400 



[Jan. 15, 



supposition that the cells would afford a large amount of surface to 

 the action of the gas. 



The serum, after being whipped, was permitted to stand for 

 twenty-four hours, that any fibrin which it might contain, and which 

 had not coagulated during the process of whipping, might do so. 



The apparatus used in all cases will be easily understood by re- 

 ferring to the annexed diagram. It consists, first, of a copper flask 

 containing black oxide of manganese, from which the oxygen was 

 slowly given off by the action of heat. 



The gas was conveyed thence by tubes into a wash-bottle, B (con- 

 taining a dilute solution of potass), for the absorption of impurities. 



From the bottle B the gas passed into the flask C, which con- 

 tained the defibrinated blood. This flask was placed in a vessel (D) 

 containing water at a temperature varying between 95 and 100 

 Fahr. ; and I had no difficulty in preserving that heat continuously 

 by a small gas-flame placed under a sand-bath. After the gas had 

 escaped from the blood, I generally passed it through a second por- 

 tion of defibrinated blood contained in another vessel (E) . 



For all these experiments pig's blood was invariably chosen, on 

 account of its richness in blood-cells. 



My apparatus being ready, and oxygen being slowly given off, the 

 whipped blood, from which every particle of fibrin had been pre- 

 viously removed, was introduced into the flask C. 



