52 COMPARISONS OF RESPIRATORY EXCHANGE. 



requires an additional weighing; furthermore, if the valve is not abso- 

 lutely tight, the whole apparatus for determining the oxygen is useless. 

 The valves also vary in their closeness of fit; occasionally one is found 

 which leaks slightly and again another will remain tight for a number 

 of months. It is also sometimes difficult to obtain a collar which 

 fits closely against the valve opening of the cylinder. 



The meter method has an advantage in that the rate of admission 

 can be noted and a leak detected while the experiment is in progress. 

 Furthermore with the meter a large cylinder of oxygen, i. e., with a 

 capacity of 100 cubic feet, may be used, this supply being sufficient 

 for a period of several months without renewal. Among the disadvan- 

 tages is the fact that occasionally the noting of the number of liters 

 used is inadvertently omitted. The operator, in looking over the other 

 factors of the experiment, may discover this omission, but the results 

 may be of such a character that the addition of 1 liter may or may not 

 correct the evident error. Several attempts have been made to avoid 

 this error by providing an automatic recording attachment. This has 

 been in most instances electrical. The pointer attached to the moving 

 drum of the meter is provided with a short rod at right angles to it, 

 so that when passing a contact at the top of the meter a circuit is 

 closed. Several different kinds of contact have been inserted in the 

 top of the meter, but none of them has as yet proved absolutely 

 reliable and they can not be recommended. With an electrical record- 

 ing device, the full amount of oxygen to be supplied must be admitted 

 during the experimental period, as otherwise the record will not give 

 the true value. Another method for preventing this error of omission 

 has been instituted by Mr. H. L. Higgins, of the Laboratory staff. 

 Instead of admitting the oxygen at such a rate as to equal the consump- 

 tion of the gas by the subject, he allows the volume of the apparatus 

 to diminish gradually for the first 3 or 4 minutes, and then admits 

 quite rapidly 1 liter of oxygen. At the end of the seventh or eighth 

 minute the process is repeated and again at the end of the tenth or 

 twelfth minute. If this routine is adhered to, there is no danger of 

 omitting the recording of a liter. The only disadvantage is that dur- 

 ing the time of admitting the gas rapidly there is liable to be a distor- 

 tion of the respiration record. Occasionally, through oversight, oxygen 

 has been admitted to the meter when the exit pipe to the apparatus 

 was closed. This caused such a pressure inside the meter that the glass 

 face was blown out. Recently, at the suggestion of Mr. L. E. Emmes, 

 of this laboratory, a device has been used which prevents such an acci- 

 dent. 1 A third glass tube is inserted in the moistener with the lower end 

 below the level of the water in the moistener and the upper end above 

 the level of the water in the water-bath. When pressure accumulates, 

 this acts as a safety valve and allows the release of the gas before suffi- 

 cient pressure can be accumulated to cause damage. 



'See page 47. 



