Honey consists of the saccharine substance collected by the bee (Apis mellifica) from the nectaries of flowers, and deposited by them in the cells of the comb. “Virgin honey” is the product of hives that have not previously swarmed, which is allowed to drain from the comb; the inferior varieties being obtained by the application of heat and pressure. As a result of the peculiar conditions of its formation, honey constitutes a rather complex mixture of several bodies; indeed, its exact composition is a matter of some doubt. The chief ingredients are levulose and dextrose, accompanied by a small amount of cane sugar, and inconsiderable proportions of pollen, wax, and mineral matter. According to Dubrunfaut and Soubeiran,[60] genuine honey contains an excess of levulose mixed with dextrose and some cane sugar. In the course of time the latter is gradually converted into invert sugar, and a crystalline deposit of dextrose forms, the levulose remaining fluid.

The following analyses made by J. C. Brown[61] and E. Sieben,[62] show the general composition of pure honey:—

Barth has examined several varieties of genuine honey with the following results:—

W. Bishop[63] obtained the following figures from the examination of honey of known purity:—

The substances said to be employed in the adulteration of honey are water, starch, cane sugar, and glucose-syrup; the last mentioned is undoubtedly most commonly used. Hager[64] states that, by treating corn starch with oxalic acid, a product is obtained which, on standing two or three weeks, acquires the appearance and taste of genuine honey; and samples of commercial honey not unfrequently wholly consist of this or some other form of artificial glucose. The season for the collection of honey by bees is a limited one, and any existing deficiency in their natural source of supply is sometimes remedied by placing vessels filled with glucose near the hives. Occasionally the bees are also supplied with a ready-made comb, consisting, at least partially, of paraffine. It has been asserted that in some instances, this factitious comb is entirely composed of paraffine, but the writer is informed that, if the sophistication is practised to a proportion of over 10 per cent., the bees do not readily deposit the honey in the comb.

Owing to the complex composition of honey and to the rather incomplete character of the analyses of the genuine article at hand, the detection of some of the forms of adulteration resorted to is a matter of considerable difficulty. The presence of starch is best recognised by the microscopic examination of the honey. This will likewise reveal the absence of pollen, which may be regarded as a certain indication of the spurious nature of the sample. There appears to exist a difference of opinion in regard to the presence of cane sugar in genuine honey, but it may safely be accepted that the detection of a considerable proportion of this substance points to its artificial addition. In all cases of suspected adulteration with cane sugar or glucose, the determination of the sugar present by means of the polariscope and by Fehling’s method (both before and after inversion) is indispensable. It is commonly stated that unsophisticated honey polarises to the left, and that a sample possessing a dextro-rotary action is necessarily contaminated with glucose or cane sugar; but, while in the great majority of cases this is doubtless the fact, it is equally certain that honey of known purity has been met with which polarised to the right. Upon the inversion of honey containing cane sugar, the dextro-rotation is changed to a levo-rotation.

According to Lenz,[65] the specific gravity (at 17°) of a solution of 30 grammes of pure honey in exactly twice the quantity of distilled water is never less than 1·1110, a lower density indicating adulteration with water. Hehner[66] states that the ash of genuine honey is always alkaline, whereas that of artificial glucose is invariably neutral. The proportion of phosphoric acid present in honey varies from 0·013 to 0·035 per cent., which is considerably less than the proportion contained in starch sugars. Honey contaminated with starch sugar will generally show about 0·10 per cent. of phosphoric acid, and artificial honey, made from cane sugar, will usually be free from the acid.

The addition of commercial glucose may often be detected by the turbidity produced upon adding ammonium oxalate to a filtered aqueous solution of the sample; this is due to the presence of calcium sulphate, a common impurity in the commercial varieties of glucose. If the glucose employed for admixture contains much dextrine, as is very often the case, this fact can be utilised in its detection as follows:—2 c.c. of a 25 per cent. solution of the honey are introduced into a narrow glass cylinder, and 0·5 c.c. of absolute alcohol is cautiously added; with pure honey, the point of contact of the liquids will remain clear or become so upon allowing the mixture to stand at rest, whereas in presence of artificial glucose a milky turbidity will appear between the two strata. Genuine honey may, it is true, contain a small proportion of dextrine and exhibit a slight cloudiness when treated with alcohol, but the difference in the degree of turbidity caused is very considerable, and sufficient to render the test of service.

The test may also be applied by dissolving 20 grammes of the suspected honey in 60 c.c. of distilled water and then adding an excess of alcohol. Under these circumstances pure honey merely becomes milky, while, if commercial glucose is present, a white precipitate of dextrine is formed, which can be collected and weighed. If the sugar in the sample is determined by Fehling’s solution, both before and after inversion with a little sulphuric acid, and an estimation of the amount of dextrine present is made by precipitation with alcohol, it often occurs that the quantity of the latter substance is proportional to the difference between the amount of sugar found.

According to the late investigations of Sieben,[67] fairly satisfactory methods for the detection and determination of glucose syrup in honey are based upon the following facts:—

1st. When genuine honey undergoes fermentation, the substances which remain undecomposed, are optically inactive. Glucose, or starch syrup, on the other hand, leaves a considerable amount of dextrine, which is strongly dextrogyrate. The test is made by dissolving 25 grammes of honey in about 160 c.c. of water, and adding 12 grammes of yeast (free from starch). The mixture is allowed to ferment at a moderate temperature for two or three days, after which aluminium hydroxide is added, and the liquid made up to 250 c.c. and then filtered. 200 c.c. of the filtrate are evaporated to a volume of 50 c.c., and a 200 mm. tube is then filled with the concentrated solution and examined by the polariscope.

2nd. The substances remaining unaffected by the fermentation of pure honey are not converted into a reducing sugar by boiling with dilute hydrochloric acid, as is the case with those obtained from starch syrup under the same circumstances. 25 c.c. of the solution employed for the polarisation test, as just described, are diluted with an equal volume of water, 5 c.c. of strong hydrochloric acid added, and the mixture is placed in a flask and heated for an hour over the water-bath. The contents of the flask are neutralised with potassium hydroxide, then diluted to a volume of 100 c.c., and the proportion of reducing sugar estimated in 25 c.c. of the solution. Honey containing different proportions of starch sugar gave the following percentages of reducing sugar:—

3rd. If the cane sugar originally present in genuine honey has been changed into invert sugar, and the honey solution is boiled with a slight excess of Fehling’s reagent, no substances capable of yielding sugar when treated with acids will remain undecomposed. Starch syrup, when subjected to this treatment, yields grape sugar in about the proportion of 40 parts to every 100 parts of the syrup used. The test is applied as follows:—14 grammes of honey are dissolved in 450 c.c. of water, and the solution is heated over the steam-bath with 20 c.c. of semi-normal acid, in order to invert the cane sugar present. After heating for half an hour, the solution is neutralised, and its volume made up to 500 c.c. 100 c.c. of Fehling’s solution are then titrated with this solution, which may contain about 2 per cent. of invert sugar (in case the sample examined is pure, from 23 to 26 c.c. will be required); 100 c.c. of Fehling’s reagent are next boiled with 0·5 c.c. less of the honey solution than was found to be necessary to completely reduce the copper. The reduced liquid is then passed through an asbestos filter, the residue washed with hot water, the filtrate treated with a slight excess of concentrated hydrochloric acid, and the solution heated for one hour on the steam-bath. Sodium hydroxide is now added, until only a very little free acid remains unneutralised, and the solution is made up to 200 c.c. Upon well shaking the cooled liquid, a deposit of tartar sometimes separates. 150 c.c. of the filtered solution are finally boiled with a mixture of 120 c.c. of Fehling’s reagent and 20 c.c. of water, and the proportion of grape sugar estimated from the amount of metallic copper obtained. (See p. 111.) When pure honey is submitted to the preceding process, the copper found will not exceed 2 milligrammes. The quantities of copper obtained when honey adulterated with various proportions of starch sugar was tested were about as follows:—

The tabulation on p. 127 exhibits the results obtained by the application of the foregoing tests to adulterated honey.[68]

The detection of paraffine in honeycomb is easily accomplished. Genuine bees’-wax fuses at 64°, paraffine usually at a lower temperature. The latter is not affected by treatment with concentrated sulphuric acid, whereas bees’-wax is dissolved by the strong acid, and undergoes carbonisation upon the application of heat. The amount of potassium hydroxide required for the saponification of one gramme of bees’-wax, as applied in Koettstorfer’s method for butter analysis (p. 71), widely differs from the quantities consumed by Japanese wax and paraffine. Mr. Edward W. Martin has obtained the following figures:—

18 out of 37 samples of strained and comb honey, examined in 1885 by the Mass. State Board of Health, were adulterated with glucose and ordinary syrup.