102 Testing Milk and Its Products. 



119. Latitude of variation. In order to determine 

 whether or not a sample of milk is skimmed or watered, 

 or both skimmed and watered, the per cents of fat and of 

 solids not fat in the sample must be ascertained, and if a 

 control -sample can be secured, these determinations for 

 both samples compared. The proper latitude to be 

 allowed for the natural variation in the composition of 

 milk differs according to the origin of the milk; in case 

 of milk from single cows, the variations in fat content 

 from day to day may exceed one per cent., although under 

 ordinary conditions the per cent, of fat in most cows' milk 

 will not vary that much. The content of solids not fat 

 is more constant, and rarely varies one-half of one per 

 cent, from day to day with single cows. Cows in heat or 

 sick cows may give milk differing considerably in com- 

 position from normal milk. 1 



120. Mixed herd milk is of comparatively uniform com- 

 position on consecutive days, and as most milk offered 

 for sale or delivered to factories is of this kind, the task 

 of the milk inspector is made considerably easier and 

 more certain on this account. Daily variations in herd 

 milk beyond one per cent, of fat and one-half per cent, of 

 solids not fat, are suspicious and may be taken as fairly 

 conclusive evidence of adulteration. This is especially 

 true in case the control-sample shows a comparatively 

 low content of fat or solids not fat (155). 



121. Legal standards. Where a control -sample cannot 

 be taken, the legal standards of the various states for fat 

 or solids in milk are used as a basis for calculating the 



i Blythe, Foods, their Composition and Analysis, London, 1882, p. 246 

 et seq. 



