320 THE FEEDING OF ANIMALS 



cow's milk, excluding samples of imusual character, 

 according to a compilation by Van Slyke of 5,652 Ameri- 

 can analyses is as follows: 



The variations in the composition of cow's milk are 

 large, the proportion of water ranging, mider perfectly 

 normal conditions, from 84 to 89 per cent, with occa- 

 sional analyses entirely outside these limits. The chief 

 known causes of such variations are breed, individuality, 

 period of lactation, and nervous disturbances. There 

 are material daily fluctuations as well, for which no 

 reasons can now be assigned. These changes are mostly 

 in the proportions of water and total solids, for the com- 

 position of the solids, that is, the relative proportion of 

 proteins, fats, and sugar, is remarkably constant with 

 the same animal. The effect of breed in cows is illustrated 

 by averages shown in Par. 362. These variations and 

 those due to other causes are important in considering 

 the relation of milk formation to nutrition, because the 

 food expense of milk is determined, other things being 

 equal, not by the volume but by the milk solids elabo- 

 rated, for which reason the draft upon the supply of 

 nutrients, water excepted, is greater for the secretion of 

 100 quarts of Jersey milk than for the same quantity 

 of Holstein milk. In studying the economy of milk pro- 

 duction, therefore, we should consider the relation of 

 food to milk solids and not to milk volume. 



419. Milk secretion. — ^There is no milk in an animal's 

 food, that is to say, hay and grain contain no casein, 

 butter-fat, or milk-sugar. They do contain nutrients, 

 which, when subjected to the vital processes of the animal. 



