52 AGRICULTUEAL EXPERIMENT STATION. [Jan. 



meal. In Series III. a, 2.91 pounds of protein (N.R. 

 1 : 4.4) were fed, and in Series III. h, 3.76 pounds of pro- 

 tein were fed, being relatively large amounts of protein with 

 correspondingly narrow rations. 



Strictly speaking, these series, so far as the yield of milk 

 is concerned, can be compared only with themselves, but not 

 with each other, for Series I. a and h was in operation dur- 

 ing the latter part of January and the first half of February, 

 Series II. a and h was completed al)out the end of March, 

 and Series III. a and h about May 1. During this time the 

 cows would naturally shrink some in their milk yield. In 

 order to overcome so far as possible the natural milk shrink- 

 age in comparing the effect of the rations in the three series 

 on the total yield from the herd, the following plan was 

 adopted: take, for example, Series I. a and h. Cows 1, 2 

 and 3 were fed the narrow ration, Series \. a, during the 

 two weeks that cows 4, 5 and 6 received the wide ration. 

 Series I. h ; and then the operation was reversed, cows 1, 2 

 and 3 receiving the wide ration, Series I. &, while cows 4, 

 5 and 6 received the narrow ration, Series I. a. This plan 

 was continued throuo:h the entire series. 



History of Cows. 



The cows were six in numl^er; during the last period 

 Hattie was substituted for Sarah. 



