
A STUDY OF RATIONS FED TO MILCH COWS. 65 
suggested. Lhe feed was gradually changed to the suggested 
ration with these three herds, and after four weeks from the close 
of the first test another twelve-days’ test was made with the new 
ration. 
In 1894-95 four herds were studied on the same plan as in the 
longer studies made the previous winter, except that the length 
of time between the two tests, on the same herd, was shortened to 
two weeks, 
[In 1895-96 two herds were studied on the same plan as those 
of the previous winter, except that the time between tests was 
reduced to nine days. In one of these cases the herd was fed a 
very large ration of protein with an unusually narrow nutritive 
ratio. 
In 1896-97 four herds were studied on the same general plan 
as those of the three previous winters, and the time between the 
tests was two weeks. In one of these cases, however, the herd was 
fed in accordance with the milk flow. The average ration was 
very heavy in protein, and had an unusually narrow nutritive 
ratio. 
GENERAL DEDUCTIONS AND CONCLUSIONS. 
(z.) In these forty-five tests on thirty-two distinct herds the 
cost of producing milk and butter depended largely upon the kind 
of cows and their condition as regards time from calving, and the 
kind of ration fed. Many of the individual cows in the tests were 
not returning the cost of feed. The average yield of milk for the 
forty-five separate tests ranged from 13.2 pounds to 23.4 pounds 
per day, while the average yield of butter ranged from seven-tenths 
of a pound to 1.33 pounds per day. This means that the herd 
giving the largest flow of milk was producing 80 per cent. more 
than the one giving the smallest flow, while the herd producing 
the most butter was giving go per cent. more than the one giving 
the smallest yield of butter. Cows calving in the fall are most 
profitable for general dairying, as the price of butter ts about 4o 
per cent. higher during the winter months than during the sum- 
mer months. One of the first things our dairymen need to do ts 
to make a closer study of the individual animals of thetr herds, 
and to reject the unprofitable ones. 
(2.) In these tests the cost of the rations depended largely upon 
the proportion of the cheaper coarse fodders like corn silage, corn 
