THE INTENSITY FACTOR IN VISION 



W. A. H. RUSHTON 



Trinity College, Cambridge; 

 Visiting Scientist to National Institutes of Health, Bethesda, Maryland 



In this assembly of distinguished photochemists the fundamental 

 visual problem which I wish to raise is "How does the eye use so 

 effectively the quanta caught by the visual pigments of the rods 

 and cones?" The question of the spectral sensitivity of vision is 

 trivial, for at the chemical level it amounts simply to the statement 

 that quanta which are not absorbed will not be seen. This expected 

 relation was established with fair accuracy more than 50 years ago 

 by Koenig (5) and by Trendelenburg (11), who showed that the 

 spectral sensitivity of twilight vision corresponded to the action spec- 

 trum for the bleaching of rhodopsin. Color vision, to be sure, is no 

 trivial question, but here the problem lies in the organization of 

 nerve signals already generated. 



What I should like to ask you is, "How are these signals generated?" 

 For my part I have no answer, but information we have (some of it 

 new) about the relation of the nerve signal to the light intensity, 

 about the site where light- and dark-adaptation occur, and about the 

 relation between the light threshold and the amount of pigment 

 bleached. This information is here distilled into a single mathemati- 

 cal expression. Any theory of how the nerve signal is generated should 

 satisfy this expression. And if it does, it will go a long way towards 

 satisfying the relation between light intensity and vision. 



I. The Response to Light 



(a) The Eye at Full Sensitivity 



The celebrated experiments of Hecht, Shlaer, and Pirennc (4) 

 measured the least number of quanta which have to be absorbed by 

 rhodopsin in the human eye for the light to be seen. They em- 

 ployed two methods. In one they measured the actual energy falling 

 upon the cornea (through an artificial pujjil) when a flash in optimal 

 conditions could just be seen, and made an estimate of the fraction 

 (10%) which was absorbed by rhodopsin. In the other method, they 



706 



