260 Bayliss and Starling, 



of the heart perfectly, the adhesion of the clay to the surface of the 

 heart being sufficiently great to overcome any tendency to jerking off 

 produced by the momentum of the electrodes. They have a low re- 

 sistance and are nonpolarizable and equipotential. 



I. Excitatory wave in ventricle. 



To determine whether there is a wave of negativity in the ven- 

 tricle we led off from the various points of the outer surface of the 

 ventricles. The electrodes were in most cases placed on the fi'ont of 

 the right ventricle near the base, and on the apex of the left ventricle. 

 We also led off from base and apex of left or right ventricle alone, 

 and jfrom points situated between apex and base. We find that in 

 animals whose hearts are in as normal a condition as possible, the 

 variation is always diphasic, and sheivs negativity of the hase pre- 

 ceding that of the apex. The result is the same, whether the peri- 

 cardium be intact or opened, or from whatever points of the heart's 

 surface we lead off. (Figs. 1, 2, and 3. Plate XV). 



This result is opposed to the majority of the results obtained by 

 Waller as mentioned above. In our earlier experiments we obtained 

 similar results to his, i. e. apex negativity preceding base negativity, 

 though this was by no means an invariable result, our curves being 

 often triphasic indicating base negativity followed by apex negativity 

 and this again by base negativity. During this period of our investi- 

 gation we were using air at the temperature of the room for artificial 

 respiration; we thought, however, that the animal would remain in a 

 more normal condition and the circulation be better kept up if we used 

 warmed air, and for this purpose the air blown from the bellows was 

 made to pass through a spiral tube surrounded by boiling water before 

 entering the trachea. From the time that this proceeding was adop- 

 ted we have always been able to rely upon getting a diphasic vari- 

 ation from the ventricle and always showing basal negativity preceding 

 apical negativity. The conclusion we draw is, therefore, that in tlie 

 mammalian heart as in the frog's heart the electrical change travels 

 in the form of a wave from base to apex. 



It was some time before we discovered the cause of the sudden 



