THE SPINAL CORD AS A REFLEX CENTRE 381 



Latency of knee-jerk ..... 5 -So-* 



Afferent endings . . . . 0-40- 



Nerve conduction . . . . .1-4 

 Motor endings and action current . .1-3 



-3-1 



Synapse time .... 2-2o- 



In this case the shortest latency determined for nerve-endings has been 

 deducted from the shortest latent period obtained from the knee-jerk in the 

 spinal cat. On the other hand, some decapitated preparations have been 

 found to present considerably longer latent periods, e.g. 11 and 12o-. Th's 

 variation in the latent period supports the view that the knee-jerk is a reflex 

 of which the synapse time is very short, about 2o-, and that in certain cases 

 there may be increased delay in the spinal cord. When the latencies of the 

 knee-jerk and the homonymous flexor reflex are compared by the electrical 

 method, it is found that the latter is roughly double the former, the average 

 latency of the knee-jerk in the spinal cat being 6'6o-, and of the homonymous 

 flexor reflex 13-2(r. Jolly suggests that this difference may be due to the fact that 

 the knee-jerk mechanism involves only one spinal synapse or set of synapses, 

 while the flexor reflex may involve two. In these estimates the rate of 

 conduction in mammalian nerve has been taken at 120 metres per second. 



* a- = -001 sec. 



