x ELECTROMOTIVE ACTION IN NERVE 299 



to that from which it is coming as in Grottlmss' theory of 

 electrolysis. Du Bois-Eeymond further assumes that this dis- 

 position in the direction of the current is not confined to the 

 intrapolar tract, but extends in a diminishing degree to the extra- 

 polar regions also, by which he explains the electrotonic incre- 

 mental current. Since this interpretation stands and falls with the 

 theory of pre-existing electromotive force in the nerve, which may 

 now be regarded as disproven, we need not enter upon it in 

 detail ; and may turn to those experiments by which Matteucci, in 

 1863, indicated the true physical explanation of galvanic electro- 

 tonus (39). He found regular differences of potential in stretched 

 metal wires (platinum) soaked in a conducting fluid, when any 

 part of the wire was traversed by a constant current. At 

 every point of the extrapolar region there was between each pair 

 of points led off to the galvanometer a current homodromous with 



FIG. 214. 



the primary (polarising) current, which was weaker in proportion 

 as the point tested was more remote from the polarising tract. 



Later on, the same phenomenon was thoroughly investigated 

 by Hermann (39), who gave a complete theoretical explanation of it; 

 showing it to be due, not as Matteucci at first thought to the 

 consequences of spread (by diffusion) of the electrolysis occurring 

 at the electrodes, but to a special case of polarisation (" secondary 

 polarisation"}. If a current is led into the moist sheath of a wire 

 (Fig. 214) at two points, it depends, as Hermann showed, essentially 

 upon the polarisability or unpolarisability of the combination how 

 far the current will diffuse in the sheath of the metallic core. 

 Matteucci stated that an amalgamated zinc wire, of which the 

 sheath is moistened with zinc sulphate, gives no extrapolar 

 differences of potential ; and this was confirmed by Hermann. It 

 is, in fact, easy to see that the current, under these conditions, will 

 enter or leave the metal core at the actual electrodes and their 

 immediate vicinity only, since the lines of current rapidly 

 diminish in intensity, with increasing length, owing to the 



