INVESTIGATIONS ON THE METABOLISM OF MILCH COWS. 811 



ainids, alkaloids, and other chemical bodies of little importance in 

 nutrition. The nitrogen-free material includes fat, soluble carbohy- 

 drates, and crude fiber. The methods employed in determining these 

 substances are well known. Those employed by the writer are de- 

 scribed at length by Konig in his book on agricultural and commercial 

 analysis. 



The estimation of water is of especial importance. If the material 

 is dried in the air at 105 to 110° C, the results may be too high, owing 

 to the expulsion of too much water and in some cases to an oxida 

 tion of the fat. The water driven off under these conditions may be 

 of 4 kinds: (1) The hygroscopic water on the surface of the substance 

 and in the capillary spaces or the residue from the former cell fluid, 

 (2) water which held the same relation to the separate constituents of 

 the cell as the water of crystallization of an inorganic salt, (3) water 

 formed at this temperature by the union of oxygen and hydrogen 

 within the organic molecules, and (4) water formed by the action of the 

 oxygen of the air on the hydrogen of the substance. It is possible 

 also that some of the carbon may be oxidized to carbon dioxid. 



The term "dry matter" can properly be applied only to substances 

 from which water included under 1 and 2 has been removed. It has been 

 the writer's practice to dry all materials in a current of dry illuminating 

 gas. Drying need not be continued longer than 21 hours at 70° to 

 attain constant weight, provided a sufficient quantity of gas can be 

 drawn through the apparatus without increasing the pressure. JMauy 

 substances — oil cakes, for instance — lose material other than water when 

 heated to 70°. In such cases the drying must be continued for more 

 than 24 hours at from 40 to 50°. The receptacle in which the substance 

 to be dried is weighed consists of 2 parts which fit together very closely, 

 so that moisture is not absorbed during weighing. It is a matter of 

 the utmost importance in metabolism experiments to estimate the dry 

 matter in food, urine, feces, milk, etc., by a uniform method. Urine and 

 milk may easily be dried in 3 days over a sufficient quantity of sulphuric 

 acid (100 times as much as the dry matter), if mixed with kaolin, 

 puniiee stone, or sand. 



Sampling for analysis is a matter of great importance. This is com- 

 paratively simple with milk, urine, and ground concentrated feeding 

 stuffs, such as cotton-seed meal or ground peanut cake. It is more 

 difficult in the case of feces and such materials as whole or coarsely 

 ground grains. Two samples at least should be taken of such materials. 

 Sampling is most difficult in the case of coarse fodders, and it is advis- 

 able to take 3 samples. In sampling hay, straw, etc., the material may 

 be divided into bundles, and. each bundle passed through a cutting 

 machine, like portions being retained from each bundle. These portions 

 should be thoroughly mixed and samples for analysis taken from the 

 mixture. 



Suppose 6 metabolism experiments are to be made with 2 cows. 

 Each experiment requires a mouth. The 2 cows in G months would 

 17950— No. 9 2 



