15 
carbon dioxid as determined in the sample drawn by the meter pump was 
slightly nearer the theoretical than the corresponding values obtained 
by the aspirators. Variations in the amounts of water as determined 
in the samples drawn by the meter pump were a little greater than 
those for carbon dioxid. The comparative results obtained by the two 
methods in test experiments Nos. 9-12 are shown in the following table: 
Table 2. — Comparison of determinations of carbon dioxid and water in the ventilating air 
current by means of the aspirators and of the meter pump. 
-. 
Dura- 
tion. 
Carbon dioxid. 
Water. 
- 
Theo- 
retical 
amount 
from 
alcohol 
burned 
in lamp. 
Determined — 
Theo- 
retical 
amount 
from 
alcohol 
burned 
in lamp. 
Determined — 
s 
By meter pump. 
By aspirators. 
By meter pump. 
By aspirators. 
h. m, 
Grams. 
Grains. 
P. ct. 
Grams. 
P. ct. 
Grams. 
Grams. 
P.ct. 
Grams. 
P. Ct. 
9 
35 55 
1, 206. 9 
1,199.2 
99.4 
1,198.9 
99.3 
809.3 
830.6 
102.6 
807.9 
99.8 
10 
35 44 
1, 159. 8 
1.163.9 
100.3 
1,087.0 
93.7 
772. 5 
773.5 
100.1 
739.7 
95.8 
11 
35 00 
1,304.2 
1,293.7 
99. 2 
1.287.9 
98.7 
869. 8 
878.2 
101.0 
877.8 
100.9 
12 
27 56 
1,075.4 
1,067.7 
99.3 
1,083.2 
100.7 
717. 2 
705.7 
1 
98.4 
705.0 
98.3 
There was on the whole but little difference between the results 
obtained by the two methods of sampling the ventilating air current, 
except in the determinations of water in No. 10. Since the meter 
pump theoretically should give a more accurate sample than the 
aspirators, and since the experimental evidence shows that the results 
with the pump were at least as accurate and sometimes more so than 
those with the aspirators, the determinations of carbon dioxid and 
water in the sample drawn by the meter pump were employed in the 
calculation of the results of all metabolism experiments made subse- 
quent to metabolism experiment No. 11, although check determina- 
tions by means of the aspirators were continued during several experi- 
ments. In most of the experiments the discrepancies between the two 
methods were so insignificant as to be within the limit of experimental 
error, and show that the determination of carbon dioxid and water 
may be made with a satisfactory degree of accuracy by either of the 
methods. 
EXPERIMENTS WITH MEN. 
The general plan of the experiments here reported in detail was 
much the same as that followed in earlier experiments, the results of 
which have been published in former bulletins of this series. a 
For the most part the experiments here reported were made in 
series, each series comprising three or more experiments which fol- 
lowed one another without intermission and without the subject 
a TJ. S. Dept. Agr., Office of Experiment Stations Buls. 44, 63, and 69. 
