312 



SCIENCE. 



[N. S. Vol. VIII. No. 193. 



long wave-length diminishes continually, 

 while that of the shorter wave-lengths in- 

 creases continually, from the center of the 

 visual field to its circumference. The con- 

 ditions under which von Kries worked, 

 however, were so different from mine that 

 I cannot regard my results so far as neces- 

 sarily invalidating his. If his results are 

 confirmed, they show that the sensation of 

 white in the normal eye is not completely 

 determined by the twilight sensation, or 

 that of the totally color-blind. It contains 

 elements derived from, or connected with, 

 the mechanism producing the sensation of 

 color, even in those portions of the retina 

 where no color-sensation exists. 



I have discussed these two theories some- 

 what at length, because our growth in 

 knowledge of the facts of color- sensation 

 has been conditioned largely by their ex- 

 istence. The enormous amount of work 

 which has been done on the vision of the 

 color-blind, on color-vision by varying illu- 

 mination, on peripheral color-vision, not to 

 mention researches upon more purely sub- 

 jective phenomena, has been largely sug- 

 gested by aspects of one or the other of 

 these theories, or undertaken with a view to 

 testing portions of them, and there has 

 seemed no better method exhibiting the re- 

 sults of these researches than by placing 

 them in connection with the hypotheses 

 they were intended to test. I need hardly 

 add that I have been greatly aided in this 

 summing up by the polemical writings 

 emanating from the hostile schools. 



In this respect, at least, the two theories 

 have been eminently useful, and have ful- 

 filled one of the chief requirements of a 

 scientific theory — that its explanations can 

 be tested by experiment. The earlier forms 

 of color-theory suggested by Newton and by 

 Young were hardly such. So long as the 

 specific effect was conceived to be entirely 

 in the central organ, to which the nerves 

 merely communicated the vibrations of 



light, there was little upon which to base 

 experimental work. Helmholtz, by ascrib- 

 ing the specific activity to the nerve-ending 

 itself, made it necessary to describe this 

 activity in some definite way, which could 

 then be tested. The very simplicity of the 

 conceptions of Helmholtz and Hering, at 

 first the apparent guaranty of their truth, 

 has proved their greatest value, but also 

 their greatest difficulty. 



It is not to be wondered at that later 

 theorists have attempted to modify one or 

 the other of these hypotheses instead of 

 starting anew. Many such attempts have 

 been made in the last few years, but few 

 have attained more than a passing notice, 

 and none any general acceptance. One or 

 two, however, are of considerable intrinsic 

 interest, and may command attention for a 

 brief period. 



Ebbinghaus attempts to advance a step 

 upon the older theories by assigning to a 

 particular retinal substance the function of 

 color-stimulus. He finds this substance in 

 the so-called visual purple, which was 

 studied with great care by Kiihne more 

 than twenty years ago. This remarkable 

 substance is reddish purple in its normal 

 condition. On exposure to light it bleaches 

 rapidly, passing through a series of tints 

 until it becomes yellow. On farther expo- 

 sure to light it becomes colorless, but in the 

 dark regains its original purplish tone di- 

 rectly without passing again through the 

 series of changes in color. The color-stim- 

 ulus is ascribed by Ebbinghaus to the ab- 

 sorption of light by the visual purple, and 

 the character of the light-sensation is 

 directly dependent on the color of the light 

 absorbed, that is, upon the physical proper- 

 ties of the substance. 



The purple substance, which is changed 

 by the action of light into the ' visual 

 yellow,' is identified by Ebbinghaus, in its 

 two stages, with the ' yellow-blue ' sub- 

 stance of Hering. In its first stage it gives 



