September 9, 1898.] 



SCIENCE. 



315 



should not exist, and the color of any light 

 bright enough to affect this portion of the 

 retina at all will at once be recognized, 

 von Kries declares this to be a fact. Two 

 colors, equally bright in strong light, will 

 remain so at all illuminations if their image 

 falls entirely on the fovea ; but, if not, the 

 color which is of the shorter wave-length 

 ■will in general be the brighter. 



Vision by strong light, and color-vision, 

 since both are possessed by the fovea, must 

 be effected by the mechanisms of that re- 

 tinal area, and these sensations von Kries 

 attributes to the cones, which are supposed 

 to be furnished with a tri-chromatic color- 

 apparatus, and to afford the sensations of 

 color and a compound sensation of white. 

 If objection is made to the compound 

 white the details of this latter apparatus 

 might be varied, might even approximate 

 that of Hering's theory, without affecting 

 the importance of the hypothesis, the es- 

 sence of which is the twofold nature of the 

 sensation of brightness. 



Such a theory explains easily the fact 

 that grays compounded from different pairs 

 of complementary colors, and equally bright 

 in ordinary light, cease to be so in faint 

 light. They are equalized at first by the 

 cone-apparatus, and are seen in the faint 

 light chiefly by the rod-apparatus, in which 

 the scale of brightness is entirely different. 



G. E. Miiller makes the acute suggestion 

 that the visual purple may not be a visual 

 substance at all, properly speaking ; but, 

 while concerned chiefly with the phenomena 

 of adaptation, may act also as a sensibilisator 

 — to borrow a photographic term — for the 

 white-sensitive substance, increasing its 

 susceptibility in faint light. This modifi- 

 cation of von Kries' hypothesis is perhaps 

 simpler than the original and equally satis- 

 factory. 



Still another hypothesis for separating 

 the white from the color-sensations is that 

 the sensation of white, from an evolutionary 



standpoint, was developed earlier than the 

 sensations of color, and that the mechanisms 

 of the latter are to be regarded as evolved 

 from that of the fundamental sensation, and 

 as modifications of it. Upon this idea Mrs. 

 Franklin has founded her ingenious theory 

 of light-sensation. Abney has made a simi- 

 lar suggestion, but in general terms only. 



Such is a brief and hasty summary of the 

 progress of color- theory. We may well ask 

 for the result. In the general shifting, what 

 views have maintained or gained a footing ? 

 A few, I think, are fairly well established. 



1. The number of color- sensations is 

 small, and all color-theories positing a large 

 number are to be distrusted. If experi- 

 mental work is of any value whatever, it is 

 certain that all light-sensations, for all pur- 

 poses, may be expressed by a small number 

 of variables. The Young-Helmholtz^ theory 

 demands three. Hering's requirements, as 

 Helmholtz has shown, may be expressed in 

 terms of three, although the number of fun- 

 damental color-sensations, using color in its 

 ordinary sense, is four. Such theories as 

 those of von Kries and Mrs. Franklin re- 

 quire four variables, such as that of Ebbing- 

 haus five. The introduction of a much larger 

 number is gratuitous and unnecessary. 



2. Out of this number of variables at 

 least one is to be alloted to the white- 

 sensation, or that which is closely akin to 

 it, the sensation of brightness. It is no 

 longer possible to think of white entirely as 

 a compound sensation, however it may be 

 compounded physically. It is unnecessary 

 to recapitulate the arguments for this state- 

 ment, drawn largely from the three forms 

 of total color-blindness. 



3. "White, however, can hardly be thought 

 of as an entirely independent sensation. 

 The phenomena of vision by faint light, the 

 facts of peripheral vision, show that, under 

 certain circumstances, color-sensations con- 

 tribute their quota to the colorless one, and 

 in differing amounts at differing brightness. 



