ESTIMATION OF FAT. 221 



into the tins ; 5 grammes are taken from each and placed in 

 the dishes, but before the milk is poured back into the bottle 

 or otherwise emptied from the tin, the temperature and specific 

 gravity should be taken ; the remaining samples are then simi- 

 larly treated in their proper order. The drying and weighing 

 are performed as before. 



Rapid Methods. — A pipette is used which delivers 2"5 

 grammes of milk, and the milk is run into flat-bottomed 

 porcelain basins about 3 inches in diameter. The numbers are 

 marked on the basins with copper paint, which paint, on ignition, 

 forms an indelible blue mark. Before the basins are filled, 

 Stokes recommends that two or three drops of a 10 per cent, 

 solution of acetic acid in alcohol be sprinkled over each. The 

 alcohol spreads itself over the surface of the milk, and the acid 

 precipitates the casein. Revis adds 1 c.c. of acetone. Under 

 these circumstances, drying proceeds very rapidly. The basins 

 are placed on the water-bath till apparently dry, a matter of 

 a few minutes only, and are further dried for about an hour in 

 the water-oven or air-bath. They are then weighed as before. 

 The diiTerence between the weight of the basin with the residue 

 and the weight of the basin alone, multiplied by 40, gives the 

 percentage of total solids. 



In the author's experience Stokes' method has a tendenc}- to 

 give results slightly above the truth, but according to Yarrow the 

 difference does not exceed 0"08 per cent. ; it has, however, the 

 advantages of rapidity and of requiring very little attention. 



Estimation of Pat. 



For the estimation of fat in a rapid manner, with an accuracy 

 sufficient for milk control, a centrifugal method must be used. 

 The Le£Emann-Beam, and the acido-butyrometric methods, will 

 be described in detail. 



The Leffmann-Beam Method, and Modifications. — LefF- 

 mami and Beam, realising that the time of whirling necessary in 

 Babcock's method, which consisted in treating the milk with an 

 equal volume of strong sulphuric acid, and separating the fat 

 by centiifuging, was a serious objection, experimented with 

 a view to shortening this. They finally decided on the use 

 of amyl alcohol as a means of assisting the fat to rise, and 

 were thereby enabled to reduce the time of whirling.* The 

 method is usually employed in conjunction with the Beimling 

 machine. 



• It was stated in the first edition that the same idea was independently 

 worked out at the Vermont Kxperiment Station ; but it appears that this 

 was not correct, ard was based on a misunderstanding. 



