Y. Henderson and A. H. Smith 4] 
ber of holes bored in the sides of the block of wood into which 
diffusion tubes can be inserted. The rate of revolution should 
be not more than 40 or 50 times a minute. During the rotation 
the contents of the tube spread in a thin film along the walls 
and allow complete diffusion of the liberated oxygen into the air 
of the tube. © 
The tube is next connected glass to glass with the gas analyzer 
by a short piece of heavy rubber tubing in the manner shown in 
Fig. 3. . The mercury bulb of the analyzer is lifted until mercury 
runs out through the top of the analyzer and fills the capillary 
tube of the blood gas diffusion tube. The lower end of the dif- 
fusion tube is placed in a beaker of water and the stopper is 
Fig. 2. Rotating apparatus. 
withdrawn under water. The beaker is raised or lowered until 
the levels of the fluid inside and outside the tube are the same, 
and the volume of the air in the tube is read. The stop-cock at 
the top of the diffusion tube is then turned and a sample of the 
air within the tube is drawn over into the analyzer and analyzed 
for oxygen. 
The residual gas from this analysis is nitrogen. As the per- 
centages of oxygen and nitrogen in atmospheric air are known, 
it is easy to calculate from the volume of this residual nitrogen 
the exact amount of oxygen which would have been found by the 
analysis if no oxygen had been given off by the blood. Todo 
this multiply the residual nitrogen by ae or 0.265. The vol- 
ume of oxygen thus calculated to have been in the air is sub- 
