46 COMPARISONS OF RESPIRATORY EXCHANGE. 



turned at the end of a normal expiration and the subject begins inspir- 

 ing from the apparatus. Oxygen is admitted either continuously or 

 intermittently. If the meter is used, the movement of the pointer is 

 recorded each time it passes the zero-point of the meter. At the end 

 of the experiment the valve is turned as with the older type of appa- 

 ratus. After running a few minutes, oxygen is admitted into the 

 apparatus until the spirometer is at the same level as at the beginning 

 of the experiment. For convenience this admission of oxygen may be 

 omitted in actual practice, care being taken to read the height of the 

 spirometer and then to correct for the actual difference in level between 

 the beginning and end of the experiment. A reading of the ventilation 

 adder is also taken at the end of the experiment and noted on the 

 record sheet. During the latter half of the experimental period the 

 completeness of absorption of the carbon dioxide is tested, as with the 

 older apparatus, by deflecting a portion of the air-current through a 

 solution of barium hydroxide. 



OXYGEN SUPPLY FOR THE UNIVERSAL RESPIRATION APPARATUS. 



In connection with the direct determination of the oxygen consump- 

 tion of the subject it is necessary to admit the oxygen in such a manner 

 that it can be easily and accurately weighed or measured. It is also 

 necessary to have the supply free from carbon dioxide and water- vapor, 

 or to make some provision for removing these gases. In the earlier 

 experimenting, oxygen was admitted from a small cylinder containing 

 about 150 liters of the gas. As the kind of oxygen first purchased 

 contained both carbon dioxide and water, the cylinder was provided 

 with tubes for the removal of these impurities. A rubber bag attached 

 to a tee which was connected with the valve prevented any sudden 

 escape of the gas through the tubes when the valve was opened. 



These small cylinders were used for some time, but there were a 

 number of disadvantages in connection with their use. It was necessary 

 to make sure that the bag was absolutely deflated each time that the 

 cylinder was used, and that the connections on the carbon-dioxide and 

 water-vapor absorbers were absolutely tight. These latter parts, being 

 fragile, were easily broken, and whenever such a break occurred the 

 determination of the oxygen was lost for that particular experiment. 

 The fitting of the purifying apparatus to the cylinders also required 

 considerable time. 



For a brief period in the early development of the apparatus an 

 oxygen generator was used which furnished oxygen by the generation 

 of the gas from the action of water on sodium peroxide. A tin can 

 containing fused sodium peroxide was held in the bottom of a container 

 by means of springs. Holes were punched in the top and bottom of 

 the can to allow the admission of water. The can of sodium peroxide 



