JOURNAL OF AMETItAL RESEARCH 
Vol. XXX Washington, D. C., March 1 , 1925 No. 5 
A COMPARISON OF DIRECT AND INDIRECT CALORIM¬ 
ETRY IN INVESTIGATIONS WITH CATTLE 1 
By Max Kriss 
Associate in Animal Nitrition, Institute of Animal Nutrition of Pennsylvania. 
State College 
INTRODUCTION 
From the time of the construction 
of the first respiration calorimeter by 
Lavoisier (9, p. 355) 2 until Rubner 
(10 p. 135) demonstrated the appli¬ 
cability of the law of conservation of 
energy to the vital processes of the 
animal body, the interest of experi¬ 
menters with animal calorimeters 
centered mainly in the question as to 
whether the products of respiration 
fully represent the production of heat 
by the animal. 
With the establishment of the 
validity of this conception began a 
new era in respiration calorimetry, 
which has been marked by the per¬ 
fecting of apparatus and the standard¬ 
ization of methods. Since, however, 
the respiration calorimeter is a compli¬ 
cated and costly apparatus, both in 
construction and use, methods have 
been evolved for the indirect deter¬ 
mination of the heat production, by 
the use of a respiration apparatus, 
without the calorimetric features. 
For the purpose of indirect heat 
estimation there are now in use two 
principal methods, with modifications 
of detail, to meet special conditions 
as to kind of animal, length of experi¬ 
ment, equipment available, accuracy 
sought, and the particular problem 
in hand. 
Briefly and generally, one of these 
methods (2) involves the estimation 
of the Amounts of protein, carbohy¬ 
drates, and fat oxidized in the body, 
and the computation of the heat 
production from the calorific values 
of these nutrients. 
The amounts katabolized are com¬ 
puted from the oxygen consumed and 
carbon dioxide produced, and the nitro¬ 
gen, carbon, oxygen, and hydrogen of 
the urine, these computations being 
based on the respiratory quotients (pro¬ 
portions of carbon dioxide produced to 
oxygen consumed) for the nutrients 
oxidized, these quotients being approxi¬ 
mately 0.7 for fat, 0.8 for protein and 
1.0 for carbohydrates. 
The other method depends on the 
balances of nitrogen and carbon in 
the animal body, and the determination 
of the potential energy of the feed and 
excreta. By this method the heat 
production is measured by deducting 
from the gross energy of the feed the 
potential energy of the excreta and 
the energy equivalent of the body 
tissue gained (or adding the energy 
equivalent of tissue lost). 
The more fundamental consider¬ 
ations as to energy metabolism having 
been established, a question of much 
importance is the comparative accuracy 
of the expensive and time-consuming 
direct-heat estimation by means of the 
respiration calorimeter and the rela¬ 
tively simple indirect-heat estimation 
by means of the two general procedures 
outlined. 
As a contribution to the under¬ 
standing of this problem the writer 
presents the following consideration 
of the possibilities of error in each, 
and the extent to which these errors 
affect results, the comparisons being 
based on the experimental work of 
this institute. 
The respiration calorimeter at the 
Institute of Animal Nutrition, which 
has been used for more than 20 years 
in metabolism experiments with cattle, 
is, as the name implies, both a calori¬ 
meter and a respiration apparatus. 
In each experiment heat production 
has been determined directly, and also, 
for comparison, by computation from 
the balance of matter and energy, as 
in the second of the indirect methods 
referred to above. This method is 
essentially the same as that used 
by Kellner ( 7 ) in his extensive re¬ 
searches on cattle. 
1 Received for publication June 11,1924; issued May, 1925. This study was undertaken at the suggestion 
of Dr. E. B. Forbes, to whom the writer is especially indebted for kindly criticism and assistance in revising 
the manuscript. He is also under obligations to the other members of the staff of the Institute for many 
valuable suggestions in connection with this work. 
2 Reference is made by number (italic) to “Literature cited, ” p. 406. 
Journal of Agricultural Research, 
Washington, D. C. 
19974—25f-1 
Vol. XXX, No. 5. 
Mar. 1,1925. 
Key No. Pa.-13. 
(393) 
