55 



portion of protein from animal food in the. diet of this subject was 

 larger than in tliat of either of the others, as shown by the data in 

 Table 28. 



The coefficients of digestibility of protein, Avhen corrections are 

 made for metabolic products in the feces, are of course larger than 

 those found by deducting the total feces from the total food consumed, 

 as shown by the data in Table 25 (page 52). It should be observed 

 that the corrections for metabolic products obtained by the two 

 methods followed differ somewhat, the digestibilit}^ being higher when 

 corrections are introduced in accordance with results obtained by the 

 IDcpsin method. This is jDerhaps accounted for b}^ the fact that treat- 

 ing tlie feces with pepsin would undoubtedly remove protein of undi- 

 gested food material as well as metabolic products, while treatment 

 Avith solvents (ether, alcohol, and limewater) would not affect the 

 protein of undigested residue. However, too little information is 

 available at present for a satisfactory discussion of this i)oint. 



SOURCES OF THE NUTRIENTS. 



Since numerous investigations have shown that animal foods are 

 more completely digested than those of vegetable origin, it was 

 thought interesting in these experiments with a man on a mixed diet 

 to determine as nearly as possible the sources of tlie different nutri- 

 ents — that is, the proportion derived from animal and vegetable 

 foods. Several of the food materials were of a mixed cliaracter, so a 

 somewhat arbitrary classification was necessary ; thus, in the case of 

 codfish hash and vegetable soup it was assumed that all the carbo- 

 hydrates and a part of the protein and fat were furnished by the 

 potatoes used in making them, the amount of protein and fat being 

 calculated from the average values for these nutrients in boiled pota- 

 toes. The rest of the protein and fat was assumed to be of animal 

 origin. Similarl}^ in the case of mince pie, two-thirds of the jjrotein 

 and fat were considered as of animal origin, and the remaining pro- 

 tein and fat and all the carbohj^d rates were assumed to be of vege- 

 table origin. The larger part of the fat of baked beans Avas not 

 derived from the beans themselves, but from the pork added. This 

 amounted on an average to nearly IG per cent of the total fat of the 

 food, and in the accompanj'ing table is classed with the animal fats. 

 While this apijortionment is only a rough approximation it seemed to 

 be the best that could be devised and is sufficiently accurate for the 

 purpose. The data regarding the source of the different nutrients are 

 summarized in the following table : 



