TESTING 255 



Barnhart^ found that there is a variation in the moisture- 

 content, ranging from 0.1 to 1.0 per cent, between differ- 

 ent samples representing the same butter. The average 

 variation was about 0.5 of one per cent. This serves 

 to show that the operator should be careful to take many 

 portions in securing a sample. 



230. Testing. — The following are steps in testing 

 butter for moisture : 



1 . This is the same as step 1. in testing butter for milk- 

 fat, par. 228. 



2. Weigh the desired amount of butter from this 

 prepared sample into a special cup that has been thor- 

 oughly dried. The usual quantities are 10 or 20 grams. 

 Under the average conditions, it is best to weigh 10 grams 

 in the light aluminum cup which is about 2^ inches high 

 by about 2| inches in diameter at the top. Such a cup 

 heats and cools quickly. The best scale for the creamery 

 is the moisture-test torsion balance. 



3. Place the cup in an oven, such as a hot water or 

 electric oven, or put it on an asbestos pad over an alcohol 

 or gas flame, or hold it over a low naked flame. In the 

 average creamery, a low flame in an alcohol burner is the 

 most satisfactory. This flame may be naked or an 

 asbestos pad may be used, as conditions warrant. It is 

 best not to shake the sample very much when being heated. 

 If shaken too much, it is more likely to effervesce or bubble 

 over than when shaken only occasionally. 



4. The end point has been reached in the heating pro- 

 cess when the color becomes chocolate brown. The casein 

 on the surface should be in small portions at this stage and 



' Lee, Carl E., Hepburn, N. W., Barnhart, Jesse M., A 

 Study of Factors Influencing the Composition of Butter, Univ. 

 lU. Agri. Exp. Sta., Bui. 137, p. 314, 1909. 



