62 C. J. CONNOLLY. 



the various color stimuli for several weeks. The water was re- 

 newed daily, and the temperature never varied more than a few 

 degrees, remaining between about u and 15 C. so that the speci- 

 mens were at all times under the same environmental conditions 

 except that of color. In each jar there were 5 fishes each about 

 8 cm. in length. On the second days it was noticed that the 

 shade was distinct for the fishes in each of the colored boxes, the 

 sequence from light to darker in shade being yellow, red, green 

 and blue. \Yhen these experiments were repeated, it was observed 

 that these differences in shade could be recognized within three 

 or four hours. On the sixth day there was also noticed a distinct 

 color for each group of fishes in the four colored backgrounds. 

 The group in the yellow box were decidedly yellow ; those in the 

 red showed a pink color; those in the green a more pronounced 

 green than that sometimes shown by normal specimen, the region 

 above the eyes revealing this tint in a striking manner ; while 

 those in the blue box were of a gray, slate-blue color. The adap- 

 tation in color increased during the second week when they reached 

 their maximum degree of simulation and maintained it as long 

 as the experiments continued, namely, for about six weeks. This 

 was approximately the maximum time that the specimens could be 

 kept in normal health under the artificial conditions in the labora- 

 tory and following this they gradually died. The tests were re- 

 peated many times between November 1922 and June 1923 with 

 different sizes of fishes ranging in length from 5 to TO centimeters 

 and uniform results were obtained. Fundulus simulates the back- 

 grounds after a prolonged stimulation and the effect produced can- 

 not be due to intensity of light alone. 



For purposes of comparison, fishes were also placed in three 

 other boxes without color ; one lined with dull white paper, a 

 second with neutral gray and a third with dull black paper. Those 

 in the white box showed the pale tint of light-adapted fish; those 

 in the black box a very dark shade, while those in the gray showed 

 an intermediate shade. On the other hand the fishes in the colored 

 boxes each showed a color distinct from the others and from those 

 in the uncolored boxes. The effect, then, cannot be due to inten- 

 sity of light reflected on the background, but must be due to the 

 color stimuli. Otherwise there is no explanation of the fact 



