11 • 
by thermal a tests and electrical tests that the measurement of heat by 
the apparatus was not correct. An electrical test made in the early 
part of December. 1899. showed a loss of over 3 per cent of the heat 
put into the apparatus. A thermal test made at this time showed that 
there was trouble with the thermo-electric elements. b The difficulty 
apparently lay in a slight short circuit of these at some point, which 
interfered with the accuracy of the measurements. 
Just when the short circuiting occurred it is impossible to say. The 
test experiment preceding metabolism experiments Nos. 23 and 24 
showed the apparatus to be in a satisfactory condition. Test experi- 
ments made shortly after the close of these experiments showed a dis- 
crepancy in the heat measured as compared with that theoretically 
given off. There is nothing to indicate whether the trouble occurred 
before, during, or at the close of this series of metabolism experi- 
ments or in the short interval immediately following. The agreement 
of the theoretical energy of the material oxidized in the body and that 
measured by the calorimeter in metabolism experiments Nos. 23 and 
24 would seem to indicate that the trouble did not occur until the close 
of the series. Had it occurred before or during these experiments 
the heat as actually determined by the apparatus would doubtless have 
been about 3 per cent smaller than the amount which would have been 
measured had the apparatus been in its usual condition. This would 
make the heat measured by the calorimeter much -mailer in amount 
than the theoretical heat given oft by the oxidation of material in 
the body of the subject, the deficiency being larger than has been 
found in previous experiments. 
Since it is impossible to determine just when the trouble arose, the 
details of experiments Nos. 23 and 24 are here given with the values 
actually obtained, and the supposition is made that the trouble in 
the apparatus occurred after the close of the experiments. The data 
are. however, given by which any correction which seems fitting can 
be applied. After test experiment No. 15. some repairs were made 
in the apparatus, following which came test experiment No. 16. in 
which ethyl alcohol was burned in the respiration chamber. In 
this test the results obtained were very near the theoretical values 
for carbon dioxid and heat, but nearly 4 per cent more water was 
found than should have been obtained from the quantity of ethyl 
alcohol burned. It was thought best, however, not to delay further 
the metabolism experiments of 1S99-1900. therefore no other test 
experiments were made until April. 1900, at the close of metabolism 
experiment No. 34. At this time test experiment No. 17 gave results 
very similar to those obtained in the previous test experiment, and 
indicated that the measurements of water in the test experiments were 
a U. S. Dept. Agr., Office of Experiment Stations Bill. 69, p. 15. 
b U. S. Dept. Agr., Office of Experiment Stations Bui. 63, p. 17. 
