434 BlgcMihility of Palm Kernel Cake, etc. 



digestion coefficients must inevitably be affected with error tending 

 to give low results. The composition of these faecal products of meta- 

 bolic origin is such that the error falls with special weight upon the 

 protein and "ether extract." Repeated attempts have been made to 

 devise a satisfactory method of discriminating between the metabolic 

 products and the undigested food residues in the faeces with the result 

 that, although it remains impossible to make any reliable correction 

 of the "ether extract" figures, it is possible to make the necessary 

 correction in the case of the protein, with apparently a fair degree of 

 accuracy. This point is dealt with more fully in a later section of the 

 paper and for the present it will suffice to indicate that according to 

 the investigations of Stiitzer, Kiihn, Kellner and others the digestibihty 

 of the protein of foods can be more accurately measured either by direct 

 treatment of the foodstuff rii vilro with an acid (HCl) solution of pepsin, 

 or by subjecting the faeces obtained in the animal experiment to similar 

 treatment. In the latter case it is assumed that the pepsin brings into 

 solution the "metabolic protein," leaving the undigested food-protein 

 intact. Recent investigations by Morgen(5) at Hohenheim indicate that 

 the pepsin treatment alone removes only about two-thirds of the 

 "metabolic protein" and that complete removal could only be effected 

 by a supplementary digestion of the residue with trypsin. Until further 

 confirmation of these results is obtained we prefer to retain the simple 

 pepsin digestion as our basis in correction of the protein data, and the 

 results so obtained are included in our tables. 



German investigations have revealed a fairly close proportionality 

 between the amount of organic matter digested and the amount of 

 "metabolic protein," as measured by pepsin solubility, in the faeces. 

 Thus Kellner(ii) found 0-3 to 0-5 gm. (mean, 0-4 gm.) of pepsin-soluble 

 nitrogen to be present in the faeces per 100 gms. organic matter digested. • 

 Pfeifier(7) obtained a similar average in experiments with pigs but 

 a higher figure with sheep, viz., 0-37 — 0-76 gm. N (mean, 0-51). Later 

 investigations have given variable results, e.g. Katayama(8) found 

 0-6 — 0-78 gm., and Morgen(9) an average of 0-57 gm. The latter found 

 the proportion to vary with the nature of the food, increasing to 0-82 

 with coarse fodder and even to 1-20 in the case of special rations con- 

 taining aqueous extracts of plants. In his most recent experiments (5) 

 he obtained an average of 0-64 gm. pepsin-soluble faecal nitrogen per 

 100 gms. digested organic matter, and adding one-third to this for 

 "metabolic protein" not dissolved by pepsin arrives at the con- 

 clusion that the most probable correction factor for the digestion of 



