80 EXPERIMENT STATION. [Jan. 



Each analytical determination in the above table was from a 

 five-day composite, taken as previously described. The total 

 solids were determined by drying on sand ; the fat by extracting 

 the sand residue with water-free ether, and also by the Bab- 

 cock method; the ash by evaporating the milk in a platinum 

 dish with nitric acid and burning the residue at a low red heat ; 

 the nitrogen by the Kjeldahl-Gunning method; the lactose with 

 the aid of Eehling's solution and the Low " zinc-acetate " 

 method. 



In the second and third periods the partial composition of 

 the milk was ascertained during the preliminary period in 

 order to note any immediate change resulting from the feeding 

 of soy bean meal or soy bean oil. 



In the first period Herd I. produced milk with a slightly 

 higher percentage of total solid matter than did Herd 11. ; this 

 evidently was due to the higher fat percentage in the former 

 milk (5.38 against 5.14). The percentage of ash, nitrogen and 

 lactose were nearly the same in each case. It may be said 

 that the chemical composition of the milk from each of the 

 two herds of 6 cows each, when fed the same ration, was sub- 

 stantially alike. 



The addition of the soy bean meal (average of 2.3 pounds 

 per cow daily in place of a like amount of normal grain ra- 

 tion) did not appear to have any noticeable effect upon the 

 total solids or fat percentages of the milk produced by Herd 

 II. during the preliminary period of two weeks. During the 

 period proper, lasting four weeks, the average percentages of 

 fat and solids not fat in case of both herds showed only slight 

 differences from those in the first period,^ and the same may be 

 said of the lactose percentages. The ash and nitrogen per- 

 centages of the milk yielded by Herd I. were quite similar to 

 those in the first period, while with Herd 11. the ash showed a 

 slight decrease and the nitrogen a very slight increase. One, 

 however, would not consider it wise to attribute these variations 

 directly to the modification of the grain ration. 



1 During the week of February 16 to 21 tlie fat iicrccntapcs as determined by the gravi- 

 metric method increased perceptibly, a change which cannot be accounted for. In most 

 cases the gravimetric method gave lower results than the Babcock, while in this instance 

 the reverse was the case. 



