COLOR CHANGES AND ADAPTATION IN FISHES. 221 



In lower light intensity the fusion rate was lower. According to notes taken at 

 6 p. m. on the third day of the experiment, September 12, the rate was found to be 

 about 4,800 per minute, and the following day the same rate was obtained between 

 II a. m. and i p. m. by reducing the light intensity by means of an opaque screen. 

 When the screen was removed, without changing the rate of rotation of the disk, the 

 fish could be seen to become almost at once distinctly more conspicuous. When the 

 light was thus increased the dark spots and patches became markedly darker, and when 

 it was again decreased b}' replacing the screen they rapidly faded again. With full illu- 

 mination, moderately strong indirect sunlight, the fusion rate was found during this 

 period to be approximately 6,400 per minute. 



The fusion rate for the human eye in all of these tests was apparently the same 

 as that for the eye of the fish. At 50 revolutions per minute and below the sectors 

 appeared nearly, if not quite, as black and white as they did when the disk was not 

 rotating, although they appeared considerably narrower. As the rate of rotation 

 increased, the sectors appeared to become narrower and narrower, and to become more 

 and more nearly alike in shade, until they finally merged entirely and the disk assumed 

 a uniformly gray shade. This occurred at about 200 revolutions per minute in strong, 

 diff'used light, and at a lower rate in weaker light, just as in the case of the fish. In 

 other words, the disk appeared uniformly gray whenever the fish became maximum 

 gray. Under no circumstances did the fish become maximum gray before the sectors 

 on the disk had entirely disappeared. 



The results obtained with the other three specimens were in all essentials like 

 those set forth above. Two of them, however, did not appear to become quite as gray 

 over the disk rotated fast enough to appear uniformly gray to the human eye as they 

 did on a gray bottom. In these individuals, however, the changes in the skin pro- 

 ceeded so slowly that it was exceedingly diflacult to detect slight changes which may 

 have occurred in the shade and the pattern. Consequently, there is some doubt con- 

 cerning these results which, if valid, would indicate that the fusion rate for some speci- 

 mens may be higher than it is for man. 



SUMMARY. 



1. The evidence presented above indicates that fishes have color-vision and that 

 the spectrum for them is, objectively, essentially the same as it is for man. 



2. It indicates that they do not recognize dots 0.5 mm. in diameter as individuals, 

 but that they do recognize dots i mm. in diameter and that they distinguish between 

 dots 2 mm. in diameter and dots 3 mm. in diameter. In this respect vision does not 

 appear to be as keen in fishes as it is in man. 



3. It indicates further that fishes recognize differences in form, that the difference 

 in the reaction to a crab and a minnow is on the basis of difference in form. 



4. It also indicates that the fusion rate of images on the retina is essentially the 

 same for fishes as it is for man, and consequently, that motion as a factor in vision is 

 approximately the same for both. 



97867°— vol 34—16 15 



