258 Baylìss and Starling, 



The intention with which we commenced the present work was 

 three-fold: 



I. To discuss whether there is evidence of a wave of contraction 

 in the ventricle, and to determine its course and time-relations if present. 



II. To obtain evidence as to the nature of the transmission of 

 contraction from auricle to ventricle, chiefly by measurements of the 

 time taken by the excitatory process in travelling from auricle to 

 ventricle and from one point of the ventricle to another. 



m. To examine Frédéricq's view on the tetanic natiu'e of the 

 ventricular contraction. 



Method. 



Our experiments were commenced on cats, but we soon found 

 them unsatisfactory, the heart appearing to be more vulnerable than 

 that of dogs, to wliich we afterwards confined our attention. In all 

 cases the animals were anaesthetized with a large dose of morpliia 

 (2 — 4 gi'ains) and chloroform. Tracheotomy was performed and arti- 

 ficial respiration earned on from the moment the chest was opened; 

 sometimes curare was given, but as a rule it was not found necessary. 

 A median incision was then made in the front of the chest, the ster- 

 num divided in the middle line by bone-forceps, and the two halves 

 of the thoracic wall puUed asunder by hooks so as to give complete 

 access to the heart. In opening the chest in this way, if one keeps 

 to the middle une accurately, there is practically no bleeding. The 

 pericardium was then opened by scissors and stitched to the thoracic 

 wall on each side; in tliis way the heart was supported below and 

 no longer aifected by the movements of the lungs. 



In some cases the vagi were divided before opening the chest, 

 and prepared for excitation. As a rule the slowing of the heart pro- 

 duced l^y the morphia exciting the medullary inhibitory centre was 

 found advantageous, — hence the vagi were usually left intact. Occa- 

 sionally one of the carotids was connected with a manometer to re- 

 gister the heart beat, but in most cases the beats were registered by 

 a small tambour pressing on the ventricle through the pericardium at 

 the side. The heart being now exposed, two points of its surface 

 were connected by nonpolarizable electrodes to the terminals of the 



