214 ANNUAL REPORT OF THE Off. Doc. 



Do we always work as well one dny as another whether we feel well 

 or ill? Though in the best of health do we do the same amount of 

 work each day? Why should we expect a cow to do the same day 

 after day? Her work is expressed by the milk she makes, and, 

 largely, by the per cent, of fat she puts into that milk. We should 

 rot expect of her what we ourselves cannot do. 



STATE OF LACTATION. 



The stage of lactation is another reason wny Smith's milk test 

 differs from that of Jones'. It is well known that cows tend to belter 

 the quality of their milk as they progress in lactation. Investiga- 

 tion has shown that cows differ greatly in this matter. Some vary 

 but slightly and others largely as the}' pass from freshness to strip- 

 ping. A farrow cow goes dry giving milk but little richer when she 

 came in; a pregnant cow going dry usually gives considerably richer 

 milk than when she came in. Experiment has shown, moreover, that 

 on the average the increase from calving to drying-off approximates 

 1.25 per cent, fat, that is to say a milk testing four per cent, at calv- 

 ing may test 5.25 per cent, of fat at stripping. Smith's milk may be 

 made largely by strippers, wiiile Jones' cows may be mostly fresh in 

 milk. 



It is now generally understood that the quality of the milk of the 

 &ame herd varies decidedly from day to day, from milking to milking, 

 and that, in order to represent correctly the weekly or monthly 

 quality, it is necessary to take a composite sample. It will some- 

 times happen, however, that even when composite samples are used 

 teste may vary one week with another, fifty, sixty, or seventy points. 

 I believe it is the duty of the creamery management in such case to 

 verify the result by retest. Many patrons have an exaggerated idea 

 as to this matter. For instance, a few years ago a creamery patron 

 told me that he was being defrauded, that his test at the creamery 

 one month was 3.90 and the next month 3.85. These five points, 

 0.05 per cent., seemed to hira enormous. No operator can take the 

 same test in the same Babcock bottle and always read it twice alike. 

 Two-tenths of one per cent, is not a wide difference between two 

 t(?sts, and three-tenths of one per cent, as between one month and an- 

 other, even when the cow's are in scant flow, is hardly a wide enough 

 variation for cavil; more than that is of importance. But, as I shall 

 say later on, one should not growl but investigate. 



MANAGEMENT OF THE CREAMING DEVICES. 



Milk is creamed nowadays either by shallow setting, deep setting 

 rr centrifugal means. The former old-styled and inadequate method 

 is not followed in co-operative dairying and may be dismissed from 



