking) 
 "It seems every now and then that I just poke a flat surface, and it looks 
 like a flat surface." A (2 BL, poking) "It's hard for me to think that I'm 
 poking the same thing that I see, but when I do get that reaUzed, aU of a 
 sudden, just in a flash like that, it's a surface behind the screen." 
 
 In this experiment the cues to localization were both 
 tactual and auditory. The Os were guided not only by sensa- 
 tions in the hand and arm, but also by sounds produced when 
 the rod struck the cardboard. It therefore seemed advisable to 
 see whether sound alone was an adequate cue to the change of 
 film to surface. 
 
 Immediately after the color had been exposed, E tapped 
 without jar, either on the back of the cardboard to which the 
 color was pasted, or else on another piece of cardboard of the 
 same kind, held in the hand before or behind the colored stimu- 
 lus. The taps were mostly given in pairs, with an interval of 
 o.g sec. between the two taps of each pair and an interval of 
 
 2 sec. between pairs. The length of the exposure varied with the 
 number of taps given. The usual mmiber was 5 pairs, requir- 
 ing an exposure time of 15 sec. The total exposure time seem- 
 ed to make no difference to the general results. 
 
 At first, again, the Os found it impossible to link up the sound 
 and the color. The two experiences seemed to be disparate and 
 unrelated. 
 
FILM, SURFACE, BULKY COLORS AND INTERMEDIATES 467 
 
 Br (i BG, tap 2)'" "All I can say is that I couldn't bring the two ideas 
 together. The color is much nearer than the sound. The hollowness of 
 the sound gave an idea of solidity, which was not borne out by the filminess 
 of the color. The sound was from cardboard, distinctly objective." A 
 (2 BL, tap 10, in front of color screen) "I can't reaUze that it's the same 
 thing. Seems as if you're tapping behind the color. Your tapping doesn't 
 affect the color. I'm looking at the color, and you're tapping away off, but 
 they have no connection for me. I can't get them connected." Br (i BG, 
 tap 2) "No realization. I saw the color as nearer than the sound." Bi 
 (i R, tap 10) "The auditory localization is very compUcated. I'm afraid 
 I'm pajrmg too much attention to auditory locaUzation and not realizing 
 that the sound comes from the colored stimulus." H (i BG, tap i) "No 
 change. I never localized the tap so far as fixation was concerned." H 
 (i BG, tap 10, in front of color screen) "I didn't succeed in bringing the 
 tapping and the color into relation." A (x BG, tap 10) "The color is in the 
 focus of attention aU the time, and I hear the tapping as an incidental 
 noise." 
 
 Again, however, when the Os had had five or six hours' 
 practice in these auditory localizations, the colors were changed 
 from film to surface as they had been by the other two methods. 
 
 H (2 BL, tap 10) "It's hard to fix the distance of those taps and hard 
 to fixate the point with the eyes. However, there was some shift over 
 to the surface thing, i. e., it became definitely localized 20 cm. behind the 
 back screen, and definitely bidimensional." H (i BL, tap 10) "It shifted 
 to surface with the first pair of taps. That was maintained till between 
 the second and third pair. Then I went over into a stare, and the thing 
 became pretty positively filmy. After the third tap, I shifted back to 
 fixation of the place where the taps came from and it became definitely 
 surface with Ughts and shades,'" and nearly papery." H (i BL, tap 10) 
 "I tried to fixate a point just behind the back screen, but this side of where 
 the tapping took place. I maintained that till after the second pair of taps. 
 It seemed to throw the color back a little, but otherwise it remained pretty 
 good film. Then I fixated the point of tapping and got the surface thing." 
 H (i BG, tap 10, behind color screen) "The color tends to fiuctuate be- 
 tween taps. The first pair of taps made no difference. With the second 
 pair, the color definitely retreated. All of a sudden you find it back. You 
 didn't see it go back. Bidimensional. About 20 cm. behind second screen. 
 
 Certainly a surface, opaque and soUd The transformation to surface 
 
 always came with (or just following) the pair of taps." H (2 BL, tap 2) 
 "After the second tap came a readjustment of fixation, and it became 
 definitely surface." H (i BG, tap i) "It certainly does shift over to surface 
 with steady fixation, when the eyes are set to fixate a point where the sound 
 is localized. The shift came just after the tap, and then I tried to locaUze 
 it in terms of the screen, but as soon as I tried to see where it was relative 
 to the screen, it became cloud. Then when I returned to the fixation of the 
 point tapped, it became surface again." Br (i BG, tap 2) "Yes, they 
 seemed to come from the same surface that time." A (i BG, tap 10) "It 
 seems that after I get fiixed for realizing that back screen and I hear you 
 strike it, I'm looking at the color and I'm conscious that it's there, but the 
 
 2»The figure after the word "tap" indicates the number of times that 
 E tapped during the exposure preceding the report quoted; thus "tap 2" 
 means that a single pair of taps was given. 
 
 "There were of course no Ughts and shades; yet the appearance of sur- 
 face occasionally led to their report. So one reported the weave of a 
 cloth, when in fact he was looking at a disc of colored paper. He had, it 
 is right to say, been observing cloths during several preceding experiments; 
 and the phenomenon was evidently an ordinary tied image. 
 
468 MAB*IN 
 
 tapping is in the focus of attention. But then, after you've stopped tap- 
 ping, the color gets into the focus of attention again. In that shift I see the 
 color as back behind the back screen, in two dimensions, just a flash of it. 
 Then it goes back to what it was before." A (i R, tap lo) "I got a flash of 
 it recedmg from the back screen and being a perfectly flat surface right 
 after you tapped." Bi (2 BL, tap 10, in front of the color screen) "Once 
 I saw a surface when you tapped. Then again I didn't. Then you tapped 
 again, and I did see a surface. Between taps the color was filmy." Bi 
 (i BG, tap 10) "For the most part the color was in front of the tapping. 
 Just once I seemed to get a fair surface. I think it came partly by way of 
 fixation." M (2 BL, tap 10) "The color changes from film to surface 
 with each pair of taps, but returns to film between pairs." 
 
 Thus by giving localization through three different sense de- 
 partments we had changed to surface a color that would other- 
 wise have been filmy. In all these experiments, therefore, we 
 are referred, as we were in Series I., to the predominating in- 
 fluence of localization. We have been able to change a film 
 color into the surface color of a determinate object by introduc- 
 ing motives to localization, and the result has been in principle 
 the same, whether the motive was visual, auditory-tactual, or 
 auditory. 
 
 Series III. Observations of Familiar Objects 
 
 Series III was designed to answer the second question pro- 
 posed in our introduction: "Can it be shown experimentally 
 that the object-consciousness is sufiicient of itself to ensure the 
 perception of surface color?" If this were so, the knowledge that 
 one is really looking at a famihar object should enable — ^nay, 
 even compel — one to see surface, and prevent the reduction of 
 surface to film. 
 
 Two forms of double screen were used, one of which was 
 identical in essential details with that previously described, ex- 
 cept that a single-sized opening of 9 cm. diam. was used. sat 
 at distances of 1.5, i, and 0.5 m. from the front screen. No head 
 rest was used. could move the head and eyes freely. No in- 
 structions were given with regard to fixation. Most of the Os 
 foimd it easiest to fixate on the edge of the hole in the back 
 screen.'''' 
 
 The stimuH were famihar objects: a board, a fur neck-piece, 
 a strip of felt, a satin cushion, a woolen sweater, aprons, coats, 
 and cloaks. E placed the stimulus in such a position that it 
 entirely filled the hole, and also projected above the upper edge 
 of the double screen. In direct vision could not have seen the 
 stimulus above the screen and in the hole at the same time; but 
 by shght movements of the head and eyes it was possible to 
 make swift comparisons of the two. Under these conditions, the 
 Os never described the color in the hole as ordinary surface color, 
 except when they also reported some detail of structure which 
 offered a fixation point. 
 
 «2Katz, op. cit., 38. 
 
FILM, SURFACE, BULKY COLORS AND INTERMEDIATES 469 
 
 The other form of double screen was used only in this series. 
 It was designed to compel parallel vision. 
 
 Like the regular form, it was made of grey cardboard fastened to 
 wooden frames. In each screen were cut two holes of 1.5 cm. diam. The 
 wooden frames carrying the cardboard screens were then clamped at the 
 front of a table, and sat as close as possible and looked directly into the 
 front screen, as into a binocular telescope. Since the interocular distance 
 differed sUghtly for different Os, different double screens constructed on 
 the same principle were used, and the precise distances between the holes 
 were chosen empirically to suit the Os. 
 
 The stimuli were small familiar objects: soap, chocolate, 
 paper, and the like. The stimulus was first handed to 0, and he 
 was allowed to examine it as he pleased. Then E placed it be- 
 hind the double screen. did not know how far from the double 
 screen the object was to be placed. With this exception, the Os 
 worked with full knowledge of the apparatus and manipulations. 
 They therefore had "the consciousness of an object upon which 
 the colors arise," which Katz regards as the essential condition 
 for perceiving surface. Yet, under the above conditions of ex- 
 perimentation, surface colors were never reported. 
 
 III. Bulk to Film 
 Series IV. Episcotister Before Surface With Fixation Point 
 
 We planned Series IV as a direct parallel to Series I. We in- 
 tended, that is, to begin with a positively bulky color, and to 
 carry this over by intermediates, phenomenological or psycho- 
 logical as the case might be, to film color. Bulky colors were 
 secured without the use of actually btilky stimuh by means of 
 colored episcotisters.^' When an object is seen through a 
 rapidly rotating episcotister, it appears as if seen through a 
 haze of color. Under certain conditions, this haze expands to a 
 positive bulk. 
 
 The episcotister used in these experiments was made in two identical 
 parts, each of which was constructed from a disc of thin but very stiff card- 
 board, 51 cm. in diam., by cutting out 6 sectors of 30° each, leaving 6 
 equally spaced sectors of 30° each. These were covered on the one side 
 with blue paper and on the other with reddish orange paper. These two 
 colors were chosen for convenience. Any other colors might have been 
 used. In cutting out the sectors a cardboard margin, 2 cm. wide, was left 
 on the outer edge, and a cardboard disc, 12.5 cm. in diam. at the center for 
 support. The episcotister was mounted on an electric motor, operated by 
 pedal. By proper combination of the two pieces, it was easy to use any 
 amount of color from 180°, which gave a sUght, vague, barely perceptible 
 haze in front of the background, to 360°, which of course was no longer an 
 episcotister, but a soUd color disk. It was possible to use either pure blue 
 or pure red-orange or any desired mixture of the two colors, by using the 
 blue side of the one disc and the red-orange side of the other. Other dif- 
 ferences in color could be obtained by varying the color of the background. 
 The backgrounds used were whole sheets of Milton Bradley colored papers 
 tacked to the wall of the dark room. The table, carrying the double-screen 
 and the episcotister, was so placed that the back screen was 65 cm. from 
 
 "Ibid., 310. 
 
470 MARTIN 
 
 the colored background. The episcotister could be moved to any desired 
 position between the screen and the background. The exact position 
 seemed, so far as our results were concerned, to make no difference. Dis- 
 tances actually used were lo, 15, 20, 25, 30, 35, 40, 45, and 50 cm. from the 
 back screen. Even with the two extremes, no consistent difference in re- 
 ports could be discovered. As one remarked: "I don't seem to get any 
 variation in the experience, however you move things behind." In the 
 latter part of the series, the distance between the episcotister and the back 
 screen was kept at 20 cm. 
 
 The Os were given tjrpewritten instructions, similar to those used in 
 preceding series. The exposure time was 5 sec. 
 
 Under these conditions, with screen 3 (8 cm. hole) appear- 
 ances precisely similar to those of Groups B, C, D of Series I 
 were secured,''* an increase in the amount of color in the episco- 
 tister having precisely the same effect as the substitution of 
 smaller holes had had in the series without fixation points. A 
 further increase in color, however, did not result in the im- 
 mediate appearance of film, as it should have done on the 
 analogy of Series I. Two other stages, which we have designated 
 as e and f respectively, were found. In Group e, the last vestige 
 of surface is gone, and the fixation point, whose contours are by 
 this time badly blurred, is seen floating deep in a thick substan- 
 tial mist. 
 
 Bi (3 O, bk, 336 0)^"ThB film is well this side of the fixation point and 
 definitely behind the screen. It hints at dimensionality. The surface has 
 been out for quite a while. I didn't realize it was gone." Bi (3 O, bk, 330 
 O) "Now the fixation point is away back in a dense tridimensional orange. 
 It might be daubed on the back of it for all I could tell, only the film 
 wouldn't support an object." H (3 O, bk, 240 O) "Uncertain locaUzation. 
 Curiously, the spot seems to be behind the screen. The color might be 
 this side of the spot. The color seemed more indefinite than the spot. Very 
 fuzzy, felty, soft. No particular objective reference. Cloudy. Trans- 
 lucency is suggested." H (3 O, bk, 330 O) "The spot always seems to be 
 back in the color, like something indistinctly seen through haze. The haze 
 is simply out there, and that is all you can say. The haze is around the 
 spot and in front at the same time." Br (3 O, bk, 330 0) "Fairly dense 
 haze. The fixation point is floating somewhere in the haze. The position 
 of the whole thing seems moved toward the hole, but not as far back as 
 when seen on the surface." Br (3 O, bk, 336 O) "The haze was still denser. 
 The fixation point came and went. Very indistinct. It might be located 
 anywhere within the haze." Br (3 BL, bk, 270 BL) "The dense haze fills 
 all the space up to the hole. The spot is somewhere behind that blue haze 
 or immersed in it, so that you can see the blue in front of it." A (3 BL, 
 bk, 342 BL) "Dark blue mist. I could see the fixation point through it a 
 good way back. I couldn't see any surface." A (3 O, bk, 339 O) "I looked 
 into an orange mist that came up to the back screen, and far back in the 
 mist I could see the fixation point very vaguely. I don't think the mist 
 looked as if it went farther back than the fixation point, but I'm not cer- 
 tain."A (3 O, bk, 348 O). "Looking into orange mist. It came up to the 
 
 "For facility in cross comparison, we shall continue to refer to these 
 by the same letters as in Series I, but shall use small letters instead of 
 capitals, so that the two series may be kept distinct. 
 
 ^The second number in the parentheses indicates the number of de- 
 grees of color in the episcotister; the letter following it, the color used on 
 the episcotister. 
 
FILM, SURFACE, BULKY COLORS AND INTERMEDIATES 471 
 
 back screen, and I don't know how far it extended; not very far. Fiery 
 kind of mist. It looked as if it might have been made up of thousands 
 of little particles. I could make out the fixation point, somewhere in the 
 mist. Its outline wasn't distinct." M (3 BL, bk, 210 O) "Fixation point 
 barely visible through a duU pink tridimensional haze. The fixation 
 point is indefinitely objective and bidimensional." 
 
 In Group f, the fixation point is no longer an object, localized 
 with difficulty in the midst of a thick haze. It has become a 
 film, as positively indefinite in localization as the color around it, 
 which approximates the films obtained by the reduction of sur- 
 faces, the only difference being that it is slightly more suggestive 
 of tridimensionaUty. 
 
 H (3 BL, bk, 180 BL) "At first glance it's the old indefinite cloudy 
 thing. Then I got my fixation point, but couldn't hold it. There was a 
 good deal of fluctuation. Localization was not very certain, but there was 
 a tendency to put it behind the back screen. Bidimensional, but mi^ht 
 easily have been tridimensional. No objective reference. Strong hint 
 of translucency, but yet I couldn't actually say that I saw into it; more 
 of the cloudlike effect. The fixation point shared the character of the color." 
 H (3 BL, bk, 300 BL) "I saw the fixation point dimly all the time. No dif- 
 ference between the spot and the blue. Indefinite localization. Strong hint 
 of tridimensionaUty. Strong hint of translucency, and yet I couldn't see 
 into it. For the most part I haven't been able to focus on the fixation 
 point." Br (3 O, bk, 330 0) "I saw the fixation point as a very faint grey 
 filmy spot, not flat. It had a film in front." Br (3 O, bk, 342 O) "Film, 
 perhaps a Uttle denser. I see the fixation point as a grey nebulous mass in 
 the center." Br (3 O, bk, 348 O) "Very dense film. Slight greyishness in 
 the center. It didn't have any definite outlines, and didn't appear like a 
 fixation point at all. Just a bit of grey film." A (3 O, bk, 339 O) "I could 
 see the fixation point very vaguely. I don't think the mist went back any 
 farther than the fixation point, but I'm not certain. I can't say whether 
 the fixation point is standing out in the mist or whether it's behind it." 
 M (3 BL, bk, 180 O & 150 BL) "The fixation point is barely visible in a 
 mass of red fog, which is tridimensional. The fixation point is filmy, in- 
 definitely locahzed and non-objective." Bi (6 BL, bk, 324 BL) "Fixation 
 point is back again just a little. I can make it out with some certainty 
 now, just a little dark spot. It's filmy, too." Bi (6 BL, bk, 327 BL) "Able 
 to see the fixation point about a third of the time. It's also filmy." Bi 
 (3 BL, bk, 180 O & 30 BL) "Fixation point and fixation screen are both 
 films now, but not very far removed from surface. All the color is close 
 together in a thick plane, no texture." 
 
 When the amount of color in the episcotister was increased 
 so much that the fixation point was no longer visible, the result- 
 ing appearance was a film, practically identical with the films 
 found in Series I. One called attention to this fact. 
 
 Bi (6 BL, bk, 330 BL) "Fixation point gone. It's a great film, filling 
 up the whole area behind the cardboard screen. As far as I can tell it 
 doesn't look any different from the plane colors of the other series." 
 
 Here, as in Series I, there were of course a few equivocal 
 cases, which seemed to fall midway between two of our groups. 
 The following may serve as examples of a suggested intermediate 
 between e and f . 
 
 Br (3 O, bk, 180 BL & 150 O) "Pink film. The fixation point floats 
 in the film. Fairly dense. Pinkish right up to the hole. The spot is filmy, 
 but seems to lie right back in the film." H (3 BL, bk, 180 BL) "The fixa- 
 
472 MABTIN 
 
 tion point was definitely localized at 30 cm. or more behind the screen, 
 probably bidimensional. No objective reference. Opaque, soUd. The 
 color was different. I know little about the color. Not definitely local- 
 ized except behind screen. So far as I can say, it was bidimensional. I 
 don't know that I saw it dimensionally at all. No objective reference. 
 Opaque; whether solid or not I don't know." 
 
 The dependence of all these appearances upon fixation is il- 
 lustrated by many reports, of which the following may serve as 
 examples. 
 
 A (3 BL, bk, 306 O) "The fixation point is back. I can't see it on a 
 surface. When I see the fixation point, the mist extends lots farther back 
 than when I don't see it." A (3 BL, bk, 312 O, immediately following the 
 previous report) "At first I looked into orange mist, which had a little 
 tinge of lavender, extending back an inch or so. Then I looked carefully 
 and saw the fixation point, and the mist spread away back." Br (3 BL, 
 bk, 300 BL) "No fixation point visible, just dense blue film. I can't see 
 through it as far as when I looked through it to the fixation point." Br (3 
 BL, bk, 321 BL) "When my fixation comes front, the color becomes very 
 dense and occupies all the space between my fixation and the screen; 
 when 1 maintain fixation, I get what I described last time, i. e., approximate 
 surface behind at the fixation point, with dilute film out in front, which 
 came toward the screen, but not to it. The surface was bidimensional; 
 the rest, tridimensional." Bi (3 BL, bk, 324 BL) "No fixation point visible. 
 I could make out an approximate surface back where the fixation point 
 would have been if I comd have seen it. The rest was thick haze, uncer- 
 tainly localized." H (3 O, bk, 336 O) "If I look simply at the color I get it 
 indefinitely localized, bidimensional, hinting at tridimensionaUty; sort 
 of a rudimentary tridimensionaUty. No objective reference. Pretty much 
 the cloud. A little hard, maybe. Hints at translucency. The spot is 
 simpler a darker bit in the color. If I look for the spot, then I seem to be 
 able to throw the spot back." 
 
 A short supplementary series was tried in which the back- 
 ground, instead of being a plain sheet of colored paper, was a 
 sheet of cardboard bearing the printed symbols PSYCHOL- 
 OGY I B and a rough drawing of an arrow. Screen opening 
 no. 3 was used, and the amount of color in the episcotister was 
 varied from 180° to 330° by steps of 30°. The resulting ap- 
 pearances were surfaces obscured by hazy or bulky colors, 
 which varied in depth and density according to the amount of 
 color in the episcotister. The more color in the episcotister, the 
 denser of course was the haze. The apparent depth or degree of 
 tridimensionaUty increased up to about 270° and then dimin- 
 ished as more and more color was added. When the letters could 
 no longer be seen the resulting appearance was the familiar film. 
 
 Series V. Episcotister Before Surface Without Fixation Point 
 
 Series V was undertaken as a forlorn hope to see whether our 
 practised Os might possibly report the impression of bulky color, 
 as in stage e of the preceding series, even though the fixation 
 point had been removed. The arrangement, therefore, was iden- 
 tical with that of Series IV except that the colored backgrounds 
 were uniform over their entire surface. 
 
FILM, SUKPACB, BULKY COLORS AND INTERMEDIATES 473 
 
 The attempt was partially successful. Usually, it is true, the 
 reported color was almost identical with the films obtained from 
 the reduction of surface colors; like other films, it was non- 
 objective, unlocalized, and pretextural; but it often seemed to 
 hint more strongly at tridimensionahty; and sometimes tri- 
 dimensionality of uncertain amount was positively ascribed to 
 it. 
 
 Br (3 O, 180 O) "Red bulk rather than red film, like a red liquid." 
 Br (3 O, 180 O) "Dense red film, redder than before. It's almost as if I 
 were looking through a red light or into a brilliant flame. Luminosity." 
 H (6 O, 180 O) "Too luminous for cloudy; too much like a glowing ember 
 where you only see the glow. No other perceptive motive there. It tends 
 toward a glassy or liquid bulk, though it isn't that." Bi (3 O, 270 O) 
 "Thicker now and more nearly tridimensional than before." A (3 BL, 180 
 BL) "It comes up to the back screen and seems to come out just a little 
 bit at the middle; just floats out like mist." H (3 O, 180 O) "Suggestion 
 of tridimensionality. Hint of depth. Sometimes part of the field was 
 almost bulky." H (3 O, 279 BL) "For the most part behind the screen, 
 but not very definiteljr localized. Tridimensional for the most part. Trans- 
 lucent, like bulky liquid tending slightly toward gaseousness. It definitely 
 had depth." M (3 O, 180 O) "Very dense haze. Comes up to the back 
 screen and extends back a short distance." M (3 BL, 330 BL) "Sta- 
 tionary blue haze, doubtfully tridimensional, suggestion of translucency. 
 Perhaps lighter in front and dark behind. Indefinitely localized." 
 
 Associated with the tendency toward bulkiness, the Os some- 
 times reported a sort of incomplete objectivity. 
 
 H (3 O, 180 BL) "For the most part tridimensional, although the 
 tridimensionality increased and decreased, with a wavelike effect. Ob- 
 jective reference, but not specific. It didn't come as an object, but you 
 could make it into an objective something." 
 
 Thus in Series V we seem to have found something like true 
 psychological intermediates between film and bulk. Equivocal 
 appearances tending more or less toward tridimensionahty were 
 reported from time to time throughout the series. Whereas, 
 under conditions of binocular vision, film changes into surface at 
 a single step by way of superadded locaUzation, film seems to. 
 pass into bulk by insensible degrees, as if by mere extension of 
 that suggested but unreaUzed tridimensionality which all Os, 
 those of Katz as well as our own, have remarked.^^ We beKeve, 
 nevertheless, that these intermediates are still nothing more than 
 phenomenological. It must be remembered that the bulk in our 
 bulk-to-film series was itself only genericaUy objective, an affair 
 of haze or mist in general, and not specifically objective like the 
 surface color of paper or cloth; nor was the background so 
 marked or shaped as to suggest a determinate object seen 
 through an intervening haze; it was mere expanse of color with 
 a central fixation point. Even in stage e, may report entire 
 lack of objective reference, especially at an early stage of train- 
 ing. Hence it is probable that we are here in presence of percep- 
 tive formations whose likeness to any' common object of percep- 
 
 ^Katz, op. dt., 7. 
 
474 MARTIN 
 
 tion is markedly less than that in the formations named b, c, d. 
 This conclusion is borne out by reports of vague or general ob- 
 jectivity, reports which increase in frequency as the training of 
 the Os progresses. 
 
 For the reader's convenience we insert here a- brief table of thephe- 
 nomenological results of our two main Series. 
 
 SERIES I 
 
 A 
 
 Soft surface 
 
 SERIES IV 
 
 B 
 
 Veiled surface 
 
 b 
 
 C 
 
 Fogged surface 
 
 c 
 
 D 
 
 Immersed surface 
 
 d 
 
 
 Substantial haze 
 
 e 
 
 
 Cloudy haze 
 
 f 
 
 
 Film 
 
 
 A, B =b, C =c, D = d, and e are without doubt perceptive formations. 
 We believe, as has been said above, that f is also psychologically a per- 
 ception; in spite of the unfamiliarity of colored hazes in everyday Ufe, 
 the Os tended more and more, as observations went on, to report a vaguely 
 objective reference. The fact, however, that any degree of doubt or hesi- 
 tation can arise at this point supports our conclusion that the ultimate 
 film is more suggestive of tridimensionality, less positively bidimensional, 
 than Katz has made it. 
 
 In connection with these bulky hazes, a rather curious phe- 
 nomenon seems worthy of remark. Some of the Os, most fre- 
 quently Br, reported not only positive tridimensionality, but 
 also differences of hue, between the front and back regions of the 
 haze.^' 
 
 Br (6 R, 300 BL) "Dense blue film with the faintest tint of purpUsh- 
 ness in the front and dense blue toward the back." Br (6 R, 270 BL) 
 
 "Deep purple, pinkish toward the front In all these the film begins 
 
 at the screen and goes back from there. Usually the front is filled in with a 
 lighter film, misty." Br (3 YG, 180 BL) "Greenish film, whitish-yellowish 
 ^t the front and green farther back. It starts at the back of the back screen 
 and goes back a long distance, almost to infinite distance. It's the sort of 
 haze that I've seen heaps and heaps of times at sea." 
 
 One of the byproducts of Series IV and V was the discovery 
 that bulky films could be produced with somewhat larger holes 
 than the films of Series I. The difference made by the use of the 
 episcotister is well illustrated by the following pair of reports, 
 taken successively in the same observation hour. The first was 
 taken without episcotister in front of color, but with motor run- 
 ning, so that there was no auditory cue to the difference. The 
 second was the same as the first, except than an episcotister with 
 270° of orange was between the color and the back screen. 
 
 2'It will be remembered that theOs were not instructed to report upon 
 hue. Hence these observations were spontaneous. It is possible that the 
 other Os might occasionally have noted these differences, if the instruc- 
 tions had called for them. 
 
FILM, 8UKPACE, BULKY COLORS AND INTERMEDIATES 475 
 
 H (6 O) "Quite definitely localized, bidimensional, fairly definite 
 objective reference, in so far as it was a smooth surface (paper or anything 
 of the kind). Solid, opaque." H (6 O, 270 O) "Very indefinitely locaUzed. 
 I don't know whether I can even say it was back behind the back screen. 
 Bidimensional, I guess. There were moments when it seemed almost 
 tridimensional. Hint of translucency. The term 'cloudy' is not ade- 
 quate, 'filmy' is not adequate. It's just a bit of luminous color." 
 
 Series VI. Control Series 
 
 The best guarantee of the validity of experimental results is 
 agreement among a number of trained Os. Our quotations have 
 shown that there was in fact a very high degree of agreement. 
 Series VI was planned as a further test of reliability. To see 
 whether the use of the episcotister carried with it any imaginal 
 suggestion which might have exaggerated the differences be- 
 tween the film colors of Series I and the bulky colors of Series 
 IV, we turned the episcotister in such a way that it did not come 
 in front of the colored background at all. The episcotister was 
 left on the motor, so that it might run at the same speed and 
 produce the same sound. The shadow of the episcotister fell on 
 the wall some distance from the colored paper, so as to have no 
 effect on its appearance. Since the colored backgrounds were 
 tacked to the wall, they could not be jarred or in any way af- 
 fected by the motor on the table. Hence it made no objective 
 difference in the color whether the motor was running or 
 not. The only difference between exposvu"es during which the 
 motor was running and those for which it was turned off was the 
 sound. It was assumed that the sound might act as a suggestion 
 to the Os that they were seeing the same sort of thing as in 
 Series IV and V. Between exposures E moved the motor about 
 so as to suggest, by sound again, that the episcotister was being 
 put in front of the color or removed. No verbal suggestion was 
 given. 
 
 The screen openings used were 3 and o (8 and 2 cm. respec- 
 tively) : screen 3 because it was that most frequently used in the 
 series to be tested, and screen o because it had invariably given 
 film in Series I. With Os A and Br screen 6(14 cm.) was ako 
 used for purposes of comparison. 
 
 In every case the Os immediately or almost immediately dis- 
 covered that the two experiences were identical. Br and A were 
 absolutely unaffected by the suggestion. H, as was shown by 
 the cautious andhesitating nature of his reports, was appa ently 
 affected by the suggestion in the first two observations, but in 
 the third he settled down to objective report, and thereafter 
 found no difference between the appearance of the color when 
 the motor was running and when it was not. Fifteen subsequent 
 trials failed to show any the least trace of suggestion. Similarly, 
 Bi showed a possible shght effect of the suggestion in the first 
 three trials; but even here the differences reported were scarcely 
 
476 MARTIN 
 
 greater than those occurring in the course of a normal series; and 
 his reports in 15 subsequent trials were entirely free from eviden- 
 ces of suggestion. This record affords objective proof of the 
 reliabihty of the Os. 
 
 IV. The Monocular Film 
 (Series VII. Holes of Various Sizes 
 
 In this series the apparatus and procedure were identical with 
 those of Series I, except that, instead of making all observations 
 binocularly, the Os were asked to alternate binocular and mono- 
 cular observations and to compare the two.^* The reports of all 
 Os showed that with monocular observations size of hole makes 
 less difference than with binocular. In every case the color ap- 
 pears as an almost bidimensional film, which tends to he in the 
 same plane as the back screen. In having less suggestion of 
 tridimensionaUty about it than the binocular fihns of previous 
 series, these monocular films are closer to the typical film as Katz 
 conceives it. The following comparative observations illustrate 
 this point. 
 
 H (°, 1) ^' "With one eye the two holes seem very much in the same 
 
 plane The color was just color. Non-objective and milocaUzed. 
 
 It was certainly behind the cardboards, but I couldn't tell more than that. 
 It tended to come up close against the card. Dull, opaque, fuzzy, soft. 
 SUghtly suggestive of translucency; not much. Neither bidimensional nor 
 tridimensional. Bidimensional; spread is clear enough, but there is nothing 
 definite about the third dimension. Tridimtensionality is suggested rather 
 in the softness, the fuzziness, the sponginess of the tlung than in anything 
 else." H (°, r) "Little difference between the two eyes. The texture was 
 perhaps a little less fuzzy and soft, but still suggested the third dimension." 
 H (°, b) "Funny how that changes! With two eyes, there is a definite 
 space between the screens. The second was definitely 20 cm. behind the 
 first. The color was stiU non-objective, locahzed at or behind the second 
 screen. More certainly tridimensional, though not in any determinable 
 amount. The texture was more translucent, more motile, as though there 
 were liveness in the color. I think that's probably what we have called 
 luminosity. More dense, glass-like; not clear, transparent glass, but a 
 Uttle more solid." Bi (°, r) "Equally filmy, but pretty much up to the 
 hole. I tried to force my eye back in it, and it didn't work somehow. It 
 left me with something that was not deep in the third dimension, but not 
 surface." Bi (°, 1) "I have a feeling that I ought to be able to force that 
 film to be deep, but I don't beUeve I can. I can pierce it a bit, but it stays 
 pretty much a flat, dense film at the hole." M (°, b) "Indefinitely local- 
 ized, except somewhere behind the back screen. Non-objective. Soft, 
 loose, penetrable texture. DimensionaUty uncertain, but tridimensionaUty 
 is at least strongly suggested." M (°, r) "Non-objective, just a patch of 
 color in the hole. Soft, thin, filmy. Probably bidimensional, though here 
 (i.e., in monocular observations), where differences of localization are not 
 
 *'We took our cue for this monocular series from Katz' suggestion that 
 monocular observation of ordinary objects affords a mode of appearance 
 intermediate between fihn and surface: op. cit., 9. 
 
 *°Small letters after the comma in the parentheses indicate the mode 
 of observation. Thus 1 means "with the left eye;" r, "with the right eye;" 
 and b, "with both eyes." 
 
FILM, SURFACE, BULKY COLORS AND INTERMEDIATES 477 
 
 SO noticeable, dimensionality also seems equivocal. All things seemshoved 
 up more nearly into the same plane. The front screen, the back screen, 
 and the color all lie, or at least look as if they might He, in the same plane." 
 Br (2, b) "Soft, filmy brown medium. LocaUzed near the hole and yet it 
 seems to go back a bit from that. Dense, cloudy appearance." Br (2, r) 
 "Same, except that it's like a plane, even, fine-textured cloth at the hole; 
 really a film, but I can imagine a cloth like that; one of those soft georg- 
 ettes." Bi (2, b) "I have a suspicion that there's a deeper film with two 
 eyes than I saw with a single eye. The one with both eyes is like a fog; 
 the one with one eye is like a rotating disc." Bi (2, r) "I looked for the 
 deep character that I thought I saw before. It is not there with the single 
 eye, though what I saw was certainly not a surface. It still had a filmy 
 character." A (2, b) "I see a brown misty thing up in the hole and behind 
 it, too." A (2, r) "I can't sajr whether it looked up at the screen or be- 
 hind, but it wasn't misty-looking as it was when I saw it with two eyes. 
 More surfacey. Not a slick surface. More like blotting paper." 
 
 Thus it is evident that monocular observations with the 
 small holes gave a flatter, more nearly bidimensional color than 
 did the binocular observations. 
 
 With the larger holes, the reverse proved true. Although the 
 object could now be easily identified, it did not appear to have so 
 sharply defined a surface as in the binocular observations. 
 
 M (9, b) "Ordinary cloth, definitely locaUzed behind the screen." 
 M (9, r) "Positively cloth, but greatly softened, and not a sharply defined 
 surface. Velvety. Not much space between fore and back screens or be- 
 tween back screen and cloth. All are in nearly the same plane. Less cer- 
 tainty of locahzation than when viewed binocularly; or rather, less posi- 
 tiveness of difference in locahzation. The very question of locahzation 
 seems out of place in these monocular observations." H (9, r) "Little dif- 
 ference monocularly between this and the last (hole no. 2, 6 cm.). Opaque, 
 softened, a httle more certainly behind the screen when I use the side of 
 the hole as a reference point, but not when I look straight at it." H (9, b) 
 "Fairly sharp surface. No hint of softness. Certainly a colored something. 
 Cloth, or it might be a colored wall. Localized very definitely behind the 
 second screen at a distance of 20 cm., more rather than less. No suggestion 
 of depth, just an opaque, soUd substance." Bi (8, b) "I get a film that goes 
 all the way back to the background, but I can nearly see the cloth, if it is 
 cloth. The cloth was away back by the wall. The haze was everywhere 
 between the cloth and the hole. Part of the time it seemed equally dense 
 throughout, and part of the time it seemed thinner up in front and denser 
 behind." Bi (8, r) "Dense film. Thick plane at the screen. It huddles up 
 to the front. It stays up. I can't force my gaze back through it." Br 
 (8, b) "Surface, some distance behind second hole. Flat piece of cloth." 
 Br (8, r) "Film, which is difiicult to distinguish as film, because it seems 
 that flat plane, lying on the back hole." A (9, b) "Back behind screen, 
 8 or 9 in., and now it looks with two eyes very much as it did with one 
 a while ago, although it has more quickness and hveness." A (9, 1) "Right 
 up in the screen, and a httle duller than before. By dull I mean that it 
 hasn't that quick, lively look." 
 
 From these and similar observations it is evident that, with 
 monocular observations, the largest holes hke the smallest gave 
 an appearance of filmy texture. In fact one (H) remarked: 
 "The one-eye appearance is nearly uniform for the whole range 
 of sizes." 
 
478 MAKTIN 
 
 For the sake of completeness we append a few typical obser- 
 vations taken with intermediate sizes of holes. 
 
 Br {^, b) "More filmy and behind the back hole. I couldn't say 
 whether it was really bulky. Indefinitely localized. No definite front 
 surface." Br (4, 1) "It looked a little more flat and seemed more a plane, 
 but not as definitely a plane as when I saw it binocularly through those 
 large holes." Br (4, r) "Same. Filmy, soft, fuzzy appearance, but not so 
 thick as when I saw it with both eyes." Br (6, b) "Filmy, thick, dense, 
 foggy." Br (6, r) "Flat, even, filmy thing; really like a spectral color hung 
 in space." Br (6, 1) "Same. More plane and even than with both eyes." 
 M (4, b) "Brown surface, seen dimly and localized with difficulty behind 
 a brown medium of increasing density which Ues somewhere behind the 
 back screen. The surface is cloth." M (4, r) "Brown cloth, but very soft, 
 thin, and filmy. The brown might be at, or even slightly in front of, the 
 screen. Certainly there is not much if any difference of distance between 
 the screen and the color, or between the two screens; but the screens 
 themselves are not localized with any degt-ee of precision. Not so much 
 suggestion of tridimensionality as with two eyes, but the surface character 
 is not positive either." 
 
 Without stopping to multiply illustrations we may therefore 
 say that, throughout our whole range of holes, including those 
 that were large enough to permit a recognition of the object, 
 monocular observations gave appearances which in texture, 
 dimensionahty, and indefinite locahzation approximated fairly 
 closely to the Katz film. 
 
 Series VIII. Monocular Observations with Episcotister 
 
 Obviously the next step was to find out what happens to 
 bulky colors under monocular observation. For this purpose we 
 used an experimental arrangement identical with that of Series 
 IV. In the episcotister we used dark blue, orange, and various 
 combinations of the two. The background was a sheet of dark 
 blue paper with a black fixation spot in the middle. As in 
 Series VII, monocular and binocular observations were alter- 
 nated for purposes of comparison. It was found that, whereas 
 binocular observations with a suitable amount of color in the 
 episcotister gave good bulky color, monocular observation re- 
 duced this to a filmy appearance. 
 
 H (3 BL, bk, 180 BL, b) "The spot was definitely 25 or 30 cm. behind 
 the back screen. It tended to give considerable depth to the color. The 
 color was transparent, dense, glassy, flinty. It was still non-objective. 
 I don't know whether the color filled all the space between the screen and 
 the dot, but it tended to have that meaning." H (3 BL, bk, 180 BL, r) 
 "More nearly flat in form. Non-objective, but it suggested some kind of 
 soft surface. Quite opaque. No translucency or transparency. The color 
 was localized back at the dot. I don't think the color or the dot either was 
 quite as definitely locaUzed as the dot was before (i. e., binocularly), but 
 the color was with the dot." H (3 BL, bk, 180 O, b) "The fixation point 
 was definitely 25 or 30 cm. behind the back screen. The color itself was 
 certainly non-objective and pretty definitely tridimensional; a mass, al- 
 most a body of it, ■extending from the region of the screen back to the dot. 
 I couldn't be sure whether it extended clear to the dot or not. Trans- 
 lucent, soft, not much glassy." H (3 BL, bk, 180 O, r) "The color and dot 
 
FILM, SURFACE, BULKY COLORS AND INTERMEDIATES 479 
 
 seemed to lie together and tended to be localized nearer the screen, but 
 there was not absolute definiteness of localization. Less massiveness to 
 the whole experience, but yet not a definite surface with objective refer- 
 ence. Soft, fuzzy, suggestive of tridimensionality but not in the same 
 sense. The whole thing is dead, static, more like a soft surface. No sug- 
 gestion of translucency or transparency." A (3 BL, bk, 270 BL, b) "I saw 
 tridimensionahty that time. Thick fog came up to the hole. I could see 
 into it. I couldn't tell how far it extended. The fixation point was in it 
 somewhere." A (BL, bk, 270 BL, r) "Bidimensional, as if I were looking 
 at something that let the light through; but I saw it in two dimensions. 
 Like thick colored glass that's soft-looking, but you don't see anything but 
 the front of that glass; Uke glass in church windows. If you didn't know 
 it was glass you'd hardly call it glass. Not transparent, but translucent." 
 A (3 BL, bk, 270 BL, b) "I couldn't help seeing it as tridimensional. I 
 tried to hold it in the same place as with one eye, but I couldn't. The two 
 things don't look anything alike." Br (3 BL, bk, 210 BL, b) "Bulky, 
 dense medium. The fixation point was buried in it somewhere." Br (3 
 BL, bk, 210 BL, r) "Flatness, filminess, rather than surface, and yet it's 
 all in one plane, or almost all in one plane. With two eyes you seem to be 
 looking into something almost glassy-like. This is Uke the blue of the sky. 
 This has the appearance of being penetrable, but you don't have that im- 
 pression of seeing space that you have with the other." Bi (3 BL, bk, 180 
 BL, b) "Hazy smokiness which goes clear back to the fixation poirt. It's 
 open and loose. I didn't see it come clear up to the screen I didn't see 
 it fall short either." Bi (3 BL, bk, 180 BL, 1) "This tended with one eye 
 to be flat. The fixation point tended to come up to the screen. Anyway 
 they got crowded together." M (3 BL, bk, 270 BL, b) "Fixation point 
 badly blurred and localized with difficulty in a dense, semi-transparent 
 fog, which began at or near the back screen and went back an uncertain 
 distance." M (3 BL, bk, 270 BL, r) "Color and fixation point are both 
 filmy, and seem to lie in approxiinately the same plane as the screens. 
 Both screens seem to he in very nearly the same plane. No part of the 
 total experience is locaUzed with any degree of precision. The color ap- 
 pears soft but not actually penetrable. It is non-objective." 
 
 Hence it is evident that all the color appearances which we 
 have here studied, when viewed monocularly, take on a more or 
 less filmy appearance. They become indefinitely localized, pre- 
 textural, predimensional; and except where familiar configura- 
 tion carries with it the meaning of a familiar object, as in the case 
 of the screens themselves or the brown cloth as seen through the 
 largest holes, they are also non-objective. 
 
 V. Conclusions 
 
 1. Our experimental results indicate that the simple visual 
 datum is a "film," predimensional but tending more strongly to- 
 ward tridimensionality than Katz' bidimensional Fldche. The 
 film is also pretextural, non-locahzable, and non-objective. It 
 has an attributive character of spatial spread or diffusion, to 
 particularize which we must have recourse to quahfied percep- 
 tive analogy. 
 
 2. There are no psychological intermediates between fihn 
 and surface. Our results show a sharp break between the true 
 film and the immersed surface, the haziest of the surfaces, a 
 
480 MARTIN 
 
 break which involves a complete change of attitude. The local- 
 ization of the color in the case of the surface is the determining 
 factor in the shift from film to surface. 
 
 3. Between film and bulky color the shift of attitude is 
 more tentative and cautious, and the transition is in a sense more 
 gradual. There are, nevertheless, no true psychological inter- 
 mediates. 
 
 4. There is a wealth of phenomenological intermediates, 
 any one of which might have been stabiUzed as a perception if 
 our everyday experience had required such stabilization."* 
 
 '"We make no claim to have reviewed the whole series of intermedi- 
 ates. It is possible that [another approach to the general problem, and 
 in particular the bringing to bear upon the film of strong perceptive motives 
 other than that of localization, might lead to another classification and 
 might reveal new phenomenological modes. We can say, however, that 
 in a good deal of casual experimenting, such as naturally suggested itself in 
 the course of an investigation like the present, we have found no 'leads' 
 toward a further variety of visual appearance at the level of our enquiry. 
 
^^y^%: 
 
 f»"t 
 
 
 -^c 
 
 
 
 V '*- ,„ -*? 
 
 .*-v^*^- 
 
 *j^f-v:^'*^' 
 
 Ni'M 
 
 lIS 
 
 ^^T-' 
 
 
 r«^