If pure vanilla extract be evaporated to about one-third its volume the rosins become insoluble and settle to the bottom of the dish. Artificial extracts remain clear under the same conditions. In examining vanilla extract the character of these rosins is studied. For this purpose a dish containing about an ounce of the extract is placed on a teakettle or other vessel of boiling water until the liquid evaporates to about one-third or less of its volume. Owing to the evaporation of the alcohol the rosins will then be insoluble. Water may be added to restore the liquid to approximately its original volume. The rosin will then separate out as a brown flocculent precipitate. A few drops of hydrochloric acid may be added and the liquid stirred and the insoluble matter allowed to settle. It is then filtered and the rosin on the filter paper washed with water. The rosin is then dissolved in a little alcohol, and to 1 portion of this solution is added a small particle of ferric alum, and to another portion a few drops of hydrochloric acid. If the rosin be that of the vanilla bean, neither ferric alum nor hydrochloric acid will produce more than a slight change of color. With rosins from most other sources, however, one or both of these substances yield a distinct color change. For filtering, a piece of filter paper should be folded once through the middle and again at right angles to the first fold. It may now be opened with one fold on one side and three on the other and fitted into a glass funnel. When the paper is folded in this manner the precipitated rosins may be readily washed with water. When the washing is completed the rosins may be dissolved by pouring alcohol through the filter. This work with the rosins will require some practice before it can be successfully performed. It is of considerable value, however, in judging of the purity of vanilla extract.

Lemon Extract

By lemon extract is understood a solution of lemon oil in strong alcohol. In order to contain as much lemon oil as is supposed to be found in high-grade extracts the alcohol should constitute about 80 per cent of the sample. The alcohol is therefore the most valuable constituent of lemon extract, and manufacturers who turn out a low-grade product usually do so because of their economy of alcohol rather than of lemon oil. Owing to the fact that lemon extract is practically a saturated solution of oil of lemon in strong alcohol the sample may be examined by simple dilution with water. A tea-spoonful of the oil in question may be placed in the bottom of an ordinary glass tumbler and 2 or 3 teaspoonfuls of water added. If the sample in question be real lemon extract the lemon oil should be thrown out of solution by reason of its insolubility in the alcohol after its dilution with water. The result is at first a marked turbidity and later the separation of the oil of lemon on the top of the aqueous liquid. If the sample ! remains perfectly clear after the addition of water, or if a marked turbidity is not produced, it is a low-grade product and contains very little, if any, oil of lemon.