504 



Dr. G. J. Burch. On Light- 



[Apr. 19, 



strength of the response, reaches its maximum under the action of smaller 

 quantities of the exciting substance. Consequently the strength of the 

 resulting sensation, when reduced by the shunt factor, will be represented by 

 a more or less flat-topped curve. 



From various data I judge that in artificial colour-blindness of quite 

 moderate degreee, a shunt ratio of 1 : 100 before and after the exposure of the 

 eye to light is well within the mark. I described in 1897 an experiment 

 bearing upon this subject which seems to have escaped notice.* 



"When the spectrum is viewed by intermittent light of great intensity, the 

 flash ratio being 1:3 or 1:4, as soon as the eye gets accustomed to the 

 strong light it is seen that the continuity of the colours is gone, and that 

 there are now visible four bands of strong colour, viz., red, green, blue, and 

 violet, upon a pale but brightly illuminated ground. Between the intense 

 red and the rich green is a space of a colour between yellow-ochre and raw 

 sienna, but very pale. Between the green and the bright blue is a region of 

 pale greenish-blue, passing into pale steel-blue. Beyond the bright blue is a 

 pale lilac space. Farther than this cannot be seen while the red end of the 

 spectrum is in the field, because of the overpowering intensity of the light in 

 the neighbourhood of the yellow-green. But by shifting the prisms so that 

 the &-lines come on the extreme left, while the G-line occupies the middle of 

 the field, the region about H and K appears of a deep rich violet of great 

 intensity, contrasting strongly with the pale lilac between it and the blue. 



I suggest the following explanation. Owing to the periodic intervals of 

 darkness during which the visual substance collects, the rate of formation 

 of exciting substance is maximum over the whole range of the resonance. 

 But the light is so strong that the shunt ratio is very high. Consequently, 

 where two colour-sensations overlap, as in the yellow, we get the pale colour 

 of the binary blend, and on either side, where only one colour-sensation is 

 effective, the full rich tone proper to it. 



The question as to how and by what mechanism the shunt function works, 

 and where that mechanism is situated, is of the greatest interest and 

 importance. 1 have already mentioned that I do not hold with those who 

 refer it to the central organ. For one thing, it would add to the complexity 

 of the optic nerve, as I have pointed out in my paper on u Areal Induction."f 

 And, for another thing, it would leave the end organs in the retina unpro- 

 tected. I should therefore look for the shunt mechanism in the retina itself. 



If Charpentier's bands and Purkinje's recurrent images are to be regarded 

 as due to spasmodic applications of the shunt factor, they would seem to 



* ' Journ. Physiol.,' vol. 21, p. 431. 

 t ' Eoy. Soc. Proc.,' vol. 69, p. 125. 



