Wi .1. 8KOWNL1K HENDERSON, F.I.C, 4ND L. A. MESTON' 177 



ice chest) until required. By this means the freezing 

 of each sample is started close to 0° 0. While the samples 

 are cooling, the. freezing mixture is prepared. The ice is 

 conveniently prepared from the ice block by means of the 

 usual ice plane. Alternate layers of ice and salt (3 : I) are 

 added until the jjorcelain beaker is filled. When about 

 half filled, an empty " milk " tube is put into the middle 

 of the beaker and th(^ mixture packed aronnrl tliis tube. 

 On removing the empty tube when the beaker is filled, there 

 is no difficulty in inserting the tube containing the milk 

 sample. After filling the beaker, the shelf is put in position 

 on the rests and the beaker j^ut on it. A tube (containing 

 a sample is then put through the hole ia the top of the 

 stand into the freezing mixture. Th(? indiarubber cork 

 is inserted, carrying the tliermometer and thcs stirrer,. 

 The thermometer litis been so adjusted that the bottom 

 of the thermometer is about 2 c.m. from the bottom of the 

 tube when the cork is in position. 



The stirrer is iu>w worked u|> and down ccjutiiiuously 

 at the rate of from one to two complete movements per 

 second to prevent the formHtion of a solid block of frozen 

 milk in the tube as the temperature falls. As a rule the 

 mercur\ will ta])idly fall to a point below the true freezing 

 point of the milk (surfusion of the milk), and then rapidly 

 rise and become almost stationary ; the high(^st point of 

 the rise after the fall will be found to be very close to the 

 true freezing point. Wlu^n the tube has become partly 

 filled with finelv b token up frozen crystals- (experience 

 with the method soon <^nables one to judge of the eorrect> 

 proportion of crystals U> ha^e, in the liquid)— the porcelain 

 beaker containing thf^ freezing mixture? is removed hy 

 withdrawing the shelf (.- and k)wering the beaker and 

 the hand put round the tube " G." so that its warmth may 

 cause a rise in tem])erature, the stirrer being worked very 

 genth until therc^ is a rise of about , 'f;,^ths on the thermometer 

 scale. The hand is now removed and the milk well stirrerl 

 so as to surround the thermometer bulb with crystals of 

 frozen milk. 



'J'he stirring is stopptnl and th(; temperature obserx ed- 

 the mercury will sh^wly fall, and when it becomes stationary 

 the reading is taken, but sho\dd not be taken as f^nal nnlessi 

 it remains constant for at least two minutes. 



M 



