INHIBITORY INTERACTION IN THE RETINA 



279 



Table 2. Summation of inhibitory effects in a test receptor (deficit in number of 



impulses over a 1 sec period) produced by two closely spaced retinal regions 



(a and b) when illuminated steadily, compared with the summation of their 



effects when illuminated by brief flashes 



part of this paper. But when short flashes were used, the deficits in X's 

 discharges showed no evidence of interaction between A and B: the com- 

 bined effect equalled the sum of the separate effects. This we interpret to 

 mean that the mutually exerted inhibitory interaction of A and B on each 

 other did not have time to act before the bursts of impulses from them had 

 been completed. Evidently, the process of combining the effects from the two 

 receptor groups is able to operate linearly for short flashes as weU as for 

 steady illumination. Our assumption is strengthened, that the combination 

 of inhibitory influences always takes place by simple addition and one need 

 only take into account mutual interaction to explain all the effects observed 

 under both the steady and the transient conditions. 



Complex transient effects are to be expected when a receptor's activity 

 lasts for a long enough time to be affected in turn by the modifications it 

 produces in the activity of its neighbors. The principles, however, can be 

 demonstrated in a rather simple experiment. The response of an ommatidium 

 to turning on a small spot of light confined to its facet was compared with that 

 obtained when the stimulus spot was enlarged to include a number of near 

 neighbors. Figure 25 illustrates the difference. When the receptor was illu- 

 minated alone, the initial peak in its discharge was followed, as the receptor 

 adapted, by a monotonic decrease in frequency to a lower level which was 

 then maintained steadily. When a larger area was illuminated, surrounding 

 and including this same ommatidium, the initial part of the discharge was the 

 same, but just after the peak there was a sudden drop in frequency — a "silent 

 period" — after which the discharge was resuined, but at a lower level than 

 when the light was confined to a single receptor. We are familiar with the 

 lowering of the steady level of a receptor's response when it is one of a large 

 group of interacting units. The "silent period" in this experiment we interpret 

 as the result of the transient in the inhibitory influences from the neighboring 

 elements, reflecting the initial peak in their responses, acting, after a delay, 

 on the receptor unit whose activity we were observing. It is no different from 



