2. If ammonium hydroxide be added to the suspected sample to distinct alkaline reaction, then a little ammonium sulphide and the liquid filtered, the filtrate from genuine wine will possess a green tint, whereas that obtained from artificially coloured wine will exhibit other colours, such as red, blue, violet, or brown.

3. 100 c.c. of the wine are evaporated to about one-half of the original volume, ammonium hydroxide added to alkaline reaction, and the liquid thoroughly shaken. Ether is then added, and the mixture again well shaken. It is next introduced into a separator, and allowed to stand at rest until the ether has risen to the surface, when the lower stratum is drawn off, and the residual ether washed by agitation with water, which is subsequently removed. The ethereal solution is now transferred to a flask connected with a Liebig’s condenser, a piece of white woollen yarn introduced into the liquid, and the contents of the flask distilled at a gentle heat: in presence of the smallest amount of fuchsine, the wool will acquire a very perceptible reddish hue.

4. A slight excess of ammonium hydroxide is added to 50 c.c. of the wine, a piece of white woollen fabric introduced, and the liquid boiled until the alcohol and ammonia are expelled. By this treatment it will be found that most aniline colouring matters, if present, become attached to the wool. Their presence can be corroborated by removing the fabric, washing and pressing it, and then dissolving it, with constant stirring, in a hot solution of potassium hydroxide. When solution has taken place, the liquid is allowed to cool, and one-half its volume of alcohol is added, then an equal volume of ether. The mixture is vigorously shaken, and, after remaining at rest for some time, the supernatant ethereal solution is removed, introduced into a test-tube, and a drop or two of acetic acid added. In presence of fuchsine, its characteristic colour will now become apparent. Methyl violet and aniline blue are separated by an analogous process.

5. Logwood and cochineal may be detected by agitating 100 c.c. of the suspected wine with manganic peroxide, and filtering. The filtrate afforded by pure wine will be colourless.

6. In Dupré’s process,[105] cubes of jelly are first prepared by dissolving 1 part of gelatine in 20 parts of hot water, and pouring the solution into moulds to set. These are immersed in the wine under examination for 24 hours, then removed, slightly washed, and the depth to which the colouring matter has permeated is observed: pure wine will colour the gelatine very superficially; the majority of other colouring principles (e.g. fuchsine, cochineal, logwood, Brazil wood, litmus, beetroot, and indigo) penetrate the jelly more readily and to a far greater degree. Dilute ammonium hydroxide dissolves from the stained cake the colouring matter of logwood and cochineal, but not that derived from fuchsine or beetroot.

7. The colouring principle of genuine wine when subjected to dialysis, does not pass through the animal membrane to any decided extent, while that of logwood, cochineal, and Brazil wood easily dialyses.

8. Many of the foreign dyes added to wine are precipitated by a solution of basic plumbic acetate. The precipitate obtained upon treating 10 c.c. of the sample with 3 c.c. of this reagent is collected on a filter and washed with a 2 per cent. solution of potassium carbonate, which dissolves cochineal, sulphindigotic acid and aniline red. The latter is separated upon neutralising the solution with acetic acid, and shaking with amylic alcohol, which, in its presence, will acquire a rose colour. The liquid is next acidulated with sulphuric acid, and again agitated with amylic alcohol, by which the carminamic acid, originating from cochineal, is isolated. Any remaining indigo (as well as the carminamic acid) is to be subsequently identified by means of its spectroscopic reactions. Upon treating the portion of the plumbic acetate precipitate which remains undissolved by potassium carbonate with a dilute solution of ammonium sulphide, the colouring matter of pure wine and of logwood is dissolved. If, in presence of logwood, the original sample is shaken with calcium carbonate mixed with a little calcium hydroxide solution and filtered, the filtrate will exhibit a decided red tint, but, if the wine treated be pure, little or no coloration will be produced.

9. An artificial colouring for wine, known as rouge végétale, is not uncommonly employed. According to Amthor,[106] its presence can be recognised as follows:—100 c.c. of the wine are distilled until all alcohol is removed. The residual liquid is strongly acidulated with sulphuric acid, and agitated with ether. Some woollen yarn is next introduced into the ethereal solution, which is then evaporated over the water-bath. In presence of rouge végétale, the wool will acquire a brick-red colour, which turns violet upon treatment with ammonium hydroxide.

10. Cauzeneuve and Lepine[107] state that acid aniline red, “naphthol-yellow S,” and roccelline red are harmless, whereas safranine and ordinary Martius’ yellow are decidedly poisonous.

The presence of “Bordeaux red”[108] is recognised by first adding sodium sulphate to the suspected wine, then a solution of barium chloride: the artificial dye is carried down with the precipitated barium sulphate, from which it can be extracted by means of sodium carbonate solution. The brownish-red liquid thus obtained acquires a deep red colour if acidulated with acetic acid, which it readily communicates to silk upon boiling. Natural red wine fails to produce a coloration under the same circumstances.

For the detection of the presence of artificial colouring matter the following process is used in the Municipal Laboratory in Paris:—Preliminary tests are made—

1st. By soaking pieces of chalk in an aqueous solution of egg-albumen; these are dried and applied for use by dropping a little of the wine upon them, and noting the coloration produced. Natural coloured wine usually causes a greyish stain, which, in highly coloured varieties, may verge to blue.

2nd. Baryta water is added to the wine under examination until the mixture acquires a greenish hue, after which it is shaken with acetic ether or amylic alcohol. If the wine be pure, the upper layer remains colourless, even after acidulation with acetic acid; whereas, in presence of basic coal-tar dyes, such as fuchsine, amidobenzole, safranine, chrysoidine, chrysaniline, etc., characteristic colorations will be obtained.

3rd. A few c.c. of the sample are made alkaline by the addition of dilute potassium hydroxide, some mercuric acetate added, and the mixture agitated and filtered. With pure wines, the filtrate is colourless; in the presence of acid coal-tar derivatives, it is red or yellow.

The general character of the artificial dye contained in the wine having been ascertained by the foregoing tests its more precise nature is determined as follows:—

In case the foreign colouring is basic, the supernatant layer obtained in the second test is separated, and divided into two portions; one portion being evaporated with pure woollen yarn, the other with filaments of silk. The dyed threads are then subjected to the following tests:—

(a) Rose-aniline or safranine affords a red coloration; safranine usually attaches itself only on silk.

(b) Soluble aniline violet produces coloured threads which become green upon treatment with hydrochloric acid, the primitive colour reappearing upon dilution with water.

(c) Mauve-aniline gives a colour which turns blue upon addition of the acid.

(d) Chysotoluidine causes a coloration which is only slightly affected by the acid, but which is discharged upon boiling with zinc powder; upon protracted exposure to the air the colour reappears.

(e) Chrysoidine and Amidonitrobenzole produce yellow colours, the former turning poppy-red if treated with sulphuric acid, the latter, scarlet. A general characteristic of dyes, similar to rose-aniline, is that they are decolorised by treatment with sodium bisulphite.

If the presence of an acid coal-tar dye is indicated by the third preliminary test, the following special methods of procedure are employed:—

Two portions of the wine are saturated respectively with hydrochloric acid and with ammonium hydroxide water, and each portion is strongly agitated with acetic ether. The ethereal layers are removed by means of a pipette, then mixed together, evaporated to dryness, the residue obtained treated with a drop of concentrated sulphuric acid, and observations made of the colour obtained:—

The method employed in the Paris Municipal Laboratory for the detection of dried fruit wine, or of added commercial glucose, is substantially the following:—A little beer-yeast is added to 300 c.c. of the suspected wine, and the mixture is allowed to undergo fermentation at a temperature of about 30°. When the fermentation is completed, the filtered liquid is introduced into a dialyser, the outer water of which is automatically renewed. The process of dialysis is continued until the outer water ceases to show a rotary effect when examined by the polariscope, after which it is neutralised with calcium carbonate and evaporated to dryness over the water-bath, with constant stirring. The residue obtained is treated with 50 c.c. of absolute alcohol and filtered, the insoluble matters being twice washed with 25 c.c. of alcohol. The alcoholic filtrates are next decolorised by means of animal charcoal, and evaporated to dryness, and the solid residue is dissolved in 30 c.c. of water and polarised. Genuine claret, when tested in this manner, fails to exhibit a rotary power, or is but slightly dextrogyrate, whereas fruit wines, and those containing artificial starch sugar, strongly rotate respectively to the left or to the right.

The following are some of the conclusions arrived at by a commission, appointed by the German Government, to inquire into uniform methods for wine analysis, and establish standards of purity for genuine wine.[109]

(a) After deducting the non-volatile acids, the extract in natural wine should amount to at least 1·1 gramme per 100 c.c.; after deducting the free acids, to at least 1 gramme per 100 c.c.

(b) Most natural wines contain one part of ash to every 10 parts of extract.

(c) The free tartaric acid should not exceed 1⁄6th of the total non-volatile acids.

(d) The relation between the alcohol and glycerine varies in natural wines between 100 parts alcohol to 7 parts glycerine, and 100 parts alcohol to 14 parts glycerine. These proportions do not apply, however, to sweet wines.

(e) Genuine wines seldom contain less than 0·14 gramme of ash, nor more than 0·05 gramme of sodium chloride per 100 c.c.

According to the analyses of Moritz, the maximum and minimum relative proportions of the constituents of natural wine are as follows:—The extract (after deducting the free acids) ranges from 1·10 to 1·78 per cent.; the proportion of ash to extract varies from 1 : 19·2 to 1 : 6·4; that of phosphoric acid to ash ranges from 1 : 12·3 to 1 : 10·49; that of alcohol to glycerine, from 100 : 12·3 to 100 : 7·7.[110] From the investigations of Dr. Dupré, it would appear that in genuine unfortified wines, the amount of alcohol present varies from 6 to 12 per cent. by weight. A wine containing less than 6 per cent. would be unpalatable, and more than 13 per cent. cannot well be present, since natural grape-juice does not contain the quantity of sugar requisite for the production of a greater amount of alcohol; moreover, an excess of this proportion would retard, if not entirely stop, the process of fermentation. Pure wines contain a greater proportion of volatile than fixed ethers, but in fortified wines the reverse is frequently the case. In natural wines, which are not over a few years old, the sugar present rarely amounts to 1 per cent., generally it is much less. Fortified wines, in which fermentation has been checked by the addition of alcohol, often contain 5 per cent. of sugar; champagnes usually show from 4 to 10 per cent., and, in some liqueur wines, a maximum of 25 per cent. has been found. In natural wines, the total dry residue generally ranges from 1·5 to 3 per cent., while in fortified wines the addition of sugar and other substances may increase its proportion to 10 per cent., or even more. At the Paris Municipal Laboratory the following standards are adopted: The amount of added water in all wines, not sold as of a special or abnormal character, is calculated on a basis of 12 per cent. of alcohol (by volume) and 24 grammes of dry extract per litre. The proportion of potassium sulphate in unplastered wines must not exceed 0·583 gramme per litre. The use of salicylic acid is prohibited.