190 Testing Milk and Its Products. 



salt, etc., in the manufacture of butter. Considering 

 first the losses of fat in skim milk and butter milk, the 

 separator will usually, when run at normal speed and 

 capacity, leave the same per cent, of fat in skim milk, 

 whether rich or poor milk is skimmed. An exception 

 to this may be found in separating rich milk having 

 large fat globules, or milk from fresh milkers, in either 

 of which cases the large size of the fat globules occa- 

 sions a more complete separation of fat by the centri- 

 fugal force. But generally speaking, the statement 

 holds good that the total loss of fat in separator skim 

 milk is a factor of the quantity of milk run through the 

 separator, rather than of its quality. It follows from 

 this, however, that the relative losses of fat in skim milk 

 will vary to some extent according to the quality of the 

 milk separated. Selecting two extremes in the quality 

 of milk, 2.5 and 6.0 per cent, of fat, there will be found, 

 say .1 per cent, of fat in the skim milk from either lot, 

 provided the separator is not unduly crowded, and the 

 separation is conducted under normal conditions in each 

 case. But .1 per cent, fat makes 4 per cent, of the total 

 fat in the poor milk ( ii^| - =4) , and only 1.7 per cent, 

 of that in the rich milk. It takes 4000 Ibs. of the 2.5 

 per cent, milk to furnish 100 Ibs. of fat, and only 1666 

 Ibs. cf the 6 per cent, milk ; in skimming the poor milk, 

 a less of .1 per cent, of fat is sustained in the skim milk 

 from 4000 Ibs. of milk (3.4 Ibs. fat), while in the rich 

 milk n similar loss is sustained in the skim milk from 

 only 1666 Ibs. of milk (1.4 Ibs. fat). 



