4io 
Journal of Agricultural Research 
Vo!. IX, No. xa 
CONCLUSIONS 
The apparent digestibility of the protein of com, based on the total 
nitrogen of the feces is about 75 per cent. On account of the existence 
in the feces of nitrogen of metabolic origin we know that the real digesti¬ 
bility is higher. The acid-pepsin method makes it appear that the real 
digestibility of the protein of corn is about 92 per cent, and the pepsin- 
pancreatin method about 96 per cent. Jordan's method gives appreciably 
lower figures, averaging 86 per cent. 
The acid-pepsin method indicates that 70 per cent, the pepsin- 
pancreatin method 84 per cent, and the Jordan method 46 per cent of the 
nitrogen of the feces from corn is of metabolic origin. 
All of the methods make the nitrogen of blood albumen appear more 
than completely digestible, even the apparent digestibility being over 
100 per cent; thus, the feeding of blood albumen with corn seems to 
increase the digestibility of the com protein to an extent more than 
sufficient to offset the incompleteness of digestibility of the protein of 
this supplement. 
With skim milk the apparent digestibility varies from 95.97 to 104.44 
per cent, the average being 99.15. With the acid-pepsin method three 
out of the five figures average 99.35. In previous work 1 five estimations 
by this method averaged 99.12. With the pepsin-pancreatin method the 
results were lower than with the acid-pepsin method. These low results 
on the supplementary food are reciprocals of the high results on the basal 
ration of corn. 
The proteins of skim milk are made to appear more nearly completely 
digestible by the acid pepsin method than by the pepsin-pancreatin 
method or by the Jordan method. 
With egg albumen the results varied considerably, but all were high. 
It would appear that raw, commercial, dried egg albumen is almost 
perfectly digested by swine. 
Important inaccuracy seems to be inevitable in any determination of 
digestibility of supplementary foods in the usual way, by difference; and 
no T>ther method seems more satisfactory. This applies equally to 
computations of real digestibility, and of apparent digestibility (based 
on total nitrogen of the feces). 
The digestion coefficients for protein involved in the feeding standards 
of our reference works on animal production assume that the nitrogen of 
the feces is entirely an indigestible food residue. The rough measures 
afforded by the results of this study indicate that, as applying to the 
digestive capacities of swine, this assumption underestimates the diges¬ 
tibility of protein by about 20 per cent. 
By way of interpretation of the individual variations in the digestion 
coefficients we would record the fact that pig 1, in Period II, manifested 
Forbes, E. B., Beetle, F. M., and others. Op. cit. 
