12 



Mr. G. J. Romanes on the 



[May % 



it enters the muscle) is in all cases attended with a marked increase of 

 sensibility to stimuli of short duration ; i. e. stimuli of much shorter 

 duration are able to evoke responsive contractions in the muscle after 

 cutting than are required to do so before cutting. At first, therefore, it 

 seemed that this experiment was confirmatory of the hypothesis which it 

 was designed to test. This, however, is not so ; for it was observed that 

 the increased sensitiveness in question was only shown when the femoral 

 end of the muscle rested on the kathode, while it was scarcely, if at all, 

 apparent when this end rested on the anode. This fact, of course, led to 

 the inference that the augmented excitability to stimuli of short duration 

 had reference, not to the opening, but to the closing excitation. Accord- 

 ingly I fitted up an appropriate arrangement of wires and keys, by which I 

 could at pleasure throw in ordinary opening and closing excitations, or the 

 closing and opening excitations of short duration. In this way it was easy, 

 by comparing in the two cases the nature of the contractions (which in 

 almost every muscle presents some idiosyncratic differences on make and 

 break), to obtain an optical proof that my inference was correct. The 

 exalted sensitiveness of the muscle after section of its nerve to stimuli of 

 short duration had reference exclusively to the closing excitation. 



This fact is of interest in itseK, but it fails to answer the question as 

 to why section of a nerve causes so disproportionate an effect on its sen- 

 sitiveness in the muscle to the excitation which is supplied by the 

 descending break. Nor have I any satisfactory answer to give to 

 this question, unless the following consideration may be deemed so. 

 Before section of the sciatic nerve, the gastrocnemius muscle is im- 

 mensely more sensitive to the ascending make than to the descending 

 break (figs. 2 and 3, left-hand lines). Consequently, when the general 

 sensitiveness of the nerve is increased by section, the increase has not 

 so much room (so to speak) for its occurrence in the one case as in the 

 other. Seeing that the minimal make contraction occurs at a point so 

 much nearer to zero of the current's intensity than does the minimal 

 break contraction, when both these minimals are reduced still further by 

 nerve-section, the latter minimal has a much wider range through which 

 it is free to fall than has the former. Of course this fact need not pre- 

 vent the lesser fall from being numerically proportional to the greater 

 one, however small the observed differences may be. The question, how- 

 ever, is as to how far a strictly numerical proportion is in this case a fair 

 one. I think we must certainly hold that the value as a stimulus of any 

 given increment of current is determined by the proportion which such 

 increment bears to the intensity of current that is required to produce 

 adequate stimulation. In other words, any given unit of electrical inten- 

 sity has more influence as an excitant if added to a current of a small 

 number of units (a weak current) than if added to a current of a large 

 number of units (a strong current). But if this is so, it follows that 

 subtraction of a unit from a strong current must have less effect than 



