Rejiex Nervous Discharge. 



277 



stretched wire dipped into a Hg pool beneath it. An electric circuit, inde- 

 pendent of the stimulating circuit, was arranged so that its current passed 

 through the electromagnet when the armature needle entered the Hg pool 

 beneath. The wire therefore continued to vibrate at its own torsional 

 frequency, the rate depending on its length and diameter and on the weight 

 and moment of the armature and cross-spar with which it was loaded. The 

 cross-spar carried, insulated from the torsion-vibrator's circuit, two fine wires 

 re-uniting the short twin paths of the primary circuit of the stimulating 

 inductorium. Each of these wires joined at its one end a fine gilt needle 

 fixed vertically at end of the horizontal spar, the needle point lying 

 close above the corresponding Hg pool. When the steel wire was 

 torsed, one of the twin branches of the primary circuit was made and the 

 other broken, and vice versa in its elastic recoil in the opposite direction. 

 With both of the twin branches of the circuit in use, the current passed 

 therefore alternately by one and the other, the speed of break (and make) 

 being identical in the two. With both of the twin branches in use, the 

 frequency of breaking (and making) of the primary circuit was, of course, 

 the double of the frequency when only one of them was in use. The 

 frequency of the torsion vibration could be readily altered within wide limits 

 by altering the length of the steel wire or by adjusting a screw- weight 

 attached to the armature carried by the wire. 



With this interrupter the frequency of the break-shocks applied to the 

 nerve could, by closing a key, be doubled and, by opening it, halved without 

 other change in the stimulation as it progressed ; we sought for the degree of 

 frequency beyond which further increase of frequency, e.g. doubling, caused 

 no further appreciable increase in the tetanic contraction. Using it on the 

 motor-nerve direct we confirmed Marey's observation that further increase 

 of stimulus frequency beyond that at which the myograph record becomes 

 steady and non-vibratory does produce increase in height and tension of 

 contraction up to a certain point. 



Turning then to the reflex preparation, and using similar retiex and similar 

 isometric records to those already mentioned our results have been as follows. 

 When the reflex contraction was in progress under a stimulation of the 

 afferent nerve at 50 per second, increase, namely doubling, the frequency 

 caused immediate marked increase in the height (tension) of the contraction, 

 which fell back to the previous height again at once on returning to the 

 previous slower mte. There was no reason to suspect any difference between 

 the making and breaking of the primary circuit as carried out by the torsional 

 interrupter in the two twin branches respectively. But to control this as a 

 possibility the observations were repeated with alternate priority of one or 



