METHODS OF INVESTIGATION. 255 
of its tissues and fluids, and that even a comparatively small varia- 
tion in the latter calls into action compensatory processes. A 
striking illustration of this is seen in the promptness with which the 
respiratory and vascular mechanism reacts to the changes produced 
in the blood by muscular activity (compare Chapter VI). It seems 
improbable, therefore, that any sufficient accumulation of the in- 
termediate products of metabolism can take place to seriously in- 
fluence the results of any but very short experiments. That the 
methods employed involve other sources of error has already ap- 
peared, but with due allowance for these it would appear that the 
results are worthy of a large degree of confidence. 
Computation from Carbon and Nitrogen Balance.—The method. 
of computing the heat production from the total excreta, as em- 
ployed by Rubner and others for carnivorous animals, we have seen 
to be inapplicable to herbivora. It, however, shades naturally into 
a third method, of genera] applicability, which consists in combining 
with a determination of the carbon and nitrogen balance by means 
of the respiration apparatus direct determinations of the potential 
energy of the food and of the visible excreta by the methods already 
indicated. Kellner has made extensive use of this method, and the 
following example, taken from his earliest. investigations,* will 
serve to show clearly the nature of the method. The ox experi- 
mented upon was fed daily 8.5 kgs of meadow hay. Respiration 
experiments showed that on this ration there was a daily gain by 
the animal of 6.2 grams of nitrogen and 127.2 grams of carbon, 
equivalent to 37.2 grams of protein and 140.8 grams of fat, the 
potential energy of which can be computed from the data on p. 244. 
From determinations of the heats of combustion of food, feces, 
and urine, assuming the combustible gases excreted to consist only 
of methane, the balance of energy is computed as in the table on 
p. 256. 
Having included under the head of outgo all the known forms 
in which potential energy as such may be disposed of, the balance 
of 14,819.5 Cals. is regarded as having been liberated as kinetic 
energy, and, since no external work was performed, to have taken 
finally the form of heat. Short of an actual calorimetric experi- 
* Landw. Vers. Stat., 47, 275. 
+ The figures are the corrected ones given in Landw. Vers, Stat., 53, 9. 
