METHODS OF INVESTIGATION. 241 
viz., a determination of the heat of combustion. An element of 
uncertainty, however, which is ordinarily not met with in the case 
of the food, arises from the ready decomposability of the excretory 
products, which is liabie to result in a loss of energy during the 
drying necessary to prepare them for combustion. The urea of 
the urine, in particular, is very readily converted into the volatile 
ammonium carbonate. Comparative determinations of nitrogen 
in the fresh and in the dried urine will show the amount of nitrogen 
lost in drying, and on the assumption that only urea is decomposed 
the loss of energy can be readily computed from the known heat of 
combustion of that substance. Atwater & Benedict * have found 
this assumption to be substantially correct for human urine, and 
the same may he presumed to be the case with the urine of carniv- 
ora. It has usually been assumed to be applicable also to the 
more complex urine of herbivora, although without, so far as the 
writer is aware, any experimental proof. : 
A greater or less loss of nitrogen has also been observed in the 
drying of the feces of domestic animals, particularly of horses and 
sheep, but the nature of the material decomposed has not yet been 
investigated, and the same is true of the possible decomposition of 
non-nitrogenous materials in both urine and feces. Atwater & 
Benedict (loc. cit.) found the loss of nitrogen from human feces to 
be insignificant. 
ComputTaTION or ENERGY.—The computation of the energy of 
the visible excreta is much less satisfactory than in the case of the 
food on account of our inferior knowledge of the proportions and 
chemical nature of their ingredients. 
The Urine.—Formerly the urine was assumed to be substan- 
tially an aqueous solution of urea, and numerous computations of 
its energy content were made on this basis, particularly in connec- 
tion with estimates of the metabolizable energy of the proteids, 
while the same method has been applied also in estimating the 
metabolizable energy of feeding-stuffs. Rubner + was the first to 
demonstrate the serious nature of the error involved in this assump- 
tion and to show that the energy of the urine is materially greater 
than the amount thus computed. In the urine of the dog he found 
* U.S. Dept. Agr., Office of Experiment Stations, Bull. 69, p. 22. 
+ Zeit. f. Biol., 20, 265; 21, 250 and 337; 42, 302. Compare Chapter X. 
