The following analyses, here slightly abbreviated, of liqueurs are given in Konig's work; but it should be borne in mind that liqueurs called by the same name are not always made from the same formula, and that there are sometimes different qualities of the same type.

Analyses of various liqueurs.

As regards the estimation of the essential oils in liqueurs, the method adopted officially in France is to distil the sample, make up the distillate to the original volume with water, and then determine the iodine absorption by Hubl's process.

One hundred c.c. of the liqueur are placed in a 250 c.c. distilling. flask, 10 c.c. of water added, and the mixture distilled till 100 c.c. have been collected. Of this, 50 c.c. are placed in a 250 c.c. glass. stoppered flask, and 25 c.c. of the following mixture added: -

Iodine, 50 grams, mercuric chloride 60 grams; separately dissolved in 1 litre each of 96 per cent. alcohol, and equal volumes of the two solutions mixed.

The distillate, after being well mixed with the 25 c.c. of iodine reagent, is set aside for three hours at a temperature of 18°. A 'blank' experiment is carried out at the same time on alcohol containing no essential oils.

At the end of the three hours, 10 c.c. of a 10 per cent. solution of potassium iodide are added, and the two solutions titrated with decinormal thiosulphate solution. Let N be the number of c.c. required for the pure alcohol, and n for the liqueur. Then (N - n) X 0 254 gives the quantity of iodine absorbed per litre of liqueur.

From a knowledge of the iodine value of the oil in question, the proportion in the liqueur can be calculated. Following are the quantities of iodine absorbed by 1 gram of the various essential oils: -

* Absinthe petite = Artemisia maritima, Absinthe grande = A. absinthium.

For the mixture of oils used in absinthe liqueur, Sangle-Ferriere and Cuniasse take 1.238 as the mean value of iodine absorbed per gram of the mixture.

According to A. Auguet,1 the results are too low if the iodine solution prepared in the ordinary way is used whilst fresh: the maximum values are given with a solution which is some months old. This is attributable to the development of hydriodic acid in the Hubl's iodine solution, and the addition of this acid is therefore recommended, in order to fit the iodine solution for use at once. The proportion of hydriodic acid suggested is 75 to 10 grams per litre. The sample should be allowed to stand for two hours at a temperature of about 18° after mixing with the iodine solution.

Various liqueurs tested by this method gave the following results, expressed in grams of iodine absorbed per litre of the sample: -

Alcoholic solutions of various essential oils (2 parts per litre) gave the following values for iodine absorbed: -

1 Ann. Falsif., 1906, 6, 385,

The Wijs method of estimating iodine absorption is not suitable for use with alcoholic solutions of essential oils.

According to X. Rocques,1 recounting the results of an investigation carried out for the Societe des Experts.Chimistes de France, the iodine method gives accurate determinations with aniseed and kummel (caraway) liqueurs only. Oil of aniseed absorbs 1.45 grams of iodine by the Hubl process, and oil of caraway 2.40 grams, per gram of oil. With other liqueurs, the method recommended is to saturate the distillate with salt, extract with petroleum ether, allow the solvent to evaporate at a low temperature, and weigh the residual oil. A recent French decree fixes the maximum of total essential oils in liqueurs at 0'5 gram per litre.

C. F. Muttelet2 points out that since menthol does not absorb iodine, it is well to use the extraction method just mentioned in examining peppermint liqueurs, otherwise added menthol is not detected. One gram of oil of peppermint will absorb 045 gram of iodine; hence if the iodine value of the oil extracted from peppermint liqueur by petroleum ether is determined the result will indicate whether menthol has been added. This author also remarks that liqueurs which have been coloured with caramel may yield distillates containing unsaturated (furfuraldehydic) compounds derived from the caramel. These would absorb iodine, and to that extent vitiate the results obtained by the iodine process.

Muttelet also describes the following method of carrying out the extraction process of determining the total quantity of essential oils in liqueurs. Two hundred c.c. of the sample are mixed with 75 c.c. of water and distilled, 200 c.c. of distillate being collected. Fifty grams of finely.powdered sodium chloride are then dissolved in the distillate, and the solution is extracted three times with petroleum ether (b.p. about 40°), using 10 c.c, 5 c.c., and 5 c.c. The petroleum ether solution is dried with a little anhydrous sodium sulphate, and removed to a conical weighing flask, which is fitted with a cork carrying two glass tubes. The solvent is evaporated by passing a slow current of air through the flask while the latter, in the early stages, is maintained at a temperature of about 25° to 30°. The evaporation is continued until the weight of the contents of the flask does not diminish by more than 2 mg. after continuing the air current for five minutes. Experiments with aniseed essence and Chartreuse essence showed that the loss of essential oil in estimations by this method was less than one-tenth of the quantity present.1

1 Ann. Falsif., 1916, 9, 127.134. 2 Ibid., 70.73; 134.143.

"Compounded spirits" is a general term indicating spirits which have been sweetened or flavoured, or both. It includes such preparations as gin, sloe gin, British brandy, imitation rum, liqueurs, and cordials; but the term is not confined to beverages.