A Quantum Theory of Colour Vision. 



229 



sensation ; of three fibres by the green sensation ; of four fibres by the blue 

 sensation. 



(d) The unit of luminous stimulus is the nerve discharge of one fibre. 



(e) No coloiir sensation is associated with this stimulus. 



(18) It seems difficult to question any of these statements if there is any 

 reality in the present theory. But the long-standing question respecting 

 colour sensation confronts us : Can we deal with these cerebral phenomena 

 quantitatively ; in a physical sense ? It is a question most difficult to answer. 

 Possibly it is ultimately unanswerable. I think, however, the present theory 

 gives a little to go on. 



It seems probable that the numbers of photons characteristic of each of the 

 fundamental colour sensations constitute an energy relation between them. 

 The colour sensation curves depict the amount of excitation of three cerebral 

 effects in answer to spectral stimuli. Each sensation is an accompaniment of 

 a particular form of energy stimulus, i.e., of two, of three, or of four photons 

 simultaneously discharged. To the form of the stimulus the specific character 

 of the sensation is to be referred. We must be prepared to admit that an 

 energy equivalent to the sensation must exist. Ux viihilo nihil fit. 



If we measure the maximum heights attained by the three colour sensation 

 curves of Konig* (or the two colour sensation curves in the case of dichro- 

 mates) we find them related to the frequencies of the corresponding primary 

 colours ; that is the heights are, for the red, green, and blue curves, in the 

 ratios 2:3:4 approximately. Thus we find for the maximum ordinates 

 (green and blue sensations) of the proteranope the heights 10"8 : 15'2, i.e., a 

 ratio of 3 : 4 approximately, and for the deuteranopes red and blue maxima 

 8 : 15"2, i.e., a ratio of 2 : 4 approximately. In the trichomate the three colour 

 sensations attain the stimuli values 8'2 : 11"4 : 15"2, which are in the ratio of 

 2:3:4 approximately. 



This seems to suggest that the quantitative values of the three sensations 

 stand in the ratio of the number of photons originating each sensation. 

 Herein is an energy relation between the colour sensations, if I have viewed 

 the matter rightly. 



(19) Luminosity or brightness would seem to be a sensation directly 

 dependent on the number of electrons which stimulate the nerve fibres. This 

 appears to be a necessary inference from the fact that curves of spectral 

 luminosity in general and the curves of absorption and bleaching of the visual 

 purple are in fair agreement.f 



We must specially notice the resemblance between the photopic and 

 * Ante, fig. 2. 



t Parsons, loc. cit., p. 55 ; compare figs. 1, 10, 11, 14, etc. 



