10& 
coil was made a little nearer and the muscle gave a maximal twitch* 
for example the stimulus of 20.) mm. coil distance applied at IB- 
acted as a maximal stimulus at or just after the stage of narcosis 
at which the conduction failed from outside electrode. Applying 
at B the maximal stimulus of this kind the curve of maximal muscle 
twitch was again registered as before to determine the latent period 
(Ln) of muscle twitch for this strong stimulus. The latent period 
(L'n) in case of stimulating at C with a weak current which was 
only just maximal for this unaffected part of nerve was again tested 
as before for the sake of control. The latter remained almost unchang¬ 
ed, so l' = L'n was in the most cases. 
It is self-evident that 1—1', the difference of the latent periods 
previous to narcosis is the time required for a nervous impulse 
to travel from B to C in fresh condition of nerve. Since BW=CW 
and the rate of nervous conduction is equal in all parts of nerve, 
so 
1—l' = 2T, 
in which T is the time required for the propagation of nervous im¬ 
pulse from B to the distal wall (W) of the narcotising chamber. 
On the other hand we know from the work of Dr. Hariya in my 
laboratory that at the stage of narcosis at which the suspension of 
conduction from outside electrode occurs, the rate of conduction in the 
nerve within the narcotising chamber is reduced at least 50<?4 of 
its original rate. So, after the narcosis, the impulse starting from 
B would take 2T to reach the distal chamber wall (W), admitting 
for the present, that the times required to set in motion a impulse 
at B before and after the narcosis are the same. If, therefore,, 
the strong stimulus applied at B after the narcosis had started a. 
impulse from B, then 
Ln-L'n=3T 
(I)