10 
and heat given off were determined, showed variations between the 
observed and the theoretical amounts no greater than are ordinarily 
obtained by the usual analytical methods employed in the laboratory. 
The detailed description of the electrical and alcohol test experiments 
may be found in earlier bulletins of this series. a 
Without going further into the details of the experiments it will 
suffice to say that they were conducted in the same way as the similar 
test experiments previously reported, by burning within the calorim- 
eter a known amount of ethyl alcohol in a specially arranged lamp 
and determining the amounts of carbon dioxid. water, and heat given 
off. The air current which passed through the chamber and the con- 
ditions of the experiment were very similar to those when a man was 
conhned within the apparatus. 
As stated above, the electrical tests were all made previous to the 
time of the experiments here reported. Of the other test experi- 
ments, the details of which have already been published, one. No. 9, 
was made in the late spring of 1898, at the close of the second of the 
metabolism experiments here described. During the summer of 1898 
no experiments were carried on with the respiration calorimeter. The 
first test experiment of the fall commenced November 3, 1898. The 
agreements between the determined and theoretical amounts of carbon 
dioxid, water, and heat, while not as close as had sometimes been 
found, were probably Avithin the limits of experimental error, thus 
indicating that the apparatus was in good working condition. Follow- 
ing this test came metabolism experiment No. 13, reported in this bul- 
letin, which was immediately followed by another test experiment, 
No. 11. 
About the middle of December, 189S, still another test experiment, 
No. 12, gave results which indicated that the apparatus was in satis- 
factory condition, although the measurement of water was not as close 
as might be desired. This test was followed by metabolism experi- 
ment No. 14, the details of which are given further on. The calorim- 
eter was then used for some experiments not reported in this bulletin. 
At their close test No. 13 was made, which again showed the apparatus 
to be in satisfactory condition, although the water determined was in 
excess of the theoretical. Metabolism experiments Nos. 23 and 21 
followed the completion of test No. 13, and one week after their com- 
pletion test No. 14 was made. In this latter test, while the results 
of the determinations of carbon dioxid and water given off in the appa- 
ratus agreed very closely with the theoretical, it was found that the 
heat measured by the calorimeter was 3.2 per cent less than the heat 
theoretically given off by the combustion of the alcohol. Test experi- 
ment No. 15, made in December, 1899, showed similar agreements of 
carbon dioxid and water, and disagreement of heat. It was also found 
a U. S. Dept. Agr., Office of Experiment Stations Buls. 63, pp. 38-74; 69, pp. 8-17. 
