344 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1916. 



determine. It appears, however, that with trained dogs and horses 

 there is no difficulty at all in teaching them to distinguish between 

 the saturated colors. The preference of some birds, notably the 

 Bower Birds, for objects of a certain color and the general evolu- 

 tion of color in the different species must point to an appreciation of 

 different shades. Color sensation must be appreciated by the stimu- 

 lation of waves of varying lengths. In man it varies from about 

 770 [X to 396 [A, the latter being the extreme of light registered at the 

 violet end of the spectrum. 



It would appear, if we adopt the Young-Helmholtz theory, that 

 man has a trichromatic vision, and that all the shades appreciated 

 are due to the degree in which the three classes of nerve fibers are 

 stimulated. Yellow, for example, is caused by an equal stimulation 

 of the sets of fibers for the red and green percipients. When red is 

 seen the fibers percipient of red are strongly stimulated, the others 

 only weakly. Color blindness is an interesting side study in this 

 respect, particularly when we come to the color vision of birds. In 

 man dichromatic vision appears most commonly with a blindness 

 for red or green, the violet blind being rare. In red or green blind- 

 ness the subject confuses reds and greens, and in a mixture of colors 

 including these colors other than red or green are the only ones 

 appreciated. 



NoAv, it has been shown by feeding experiments that birds are 

 blind in the violet end of the spectrum. In other words, if we accept 

 the Young-Helmholtz theory they have a dichromatic vision. Their 

 color vision would be restricted to red and green and the mixtures 

 of these colors. They would be blind to violet and to the spectral 

 violet in blue, indigo, and yellow. Such a conclusion would be dis- 

 astrous to our theory of selection in the coloration of birds, where 

 many blues and shades of blue are seen. It would mean that the 

 development of color in the evolution of the present-day bird was 

 merely incidental and apparently without reason. The flaw in the 

 reasoning probably lies in our acceptance of the Young-Helmholtz 

 theory instead of recognizing the other colors as primary. Again, 

 the conclusion obtained from the feeding experiments may be faulty. 

 The birds are fed in spectral red light and in spectral green, where 

 they pick up the grains readily; but when taken to spectral violet 

 remain still, fail to see the grains, and are to all intents and purposes 

 in darkness. 



A man color blind in red or in green, though not seeing these colors 

 as a normal person would see them, still sees the objects, but is blind 

 to the color only. His vision extends right to the red end of the 

 spectrum, though not recognizing the red there, so that the waves 

 stimulate the eye, though not giving the color sense. It is probable 

 that in birds the sight is keyed to a higher pitch than in man, and 



