of alkali naturally present in the ash is sufficient to prevent any loss of HBO, by volatilization. If this is not the case a little alkali carbonate may be added. Butter or fatty matter is best melted and shaken out with warm solution of Na2CO3; the separated alkaline aqueous liquid is then evaporated and gently calcined. The ash is then treated with definite amounts of H2SO, and methylic alcohol. One c.c. of H2SO4 is sufficient for the ash from 25 c.c. of wine. The mixture is then transferred to a small flask, washing out the capsule with 2 or 3 c.c. of methyl alcohol in successive portions. The flask is then attached to a condenser, and the contents distilled until vapour of H2SO4 appears. The alcoholic distillate is then ignited, and the flame observed against a black background; a green colour is given by HBO, in quantities of 0.1 Mgm.

Salicylic acid is detected by means of the FeCl reaction; the reagent employed should be perfectly neutral and very dilute, and a large excess should be avoided. If the substance to be tested be a liquid containing but little tannin, it is acidified with HCl and shaken out with benzene in a separator. If tannin be present in quantity, the liquid should first be precipitated with neutral lead acetate. Milk should be first coagulated by means of acetic acid, and the separated serum shaken out with benzene. Fats are melted, shaken out with warm solutions of NaHCO3; the separated alkaline liquid is acidified and shaken out with benzene. Solids are cut up small, extracted with aqueous solution of NaHCO3, and the alkaline liquid treated as above. Agitation with benzene should be performed cautiously, to avoid formation of a troublesome emulsion. The aqueous solution and benzene are best mixed by introducing them into a tube 3 cm. wide and 20 to 30 cm. long, which is then turned horizontally on its axis. Having separated the benzene extract, this is evaporated to about 20 c.c. and tested with 5 c.c. of the dilute FeCl solution, when the characteristic colour reaction will be obtained in the presence of salicylic acid.

Benzoic acid.-The substance is extracted with alcohol or ether or an alkali. After evaporation a portion of the residue is submitted to sublimation, when a crystalline sublimate and a characteristic odour may give positive results. A portion of the residue is treated with hot water, exactly neutralized, and tested with neutral Fe2Cle. The precipitate with benzoic

acid is characteristic. Another portion is heated with H2SO, and a few drops of KNO, solution when metadinitro-benzoic acid is formed; when this is treated with excess of AmOH a yellow colour becoming red with AmHS is given.

Abrastol and B-naphthol derivatives. This antiseptic is removed from liquid, such as wines, by shaking out with amyl alcohol in the presence of a slight excess of AmOH. If the solvent does not separate quickly, the emulsion may be broken down by adding to it a few drops of alcohol. It is then separated, filtered if necessary, and evaporated on the water-bath. The residue is treated with 1 c.c. of HNO3 and 1 c.c. of water, evenly moistening the whole residue therewith. The liquid is then evaporated on the water-bath to one-half. It is then transferred to a test tube, the capsule being washed with 1 c.c. of water, which is added to the first portion. To the mixture 0.2 Gm. of FeSO4 is added; when this has dissolved dilute AmOH is added drop by drop until the precipitate is permanent; 5 c.c. of alcohol and a few drops of H2SO4 are then added, and the mixture is set aside and filtered. In the case of pure wines the filtrate obtained will be colourless. If abrastol be present, a more or less red tint is evident. If salicylic acid be present, an orange instead of a red colour results, and the reaction is less sensitive. The original amylic alcohol residue is coloured violet with Fe2Cle, permanent on heating, if it contains salicylic acid; with abrastol a blue colour results with the same reagent and disappears on heating. The reaction for abrastol is completely masked by the presence of saure-fuchsin and other aniline colours. If the amyl alcohol extract is coloured, the residue is treated with dilute acetic acid, or in the case of alkali blue, with dilute AmOH, and again evaporated to dryness on a flock of white wool. On again treating the residue with water only the abrastol is dissolved. B-Naphthol is detected in a similar manner.

Formaldehyde. The following tests relating to milk are also applicable to other dietetic articles. Five c.c. of milk is mixed with 2 or 3 c.c. of colourless 1: 1000 solution of phloroglucin, and 1 or 2 c.c. of NaOH solution 10 per cent. Pure milk shows only a greenish tint and becomes semi-transparent. If formaldehyde is present, a salmon pink colour is given, which disappears in a few minutes. It is very marked with 1 of formalin in 100,000, quite evident with 1: 500,000, and distinguishable with 1: 1,000,000 by comparing with pure milk.

If 100 c.c. of the milk be distilled and the first 20 or 25 c.c. of distillate collected; on adding a few drops of phenol to this distillate and floating some of the mixture on strong H2SO4, a carmine red ring will be formed at the zone of contact in the presence of formaldehyde.

Milk containing formaldehyde treated with an equal volume of H2SO, and a few drops of FeCl solution gives a fine violet colour, especially when warm.

The above reactions are given by many aldehydes the following test of Trillat distinguishes formaldehyde. 25 c.c. of distillate, obtained from 100 c.c. of milk or other liquid, is treated in a small flask with 0.5 c.c. of dimethylaniline and 5 c.c. of 1 per cent. H2SO4; the flask is corked and kept at 50°C. for an hour. It is then transferred to a larger flask, diluted with 100 c.c. of water, and made alkaline with 5 c.c. of strong NaOH solution. The alkaline mixture is then distilled with a strong current of steam until all excess of dimethylaniline is driven off, as shown by no more oily globules passing in the distillate. The residue is then acidified with acetic acid and a portion withdrawn. To this is added a trace of PbO2 suspended in water. On boiling, a blue colour appears in the presence of formalin, and disappears on cooling, being again formed on heating. To determine the amount of formaldehyde, the whole of the alkaline distillation residue is shaken out with ether; on evaporating the solvent crystals of tetramethyldiamidodiphenylmethane CH2(CH ̧N CH3)2 are obtained, which are weighed, and from the weight the equivalent of formaldehyde is calculated.

Polymers of formaldehyde.-Formaldehyde may be met with in dietetic articles in a polymerized condition, either having become changed, or having been added in that form. In this condition, being insoluble, it will not respond to ordinary colour reactions. Under these circumstances the dimethylaniline test must be applied.

Saccharin. The liquid product or the aqueous or alcoholic extract of a solid is evaporated or distilled to remove alcohol, and precipitated with neutral lead acetate in the presence of free acetic acid. Excess of lead is removed from the filtrate with H2SO4, the PbSO, is filtered out, and the filtrate shaken out with three successive washings of ether. The ether is evaporated and the residue is taken up with 10 c.c. of 10 per cent. H2SO4, warmed on the water-bath and treated with a few drops

of saturated KMnO, solution until a permanent pink colour results. This treatment destroys all substances which might mask the taste of saccharin. The liquid is then shaken out with three successive portions of benzol, the solvent separated and evaporated. The residue, taken up with 2 c.c. of water, will give a characteristic sweet taste when 1 drop of saccharin is present. If this is found, the remainder of the aqueous solution is treated with 2 c.c. of 3 per cent. NaOH solution, and evaporated to dryness in a test tube. A thermometer is introduced to the bottom of the tube, which is then plunged in a bath of fusible metal and heated to 270°C. for one minute. The residue, after this treatment, is dissolved in 10 per cent. H2SO4, and the solution shaken out with benzine. The benzol is separated, filtered, and shaken up with 1 c.c. of neutral FeCl solution. In the presence of saccharin, a positive reaction for salicylic acid derived from the decomposition of that substance will be obtained.

Sucramine and saccharin derivatives. These bodies are salts of saccharin and do not pass into solution in the immiscible solvents used. It is therefore necessary to acidify the aqueous solutions with H2SO, before shaking out.

Dulcine, paraphenetolcarbamide, C,HÈO C,H,NH CONH, may be extracted directly with CHCl3. In the case of a liquid, PbCO3 is added, and the mixture is evaporated on the water bath to a thick paste. This is extracted with alcohol, and the solvent is evaporated off. The dry residue is treated several times with ether, which leaves dulcine in a pure crystalline condition on evaporation. It has the m.p. 173-174°C. When suspended in water and treated with a few drops of Hg2NO, solution free from HNO3, and heated for 10 minutes on the water bath, it gives a violet colour, which is deepened by adding a little PbO2. A trace of dulcine, heated with three or four drops of phenol and of strong H2SO4, then diluted with water, gives a blue colour zone when a little solution of AmOH is floated on the surface.

Aphrometry, or Determination of Frothing Power of Liquids. A. Bignon. (Nouveaux Remèdes, 23, 522.) Certain organic substances are known to give a very persistent froth on agitation, which is found to bear a very close relation to the amount of the froth-forming substance present, and to be capable of approximate determination. There being no expression to

denote this frothing property, the word aphrosity has been coined for it, and the frothing power of a liquid may be expressed in aphrometric degrees. The process is thus conducted. A known number of c.c. of the liquid to be tested is introduced into a 150 c.c. bottle, made up to 100 c.c., and well shaken for a minute exactly. The flask is then set aside, and the time being noted, the duration of the froth is recorded, the standard time being 30 minutes. If the foam breaks down before then, a fresh experiment, or a series, with more of the fluid is started, until the exact volume of liquid giving a froth persistent for 30 minutes is obtained. With some bodies, as for instance, tincture of quillaia, the smallest quantity only must be used, or a previously diluted solution of known strength. The number of c.c. of the solution taken divided into 100 gives the "aphrometric degree." For many substances, ordinary water may be used as the aphrometric liquid; but dilute acids increase the frothing power very markedly, and for delicate experiments acetic acid 15 per cent. gives better results. This is specially the case with urines. In the case of wine, two determinations may be made before and after the removal of the alcohol. The author gives the aphrometric degrees of urines, wines, beers, and a number of pharmaceutical tinctures.

Apocynum cannabinum, Constituents of, and a new Synthesis of Apocynin. H. Finnemore. (Proc. Chem. Soc., 24, 171.) Apocynin, CH10O3, has a faint vanilla odour and occurs in prisms, m.p. 115°C. It is convertible into a crystalline K-derivative, and forms methyl-benzoyl and acetyl derivatives with the m.p. 51°, 106° and 57°C. respectively. It forms an oxime, a well defined hydrazone, m.p. 126°, and a semi-carbazone, m.p. 167°C. It contains one methoxyl group. It is identical with synthetic aceto-vanillone having the constitution

The pharmacological properties of apocynin are under investigation.

The synthesis of acetovanillone from vanillic acid by Neitzel could not be confirmed, and that by Otto gave poor yields. The author has successfully synthetized it from vanillin by Grignard's method. Benzovanillin, CH2(OBz)(OMe)·CHO is