8 BULLETIN 1281, U. S. DEPARTMENT OF AGRICULTURE 
Live weight. — The cow was weighed just before entering the respira- 
tion calorimeter and immediately after leaving it at the end of the 
experiment. 
Feed and water. — The animals were fed at 6 o'clock morning and 
evening through a lock trap which is an integral part of the apparatus. 
After the animal had finished the morning feed, water was offered in 
a specially constructed basin. The details of these operations have 
been described in previous publications of the Pennsylvania Institute 
of Animal Nutrition. 
Temperature. — The rectal temperature of the animal was taken 
with a clinical thermometer about one hour before the beginning of 
the experiment and also at the end of the experiment. 
Milking. — The cow was milked twice daily, at 4.30 morning and 
evening, by the same person. To accomplish this the main door of 
the apparatus was opened and closed while the milker stepped into 
and out of the apparatus. The presence of the man in the chamber 
necessitates a correction which has been taken into account. 
Error due to opening the respiration chamber. — Since the ventila- 
tion current is drawn through the chamber the loss of gases due to 
opening the door is less than it would be were the air forced through 
or were it stationary at normal pressure. However, since the tem- 
perature of the air in the chamber is usually several degrees centigrade 
lower than the room air, an outward flow of air naturally takes place 
at the opening of the doors. This loss can not be very large under 
these conditions; but, whatever it may be, no correction is applied and 
the error is carried throughout the experiment. 
PREPARATION OF FEED AND FECES SAMPLES FOR ANALYSIS 
A sufficient quantity of hay, grain, or feces samples to contain 
from 1.5 to 2 kilograms of dry matter was weighed out and placed on 
shallow, galvanized-iron pans and allowed to dry in a steam-heated 
drying-closet at from 55° to 65° C. When the samples had become 
thoroughly dry the pans were removed and placed on shelves in the 
grinding room for several days in order that the material might 
come to approximately normal moisture content at the room tempera- 
ture. Just before the grinding, the pans with the samples were 
weighed and the loss in weight termed "loss on air drying." The 
material was then finely ground, being passed through the mill two 
or more times until the desired degree of fineness was obtained. 
This finely ground material was allowed to lie. exposed to the air 
until cooled, then carefully mixed, transferred to glass-stoppered 
bottles, and sealed. 
METHODS OF ANALYSIS 
On the ground feed and feces samples, the usual feeding-stuff 
analyses were made; and the heat of combustion was determined by 
means of a bomb calorimeter. In general, the methods of analysis 
of the Association of Official Agricultural Chemists have been 
followed. 
In the fresh feces "total nitrogen' ' was determined by the Konig 
method. 
In the fresh composite urine the specific gravity, total nitrogen, 
and carbon were determined. For the energy determination about 5 
grams of urine was weighed into a platinum capsule and dried in a 
