76 VELOCITY OF NERVOUS IMPULSE. [BOOK i. 



spot, at the moment when the point of the lever had reached 

 exactly the same point of the travelling surface as before, two 

 curves would be gained having the relations shewn in Fig. 12. 

 The two curves resemble each other in almost all points, except 

 that in the curve taken with the shorter piece of nerve, the latent 

 period, the distance a to b as compared with the distance a to b' is 

 shortened : the contraction begins rather earlier. A study of the 

 two curves teaches us the following two facts : 



1. Shifting the electrodes from a point of the nerve at some 

 distance from the muscle to a point of the nerve close to the 

 muscle has only shortened the latent period a very little. Even 

 when a very long piece of nerve is taken the difference in the two 

 curves is very small, and indeed in order that it may be clearly 

 recognized or measured, the travelling surface must be made to 

 travel very rapidly. It is obvious therefore that by far the greater 

 part of the latent period is taken up by changes in the muscle 

 itself, changes preparatory to the actual visible shortening. Of 

 course, even when the electrodes are placed close to the muscle, 

 the latent period includes the changes going on in the short piece 

 of nerve still lying between the electrodes and the muscular fibres. 

 To eliminate this with a view of determining the latent period in 

 the muscle itself, the electrodes might be placed directly on the 

 muscle poisoned with urari. If this were done, it would be found 

 that the latent period remained about the same, that is to say, that 

 in all cases the latent period is chiefly taken up by changes in 

 the muscular as distinguished from the nervous elements. 



2. Such difference as does exist between the two curves in 

 the figure, indicates the time taken up by the propagation, along 

 the piece of nerve, of the changes set up at the far end of the nerve 

 by the induction-shock. These changes we have already spoken 

 of as constituting a nervous impulse ; and the above experiment 

 shews that it takes a small but yet distinctly appreciable time 

 for a nervous impulse to travel along a nerve. In the figure the 

 difference between the two latent periods, the distance between b 

 and b', seems almost too small to measure accurately ; but if a long 

 piece of nerve be used for the experiment, and the recording- 

 surface be made to travel very fast, the difference between the 

 duration of the latent period when the induction-shock is sent in 

 at a point close to the muscle, and that when it is sent in at a 

 point as far away as possible from the muscle, may be satisfactorily 

 measured in fractions of a second. If the length of nerve between 

 the two points be accurately measured, the rate at which a nervous 

 impulse travels along the nerve to a muscle can thus be easily 

 calculated. This has been found to be in the frog about 28, and in 

 man about 33 metres per second, but varies considerably es- 

 pecially in warm-blooded animals. 



Thus when a momentary stimulus, such as a single induction- 

 shock, is sent into a nerve connected with a muscle, the following 



