22 



Art. 1.— K. Fuji 



at all. The stimulus used in these experiments consisted of tlie 

 closing and the opemng-stimulus.'^ The stimulus on the break of 

 the current must be due to the recovery from some sort of 

 polarisation, though not limited to electrical polarisation, caused by 

 the current. It is not therefore unreasonable to suppose that, 

 the closer the opening-stimulus is to the closing-stimulus, the small- 

 er is the effect of the former. ]Moreover, when we consider the 

 existence of the refractory period, we may suppo'^e the effect of the 

 opening-stimulus neglected when the breadth of the stimulus is 

 sufficiently small. Comparing ( )scillogi'ams No. o7 and No. 40 

 with No. 54, we find that the breadths of the stimuli in No. 54 

 are gi'eater than those in No. 37 and in No. 40. Examining many 

 other oscillograms, we see that, where the breadth of the stimulus 

 is gi'eat, the unsymmetry of the curve is manifest. Hence we 

 may conclude that the anomaly in No. 54 must ])e due to tlie 

 supei*position of the effect of the opening-stimulus. 



In No. 54 we see that, when we take the magnitude of the 

 latent period into consideration, the actual commencement of the 

 discharge by the opening-stimulus must be a little later than the 

 culmination of the curve. Hence assuming that the discharges 

 caused by the closing-stimuli had their origins at tlic begiiming of 

 the stimuli and the ascending branches of the curves were wholly 

 due to them, we calculated Ir h-om tlieir respective co-ordinates x-^ 

 and 2/j. These values of h^ corresponding to each curve in No. 54 

 are represented in the following table. Let us denote sucli Ir \\\ 

 hi 



Table IV. 



* In this paper, the stimulus on the growth of the stimulating current is called 

 the closing-stimulus, and the stimulus on the decay of the current is called the openhig-sthnuhis, 

 •whether the current is derived directly from the battery or is induced by an induction coil. 



