Il6 MODERN METHODS OF TESTING MILK 



placed under the .asbestos sheet, which rests on the 

 iron stand. The cup with the weighed sample in it is 

 placed on the asbestos with the lifter and allowed to 

 remain until the foamy, snow-white covering that is 

 soon formed changes to a dirty-brown color, which 

 is usually accompanied by a slightly pungent odor. 

 The dish is at once removed from the asbestos, allowed 

 to cool to room temperature and weighed. The per- 

 centage of water in the butter tested is calculated as 

 already described (p. 115). 



While the sample is being heated, it should be shaken 

 from time to time with a rotary motion in order to 

 break up the surface covering and facilitate the escape 

 of the water-vapor. So long as snow-white portions 

 remain, the evaporation is not complete. 



When the special form of scale that is recommended 

 is used, a sample weighing 20.2 grams is taken, which 

 permits direct reading on the scale-arm of the per- 

 centage of water in butter. 



The Cornell method possesses the advantages of 

 simplicity, ease of operation and reasonable accuracy ; 

 the apparatus is inexpensive and durable. 



The Ames method, devised by McKay and Bower at 

 the Iowa experiment station, differs from the Cornell 

 method mainly in that the dish containing the sample 

 of butter is heated over melted paraffin at a tempera- 

 ture of about 175 F. This permits good control of 

 temperature and yields good results. Some find the 

 use of a paraffin bath objectionable on account of 

 possible danger of catching fire. Paraffin that has 

 been repeatedly heated is apt to undergo some decom- 

 position and give off an unpleasant odor. 



