WERNER-SCHMID METHOD. 125 



Scbmid's method is not applicable to the determination of fab 

 in milk to which cane sugar has been added, a conclusion con- 

 firmed later by Dyer and Roberts, who showed that by boiling 

 cane sugar with hydrochloric acid a substance soluble in the 

 water taken up by the ether was formed. 



Stokes has pc inted out that this can be got rid of by extracting 

 the dried residue with dry ether, in which the caramel formed 

 from the sugar is insoluble. Allen recommends petroleum ether. 

 The author prefers adding to the ethereal solution an equal 

 bulk of petroleum ether, and washing with water containing just 

 sufficient ammonia to neutralise the free acid. 



Stokes later introduced a new form of tube, in which the 

 middle portion is narrowed for greater accuracy of measurement 

 of the ethereal layer (Fig. 12). 



Yarrow points out that it is absolutely necessary when using 

 Stokes' tubes to read the volume before and after 

 pipetting off a known volume at the same tem- 

 perature. He has observed that a tube which after 

 pipetting showed 24 c.c. read, after 30 minutes 

 and at a higher temperature, 3*4 c.c. a very serious 

 difference. 



Allen proposed drawing off as much ether as 

 possible, adding a further supply, drawing that off as 

 completely as could be done, and continuing washing 

 in this way till all the fat was separated from the 

 aqueous layer. 



Molinari and Stokes have both described forms of 

 apparatus in which the ethereal solution can be 

 removed completely from the aqueous layer without Fig. 12. 

 the necessity of pipetting it off. Stokes' Tube. 



Stokes' apparatus has the advantage that the globules of ether 

 and water during separation have but a short distance to travel, 

 and the separation is complete in a much shorter time than in 

 the longer tube, which was not devised for the purpose of complete 

 extraction where rapidity of separation is important. 



The author prefers diluting the milk with an equal bulk of 

 water before heating with hydrochloric acid, as there is then little 

 tendency for the formation of a fluffy-looking layer at the point of 

 junction. He finds that it is necessary to wait for ten minutes 

 at least after the ether apparently has separated from the aqueous 

 layer, in order to allow the fine globules of water to settle out of 

 the ether. 



Analysts who have used this method are generally agreed that 

 it gives results practically identical with that of Adams ; it is 

 a question whether the drawing off of an aliquot portion of 

 the ether is to be preferred to the extraction of the whole of the 







