APPARATUS AND METHODS USED IN THIS RESEARCH. 63 



ture of the water in the water jacket around the chamber was recorded 

 and, if necessary, controlled by cooling during the period until the air 

 inside the chamber, as recorded by the thermometer in the cover and 

 by the thermometer in the outgoing air, had a temperature not far 

 from 20° C. During this preliminary period, the air-current passed 

 through one of the two sets of purifiers, the other set having previously 

 been weighed and connected into position ready for use. Just prior 

 to the end of the period, air was allowed to escape through a pet-cock 

 in the system until the bell of the spirometer was about 30 mm. above 

 the lowest point; the spirometer was then about half filled with oxygen, 

 thus avoiding any possibility of there being a deficiency in the oxygen 

 content of the air. While the rate of ventilation should be approxi- 

 mately constant, i. e., about 35 liters a minute, it is practically impos- 

 sible to regulate this unless a very constant electrical current is available. 

 With the Crowell blower, as here used, about 270 revolutions a minute 

 gives a suitable ventilation. 



When the infant was perfectly quiet and the preliminary period was 

 nearing the end, records were made of the readings of the wet- and dry- 

 bulb thermometers, the thermometer in the top of the chamber, and the 

 thermometer showing the room temperature. Air was then released 

 from the system through the pet-cock until the spirometer pointer read 

 about 30 mm. Inasmuch as there is a slight tension due to the uncom- 

 pensated spirometer bell, the counterweight of the spirometer was taken 

 in the hand and the bell gradually raised until the delicate petroleum 

 manometer read zero. While the valves were turned to deflect the 

 air-current from one set of air-purifiers to the other, the manometer 

 was held at zero and the readings of the spirometer level taken imme- 

 diately before and afterward. The beginning of the new period was 

 marked on the kymograph record by the assistant, who then recorded 

 the barometer reading and the temperature of the barometer. The 

 readings on the oxygen meter having previously been taken or the 

 oxygen cylinder carefully weighed, about one-half liter of oxygen was 

 introduced or even more if the infant was very large or restless. This 

 of course raised the spirometer bell. By determining the time in which 

 the infant utilizes the half liter, of oxygen, the oxygen supply can be 

 regulated with considerable exactness, so that the spirometer reading 

 will be nearly the same at the beginning and end of each period. 



The length of a period of observation depends altogether upon the 

 muscular repose of the infant, as only quiet periods, accompanied by 

 a low pulse-rate, are of value. With a small, quiet infant, the periods 

 may vary in length from 20 to 30 minutes, but with a large infant they 

 may be as short as 15 minutes. About a minute and a half before the 

 close of the period, the dry- and wet-bulb thermometers were again 

 read, also the thermometer in the cover of the chamber, and the temper- 

 ature of the room was recorded. The manometer was then adjusted 



