WINES. 345 



dish with a spatula, and the entire mass rubbed up with a pestle to a uniform pasty 

 mass, the pestle and spatula being rinsed with a little alcohol ; in heating up the 

 alcoholic paste with lime, bumping and spurting may be avoided by careful stirring ; 

 the heating and subsequent washing out with hot alcohol is necessary, however, to 

 dissolve out the glycerine properly. Iu evaporating with both the alcoholic and tbe 

 ether-alcohol solution, all violent boiling of the liquid must be avoided, or mechani- 

 cal losses will occur. The best way is to place the vessels containing the solutions 

 inside of beakers filled with water on the bath. The clearing up of the ether-alcohol 

 solution can be hastened by energetic shaking in the stoppered flask containing it. 

 The vessel in which the ether-purified glycerine is finally weighed should have ver- 

 tical walls at least 40uim. in height. The losses which are caused by the volatility 

 of glycerine cannot be entirely avoided, but may be reduced to a minimum by a care- 

 ful observance of all the directions, even those which are apparently unimportant. 

 That the loss of glycerine is smaller by heating in a drying oven than on the open 

 water bath has been noticed in the estimation of the extract ; the choice of weighing 

 tubes also with proportionally high, vertical walls has for its object the lessening of 

 the possibility of losses in weight. 



For the estimation of the glycerine in sweet wines the following precautions should 

 be observed: Sufficient powdered lime must be added to the wine to convert the 

 whole of the sugar into its lime compound. The formation of the latter takes place 

 gradually during the heating on the water bath. ' The mass becomes at first dark 

 brown (special care is necessary to prevent its foaming over the neck of 'the flask), 

 but when the saturation with lime is complete it becomes somewhat clearer, and, to- 

 gether with the characteristic smell of the sugar-lime, a caustic odor becomes mani- 

 fest. 



If the residue obtained from the concentration of the alcoholic solution remains 

 somewhat thin even after cooling, it is not necessary to repeat the treatment with 

 lime. The purification with ether-alcohol in the way described will be all that is nec- 

 essary. 



The above described method for glycerine estimation is intended to obtain the 

 glycerine in a state of purity by its separation from all the other constituents of 

 wine, either by their volatility, by their insolubility in alcohol, or their lime combi- 

 nations, or finally by their insolubility in a mixture of one volume of alcohol with 1 

 volumes of ether. If pointed crystals appear on cooling, the presence of mannite is 

 indicated. Since the separation of glycerine in an insoluble condition in a form or 

 union peculiar to itself has not yet been accomplished, the extraction method must 

 serve for its estimation, but the latter is only useful for the conclusions which are 

 drawn from its results, when it is carried out with a strict observance of the preced- 

 ing conditions. 



Several methods have lately been proposed for the estimation of glyc- 

 erine, and it was with the hope of some of them proving more exact and 

 less tedious than the above that a somewhat hasty examination of these 

 methods has been made. 



Benedikt and Zsigmondy l published in 1885 a method for the estima- 

 tion of glycerine by its oxidation to oxalic acid by permanganate of 

 potash, precipitating the oxalic acid with calcium acetate, and deter- 

 mining it volumetrically by titration with acid. This method is also 

 claimed by Fox and Wanklyn. 2 At the time of the publication of this 

 method I made several trials of it on pure glycerine with very satisfac- 

 tory results, and Allen 8 has confirmed the accuracy attributed to it by 



'Chem. Ztg. 9, 975; Analyst 10, 206. '^Chern. Newa 53, 15. 'Analyst 11, 52. 



