OF TEMPERATURE UPON THE BEAT OF THE DOG’S HEART. 
673 
the left ventricle, then the elastic reaction due to distension of the stumps of the 
great arteries is altered. Hence, even while a heart is pumping blood freely out 
through the exit S, kept at a constant height, variation of arterial pressure, as mea¬ 
sured in the carotid, may occur to the extent of 10 millims. of mercury pressure. Such 
variations will be noticed in some of the protocols of experiments given in this paper ; 
but fortunately they in no way affect the question here considered, viz.: the influence of 
changes of temperature upon the rate of beat of the isolated heart. I have previously 
shown (2) that slow variations of arterial pressure between the limits of 30 and 150 
millims. of mercury do not in the least influence the pulse-rate of the isolated Dogs 
heart, provided venous pressure and the composition and temperature of the blood be 
kept constant. 
Venous pressure and, approximately, arterial pressure being kept uniform the 
temperature of the blood alone was altered in the experiments below described. The 
variation was effected in two ways. First by pouring a little heated (50° C.) or cooled 
(10° 0.) blood into the funnel F from which it entered the flask not in use at that 
moment, and warmed or cooled the blood already in it. Then this flask was used to 
feed the heart and the other as the recipient, by opening and closing the proper 
clamps and stopcocks. This method was rarely used, as it sometimes produced 
secondary effects, due to the comparatively sudden changes of temperature in the 
blood supplied to the heart. A more gradual and uniform alteration in the tempera¬ 
ture of the blood was secured by changing the water in the jackets around the 
Mariotte’s flasks. Some hot water and some water cooled by ice to 5° or 10° C. were 
always kept at hand during an experiment. If a series of heating observations was 
to be made, some of the water already in the jackets was syphoned oft', and replaced 
with warm, care being taken that the temperature never rose above 60° C., so as to 
avoid all risk of coagulating any of the proteids of the defibrinated blood : more hot 
water was added from time to time if necessary. To initiate a series of observations 
as to the effect of cooling, the iced water was of course employed. The best results 
were obtained when the temperature of the water in the jackets did not differ 
by more than 20° from the temperature of the blood in the flask. When either 
Mariotte’s flask was in use the rapid bubbling through its contents of the air 
entering by the tubes e and e ensured their thorough mixture. 
Having waited, then, for the death of extrinsic nerve centres, and until the 
thermometer p had during some minutes indicated a tolerably even temperature, the 
water around the Mariotte’s flasks was cooled or heated, and a series of observa¬ 
tions commenced. The initial temperature usually lay between 37° and 38°, but, 
as will be seen in the experiment protocols which follow, was sometimes higher or 
lower. Tracings of arterial pressure and pulse-rate were taken at intervals varying 
from one to five minutes. When the tracing was completed an assistant immediately 
opened a small door in the front of the warm chamber, and read off the temperature 
of the blood flowing through the heart. 
