DAIRY PRODUCTS. 89 



in the flask does not contain a trace of fat. It is, therefore, necessary 

 to assume that the emulsion has always the same proportion of ether, 

 otherwise there would be variations in the density of the clear super- 

 natant solution. This may be entirely true with those milks which per- 

 mit the etber solution to separate readily, but where the separation is 

 difficult and a great deal of time is required for it to take place, it is 

 possible that this assumption may not hold good. 



Liebermann, who having once condemned Soxhlet's method, on ac- 

 count of the slow separation of the ether solution, introduced a modifi- 

 cation of gently shaking the mixture and then pronounced it successful. 

 He says 1 : " Therefore the manner of making the first shaking with ether 

 is not immaterial. The shaking must not be violent (Soxhlet says 

 'schiitteln eine halbe Minute kraftig)/ but, nevertheless, always suffi- 

 ciently so. The light vertical blows must also be skillfully applied. In 

 fact, these are things which can only be attained by the skilled touch 

 coming from experience." 



It appears from the above, and my own experience fully convinces 

 me of the truth of it, that in order to secure a proper separation of the 

 ether fat solution, in all cases, one must be possessed of the art of leg- 

 erdemain.. 



After giving the preference to his own volumetric method Lieber- 

 mann ends his paper by the remark: "But if, on the contrary, the 

 method of Soxhlet is employed, it will usually happen that a given milk 

 must be examined twice, once to see whether the original method is 

 applicable, and again, in case of failure, with the necessary modifica- 

 tions." 



I think it will be readily agreed that such a method can find no place 

 in the examination of the milk of individual cows for each sample of 

 which a special modification of the method would have to be made. 



Schmoeger 2 calls attention to the shortcomings of Soxhlet's method 

 when milk poor in fat or skimmed milk is used. He says milk contain- 

 ing under 2 per cent, of fat must be evaporated with gypsum before it 

 will give up its fat to ether. Yet it must not be forgotten that gypsum 

 will give up something to ether, also. Schmoeger obtained from 20 

 grams of pure ignited gypsum 5mg. and from 20 grams sea sand 2mg. of 

 extract. These facts must be taken into consideration when the gravi- 

 metric comparisons of fat per cents with the areometric results are 

 made. Certain variations in the method of making these gravimetric 

 determinations have lately come into use, which promise some improve- 

 ment. Babcock 3 substitutes ignited asbestos for sand or gypsum. This, 

 in my opinion, is a great improvement. Babcock 4 also dispenses with 

 the evoprating dish and uses a tube open at both ends filled with as- 



Anal. Chem., loc. cit. 



2 Bericht ueber die Thatigkeit des Milch wirthschaftlichen Instituts zu Proskau, as 

 quoted in Zeit. Anal. Chem., 1885, p. 130. 



3 Second Ann, Kept. Bd. Control N. Y. Exp. Sta., p. 167, 



4 Loc. cit, 



