228 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1923 



The protein concentration of the rations was presumably sub- 

 optimal in all cases, so that, except for protein mixtures of good 

 quality, the growth curve of the rats, serving as experimental sub- 

 jects, was submaximal. Thus, the extent of retardation of growth 

 was taken as an index to the extent to which the quality of any 

 mixture of proteins fell below that of the best combinations previ- 

 ously discovered. This information was supplemented by observa- 

 tions of the animals throughout their reproductive period relative 

 to fertility, infant mortality, and the appearance of senility, as 

 well as observations of successive generations on the same diet. These 

 observations on the maternal functions figured largely in the final 

 judgment of the relative values of the proteins studied. 



According to these experiments, the foods studied arrange them- 

 selves in decreasing order of the biological value of their proteins 

 according to the following scheme: (1) Beef kidney; (2) wheat; 

 (3) milk, beef liver; (4) beef muscle, barley, rye; (5) corn, oats; 

 (6) soy beans, navy beans, peas. 



It will be observed that the proteins of wheat are assigned a higher 

 value than the proteins of milk or of meat, while the proteins of 

 barley and rye are placed on a par with the proteins of meat. These 

 relations are quite contrary to those indicated by other less compre- 

 hensive methods of research. One naturally raises the question 

 whether the method, used by McCollum is entirely equivalent to the 

 methods previously used. To us there is little doubt but that 

 simple growth experiments and nitrogen balance studies, conducted 

 under proper conditions, can give reliable information of the 

 chemical adequacy of proteins for the purposes of maintenance and 

 of growth. To account for failures in the proper performance of 

 the maternal functions, and for nutritive failures in the second and 

 third generations, by reference solely to the source of protein in 

 the diet, no matter how complete the diet may seem to be in other 

 factors, is, we believe, equivalent to assuming that our present knowl- 

 edge of nutritive requirements is complete. I doubt whether such 

 an assumption is entirely justified. The more favorable results ob- 

 tained with the wheat ration than with the milk or meat rations, 

 may be related to the amount of food consumed rather than to its 

 composition. 



The measurements of the biological values of proteins may be 

 combined with the measurements of protein digestibility to give 

 what may be called the " net " protein content of a food. For ex- 

 ample, if a cut of beef contains 20 per cent of protein of which 95 

 per cent is digestible, it would contain 19 per cent of digestible pro- 

 tein. Now if only 80 per cent of the digestible protein could be 



