THEORIES OF COLOUR-VISION. 1105 



also present in all the instances examined. Krause, however, finds that 

 nocturnal animals are characterised by the length of the outer limbs of the 

 rods. Krause did not pay especial attention to the presence or absence of 

 visual purple ; but there is no doubt that it is present in fishes, in nocturnal 

 animals, and in those who live mainly underground. It is very abundant in 

 the owl, and absent or scanty in birds whose habits are wholly diurnal. The 

 greater length of the outer limbs of the rods found by Krause in nocturnal 

 animals would give greater thickness in the layer of visual purple. The 

 only known instance of a nocturnal animal whose retina is devoid of visual 

 purple is the bat. Its minute rods show, according to Kiihne, only a trace of 

 a purple colour, but the experiments of Spallanzani and others have shown 

 that the bat is dependent to a very slight extent upon vision, and the imper- 

 fect condition of the rods in this animal may well be the result of disuse. 



If dark-adaptation depends primarily on visual purple, the question remains, 

 whether the changes in the pigment and cones have a similar function. Their 

 rate of occurrence is much the same as that of the bleaching and regeneration of 

 visual purple, but the fact that the changes of cones and pigment are not limited 

 to the eye acted on, is against their connection with adaptation being so close 

 as is that of visual purple. There is no doubt that one function of the 

 pigment is to supply material for regeneration of visual purple, and it can 

 probably do this quite as efficiently when situated round the outer part of the 

 outer limb of the rods, as in the dark position. In the light position also, it 

 probably acts by absorbing light and preventing too violent action on the 

 sensitive structures, and its absence will explain the photophobia of albino 

 vision, and also the photophobia which often accompanies night-blindness in 

 abnormal conditions of the retina. 



Cerebral centres for light and colour-vision. This is a subject 

 on which our knowledge is very scanty. It has been supposed l that 

 recorded cases in which there has been hemianopia for colours with 

 retained light sense, are in favour of a separate cerebral centre for colour, 

 and similar cases in which red-green blindness occurs as a hemianopic 

 defect may similarly be held to point to separate cerebral representa- 

 tion of different colours. The facts known are, however, far too few to 

 justify any definite conclusion. 



THEORIES OF COLOUR- VISION. 



Introductory. If the theory of the function of visual purple con- 

 sidered in the preceding section is correct, it still leaves the question 

 of colour-vision almost untouched. But certain anomalous features of 

 colour-vision are explained by it ; and if accepted, it will necessitate some 

 modification of, at least, one of the prevailing theories of colour-vision. 



There are two important assumptions common to these theories. It 

 is assumed that the manifold colour sensations that we experience are 

 derived from a limited number of components, these components being 

 known as primary, fundamental, or elementary colour sensations. Any 

 given colour sensation is to be regarded as arising by a process of fusion 

 of two or more primary sensations. The idea appears to have had its 

 origin in the facts of colour mixture, and it derives its strongest support 

 from colour-blindness, which seems to be most naturally explained by 

 the absence of one or more of the components. Although the idea is 

 earlier than that of specific nervous energy, the two are closely connected, 

 and the former is a necessary consequence of the latter. 



The second assumption is, that the visual process is primarily photo- 



1 Schneller, Arch. f. Ophth., 1882, Bd. xxviii. Abth. 3, S. 73. 

 VOL. II. 70 



