On the Mechanism of Chemical Temperature Regulation. 203 



refilled with oxygen, the same process was again gone through with the 

 modifications of noting the time to the nearest second when the pump was 

 stopped and the reading of the burette. The differences between the two 

 readings of the burette give the oxygen consumed in the time. After each 

 observation the temperatures of the rectum, the skin, and the water of the 

 bath in which the animal was placed, were entered in the protocol. 



As owing to the changing temperature of the animal the temperature of 



^ To Excentnc 



To Gasometer 



Fig. 1. 



the air in the lungs necessarily varied, no effort was made to keep the 

 temperature of the air in the apparatus under control. But as the volume 

 of the apparatus was reduced to a miminum — about 300 c.c. at the middle 

 stand of the burette — the error due to changing temperature or pressure was 

 very small in proportion to quantities of gas (oxygen) consumed. The pump 

 gave a circuit of about 65 c.c. for each downstroke of the pump — thirteen 

 to the minute — which proved an efficient ventilation for rabbits up to 

 2*5 kgrm. The oxygen in the apparatus was never found below 35 per cent. 



S 2 



