LIGHT OF DIFFERENT WAVE-LENGTHS BY FISH 23 



2. Wave-Length Discrimination: White- Red by Dace 



While the last two dace were failing to discriminate the 

 two white plates, a red filter without the slit was placed 

 on the side of the brighter plate. This cut off the rays 

 shorter than ^ 5890.5 The illumination of the positive 

 plate continued to be with white light through a 5 mm. 

 s^it. The results were most surprising. When the red was 

 first introduced, each of the fish went to it for the first 

 three responses. But the manner of approach was notice- 

 ably different from that to the white plates. The fish 

 swam more swiftly toward the red plate, or after approach- 

 ing darted away from it, or sidled along the edge of the 

 red illuminated area, as shown in figure 10. This sort of 

 behavior is described later in the case of " Large Sunfish," 

 in which it was somewhat more striking. But it is to be 

 noted that it took place here when the red filter had re- 

 duced the brightness of the total light to which the fish 

 were accustomed, by cutting off a considerable part of 

 its energy. It was not then a response to increased inten- 

 sity. Each fish after from twenty to thirty trials, equally 

 divided between white and red or showing a strong position 

 habit, got the " cue," and for the following thirty trials 

 gave 80 to 90 per cent white, or correct choices. When 

 ten successive correct choices had been given, it seemed 

 possible to conclude that discrimination was by means of 

 brightness. 



Hess (1909) has stated that the red rays have little effect 

 on fish. If the introduction of the red plate had greatly 

 reduced for the fish the brightness on that side, then the 

 brightness differences between the red and white plates 

 might now be greater for them than it had previously 

 been between the two white plates; it might be greater 

 for the fish than 1 to 4. This increased brightness dif- 

 ference might enable the fish to discriminate between the 

 red and white, although they had previously failed to 

 discriminate the two white plates illuminated as 1 to 4. 

 It has been noted that the fish tended to swim along a 

 plane separating two white illuminated spaces when there 



* It is to be noted that the orange as well as the red wave-lengths, are included 

 whenever the red stimulus is mentioned in this paper. 



