80 STORRS AGRICULTURAL EXPERIMENT STATION. 
and butter yield. It is impossible to say exactly how much 
this natural shrinkage in milk would have been. In animals as 
near calving as some of these were the shrinkage would have 
been large; while in the case of cows in ‘‘flush,’’ the decrease 
would have been less marked. ‘The shrinkage in butter yield 
would, of course, be less, because the milk grows richer in fat 
as the period of lactation advances. 
From the summary of the past three winters’ wark it will be 
seen that there was an increase in milk flow in five cases (herds 
A, E, F, H, and I, ) when the animals were fed a narrow ration, 
over that obtained with the wider ration, and in two other 
cases (herds B and C) the yields were essentially the same in 
both tests, although in those instances the narrow ration was 
fed four weeks after the wide. Of the eight herds which were 
fed the wide ration, followed by a narrower one, all except one 
(D) gave an increase in butter yield during the second test. 
The fact that there was more often an increase in butter yield 
(calculated from the butter-fat) than in the milk yield, during 
the period when the narrow rations were fed, would indicate 
an increase in the percentage of fat as a result of using the 
narrow rations. In some instances this was noticeably the 
case. In herd C, with no increase in milk flow, there is quite 
a little gain in butter, while in herd I the contrast is still more 
noticeable. No determinations were made of the percentages 
of the other constituents of the milk. | 
Although a shrinkage in production would naturally follow 
from advancement in period of lactation, the herds as a whole 
more than held their own when changed to the narrower ration 
from one to four weeks after the first test. The results are in 
accord with observation and experiment elsewhere in that so 
far as physiological effects are concerned narrow (nitrogenous) 
rations give larger yields of both milk and butter than do wide 
(carbonaceous) rations. 
COSTS OF THE DIFFERENT RATIONS. 
Omitting herd G and considering only the eight herds which 
_ were fed a narrower ration following a wider one, there are six 
cases where the total cost of producing 100 pounds of milk is 
less with the narrower ration, and six cases where the cost of 
one pound of butter is less. One or more nitrogenous grain 
