THE FOOD AS A SOURCE OF ENERGY. 335 
would probably be premature to attempt to decide at present. 
Pending further investigation and experience, however, it should 
be remembered that the figures on p. 332 will give, in most cases, 
too high results for the metabolizable energy of mixed rations, while 
the same thing is still more emphatically true of Rubner’s factors. 
One additional point requires mention. In discussing the 
metabolizable energy of protein it was pointed out (p. 320) that 
it is at least a plausible hypothesis that the proteids are metabo- 
lized in the herbivora substantially as in carnivora, and that the 
excess of energy in the urine is derived from the non-nitrogenous 
ingredients of the food. If we accept this hypothesis, however, 
and assume the metabolizable energy of protein (N x 6.25) to 
be in the neighborhood of 4.9 Cals. per gram, then the figures 
for the non-nitrogenous nutrients are subject to a still further 
deduction, especially in the case of coarse fodders. If we were to 
assign to the fat its full value as given, it would not be difficult to 
compute the metabolizable energy of the carbohydrates on this 
basis, and probably a set of factors could be worked out which 
would correspond to the actual results obtained with mixed rations. 
These, however, if successfully obtained, would be substantially 
identical with the results given on previous pages for the apparent 
metabolizable energy of total or of digestible organic matter, and 
it does not appear that the former would offer sufficient advantages 
over the latter to justify the labor involved in their computation. 
