TESTING 237 



bottom of the test bottle, thus crowding the Hghter con- 

 stituent, the fat, into the graduated neck; (5) careful 

 addition of water to the liquid, when forcing the fat into 

 the neck of the bottle ; (6) proper reading. 



218. Handling of composite samples. — Composite 

 samples should be held in glass bottles with ground glass 

 stoppers as mentioned above. A preservative such as 

 corrosive sublimate (mercuric chloride) should be used 

 in sufficient quantities to prevent the souring of the milk. 

 It is good practice to shake the sample each time a portion 

 is added in order to distribute the preservative through- 

 out. A rotary motion should be used in this agitation, 

 since it is desirable that a minimum amount of milk or 

 cream be splashed on the sides of the bottle. When this 

 happens, milk or cream clings to the sides of the bottle 

 above the surface of the sample and is likely to become dry 

 and leathery ; thus it is in bad condition to prepare for 

 testing. Coloring matter should be used in the preserva- 

 tive as a warning that poison is present. All specially pre- 

 pared preservatives for composite samples have this added 

 color. The mercuric chloride is of course very poisonous. 



219. Sampling sour milk and cream. — Milk so sour 

 that it is curdled cannot be sampled properly with- 

 out special precaution. Milk that is sour but not 

 curdled may be correctly sampled. However, it should 

 be tested soon after it is sampled. Sour cream is hardly 

 as difficult to sample as sour milk ; nevertheless it is more 

 readily handled when it is sweet. 



220. Sampling frozen milk and cream. — According 

 to Ross and Mclnerney,^ partly frozen milk should never 



' Ross, H. E.,and Melnerney, T. J., The Babcock Test, with 

 Special Reference to Testing Cream, Cornell Univ. Agri. Exp. 

 Sta., Bui. 337, p. 35, 1913. 



