LIGHT OF DIFFERENT WAVE-LENGTHS BY FISH 63 



could be seen in the method or in the frequency of approach 

 to either of the two plates. 



Although I obtained no evidence of a shift in relative 

 brightness of the red and blue used, or in that of red and 

 white, there was evidence with both dace and sunfish of 

 an increased sensitiveness to light after dark-adaptation. 

 If the intensities of the red and blue plates were so ad- 

 justed that fish adapted to the illumination from a tungsten 

 lamp went without hesitation to either plate and visited 

 the two plates with equal frequency, and if the fish were 

 then allowed to become dark-adapted and the plates again 

 presented, they no longer went close to the plates but 

 halted at some distance from them. Their sensitivity to 

 light had increased, but my evidence does not show that 

 the relative brightness of the plates had changed for them. 



III. REVIEW OF THE LITERATURE 



It is not uninteresting to note that the early workers 

 in this field used a variety of methods which involved both 

 (1) training of the fish and (2) their unlearned responses. 

 But the discordant results obtained are not correlated with 

 the type of method employed. 



Graher (1885) was one of the first to test the color vision 

 of fish by the method of response. He illuminated dif- 

 ferent parts of the aquarium simultaneously with lights 

 of different sorts, and he noted in which of them the fish 

 gathered. He found a difference in the behavior of Gaster- 

 osteus spinachia toward darkness as compared with their 

 behavior toward light, and a difference in their behavior 

 toward red as compared with their behavior toward blue. 



Zolotnitsky (1901) fed fish with red Chironomus larvae 

 until they were accustomed to this food. He then put 

 against the glass aquarium wall a cardboard, upon which 

 were pasted bits of wool of various colors and of the size and 

 shape of the larvae. The fish jumped most frequently 

 for the red pieces. He decided that fish could distinguish 

 red from other colors. He did not consider the brightness 

 values of the colors used. 



Washburn and Bentley (1906) attacked the problem by 

 a method which involved more exact discrimination by 



