1911.1 



PUBLIC DOCUMENT — No. 31. 



113 



The total pi'oteiu digested was calculated from the amount 

 digested daily multiplied by the number of days of the experi- 

 ment. The protein for nuiintenance was calculated from the 

 average weight of each herd, allowing .7 of a pound of digesti- 

 ble protein per 1,000 pounds live weight. The protein in the 

 milk was calculated from the actual analysis of the milk. It is 

 admitted that the above results are only approximate, being 

 secured partly from average figures, and on the basis of crude 

 in place of true protein. They indicate, however, that Herd D 

 was receiving a ration rather deficient in protein, and that Herd 

 E was receiving at least 38.4 per cent, over that required for 

 maintenance and milk. 



Milk Yield and Milk Shrinkage. 



In spite of the fact that the three cows comprising Herd 

 D received hardly sufficient protein for maintenance and milk 

 produced, they did not shrink as much during a period of 

 twenty-four weeks as did the three cows in Herd E, which re- 

 ceived substantially 38 per cent, protein in excess of supposed 

 requirciiicnts. Such a result can only be explained on the 

 ground that the animals were too few in number to give accu- 

 rate results by the grouj) method, and that individuality rather 

 than food appeared to be the controlling factor. See also Exper- 

 iment VIII. 



Experiment VIII. — 1908-09. 



This experiment was planned primarily to study the protein 

 requirements of dairy animals. It will not show the effect of 

 protein upon the chemical composition of the milk. 



Plan of the Experiment. — Inasmuch as the cows in the herd 

 calved at different times, the experiment was planned with 

 pairs of cows, i.e., each pair of cows, when ready, was started, 



