404 
Journal oj Agricultural Research 
Vol. XXX, No. 5 
and urine for analysis. Of these the 
drying of the feces-and-urine mixture is 
the most important, since it involves 
the computation of corrections, the 
bases for which are imperfectly under¬ 
stood. 
DIRECT HEAT MEASUREMENT 
The direct measurement of the heat 
given off by the animal body is accom¬ 
plished by means of the respiration 
calorimeter. Descriptions of such an 
apparatus, as used at this institute, 
and also of the methods employed, 
have appeared in several publications 
by Armsby and Fries ( 3; 4, p • 263-270; 
1; 6, p. 200-222). For the present 
purpose the more important details 
may be condensed into the following: 
The respiration calorimeter used at 
the Institute of Animal Nurtition of the 
Pennsylvania State College is of the 
Atwater-Rosa type, consisting of a 
Pettenkofer respiration apparatus, the 
chamber of which is also an animal 
calorimeter. 
The respiration chamber is built with 
a double metallic wall inclosing a 
3-inch dead-air space, surrounded, at a 
distance of 4 inches, by a wooden wall, 
and this in turn, at a distance of 4 
inches by a second wooden wall, thus 
forming two air-spaces surrounding the 
chamber. By the heating or cooling 
of these air spaces the inner wall is 
maintained adiabatic. 
The sensible heat given off by the 
animal, which constitutes the greater 
part (about three-fourths) of the total 
heat production, is absorbed by a 
current of cold water passing through 
copper pipes at the top of the respira¬ 
tion chamber, the exposure of these 
pipes to the air within the chamber 
being regulated by means of shields 
which can be raised or lowered by the 
operator. By this means the rate of 
removal of heat from the chamber is 
adjusted as required for the mainten¬ 
ance of an approximately constant 
temperature. 
The temperature of the in-going air 
is maintained equal to that of the out- 
coming. 
The temperature of the in-going and 
outcoming water is read every four 
minutes by means of two mercurial 
thermometers, graduated to 1/50° C., 
and carefully calibrated. 
The volume of water passing through 
the calorimeter is measured by means 
of two copper meters of 100-liter 
capacity. 
For each period of uniform water 
flow the product of the amount of 
water passing through the heat ab¬ 
sorbers multiplied by the average 
temperature difference equals the 
amount of heat removed from the 
chamber in the water current. 
Since a part (approximately one- 
fourth) of the total heat given off by 
the animal is in the form of latent heat 
of water vapor, this portion of heat, 
not being directly measurable by the 
calorimeter, is calculated from the 
amount of water vapor given off, 
using 0.587 Calorie per gram at 18° C. 
This heat is added to the heat removed 
in the water current to get the total 
heat emission. 
CORRECTIONS 
In order accurately to determine the 
amount of heat produced by the animal 
several corrections are made, for pos¬ 
sible errors in the readings, for heat 
measured by the apparatus but not 
coming from the animal, and for heat 
withdrawn from the chamber but 
escaping measurement. The follow¬ 
ing is the list of corrections usually 
made: 
(a) Corrections for difference of 
pressure on the bulbs of the two ther¬ 
mometers. 
(i b ) Correction for friction of water 
in absorbers. 
(c) Correction for lag in the rise or 
fall in temperature of the wetter from 
that at the inlet to that at the outlet 
of the absorber system. 
(d) Correction for change in tem¬ 
perature of the absorber system during 
the experiment. 
( e ) Correction for heat developed by 
the blades of the fan used to stir the 
air in the calorimeter. 
(/) Correction for change in tem¬ 
perature of the walls of the chamber 
during the experiment. 
(g) Corrections for heat introduced 
into the apparatus or withdrawn from 
it, in feed, drink, excreta, and vessels 
containing these materials. 
(h) Correction for the metabolism 
of the man entering the chamber dur¬ 
ing the experiment, to milk the cow. 
(i) Correction for condensation of 
water on the absorber system. 
(j) Correction for storage or loss 
of heat due to gain or loss of matter 
by the animal body during the ex¬ 
periment. 
ACCURACY OF THE DIRECT METHOD 
This list of the corrections serves 
to indicate the degree of accuracy 
which it is sought to attain. There 
are, however, two other sources of 
error which have been disregarded 
