226 ON ACCIDENTAL OR SUBJECTIVE COLORS. 



image is projected on a hlue surface ? After having regarded fixedly a small 

 circle of paper placed in the red ray, and then turned the eyes upon a larger 

 circle illuminated by the yellow ray, M. Plateau distinctly saw upon the latter 

 an image of a fine yellowish green, while, if the retina had simply become insen- 

 sible to red, it could evidently have perceived only a blackish image. 3d. The 

 apparently irregular process of decrease presented by the accidental images. 

 The following fact would suffice, it might seem, to overthrow it completely : 

 When, after having pressed his eyes in a particular manner, this pressure was 

 made suddenly to cease, M. Plateau, his eyes being closed and covered, saw a 

 large red spot bordered with green ; but as soon as he opened his eyes or cast 

 them on a white surface, these two colors changed into their complementaries ; 

 the spot became green and the surrounding border red. But here no exterior 

 light had been present to fatigue the organ and to render it less sensible to cer- 

 tain colors ; we have here spontaneous sensations of red and green, which 

 change into sensations oi green and red. 



Colors of the seco'tid class. — Theory of contrast. — The most generally ac- 

 cepted theory attributes these colors to contrast ; that is to say, to a moral 

 cause which renders more intense our perception of whatever is unlike in the 

 colors collated with one another, while it weakens the perception of what they 

 have in common; thus a small white object being projected upon a colored 

 ground — a red one, for instance — the effect of the contrast is to diminish our 

 perception of the red constituent part of that white, and to exalt, on the con- 

 trary, our perception of the complementary part, or the green. If, for example, a 

 green object be placed in juxtaposition with a violet color, contrast will enfeeble 

 the perception of the hlue, which is common to the two objects, and will enliven, 

 on the other hand, the sensation of the yellow and red, by which they differ; the 

 first of the two objects will therefore appear more yellow and the second more 

 red, which is in accordance with the law indicated by M. Chevreul. If the 

 objects finally differ in brightness, contrast will likewise intensify this differ- 

 ence, as in the case, for example, of black placed in presence of white, &c. 



Without entirely denying the influence of contrast, M. Plateau thinks that 

 this cause, though capable of increasing or even producing in certain cases the 

 subjective phenomena of the second class, is not that to which they can be col- 

 lectively attributed. His reasons are the following : 1st, how is it possible to 

 admit that a moral cause should be competent to create, in complete darkness, a 

 sensation of light purely imaginary, like that of the aureolas before mentioned ? 

 2d, in the experiments of M. Chevreul, since the effect was manifested when 

 two colors in juxtaposition had no ingredient common to them, the operation of 

 contrast is more difficult to explain. Several physicists, Young among others, 

 have attempted to extend to this second class of colors the hypothesis of a mod- 

 ification in the sensibility of the retina, by assuming that when a portion of the 

 retina is subjected to the action of any color the surrounding parts lose some- 

 thing of their sensibility for that color. This theory, indeed, attributes the 

 phenomena to a physical cause, but it is, nevertheless, wholly insufficient. 

 What, in that case, would become of the phenomena of accidental colors pro- 

 duced in darkness ? How should we explain the mauifestation, in certain cases, 

 of a tint identical with that of the colored object at some distance from the out- 

 line of that object ? &c. 



Theory of M. Plateau. — We shall, in the first place, remark with Scherffer 

 that it results from the apparent change of magnitude of the accidental images, 

 when projected on surfaces more or less distant; that these images are due to a 

 physical modification of the organ. Such would, in effect, be the phenomenon 

 produced, if it be admitted that a definite portion of the retina undergoes any 

 modification ; for, as the resulting image corresponds to a constant visual angle, 

 that image must seem to us the greater according as we refer it to a greater dis- 

 tance. We see no reason, on the contrary, why it should be so, on the hy- 



