258 COMPARISONS OF RESPIRATORY EXCHANGE. 



exactly the same manner, with no control upon the sampling. A 

 sample which was incorrectly drawn may therefore be equally divided 

 between the two burettes and yet duplicate results obtained. The 

 apparatus is very large and cumbersome and has a great number of 

 rubber connections which are liable to deteriorate, with consequent 

 leaks. It also requires the use of large samples 100 c.c. so that if 

 the analysis is not carried out immediately, a very large sample, at 

 least 200 c.c. or more, must be collected in order to have sufficient air 

 for flushing the connections when the samples are drawn. One of the 

 chief objections to the Zuntz apparatus is the fact that the analysis 

 is made over water. Practically all investigators are agreed that the 

 collection and analysis of air samples over water is to be avoided if 

 the carbon-dioxide content of a mixture of gases is to be determined 

 to less than 0.05 per cent. 



The major part of the analyses carried out in connection with this 

 research were made with the two forms of the Haldane gas-analysis 

 apparatus. 1 In the earlier comparisons, the laboratory form of this 

 apparatus was used exclusively. Phosphorus was successfully sub- 

 stituted for potassium pyrogallate as an absorbent, thus doing away 

 with the necessity for repeated raising and lowering of the mercury 

 reservoir and saving much time and labor in continuous work. The 

 phosphorus also required less frequent renewal; but on the other hand 

 it absorbed the oxygen more slowly than the potassium pyrogallate. 

 In the later experimenting the portable form of the Haldane gas-analysis 

 apparatus was used with very good success. Practically as good results 

 were obtained with it as with the laboratory form and it was much more 

 convenient to use. With both forms of the apparatus only a small 

 sample is required, i. e., 20 c.c. for the larger apparatus and 10 c.c. for 

 the portable apparatus. Smaller containers may therefore be used 

 for collecting the samples, which is of advantage when space is limited 

 and when large amounts of mercury are required. In all of the gas 

 analyses with these two apparatus it has been the routine to collect 

 the samples over mercury, so that both the collection and the analyses 

 were made over mercury. 



It must be pointed out that while apparently many people have 

 found it difficult to make gas analyses with sufficient accuracy for use 

 in determining the respiratory exchange, yet in this laboratory a 

 considerable number of individuals have been trained to use the Hal- 

 dane gas-analysis apparatus with good success. For example, one 

 young lady, who had had neither prior chemical training nor training 

 in gas analysis, was instructed in the use of the Haldane apparatus and 

 in two weeks was able to make satisfactory analyses of outdoor air and 

 of expired air. This young lady was but one of several assistants who 

 have been taught the technique in the same manner. The fact that 



'See description on p. 70. 



