FURTHER EXPERIMENTS. 223 



off the violet and blue rays. The light beneath it con- 

 tains no more )'ellow rays than elsewhere; but those 

 rays produce the impression of yellow, because the 

 yellow is not neutralized by the violet and blue. In 

 each case, therefore, there was less light in the covered 

 than in the uncovered part. 



After every five experiments I added up the number 

 of the Daphnias; and the following table gives twenty 

 such totals, each containing the result of five observa- 

 tions, making in all one hundred. 



My reason for adding one vessel in which one half 

 had an opaque cover was to meet the objection that 

 possibly the light might have been too strong for the 

 Daphnias ; so that when they went under the sheltered 

 part they did so, not for color, but for shade. I was 

 not very sanguine as to the result of this arrangement, 

 because I had expected that the preference of the 

 Daphnias for light would overcome their attachment to 

 yellow. 



The numbers were as in the following table (p. 224). 



The result was very marked. The first two columns 

 show the usual preference for light. If the covered 

 half had been quite dark, no doubt the difference in 

 numbers would have been greater ; but a good deal of 

 light found its way into the covered half. Still the 

 result clearly shows that the Daphnias preferred the 

 lighter half. The numbers were 204S in the dark to 

 2952 in the light; and it will be seen that the preference 

 for the light was shown, though in different degrees, in 

 almost every series. 



The result in the blue gives, I think, no evidence as 

 to color-sense. The numbers were respectively 2046 

 against 2954, and were therefore practically the same 



