MILK 51 



afterwards placed on a layer of extracted cotton wool in the 

 Soxhlet apparatus. Ether-extracted cotton wool is also used to 

 cover the kaolin in the cartridge. The extraction proceeds for 

 8 hours in the same manner as in Adams's method. With 

 whole milk of normal composition very exact results are 

 obtained with this kaolin method. 



To determine the amount of fat in cream by either of the 

 extraction methods, the procedure is similar to that with 

 milk except that only 5 g. of cream is taken. It is not possible 

 either to use the small wash bottle for weighing out the cream 

 in the Adams's method. This must be done in a small beaker, 

 and the contents poured over the strip of paper. The materials 

 formerly used to assist the drying, e.g. sand, powdered pumice, 

 &c., have been abandoned because they did not permit of the 

 complete extraction of the fat. The methods described above, 

 and in a greater degree those which were originally used, all 

 have the defect that they give too low results for separated 

 milk or for milk which has been subjected to energetic 

 mechanical treatment, e.g. " homogenised " milk, milk heated 

 by direct contact with a current of steam, or pasteurised milk 

 which has been stirred too violently. This has been shown by 

 the author l to be due to the breaking up of the fat globules 

 by the mechanical force, and the minute particles of fat which 

 thereby arise become imbedded in the casein on drying, and are 

 not reached by the ether during the extraction. The author 

 has also proved that the further the division of the fat particles 

 is carried, the lower are the results obtained. A longer period 

 of extraction does not improve matters either. 2 Numerous 

 investigators have confirmed the author's explanation of the 

 phenomenon. 3 



The methods about to be described do not suffer from the 

 above-mentioned defect, for in them the fat comes in direct 

 contact with the solvent, so the fine state of division of the fat 

 has no influence upon the analysis. 



1 Milch-Zeitung, 1903, No. 22. Molkerei-Zeitung (Berlin), 1903, No. 23; 

 also Milch-Zeitung, 1903, No. 31 and No. 37. 



2 Milch-Zeitung, 1904, No. 26. 



:5 L. F. Rosengren, Milch-Zeitung, 1904, No. 22; M. Siegfeld, Molkerei- 

 Zeitung (Hildesheim), 1904, Nos. 39'and 40 ; M. Henseval, Revue Gentrale du 

 Lait, Vol. Ill, p. 529 ; E. Holm, 56 te Beret ning fra Kgl. Veterinaer-og Land- 

 bohojskoles Laboratorium for Iand6konomike Forsog, 1905. 



E 2 



