THE TESTING OF MILK AND CREAM 109 



cock is used, it is of the greatest importance that no air is 

 allowed to remain in the small rubber tube at the bottom of the 

 burette. 



Burettes having a small bore can best be filled by inserting 

 the lower part in the reagent and drawing it up slowly by 

 means of the mouth or by using a rubber bulb. 



By having the liquid in both burette and bottle at zero, it is 

 an easy matter to draw the liquid of the burette slowly into the 

 test bottle. Observations may be made at any point, provid- 

 ing sufficient time is allowed for the liquid to flow down the 

 side of the burette. The correct reading is always given di- 

 rectly by the burette. If, for example, the burette reads 25, 30, 

 or 40 per cent, the reading on the neck of a calibrated cream 

 bottle should also be 25, 30, or 40 per cent. If the burette reads 

 25.5 or 30.5 and the bottle reads 25.0 or 30.0, it is evident that 

 the bottle reads one-half per cent too high. 



TESTING MILK FOR BUTTER-FAT BY THE BABCOCK METHOD 



(Hunziker) 

 Sampling the milk 



The sampling is the most important operation of the test. 

 Unless the sample is representative of the milk from which it is 

 taken, the result of the test cannot be correct. The funda- 

 mental cause of non-representative samples lies in the fact that 

 the butter-fat is lighter than the remainder of the milk con- 

 stituents. When the milk is allowed to lie in the cans undis- 

 turbed, the butter-fat rises to the surface. Unless the milk is 

 thoroughly mixed before sampling, a representative sample 

 cannot be taken. 

 Single samples of milk. 



The most accurate and reliable method of sampling is to 

 take single samples of each patron and test them daily. In 

 order to minimize the work, the sample may be pipetted from 



