Sensory Discrimination: Vision 167 



owls is somewhat longer. Watson (773), on the other 

 hand, working with a more exact apparatus, concludes that 

 the spectrum is visible to the chick and the homing pigeon 

 within the same limits as to man. A study of the electric 

 currents generated by the action of light on the eyes of 

 day and night birds gives evidence confirming the hypothe- 

 sis that the latter are color-blind : in the day birds, each 

 color gives a characteristic deflection of the galvanometer, 

 not due to its intensity, while no such differences appear for 

 the eyes of night birds (403). Breed (101, 102), using 

 colored screens through which the light was passed, and 

 offering a choice of passages differently illuminated, ob- 

 tained evidence of color discrimination in the chick. The 

 preference of the chicks for one color rather than another 

 appeared to depend on the relative brightness of the 

 colors, since it could be reversed when their brightnesses 

 were sufficiently altered. When a blue and red were found 

 between which the chick showed no preference, this was 

 taken as an indication that they looked equally bright to 

 the chick. The bird could, however, be trained to choose 

 one of these two colors ; hence the conclusion was reached 

 that it could probably react to a difference in color and 

 not merely to one in brightness. The evidence for color 

 vision in birds has lately been made practically conclusive 

 by the careful experiments of Lashley (413) on the domestic 

 fowl. He used spectral light whose intensity was accu- 

 rately controlled. The ability of the fowl to distinguish 

 red and green was demonstrated under the following con- 

 ditions, which ruled out the possibility of discriminating 

 by brightness differences, (i) Each of the lights was al- 

 ternately reduced to threshold intensity, while the other 

 remained at full intensity. (2) White light of a constant 

 intensity was substituted for each colored light in turn. 



