68 PRINCIPLES OF ANIMAL NUTRITION. 
nitrogenous organic matter of muscular tissue has a materially 
higher percentage of nitrogen, viz., about 16.67 per cent. This 
would reduce the factor for protein from 6.25 to 6.00. Kéhler’s 
samples, after extraction with ether for 480 hours, still contained 
from 0.27 to 1.61 per cent. of fat. If we assume the ash and 
fat-free substance of Lawes & Gilbert’s experiments (p. 65) to 
have still contained 1 per cent. of fat, the average nitrogen con- 
tent of the fat-free substances would be 16.38 per cent. and the 
corresponding protein factor 6.11, while the factor 6.00 would re- 
quire the assumption of a fat-content of 2.7 per cent. 
The factor 6.0 has been used by Kellner in computing the results 
of his extensive investigations upon cattle at Méckern. Strictly 
speaking, this assumes that®all the gain of nitrogen takes place 
either in the form of muscular tissue or of material.of the same 
average composition. To what, extent such an assumption is 
justified it is difficult to say. Certainly a part of the protein of the 
food is applied to the production of epidermis, hair, horns, hoofs, 
etc., consisting largely of keratins. The data regarding the com- 
position of these tissues given on p. 63 would seem to show 
that they are, on the average, richer in nitrogen than muscular 
tissue, a fact which would tend to lower the protein factor, but on 
the other hand, the amount of this growth is small as compared 
with the usual protein supply. On the whole, Kohler’s factor 
would seem to afford the most trustworthy basis of computation 
which is at present available, especially in view of its close agree- 
ment with Lawes & Gilbert’s results. 
Urnza aS A Measure or Proterp Mretasorism.—lIn the earlier 
investigations upon this subject, the urea of the urine, as deter- 
mined by Liebig’s titration method, was commonly taken as the 
measure of proteid metabolism, one part of urea equaling 2.9 parts 
of protein, while in many cases the. metabolism was also expressed 
in terms of “flesh” (muscular tissue) with its normal water con- 
tent and an average of 3.4 per cent. of nitrogen. The errors inci- 
dent to the use of this method are now generally recognized, while 
its inapplicability to herbivora was obvious from the first, and with 
the improvements in the methods of nitrogen determination, the 
latter has almost entirely replaced the old urea determination and 
