A Quantum Theory of Colour Vision. 



221 



varying energies of the photo-electrons are to be transmitted to the cerebral 

 cortex, and if such energy differences cannot be conveyed by the nerve fibre, 

 the alternatives presented to us are either to reject the theory altogether or to 

 assume that more than one nerve fibre transmit the stimuli concerned with 

 colour vision, and that it is by the number of fibres stimulated that the 

 energy is interpreted. This naturally sends us to the histology of the retina, 

 because as colour vision appears to be restricted to the cones and achromatic 

 vision to the rods, the connections of these organs with the optic nerve 

 might be expected to reveal an anatomical difference 

 in accordance with the structural features we are 

 driven to assume as existing in the cones. 



This anatomical difference is presented to us with 

 remarkable clearness. All delineators of these 

 minute organs agree in depicting the rod as connected 

 through a fibre of extreme tenuity with the deeper 

 lying nervous structures; while on the other hand, 

 the corresponding connection of the cone is, relatively 

 to that of the rod, exceedingly bulky. The accom- 

 panying figure (after Greeff ) shows these features in a 

 conspicuous manner. Obviously many such fibres as 

 lead from the rod might be contained in the cone 

 fibre. Indeed it is difficult to account for the 

 remarkable difference of structure on any other 

 suggestion. 



(3) I assume that the explanation is correct and 

 that the constituent fibres of the cone connection 

 are collected in the spindle-shaped enlargement of 

 the organ and brought to the base of the percipient 

 extremity, where their terminals are activated by 

 energy carried into the cone by photo-electrons. It 

 may be that the fibrils persist into the cone or into 

 the cylindrical percipient extremity of the foveal 

 cone. In either case the number of fibres activated depends on the kinetic 

 energy of the photo-electron. 



On this view we find a purely anatomical reason for the differing functions 

 of rods and cones. The photo-electrons liberated within the rod can transmit 

 but the one intensity of stimulus to the brain, no matter how much they may 

 differ in kinetic energy. Hence its achromatic functions and the low luminous 

 character of scotopic vision. A one-fibre stimulus is all it can send. On the 

 other hand the cone sends a number of stimuli proportional to the energy of 



s 2 



Fig. 1. 



