THE PERCEPTION OF COLOUR 



627 



Although it may be thus conchided that colour vision is a definite 

 acquisition of teleostean fishes, it is more difficult to say how far it 

 determines their conduct in comparison with other visual sensations. 

 The work which we have quoted, particularly that of Reeves (1919), 

 would indicate that brightness has a greater attraction than colour, 

 while that of Horio (1938), a Japanese investigator who combined 

 training to different colours with different forms (triangles, discs, etc.), 

 suggests that colour is a more clamant stimulus than form. It would 

 seem, therefore, that as a determinant of behaviour, the colour sense 

 takes a place intermediate between the light and the form senses. 



That it does influence conduct is obvious from certain observations. Two 

 of these may be noted. We shall see that to the male stickleback, Gasterosteus, 

 the sight of red, the colour of the belly of its rival, serves as a release of the 

 fighting resjaonse no matter what the object with which the red is associated.^ 

 Young jewel-fish (Hemichroniis himaculatus) are attracted to i-ed, the breeding 

 colour of the adults, and Noble and Curtis (1939) found that adult females 

 recognized their mates as individuals by the colour-pattern on the head : if the 

 head were painted while the rest of the body retained its natural colour, no 

 recognition was shown, but if the entire body except the head of the male were 

 covered, recognition readily occurred. 



Gasterosteus 



THE COLOUR VISION OF AMPHIBIANS 



Investigations into the colour sense of Amphibians have been 

 largely devoted to the Anurans. There is no doubt that from the 

 anatomical point of view a peripheral mechanism which could sub- 

 serve colour vision is present in the retina of the frog. The electro- 

 retinogram of this animal shows that a Purkinje shift exists between 

 the light-adapted and dark-adapted eye (Himstedt and Nagel, 1902 ; 

 Granit et oL, 1937-39), but we have already seen that this does not 

 imply the existence of a colour sense. In the functional behaviour of 

 this animal a phototactic response can be elicited to light which varies 

 with the wave-length : in one species, Loeb (1890) found a negative 

 phototactic response in which red was preferred to blue ; in two 

 other species, Torelle (1903) obtained a j^ositive response wherein blue 

 was preferred to red. It is to be noted that Cole (1910) found that the 

 phototaxis of Bana clamata varied A\'ith the temperatiu'e. These 

 observations, however, lead to no definite conclusion. Moreover, in 

 the hands of the early workers training experiments invariably gave 

 inconclusive results, probably because the learning ability of the 

 frog is practically non-existent (Yerkes, 1903 ; v. Hess, 1912-22) ; 

 but R. G. Smith (1948) found that by intensive training a response 

 could be elicited in the frog, Bana, suggesting that a discrimination 

 might be possible between red and blue ; Thomas (1953-55), on the 



^ p. 665. 



Rana 



