218 ANNUAL REPORTS OF DEPARTMENT OF AGRICULTURE. 



in body weight than when kept in stanchions. The value of the 

 increased product is less than the additional expense of bedding and 

 labor for keeping in box stalls, and this method is not to be advised 

 except where maximum production is desired regardless of expense. 



Twelve cows in alternating periods were milked so as to test the 

 effect of changing milkers. In the first and third periods each cow 

 was always milked by the same man ; in the second period never by 

 the same man twice in succession. When the same man did the 

 milking the average yield showed a very slight increase, 0.6 per cent 

 of milk and 0.4 per cent of butterfat. 



Feeding 3 pounds a day of grain above the Savage feeding stand- 

 ard caused a daily increase of 1.6 pounds of milk, 0.09 pound of 

 butterfat, and 0.32 pound of body weight. 



Cows fed a small quantity of roughage and a liberal quantity of 

 grain produced and gained more than those fed scantily with grain 

 and liberally with roughage. Since both groups were fed in accord- 

 ance with the Savage standard, it is evident that the digestible 

 nutrients in concentrates are worth more than an equal weight in 

 roughage. 



Cows were fed a definite small quantity of hay, followed by all the 

 silage they would eat, with grain sufficient to bring the total nutrients 

 to standard, and the quantity of silage taken was noted. Later they 

 were fed a larger quantity of hay, followed by their fill of silage as 

 before, and it was found that for every additional pound of hay they 

 ate almost exactly 1 pound less of silage, the total of roughage re- 

 maining the same. Since hay contains more nutrients than silage, 

 using the larger ration of hay reduces the amount of grain needed. 

 The quantity of total roughage consumed seems to be dependent on 

 individual size and nature rather than on the quantity of milk pro- 

 duced. 



Hydrolyzed sawdust, a by-product in the manufacture of indus- 

 trial alcohol, was found by experiment to contain a large proportion 

 of material that has no feeding value. This chemically treated saw- 

 dust can be used in limited quantities for dairy cows, but it possesses 

 so little nutritive value that its use is inadvisable at present prices 

 of feed and cost of treating sawdust. 



Experiments to find out whether magnesium is the element in 

 prickly pear which lowers the fat content of milk indicate that mag- 

 nesium has little or no influence in this respect. 



Tests of butterfat in the milk of all cows at the farm, made in 

 January and July, were studied. There are 166 cows with at least 

 one test in each of these months. The lowering of the fat in summer 

 as compared with winter varies from 0.02 to 0.72 per cent. 



Experiments are in progress : To ascertain the influence of feeding 

 a large quantity of hay and a small quantity of silage compared with 

 reversed quantities of these feeds; to gain information by which to 

 revise the Wolff-Lehmann standards for growing dairy 'animals ; to 

 see how much extra feed first-calf heifers need in order to make nor- 

 mal growth while giving milk; and to determine whether the fre- 

 quency of feeding hay affects the quantity consumed and the milk 

 production. 



DAIRY CATTLE BREEDING INVESTIGATIONS. 



The registered purebred dairy cattle in the five herds where breed- 

 ing projects are under way now number 366. Official records have 



