x ELECTROMOTIVE ACTION IN NERVE 279 



ill proportion with the time (Fig. 211, ng'at'e). Once 

 more, then, we find propagation of a certain alteration in the 

 nerve, the absolute rapidity of which is best determined by 

 leading off isoelectrically from two longitudinal points, since the 

 negative variation then causes hardly any interference. Bern- 

 stein reckons this transmission at 613 m. per sec. In both 

 phases of galvanic electrotonus, therefore, certain alterations 

 are propagated in the nerve from section to section, at a rate which, 

 under all conditions, is considerably below that of the excitatory 

 process. This fact is obviously of great importance to the theory 

 of electrotonic alterations. 



None of these experiments, however, seem to have taken 

 sufficient account of the time-distribution of an- and katelectrotonic 

 changes at any one point of the nerve, with a single closure. The 

 earlier- experiments of du Bois-Keymond and Pflliger showed 

 that with prolonged closure of the polarising current, both 

 the state of depressed excitability, and the corresponding altera- 

 tions in anelectrotonus at every point of the nerve, reach 

 their maximum gradually, and then slowly decline again. 

 Pfliiger (32, p. 319) frequently found no trace of altered 

 excitability, on rapidly exciting after a make twitch, the depres- 

 sion only setting in after 30 sec. 1 min. 



" As the anelectrotonus swells up at any point, and reaches 

 its maximum, so it declines again subsequently, and ebbs back 

 towards the intrapolar tract, if the closure is protracted." 



" The period of flow is less in proportion as the same current 

 is more frequently closed, or is initially stronger, so that with 

 very strong currents the anelectrotonus appears to break in 

 suddenly." 



According to du Bois-Keymond (30, p. 446, and 6, p. 255), 

 this reaction is expressed, with reference to the apparently 

 corresponding galvanic effects of anelectrotonus, by a curve of 

 the form a^a^ (Fig. 212); S being the moment of closure, 

 t l the first reading of the galvanometer. The time-distribution 

 of the katelectrotonic alterations during the passage of current 

 is very different. The katelectrotonus at any point of the nerve 

 invariably reaches its maximum (which is always lower than 

 that of anelectrotonus under the same conditions) much earlier 

 than the latter, and appears, at least as regards galvanic 

 changes, to decline steadily from the commencement of the 



