COLOR VISION IN REPTILES, BIRDS 497 



A majority of snakes are pure-cone, but the strange history which their 

 eyes seem to have had (see Chapter 16, section D) makes it anything but 

 presumptive that they have retained the color vision of their lizard an- 

 cestors. If they have color vision, it is de novo; but it is unlikely that they 

 do, since their cones are plump (Fig. 26a, p. 63) and their vision, in con- 

 sequence, is crude and unsharp as compared with other diurnal verte- 

 brates. Experimental evidence is wanting, though Kahmann several years 

 ago mentioned that his training of 'an exceptionally trainable snake 

 species' to red and blue had succeeded quite well. He has apparently 

 published no full account of this work. 



Of all the unstudied reptiles, it is the geckoes and Sphenodon which 

 offer the greatest interest. There is a large 'hole' in the Duplicity Theory, 

 which can be plugged only when we know whether such forms have 

 retained the color-vision machinery of their diurnal ancestors despite 

 their transmutation of cones into functional rods. 



Birds — No one has ever scientifically questioned that the diurnal birds 

 have color vision. Since 1863, when Krause first interpreted the multi- 

 plex oil-droplet mosaic of birds as a mechanism for hue discrimination 

 (see pp. 192-3), no doubt of a hue-perceptive capacity on the part of the 

 birds has ever had a chance to grow. 



Though color vision was assumed for decades before it was ever 

 proven by experimental work, that work has fully justified the assump- 

 tion. During the last quarter-century the researches on avian color vision 

 have not had to be wasted in controversy as to whether birds see colors 

 or not, but have been devoted directly to such matters as the determin- 

 ation of the spectral limits, the relative brightnesses of colors for the bird, 

 and the latter's capacity for hue-discrimination in different parts of the 

 spectrum. 



For years it was generally believed that the birds are blind to violet 

 and blue, the short-wave end of their spectrum greatly shortened. The 

 work of Hess up to 1912 seemingly established this beyond doubt. 

 Sprinkling rice grains in a spectrum projected upon a white floor, he 

 found that fowls would eat the rice from the red end to the junction of 

 the green and blue, but would peck no grains in the blue or violet lights 

 — allegedly, because they could not see them. The absorption of short- 

 wave light in the red and yellow cone oil-droplets was held accountable, 

 despite the fact that the many colorless cones (and the rods) should have 

 been able to record blue rice — though perhaps hazily, and not as blue. 



