INHIBITORY INTERACTION IN THE RETINA 



269 



1 2 3 



Decrease in frequency 

 produced by A acting alone 



Fig. 19. Summation of inhibitory influences exerted on a test receptor (X) by two 

 groups of receptors (A and B) at various distances from one another and from X. 

 Each of the graphs was obtained from an experiment on a different preparation. 

 In each case B refers to a group which was illuminated at a fixed intensity, A to a 

 group illuminated at various intensities. (X was always illuminated at a fixed 

 intensity.) As abscissa is plotted the magnitude of the inhibition (decreases in 

 frequency of the discharge of X) resulting from illumination of A alone. As 

 ordinate is plotted the change in frequency produced by A when it acted with B; 

 that is, the decrease in frequency produced by A and B together less the decrease 

 produced by B alone. 



In the upper graph A and B were on opposite sides of X; in the others they were 

 on the same side, in various configurations, the lowest being a case showing 

 disinhibition. (From Hartline and Ratliff, 1958.) 



(not shown in Fig. 19) then as the illumination on A was increased, it tended 

 to release itself from B's inhibition (cf. Hartline and Ratliff, 1957, Fig. 7). 

 Therefore the slope of the function was positive and in one experiment was 

 as high as 1-0 (theoretically, the slope could be greater than 1). In all of these 

 cases, the combined action of A and B was always less than the sum of their 

 separate actions. 



Returning to the case in which A and B were widely separated on either 

 side of X, we note that a number of the individual observations in Fig. 14 

 showed the combined effect of A and B sUghtly to exceed the sum of their 

 separate effects. This is a real phenomenon, greater than the scatter attri- 

 butable to experimental error. It is readily understood, for in our first approxi- 

 mation we neglected the inhibition exerted by the test receptor itself back on 

 A and B. When A and B were both illuminated they, of course, reduced the 

 activity of X more than when only one of them was active, hence X inhibited 

 each of them to a smaller degree when they were both active and in turn 

 their individual contributions to the inhibition of X were greater when 

 acting together than when acting separately. That this is indeed the correct 



