. METHODS OF INVESTIGATION. 253 
hours, but it is open to serious question whether such a period i is 
sufficiently long. 
Rubner * has made extensive use of a method substantially the 
same as that just outlined, but differing in details. The computa- 
tion is based upon the total nitrogen and carbon (determined or 
estimated) of urine, feces, and respiration for twenty-four (or 
twenty-two) hours, the feces being regarded as substantially a 
metabolic product. The oxygen consumption is not determined. 
From the results for nitrogen and carbon the proteid and fat meta- 
bolism is computed in the manner explained in Chapter III (p. 78), 
For each gram of carbon in the fat metabolized Rubner reckons 
12.31 Cals. of energy, equivalent to 9.4 Cals. per gram of fat, while 
for each gram of excretory nitrogen (urine and feces) he uses an 
energy value based on previous experiments + in which the actual 
heats of combustion of proteids and the products of their meta- 
bolism were determined. ‘These results will be considered in another 
connection (Chapter X). The resulting values for the evolution 
of energy corresponding to each gram of excretory nitrogen are: 
Fasting (mammals)..................-.. 24.94 Cals. 
i (BITS): na acre orcad een eex 24.35 “ 
Lean meat fed.. ht auiied de meaaaees 20.98) 
Extracted lean medit fos spot acc arata sng 26.66 “ 
These factors were obtained in experiments on dogs and in 
strictness apply only to carnivorous animals. By their use, espe- 
cially if average figures are assumed for some of the minor quanti- 
ties, such as the carbon of the feces and urine, the determination 
of the heat production of a quiescent animal in this indirect way 
becomes a relatively simple matter, while comparisons with direct 
calorimetric results have shown it to be quite accurate. As was 
pointed out on p. 78, however, when carbohydrates: enter largely 
into the diet the results are ambiguous, and this fact as well as 
the marked differences in the character of the excreta forbid its 
application to herbivorous animals. 
Cleavages, H: ydrations, etc.—Both the above methods of comput- 
ing the heat production of an animal assume that the gaseous ex- 
* Zeit. {. Biol., 19, 313; 22, 40; 80, 73. 
+ Ibid., 21, 250 and 337. 
