30 CORA D. REEVES 



an internal factor. A further indication of an internal 

 factor was the behavior of fish H. He jumped from the 

 aquarium once or twice and this jumping was on days 

 when the fish had approached the bright light. In its 

 native brook the male fish in the spawning season remains 

 in the brightly lighted shallows rather than hiding in the 

 dark pools (Reighard 1910). His jumping from the aquar- 

 ium may have been a positive reaction toward stronger 

 light. In spite of these modifications of behavior the 

 record indicates that the fish discriminated between the 

 red and blue when the red was very bright. This is espe- 

 cially true of the fish Md. 



b. Tests of dace Md. Blue with red decreasing in intensity. 



Since discrimination appeared to be possible with the 

 red at the maximum intensity as well as at 2/3, 1/3, and 

 1/5 maximum, Md was again tested while the slit on the 

 red side was gradually reduced in width so as to give a 

 wide range of reduced intensities. The results are shown 

 in figure 13, graph Md, at the right over the heading " Red 

 decreasing." Her percentage of red choices remained be- 

 tween 85 and 95 as long as the slit opening was more than 

 1 mm. At a slit width of about 0.9 mm. the percentage 

 of correct choices was only 60 for the first twenty trials 

 (trials 201-220). But when the slit width was kept at 

 about this value for twenty additional trials, the fish made 

 ^3) per cent of correct choices (trials 221-240). With greater 

 reduction of the light intensity of the red area the fish 

 continued to show a high percentage of positive or blue 

 responses. This happened when the red was reduced in 

 brightness so as to be duller than the blue for the human 

 dark-adapted eye, and finally so dull that it had little color 

 value to the human eye. The slit width was reduced suc- 

 cessively to 0.9 mm., 0.6 mm., and 0.4 mm. Following 

 this a photographic negative" was interposed to further 

 reduce the light and was used with slit widths of 0.6 mm., 

 0.4 mm., and 0.2 mm., successively. 



^ A photographic negative was made by passing the Hght through the blue filter 

 so that any local differences in the amount of light transmitted by the filter might 

 be duplicated on the other side. 



