NORMAL METABOLISM. 97 



"units" or "building stones" of protein molecules, but in no 

 two proteins are the materials used exactly in the same propor- 

 tions, some proteins having a preponderance of one or more and 

 an absence of others, just as in a row of houses there may be no 

 two that are exactly alike, although for all of them the same 

 building materials were available. Albumin and globulin are 

 the most important proteins of blood and tissues, so that the 

 food must contain the necessary units for their construction. If 

 it fails in this regard, even to the extent of lacking only one of 

 them, the organism will either be unable to construct that pro- 

 tein, and will therefore suffer from partial starvation, or it will 

 have to construct for itself this missing unit, a process which it 

 can accomplish for some but not all of the above list. 



It is therefore apparent that those proteins are most valu- 

 able as foods that contain an array of units which can be reunited 

 to form all the varieties of protein entering into the structure 

 of the body proteins. Naturally, the protein which most nearly 

 meets the requirement is meat protein, so that we are not sur- 

 prised to find that less of it than of any other protein has to be 

 taken to gain nitrogen equilibrium. Casein, the protein of milk; 

 although it does not contain one of the most important units, 

 namely, glycin, is almost as good as meat protein, because the 

 organism is itself able to manufacture glycin. When, on the 

 contrary, proteins such as zein from corn are given, in which cer- 

 tain units are missing, starvation inevitably ensues. But it does 

 not do so if the missing unit, which in the case of zein is trypto- 

 phan, is added to the diet. 



These most important facts have been ascertained by experi- 

 ments carried out in New Haven by Osborne and Mendel. 

 Young albino rats, just weaned, were fed on a basal diet con- 

 sisting of the sugar, fat and salts of milk to which was added 

 the protein whose nutrition value it was desired to study. The 

 rats were weighed from day to day, and the results plotted 

 as a curve the curve of growth. A gradually rising curve 

 was obtained when casein or the albumin of milk or eggs, or 

 the edestin of hemp seed, or the glutenin of wheat was fed, 

 but this was not the case with the gliadin of wheat or, as 



