270 Bayliss and Starling, 



sequence to the ventricular contraction. We were in doubt as to this 

 at first, and therefore took tracings with levers attached to auricle 

 and ventricle, and found it to be unquestionably the case; it is clearly 

 of some importance with respect to the nature of the transmission from 

 auricle to ventricle. (We were unaware at the time that we made 

 these investigations, that tliis method of producing a reversed beat 

 had been already described by Mc. William^).) 



By this means then, we replace the natural by an artificial rhythm, 

 and we know also the exact time at winch the excitatory process 

 commences. The stimulus used was an induction shock, excited by the 

 break of the primary circuit. This was performed about 3 or 4 times 

 per second by one of the contacts of a v. Frey's rheotome, the other 

 contact being arranged to short-circuit the make-shock. 



To measure the latent period of the electrical change of the ven- 

 tricular muscle, some point of the surface of the ventricle was excited 

 immediately underneath one of the leading-off electrodes. 



To measure the velocity of propagation of the wave from auricle 

 to venti icle, the base and apex of the ventricle were led off to the 

 electrometer and the stimulating electrodes applied either to an auri- 

 cular appendage or to the body of one of the auricles in front or 

 behind. 



When using morphinised dogs, it is necessary to cut both vagi, 

 otherwise the vagus excitation produced by the drug acting on the 

 cardio-inhibitory centre, lowers the excitability of both auiicle and 

 ventricle, and produces a „block" between auricle and ventricle, so 

 that either the point excited does not respond to every stimulus, or 

 the contraction, if produced, does not travel across the auriculo-ventri- 

 cular groove. 



In these experiments, in addition to the excursions of the mer- 

 cury meniscus, there were simultaneously photographed by means of 

 two chronographs the moment of stimulation and the vibrations ot a 

 tuning-fork of 100 per second. 



In oj'der to obtain more exact information of tlip time-course of 



') Journal of Physiology. Vol. IX. 1888. 



