216 



Mr. G. J, Romanes on the 



[May 3, 



of fact, the desired elimination is performed by the nervo-muscnlar tissue 

 itself. For it usually happens that a gastrocnemius presents some per- 

 ceptible difference in the character of its contraction, according as the 

 latter is given in response to make or to break of the current. Therefore, 

 by first ascertaining, with an ordinary key, the optical appearance which 

 the responses to make and break respectively present, it is not difficult 

 afterwards to recognize which of these appearances is presented by the 

 response to rapidly succeeding make and break stimuli, and so to deter- 

 mine which of these rapidly succeeding stimuli is the one to which the 

 response is given. Now I found in this way that, by making the duration 

 of contact sufficiently brief, the nerve, whether or not injured and in 

 whichever direction the current was allowed to pass, only responded to 

 the closing stimulus. Therefore, by choosing a strength of current which, 

 in each of the cases (a) and (b) before nerve-injury, was just sufficiently 

 strong to elicit a response when the voltaic stimulus was of t duration in 

 the one case and t' duration in the other, I was sure that in each of the 

 two cases the response which I obtained was given to the closing, and not 

 to the opening, excitation. Having noted the values of t and t\ I divided 

 the sciatic just where it enters the gastrocnemius, and then shortened the 

 duration of contact down to the point at which, in each of the two cases 

 (a) and (6), the muscle again only just responded to the stimulus. Let 

 these durations be respectively represented by T and T'. As before, I ascer- 

 tained that the responses had exclusive reference to the closing excitation ; 

 bo that, by recording the values of t, t\ T, T', I was able to obtain for 

 responses to stimuli of minimal duration representative numbers, such as 

 those in the former Table, which have reference to stimuli of minimal 

 intensity. It is only necessary further to state that, as different gastro- 

 cnemius muscles exhibit considerable variations in the degree of their 

 natural irritability towards voltaic stimuli of short duration, and as for 

 my purposes it was desirable to obtain a physiological, as distinguished 

 from a physical, basis whereon to institute my comparisons, in the case of 

 each muscle I began by graduating the intensity of the current down to 

 that point at which the duration of contact required to produce minimal 

 stimulation before nerve-injury was the same as it had been in my previous 

 experiments. Or, in other words, by appropriately varying the intensity 

 of the current to suit the degree of excitability manifested by each parti- 

 cular muscle before injury, I was able, notwithstanding the differences in 

 excitability presented by different muscles, to render t a constant. When 

 this was done, however, t', T, and T' were all found more or less variable 

 in different muscles — as, of course, we should expect from the analogous 

 case of responses to stimuli of minimal intensity. I therefore tabulated 

 the results yielded by twenty gastrocnemius muscles, and then calculated 

 the average duration of contact which in each of the cases (ci) and (6) be- 

 fore cutting, and («) and (5) after cutting, was required to cause minimal 



