276 Messrs. N. B. Dreyer and C. S. Sherrington, 



in response to repetitive stimulation of the motor nerve at various rates. 

 We have controlled this in the muscle we used for the reflex, tibialis anticus, 

 and in the gastrocnemius (without soleus), and find the limit with our 

 myograph at close to 65 per second. Above that frequency, and sometimes 

 a little below it, the tetanic contraction in response to motor-nerve stimula- 

 tion exhibits practically no visible undulation. 



It was noted, however, by Marey,* a number of years ago (1867), that 

 when the motor nerve is stimulated by rhythmic stimuli of progressively 

 quicker and quicker rate there is, after a frequency has been reached at 

 which the tetanic contreu3tion becomes a smooth non-vibratory line, a stUl 

 further increase in the contraction height and tension on further increasing 

 the frequency of the rhythmic stimuli. " Apres que toute vibration a disparu 

 dans le graphique, on voit la ligne tracee s'elever de plus en plus sous 

 I'influence d'excitations de plus en plus rapprochees."f This further incre- 

 ment of contraction, due to increment in number of nerve-impulses per 

 second in the motor nerve, ofifered a means of testing whether the reflex 

 centre can respond with increased frequency -rate of discharge even after its 

 afferent nerve's stimulation-rate has reached and passed beyond that pitch at 

 which in the muscle individual contraction waves cease to be visibly 

 distinguishable by the myogram. 



Marey's observations, in the description extant of them, may not have been 

 entirely free from objection ; using a rotating interrupter, as he did, the 

 individual breaks are more sudden under the quicker rotation, and the 

 individual break-shocks therefore become more powerful as individual 

 stimuli. He was, however, presumably using maximal stimuli throughout, 

 and in that case further increase in their stimulation potency would not 

 account for the increase of contraction observed. In applying the experi- 

 ment to reflex contraction, where it is less easy to be sure that the individual 

 stimuli are maximal, it would be preferable, when increasing their serial 

 frequence, not to change the speed of the individual acts of breaking of the 

 primary circuit. 



To attain this, the following plan was devised (C. S. S.). The current 

 path of the primary circuit was bifurcated for a short distance into two equal 

 branches, each including a Hg pool, the twin pools lying under the ends of 

 a horizontal spar, which was fixed transversely at its middle to a horizontal 

 steel wire stretched between rigid uprights. The wire carried a horizontal 

 armature, which could be acted on by tlie poles of a small electromagnet, and 

 from one end of the armature a vertical needle on slightly torsing the 



* E. J. Marey, ' Du Mouvement dans les Fonctions de la Vie,' p. 376. Paris, 1868. 

 t Loc. cit., p. 376. 



