236 



Testing Milk and Its Products. 



A modification of the Gray method has been proposed 

 by Mitchell and Walker of the Kingston (Ont.) Dairy 

 School, and described as the Mitchell-Walker test.' 



275. Patrick's method. Ten grams of butter are 

 accurately weighed into a 300 cc. aluminum beaker 

 (about 3 inches tall and 2 inches in diameter) ; this is 

 held by means of a hand 

 clamp over the flame of 

 the alcohol lamp or a gas 

 burner (see fig. 60) and 

 very carefuJly heated until 

 all the water is expelled. 

 The beaker is then cooled 

 by sinking it to the rim in 

 water of 50° to 60°, wiped 

 dry, and the loss in weight 

 calculated as water. If ten 

 grams of butter weighed 

 8.45 grams after heating, *^s* 

 the loss in weight of 1.55 grams represents 15.5 per 

 cent, of the weight of the sample, and the butter there- 

 fore contained 15.5 per cent, of water.^ The results ob- 

 tained by this method seldom vary more than .2 per 

 cent, from those of chemical analysis, and often less 

 than .1 per cent, when proper care in sampling and 

 weighing has been taken. 



A few points need special attention in using this 



1 Bull. 167, Dairy Branch, Ont. Dept. of Agriculture. 



^ A convenient table showing per cents of moisture in butter direct 

 when 9 to 10.15 grams are weighed out, has been published by the 

 Copenhagen experiment station (62nd report ; see N. T. Produce Rev.. 

 IflOS, p. mo). 



Fig. 60. Aluminum beaker and 

 alcohal lamp used In the Patrick 



