For the detection of saccharin in wine and other beverages, two tests are generally relied upon - (1) the characteristic sweet taste of the substance, obtained as an ether extract from the wine, and (2) the conversion of the saccharin into salicylic acid. As regards these, it may be remarked that a sweet taste alone is not absolutely conclusive, since it might be due to dulcin, another artificial sweetening ingredient. Also, since salicylic acid may be present as such, care must be taken to remove it before applying the second test. For the quantitative determination of such small proportions of saccharin as are met with in beverages, probably the best method is the oxidation of the substance by fusion with an oxidising mixture, and estimation of the resulting sulphate.

(I.) The wine does not contain salicylic acid. - One hundred and fifty c.c. of the sample are mixed with sand in a porcelain basin, and evaporated on the water-bath; 2 c.c. of syrupy phosphoric acid are added, and the mass is extracted warm with a mixture of ether and petroleum ether (equal volumes), a glass "crusher" being used to break down the solids during the extraction. The ethereal extract is poured off through an asbestos filter, and the extraction repeated until 300 c.c. of filtrate are obtained. One-third of the filtrate is taken for the qualitative tests. It is evaporated to a small bulk, and then removed to a short, wide test-tube or a porcelain crucible. A small portion is withdrawn and evaporated on a watch-glass for tasting; if the sweet taste is detected the remainder of the solution is evaporated to dryness in the tube or crucible, then mixed with 2 c.c. of a 3 per cent. solution of sodium hydroxide, and again taken to dryness. The residue is next heated in an oil-bath to 250° for thirty minutes to effect the conversion of the saccharin into salicylic acid. After cooling, the substance is dissolved in a little water, acidified with sulphuric acid, removed to a separator and extracted with ether. The ethereal extract is washed with a few drops of water to remove any traces of sulphuric acid, then evaporated to dryness and tested with a drop or two of a dilute, neutral solution of ferric chloride to obtain the violet colour due to salicylic acid.

For the estimation, the remaining two-thirds of the filtrate (= 100 c.c. of the wine) are evaporated to dryness, the residue dissolved in dilute solution of sodium carbonate, and filtered into a platinum capsule. The liquid is evaporated to dryness, and mixed with four or five times its weight of dry sodium carbonate; this mixture is then added a little at a time to an excess of potassium nitrate kept fused in a platinum crucible. After cooling, the melt is dissolved out with hot water, transferred to a beaker, covered with a watch glass to prevent loss by effervescence, then carefully acidified with hydrochloric acid, and the sulphate precipitated with barium chloride and determined in the usual manner. The weight of barium sulphate obtained X 0.7857 = weight of saccharin in the 100 c.c. of wine taken.

(II.) The wine contains salicylic acid. - The following procedure is described by Blarez1: volume. After the residue has cooled somewhat, 5 c.c. of a 5 per cent. solution of potassium permanganate are run in, little by little, in three portions and stirring after each addition. When the colour of the permanganate has disappeared and the reaction, which is shown by the production of bubbles, has terminated, the (cooled) solution is extracted with ether. For this purpose 70 to 75 c.c. of ether are placed in a separator of about 250 c.c. capacity, and 50 c.c. of the prepared wine poured in; the mixture is then shaken vigorously during two or three minutes, and this shaking is repeated two or three times. After separating, the extracted wine is run off, and the rest of the prepared wine is passed into the same ether and treated in the same way. "The ethereal solution is now allowed to separate completely, passed through a filter, and evaporated to dryness; the dry residue can now be tested as before by the taste and the conversion into salicylic acid.

Two hundred c.c. of wine are acidified with 2 grams of syrupy phosphoric acid and evaporated to a little less than half the original

1 Gayon and Laborde, " Vins," p. 261.

The presence of salicylic acid in the wine does not, of course affect the estimation of the saccharin by the sulphate process described above. An approximate estimation of the saccharin, however, can also be made by noting the depth of colour given with ferric chloride in the fusion test for saccharin carried out on wine treated with permanganate as above described, compared with the colour obtained from known quantities of salicylic acid. One hundred and thirty-eight parts of the latter are given by 182 parts of saccharin.