ILLINOIS STATE DAIRYMEN'S ASSOCIATION. ^5 



composite test. Of the 35 cows tested in this way, not one differed more 

 than one-tenth of one par centi between the composite test and the per 

 cent of fat, as shown by the several milkings. And in no case was there 

 a difference of more than one-fourth of a pound of fat for the week's 

 work; several agreed exactly. 



In further support of the reliability of the Babcock test and also of 

 the composite method. I would say that of the 1300 official tests made in. 

 connection with this work (each one of which was run in duplicate) only 

 seven of these pairs differed more than .1 per cent. Where a difference 

 of more than this occurred, the test was repeated. 



VARIATIONS OF TESTS. 



Creamery patrons are apt to demand an explanation of any sudden 

 change in the test of their milk at the creamery. Some of them seem to 

 think that the man who operates the test should be able to tell why the 

 difference occurs ,and I believe some buttermakers, rather than plead 

 ignorance, sometimes offer explanations which may seem to him to 

 answer the purpose for the time being, but which fails to apply at some 

 other time. 



The man who feeds and milks the cow is really in better position to 

 give intelligent reasons for? variations in the test, but I see no reason 

 why either one should be ashamed to own that they do not know, as it 

 is pretty certain that there are a great many more things still to be learned 

 on the subject than have been learned up to date. 



A few things which affect the richness of milk of the individual cow 

 from day to day or from week to week are pretty well established, but 

 since some of these variations of individuals are equalized in the herd, 

 we have fewer reasons for great variations in herd milk. The larger the 

 herd the more uniform should be the test from week to week, ether 

 things being equal. 



One of the Jersey cows at the Pan-American jumped around in test 

 for successive milkings as follows; 5.6, 4.9, 3.7, 5.6, 5.2, 2.5, 4.7, 3.8, 2.9 



