66 CORA D. REEVES 



toward the light and in a dark room gave this response 

 to light of very low intensity. They went to the brighter 

 of two white plates which differed in illumination as 1 to 

 1.23. Taking advantage of this phototactic behavior, he 

 found that fish gathered in the yellow-green of a spectral 

 band thrown into the aquarium. By using a piece of 

 cardboard he slowly cut off the successive portions of the 

 spectral band from the violet end until only red was left. 

 The fish gathered in whatever part of the spectrum re- 

 mained except in the extreme red. If the spectrum was 

 shortened from the red end, they continued to gather in 

 it until they were finally in the violet. Since the fish did 

 not gather in the red of longest wave-length, Hess decided 

 that it stimulated them very little. For them the red end 

 of the spectrum appeared to be shortened as it is for the 

 totally color blind. 



By means of their brightness response Hess determined 

 the relative brightness of different parts of the spectrum 

 for fish. To do this he used monochromatic light on one 

 side of the aquarium and a white light of variable intensity 

 on the other, and found intensities of the variable at which 

 the fish distributed themselves equally on both sides. He 

 decided that for fish the yellow-green was the brightest 

 part of the spectrum, that blue, yellow, and orange were 

 less bright; that red was the darkest; and that for fish 

 orange and red have little brightness value. These deter- 

 minations of brightness values of the spectrum were so 

 like those already found by others for the human color- 

 blind that Hess was led to conclude that fish are color- 

 blind. 



The first work of Hess seems to be open to criticism on 

 several grounds: 



(1) He used very young fish. The condition of the 

 retinal cones of young fish has been shown by Shafer (1900) 

 to be quite difTerent from that of adults. It is not impos- 

 sible that their color vision is different. Whether this is 

 the case must always be difficult to determine, for such 

 young fish cannot well be used in training experiments 

 because their behavior is too mechanical. They feed on 

 minute plankton animals and have perhaps little need to 



