380 CARNEGIE INSTITUTION OF WASHINGTON. 



ob\ious shortcoming of this method of procedure that the energy- 

 intake can not be controlled, for when this is large there may be a 

 relatively great protein-sparing effect. Our results scarcely give any 

 justification for the assumption that greatly increased energy intake 

 will explain the lower values at which maintenance was secured with 

 the lactalbumin food. It will be noted further that whereas mainte- 

 nance could be secured very frequently with a lactalbumin content of 

 less than 5 per cent in the diet, this was rarely the case with the other 

 proteins. With all the foods, however, very small rats, naturally 

 eating larger quantities of food per unit of body- weight, often attained 

 maintenance (without growth) on lower protein intakes than sufficed 

 for rats of larger size, e. g., above 125 gi*ams. 



Speaking broadly, in the case of either edestin or casein, it has not 

 been possible to obtain, with a protein intake of 12 to 15 milUgrams 

 per gram of rat, the degree of maintenance which is usually attained by 

 feeding lactalbumin. For edestin and casein a correspondingly efficient 

 intake seems to be about 20 milHgrams in the form of the particular 

 mixtures supplied. It is too early to make thoroughly satisfactory com- 

 parisons between the other proteins except to note that, comparatively, 

 more ghadin was required. That this is not due to defective utiliza- 

 tion in the alimentary tract in the case of any of these proteins has 

 been demonstrated by direct analyses of the feces in many instances 

 where a relatively high protein intake was found inadequate for per- 

 fect maintenance. The mixed proteins of cow's milk, as represented 

 in our milk food, containing casein together with some lactalbumin, 

 approach casein and edestin in value for maintenance, possibly being 

 somewhat more efficient, as exemplified by a very slightly lower main- 

 tenance minimum figure. The higher figures for gUadin are in accord 

 with its known exceptional character in respect to the yield of the 

 various amino-acids characteristic of proteins. 



Most of the food proteins are not so widely divergent in their general 

 amino-acid make-up as to lead us to expect wide differences in the 

 minimum amount of protein required for maintenance. In the case of 

 growth, with its greater need of protein units, the divergences of the 

 proteins may manifest themselves more conspicuously than in mere 

 maintenance. The apparently greater efficienc}'^ of lactalbumin, in 

 contrast with the other proteins recorded above, is in harmony with the 

 observations of the apparent economy of this protein as a supplement 

 to rations containing proteins deficient in one or more amino-acids. 

 Numerous data now available in respect to growth have not yet been 

 evaluated so as to permit exact statements regarding the comparative 

 value of the proteins for this function. 



These experiments, which have required a great deal of time and 

 labor, have now yielded us a mass of data which was necessary before 



