182 TWENTY-FIFTH ANNIVERSARY REPORT. 
8. EFfECT»OF AERATING: MILK BY PASSING THROUGH A SEPARATOR. 
Tainted nulk was passed through a separator and the cream and 
skin-miik mixed again before making into cheese. The yield was 
not increased. but .the quality was much improved. Milk of good 
quality: that is thus treated by a separator makes cheese of higher 
quality but lower yield. 
IV. METHODS OF PAYING FOR MILK FOR CHEESE-MAKING. 
In the summer of 1890, the Babcock test first made possible the 
determination of fat in milk by a simple method. While its useful- 
ness was apparent at once as a basis for paying for milk at cream- 
eries, it was not regarded as having any practical application in the 
case of cheese factories, because milk-casein as well as milk-fat 
must be considered. So little was then known of the quantitative 
relation of the constituents of milk to the constituents of cheese 
that no one was in position to deny or disprove the belief universally 
held. By the fall of 1892, we,had accumulated sufficient data in 
our investigations of cheese-making to demonstrate for the first time 
that for all practical purposes the fat alone in milk is a fair basis 
on which to pay for milk at cheese factories, at least much fairer 
than the old method of paying for milk solely by weight. The ap- 
plication of the results of our investigation to this question was 
fully presented in Bulletin No. 68, “ Fat in milk, as a practical basis 
for determining the value of milk for cheese-making,” and Bulletin 
No. 110, “ Milk-fat and cheese yield.” 
The main facts bearing on the question have already been pre- 
sented in the pages preceding and it will be sufficient here merely 
to recapitulate some of the more important statements. 
(1) Milk varies greatly in its composition. In paying for milk 
for cheese-making, absolute fairness can be realized in every indi- 
vidual case only by a careful determination of both fat and casein. 
But no practicable test for both fat and casein is yet known. 
(2) Cheese made from milk rich in fat is greater in yield and 
its constituents, pound for pound of cheese, possess a higher value 
than cheese made from milk poorer in fat. 
(3) When a pound of fat in poorer milk is equivalent to more 
cheese than is a pound of fat in richer milk, the difference can be 
wholly removed by adding skim-milk to, or removing fat from, the 
richer milk. The difference in composition between cheese made 
from poor and rich milk is a skim-milk difference and a skim-milk 
cheese difference. 
