( 36 ILLINOIS STATE DAIRYMEN'S ASSOCIATION. 



and 3.5, but the same cow's weekly composite tests ran fairly even for 

 successive weeks, thus: 3.95, 4.0,4.1, 3.7, 4.1, 4.4, 4.2, 4.2 and 4.0. 



In like manner the variations of individual cows from week to week 

 are partially obliterated in the test of the mixed milk of the herd. Some 

 of the greatest variations in herd tests for successive weeks ai the Pan- 

 American were as follows: Jerseys beginning with the week of August 

 oth, and for succeedings weeks was 4.32, 4. SO, 4.40. The French Cana- 

 rian July 16th and succeeding weeks, 3.59, 4.03, 3.87, 4.09. The Guernseys 

 for weeks of July 30 and August 6 went 4.42 and 4.72. 



I believe the Babcock test has been repudiated by creamery patrons 

 sometimes for showing variations from week to week no greater than 

 the above, just because they knew rot reason why the milk should vary 

 at all. But worse than this, and some times a direct result of kicking on 

 the test, is the practice of evening up the tests by a dishonest butter- 

 maker, who prefers to remove any teeming inconsistencies in the test be- 

 fore giving them out. I regret that such things as this have been known 

 to occur, and am sure that such practices as this have in one way and 

 another resulted in much lack of faith in the test even when honestly and 

 intelligently handled. 



When I find the test of a creamery running uniform from week to 

 week with a range of only .4 to .6 per cent among a large number of 

 patrons, I am suspicious that, all may not be right. 



It may be of interest to know that with every breed tested for con- 

 secutive milkings at the Pan-American, nearly every cow tested lowest 

 in the morning. Five of the breeds tested in this way (25 cows) were 

 milked three times a day, at 5:00 a. m., 12:30 p. m., and 8:00 p. m., mak- 

 ing the night period 9 hours and the other two 7% hours each. With 

 these cows the morning's- milk tested lowest, the noon milk highest, and 

 the night milk between the two. Two breeds so tested were milked only 

 twice a day at equal intervals and all of these averaged to test lower in 

 the morning. This seems to indicate that the milking following the 

 period of rest was least rich in fat, regardless of the comparative length 

 of periods. 



