PHENOMENA OF MUSCLE. 235 



diverge from one another. The tract which extends from the upper 

 end of one of the muscles to the lower end of the other is equal to 

 about half the length of the muscles themselves. This whole 

 tract was, so to speak, anodic and kathodic at the same time ; and 

 it would have been necessary to put the galvanometer electrodes 

 on the end quarters of the muscles in order to escape it. 



If this was done, however, they would get into the region of the 

 anodic and kathodic tracts in the neighbourhood of the stimulating 

 electrodes. Taking into consideration the fact that the direction 

 and strength of the polarisation-currents produced by one stimu- 

 lating current depend on its density at the individual anodic and 

 kathodic spots, and that in every muscle there must be two 

 break-stimulations, which affect sometimes a larger, sometimes 

 a smaller muscle tract (see below), it is easy to see under what 

 extraordinarily complex conditions du Bois- Raymond's experiments 

 were conducted. On this ground, also, it is not possible to deduce 

 the theoretical result of each experiment. 



According to the propositions set forth in my last treatise some 

 points appear explicable, viz. first, that very weak currents in- 

 variably only gave negative polarisation. This was inevitable, 

 for such currents produce negative polarisation at the anode as 

 well as at the kathode, hence in such cases the total result of the 

 combined anodic and kathodic polarisations of du Bois-R/eymond's 

 preparation could only be a negative polarisation-current. 



Again, it is clear why very strong currents of short duration 

 only gave positive polarisation-currents. For such currents either 

 affect the anode only, and always in the direction of positive 

 polarisation, or in any case their positive anodic action is much 

 stronger than their negative kathodic action ; hence the combined 

 result of the polarising actions can only be a positive one. 



With very weak currents, as well as with strong ones of short 

 duration, the relations are much simpler and more easily reviewed, 

 because the persistent break-stimulation is then confined to those 

 portions of the fibres immediately adjacent to the anodic spots. 

 Stronger stimulating currents of longer duration, after the break, 

 effect however a persistent stimulation spreading farther in the 

 direction of the stimulating current. As a necessary consequence of 

 this in du Bois-Reymond's preparation, the galvanometer electrode 

 which was next the stimulating anode, came within the limits of 

 the stimulated and therefore relatively negative muscle tract, when 

 the break-excitations were strong* If the galvanometer electrodes 



