78 CORA D. REEVES 



IV. DISCUSSION 

 A. Analysis of Methods Employed 



In the study of color vision in fish, investigators have 

 availed themselves of three methods — that of training 

 (formation of a food association), that of instinctive re- 

 sponse, and that of change of color of the fish with change 

 of background. The second and third of these methods 

 might, of course, be considered together since both involve 

 instinctive response, but it is convenient for purposes of 

 discussion to separate them. 



a. The method of response has been used chiefly by Reig- 

 hard, Bauer, and Hess, and results obtained by it are in- 

 cluded in this paper. No attempt is made by this method 

 to form a food association, but the fish are presented with 

 light stimuli of different intensities and qualities, and 

 their behavior is noted. Two patches of light of restricted 

 wave-length, or one of restricted wave-length and one of 

 white light, may be presented together or a considerable 

 part of the spectrum or the whole of it may be shown at 

 one time. Where two qualities of light are used, the fish 

 may show by difference in frequency or method of approach 

 or by tendency to avoid one of them, that the lights differ 

 for it. The intensity of one light is then varied (Bauer 

 and Hess). If at any intensity of the variable, behavior 

 of the fish toward the two lights is found to be the same, 

 it is concluded that their past differential responses toward 

 the lights have been due to a difference in brightness; the 

 cessation of these responses is taken as evidence of matched 

 brightness. If the fish do not, under the conditions of 

 the experiment, show differential response at this matched 

 brightness, they are regarded as color-blind. The evidence, 

 failure to respond at matched brightness, thus leads to a 

 negative conclusion. It has been pointed out that responses 

 of this sort are evanescent. The response may be given 

 when the stimulus is first presented, but may not appear 

 on repetition of the experiment: because the fish may have 

 become " adapted " to the unwonted stimulus and no 

 longer respond (see adaptation to red in figs. 13, 15 and 

 discussion p. 33). Failure to respond may also result from 



