Why, then, if it is a true sensation due to physiological causes, and not an error of judgment, which causes the contrast, does the color disappear when the outlines of the gray scrap are traced, enabling us to recognize it as an independent object? In the first place, it does not necessarily do so, as will easily be seen if the experiment is tried. The contrast-color often remains distinctly visible in spite of the black outlines. In the second place, there are many adequate reasons why the effect should be modified. Simultaneous contrast is always strongest at the border-line of the two fields; but a narrow black field now separates the two, and itself by contrast strengthens the whiteness of both original fields, which were already little saturated in color; and on black and on white, contrast-colors show only under the most favorable circumstances. Even weak objective differences in color may be made to disappear by such tracing of outlines, as can be seen if we place on a gray background a scrap of faintly-colored paper, cover it with transparent paper and trace its outlines. Thus we see that it is not the recognition of the contrasting field as an independent object which interferes with its color, but rather a number of entirely explicable physiological disturbances.
The same may be proved in the case of holding above the tissue paper a second gray scrap and comparing it with that underneath. To avoid the disturbances caused by using papers of different brightness, the second scrap should be made exactly like the first by covering the same gray with the same tissue paper, and carefully cutting a piece about 10 mm. square out of both together. To thoroughly guard against successive contrast, which so easily complicates the phenomena, we must carefully prevent all previous excitation of the retina by colored light. This may be done by arranging thus: Place the sheet of tissue paper on a glass pane, which rests on four supports; under the paper put the first gray scrap. By means of a wire, fasten the second gray scrap 2 or 3 cm. above the glass plate. Both scraps appear exactly alike, except at the edges. Gaze now at both scraps, with eyes not exactly accommodated, so that they appear near one another, with a very narrow space between. Shove now a colored field (green) underneath the glass plate, and the contrast appears at once on both scraps. If it appears less clearly on the upper scrap, it is because of its bright and dark edges, its inequalities, its grain, etc. When the accommodation is exact, there is no essential change, although then on the upper scrap the bright edge on the side toward the light, and the dark edge on the shadow side, disturb somewhat. By continued fixation the contrast becomes weaker and finally yields to simultaneous induction, causing the scraps to become indistinguishable from the ground. Remove the green field and both scraps become green, by successive induction. If the eye moves about freely these last-named phenomena do not appear, but the contrast continues indefinitely and becomes stronger. When Helmholtz found that the contrast on the lower scrap disappeared, it was evidently because he then really held the eye fixed. This experiment may be disturbed by holding the upper scrap wrongly and by the differences in brightness of its edges, or by other inequalities, but not by that recognizing of it 'as an independent body lying above the colored ground,' on which the psychological explanation rests.
In like manner the claims of the psychological explanation can be shown to be inadequate in other cases of contrast. Of frequent use are revolving disks, which are especially efficient in showing good contrast-phenomena, because all inequalities of the ground disappear and leave a perfectly homogeneous surface. On a white disk are arranged colored sectors, which are interrupted midway by narrow black fields in such a way that when the disk is revolved the white becomes mixed with the color and the black, forming a colored disk of weak saturation on which appears a gray ring. The latter is colored by contrast with the field which surrounds it. Helmholtz explains the fact thus:
"The difference of the compared colors appears greater than it really is either because this difference, when it is the only existing one and draws the attention to itself alone, makes a stronger impression than when it is one among many, or because the different colors of the surface are conceived as alterations of the one ground-color of the surface such as might arise through shadows falling on it, through colored reflexes, or through mixture with colored paint or dust. In truth, to produce an objectively gray spot on a green surface, a reddish coloring would be necessary."[26]
This explanation is easily proved false by painting the disk with narrow green and gray concentric rings, and giving each a different saturation. The contrast appears though there is no ground-color, and no longer a single difference, but many. The facts which Helmholtz brings forward in support of his theory are also easily turned against him. He asserts that if the color of the ground is too intense, or if the gray ring is bordered by black circles, the contrast becomes weaker; that no contrast appears on a white scrap held over the colored field; and that the gray ring when compared with such scrap loses its contrast-color either wholly or in part. Hering points out the inaccuracy of all these claims. Under favorable conditions it is impossible to make the contrast disappear by means of black enclosing lines, although they naturally form a disturbing element; increase in the saturation of the field, if disturbance through increasing brightness-contrast is to be avoided, demands a darker gray field, on which contrast-colors are less easily perceived; and careful use of the white scrap leads to entirely different results. The contrast-color does appear upon it when it is first placed above the colored field; but if it is carefully fixated, the contrast-color diminishes very rapidly both on it and on the ring, from causes already explained. To secure accurate observation, all complication through successive contrast should be avoided thus: first arrange the white scrap, then interpose a gray screen between it and the disk, rest the eye, set the wheel in motion, fixate the scrap, and then have the screen removed. The contrast at once appears clearly, and its disappearance through continued fixation can be accurately watched.
Brief mention of a few other cases of contrast must suffice. The so-called mirror experiment consists of placing at an angle of 45º a green (or otherwise colored) pane of glass, forming an angle with two white surfaces, one horizontal and the other vertical. On each white surface is a black spot. The one on the horizontal surface is seen through the glass and appears dark green, the other is reflected from the surface of the glass to the eye, and appears by contrast red. The experiment may be so arranged that we are not aware of the presence of the green glass, but think that we are looking directly at a surface with green and red spots upon it; in such a case there is no deception of judgment caused by making allowance for the colored medium through which we think that we see the spot, and therefore the psychological explanation does not apply. On excluding successive contrast by fixation the contrast soon disappears as in all similar experiments.[27]
Colored shadows have long been thought to afford a convincing proof of the fact that simultaneous contrast is psychological in its origin. They are formed whenever an opaque object is illuminated from two separate sides by lights of different colors. When the light from one source is white, its shadow is of the color of the other light, and the second shadow is of a color complementary to that of the field illuminated by both lights. If now we take a tube, blackened inside, and through it look at the colored shadow, none of the surrounding field being visible, and then have the colored light removed, the shadow still appears colored, although 'the circumstances which caused it have disappeared.' This is regarded by the psychologists as conclusive evidence that the color is due to deception of judgment. It can, however, easily be shown that the persistence of the color seen through the tube is due to fatigue of the retina through the prevailing light, and that when the colored light is removed the color slowly disappears as the equilibrium of the retina becomes gradually restored. When successive contrast is carefully guarded against, the simultaneous contrast, whether seen directly or through the tube, never lasts for an instant on removal of the colored field. The physiological explanation applies throughout to all the phenomena presented by colored shadows.[28]
If we have a small field whose illumination remains constant, surrounded by a large field of changing brightness, an increase or decrease in brightness of the latter results in a corresponding apparent decrease or increase respectively in the brightness of the former, while the large field seems to be unchanged. Exner says:
"This illusion of sense shows that we are inclined to regard as constant the dominant brightness in our field of vision, and hence to refer the changing difference between this and the brightness of a limited field to a change in brightness of the latter."
The result, however, can be shown to depend not on illusion, but on actual retinal changes, which alter the sensation experienced. The irritability of those portions of the retina lighted by the large field becomes much reduced in consequence of fatigue, so that the increase in brightness becomes much less apparent than it would be without this diminution in irritability. The small field, however, shows the change by a change in the contrast-effect induced upon it by the surrounding parts of the retina.[29]