34 A RESPIRATION CALORIMETER. 



acid, and several portions of concentrated acid poured in at the top, 

 each successive portion being allowed to drain out before the next is 

 added. Obviously such replenishment of acid is made only when the 

 purifying system is to be changed from an empty cylinder to a new 

 one, and, as pointed out above, at least ten cylinders can be used with 

 each charge of acid. The replenishment of the soda lime in the U tube 

 is made only when the reagent becomes exhausted, as is readily noted 

 by the whitening effect on the reagent. (See p. 29.) 



By attaching the purifying device to the cylinder itself and noting 

 the loss in weight in the system as a whole, the weight of gas used can 

 be obtained, since it corresponds to the amount of oxygen and nitrogen 

 leaving the cylinder. The quantity of oxygen consumed in the course 

 of 24 hours by a subject, varying as it does from 350 to 1,500 liters, 

 can be best determined in this way. 



By means of the balance described on page 57 it is possible to note 

 the loss in weight of an oxygen cylinder to within 10 mg. As 10 cc. of 

 oxygen weigh but 14 mg., it is thus seen that 283 liters (the contents 

 of one cylinder) can be measured with an accuracy far beyond that of 

 any gas-meter with which we are familiar. In all of the work with the 

 respiration calorimeter this method of measuring oxygen is constantly 

 used. 



In weighing, the cylinder is suspended on a wire from the balance- 

 arm by two loops of wire, one around the valve end of the cylinder and 

 the other, a much larger loop, around the bottom of the cylinder. The 

 manipulation of the cylinder and its adjustment on the balance require 

 a little care on the part of the assistant, but in spite of the use of glass 

 tubes for absorbers there has been as yet no loss by breakage during 

 weighing. 



ANALYSIS OF OXYGEN. 



Since the gas in the cylinder contains nitrogen as well as oxygen, 

 and the amount of oxygen admitted to the system is estimated from the 

 loss in weight of the cylinders, it is obviously necessary to analyze the 

 gas and determine the percentage of nitrogen. As has been stated, the 

 amount of nitrogen generally present is not far from 2.5 to 8 per cent. 

 There is also a small amount of carbon dioxide, but this is removed by 

 the soda-lime U tube attached to the cylinder. Of the three standard 

 methods for absorbing oxygen, i. <?., the use of potassium pyrogallate, 

 phosphorus, and explosion with hydrogen, the first is most readily 

 adapted to the analysis of nearly pure oxygen. The method of analysis 

 consists, therefore, of measuring a known volume of commercial oxygen 

 free from carbon dioxide, absorbing the oxygen by potassium pyrogal- 



