158 MILK [x 



regard to the determination of the percentage of fat 

 in milk by the thickness of the cream layer, it must 

 at once be said that, although this method has long 

 been popular on account of its simplicity, it is a 

 most unreliable one. An apparatus which is largely 

 used for this purpose is the Chevalier Cremometer, 

 which consists of a glass cylinder of about 20 centi- 

 metres deep and about 4 centimetres broad, possessing 

 a graduated scale which indicates the percentage of 

 fat corresponding to different layers of cream. The 

 milk is allowed to stand in this for twenty-four 

 hours, and the depth of the cream layer then read 

 off. The reason of the inaccuracy of such a method 

 of testing the fat in milk is due to the fact that, as 

 has been pointed out elsewhere, the same quantity 

 of fat in milk will, under different conditions, give 

 layers of cream of very varying depth. It may 

 be said that ordinary milk will give, on an average, 

 about 10 per cent of cream ; but so variable is this 

 that it may sometimes amount to from 30 to 40 per 

 cent, and at other times may be as low as from 4 to 

 5 per cent, and this despite the fact that the range 

 in the amount of fat present may be very trifling. 

 Experiments carried out by Kirchner exemplify 

 this. In four different samples of milk in which the 

 percentage of fat only varied to the extent of '03 

 per cent, a difference in the percentage of cream, 

 amounting to 4 per cent, was found, owing to the 



