20 
results of which are not yet ready for publication, the body temper- 
ature was measured by a specially devised electrical thermometer 
inserted in the rectum, by which differences in body temperature of 
0.02° could be detected. a 
The observations of the pulse rate were also made by the subject. 
How observations made by a person upon himself compare in relia- 
bility with those made by another person we are unable to sa}^ 
The reading and recording of weights and measurements of food 
and excretory products and the other determinations outside the 
chamber are controlled either by repeating the observations or by 
having the quantities observed and recorded by two different persons 
or otherwise. The conditions for the reading and recording of figures 
for body weight, temperature, and pulse rate by the subject in the 
respiration chamber do not permit of so careful a system of checks 
for avoiding errors in the reading and recording of the figures. For 
this reason, and because of the uncertainties regarding the value of 
the observations of temperature and pulse rate just named, the figures 
for body weight, temperature, and pulse rate are hardly as reliable as 
those for the other determinations. 
MEASUREMENT OF MUSCULAR WORK. 
In the rest experiments the subjects moved about within the appa- 
ratus as little as possible, thus avoiding all unnecessary muscular exer- 
tion. In the work experiments, however, they were engaged in active 
muscular work for eight hours each day. The apparatus used for 
measuring the work in experiment No. 11 consisted of a bicycle belted 
to a small dynamo, as in experiment No. 6. b The voltage of the elec- 
tric current generated was measured, and together with calibrations 
of the apparatus gave sufficient data for computing roughly the 
amount of work done by the subject. 
In the work experiments carried on during the winter of 1900 
(experiments Nos. 29, 31, 32, and 34) the apparatus was so modified 
as to give, it is believed, a more accurate measure of the work done. 
In these experiments the rear wheel of the bicycle was applied 
directly to a pulley fastened to the shaft of the dynamo, which was 
mounted upon a rocking base, the necessary tension being secured by 
means of a coiled spring which kept the pulley of the dvnamo pressed 
against the tire of the bicycle. This apparatus was calibrated before 
and after each experiment. Inasmuch, however, as the method of 
calibration was greatly improved in the experiments made during the 
winter and spring of 1901, further description of the apparatus and 
the methods of measuring the work done is not given in the present 
publication. 
"Arch. Physiol. [Pfliiger], 88 (1901), No. 9-10, p. 492. 
b U. S. Dept. Agr., Office of Experiment Stations Bills. 63, p. 76; 69, p. ■*.. 
