264 PRINCIPLES OF ANIMAL NUTRITION. 
which may also be given the form 
H=(F-G)-E. 
Rubner determined summarily the value of the quantity F—G 
in the second member of the last equation by the method described 
in Chapter VIII, p. 253, while the actual heat production was deter- 
mined by means of his respiration-calorimeter. | 
The quantities actually determined in these experiments were 
the weight and nitrogen content of feces and urine, the carbon 
dioxide of respiration, and the heat produced. The carbon of feces 
and urine was estimated from their nitrogen and the absence of 
combustible gases in the respiratory products was assumed. From 
the total excretion of nitrogen and carbon the amounts of protein 
and fat metabolized are computed, it being assumed that all the. 
excretory carbon is derived from these two substances. The corre- 
sponding amount of potential energy, equivalent to the expression. 
F —G in the equation above, can readily be computed from the heats 
of combustion of fat and protein. From this the potential energy 
of the excreta must be subtracted, and this Rubner virtually com- 
putes from their total nitrogen on the basis of results obtained in. 
previous experiments with similar food. 
_ A-comparison of the heat production as thus computed with that 
actually measured by means of the calorimeter gave the following: 
results: 
Length Total Heat. 
f E: i- P 
Food. 2 ent Diferoaee 
Days. Computed, Measured, 
Cals. Cals, 
‘ 5 1296.3 1305.1 +0.69 
Fasting evaa vs cae s sane « { 2 1091.2 1056.6 3.15 
Pats ceeded caveat 15 Re a Ns é 1510.1 1495.3 —0.97 
Meat TAPE ce eee ees 2492.4 2488 .0 —0.17 
eat and fat { 12 3985 4 3958.4 | —0.68 
MC AG sb. tcone’s sain edes eed opens { ; 2276.9 +1.20 
7 4780.8 4769.3 —0.24 
Gish sve o penn duc. 45 | 17406.0 | 17349.7 | —0.32 
While some of the individual experiments show not inconsider- 
able discrepancies, the averages of computed and measured heat 
