60 Art. 1 -K. Fuji. 



l)etwL'on tliu two stimuli. When tlio interval l)ocoine.s 45 0x10-* 

 sec. the second discharge seems to disappear. When it is di- 

 minished to 37GxlO-* sec. the second discharge becomes percep- 

 tible on the descending Ijranch of the first discharge, but the 

 modal latent period cannot l)e measured, although it is certain that 

 it is made smaller. AVhen the interval Ijecomes 28-9x10"* sec. the 

 modal latent period of the superposed second discharge ma_y rough- 

 ly be determined to be 190x10"* sec. Finally, the interval becom- 

 ing 8 3x10"* sec, the two discharges coalesce into a single one. 

 From this onwards, the height of the first discharge becomes greater 

 than those of the preceding, thus showing the summation of the 

 second discharge. 



Oscillogram N(.». 51 (Plate XXI.) is a series of the experinients 

 of the same type. Here we may find as the ratios of the second 

 modal latent period to the first the values of lOO, 0-98, 95 and 

 0-93 for 4th, 5th, Gtli and 7th curve respectivel}^ 



Let us now discuss the results with reference to the theory 

 of "all or none.'' As the stinuilus is sul)maximal, a })art of the nerve 

 fibres is excited by the first stimulus and the other part is not excit- 

 ed, but only receiving the sul)minimal stinuilus. Let us call the 

 former the first part and the latter the second part. ^Vhen the 

 second stimulus, whicli is also subminimal to the second part, is in 

 the refractory j^eriod and on the outside of the summation Interval" 

 so-called l)y Lucas, which must be assumed to be smaller than the 

 refractory period, the second response cannot occur. But when 

 the second stimulus enters into the summation interval it excites 

 the second part in concordance with the first stimulus, while it 

 does not excite the first part. Hence the experimental results can 

 l)e explained. For the correctness of this interpretation, it must 

 be allowed that the refractory period in this case is greater than 

 775 X 10"* sec. and the summation interval is greater than 66-3 x 10"* 

 sec. The latter numl )er is \-ery great compared with the same measur- 

 ed I )y Lucas on the motor nerve to the gastrocnemius muscle of 

 the frog which was equal to 5 x ]0-*sec. AMiether the modal latent 



Tho .Journal of Physiology, Vol. 39. p. 461. 



