antimony trichlorides, but have not yet been obtained in quantity sufficient for investigation.

A. J. G.

Organo-bismuth Compounds: Valency of Bismuth. By A. MICHAELIS (Ber., 20, 52—54).-The author has entered on a general investigation of the organo-bismuth compounds to ascertain if any undoubtedly pentavalent bismuth compounds can be obtained. The alkyl compounds are being investigated by Marquardt, whilst the author and Polis are working on the aromatic compounds. The results obtained so far, are that whilst trimethyl- or triethyl-bismuthine do not form additive compounds with bromine nor with ethyl iodide, substitution of halogen for methyl, &c., occurring, triphenylbismuthine unites with chlorine or bromine, to form stable and well characterised compounds, in which 1 atom of bismuth is united with 5 monovalent groups or atoms. The pentavalent nature of bismuth is thus rendered as certain as that of antimouy.

A. J. G.

Triphenylbismuthine and its Derivatives. By A. MICHAELIS and A. POLIS (Ber., 20, 54-57).-Triphenylbismuthine, BiPh,, is prepared by boiling an alloy of bismuth and sodium (10 per cent.) with bromobenzene and a little ethyl acetate for 50 hours in a reflux apparatus. The product is filtered, the residue repeatedly extracted with benzene, and from the united liquids the benzene first distilled off, and then, under reduced pressure, the bromobenzene. The fused triphenylbismuthine left solidifies when cooled and stirred, and is purified by repeated crystallisation from hot alcohol. It forms colourless needles or tables which resemble the crystals of triphenylarsine and stibine, and seem to be triclinic. It has a sp. gr. = 1:5851 at 20°, melts at 82°, is readily soluble in ether and light petroleum, sparingly in cold alcohol, and closely resembles triphenylstibine in its chemical properties, save that it is less stable. When gently heated with hydrochloric acid, it is completely decomposed into benzene and bismuth chloride. When treated with bismuth bromide in ethereal solution, a yellow precipitate, probably phenylbismuthine dibromide, BiPhBr2, is obtained.

Triphenylbismuthine dichloride, BiPhaCl2, is prepared by passing chlorine gas on to the surface of a solution of triphenylbismuthine in light petroleum cooled with ice, as long as a white precipitate forms. It crystallises in thick prisms, melts at 140°, is readily soluble in benzene, sparingly in ether and cold alcohol, and is not decomposed by concentrated hydrochloric acid.

Triphenylbismuthine dibromide, BiPh,Br, is prepared by mixing solutions of its components in light petroleum; it forms long, paleyellow prisms, melts at 119°, and is readily soluble in benzene, sparingly in ether and alcohol. Both the bromide and chloride are decomposed by hydrogen sulphide, triphenylbismuthine being regenerated with separation of sulphur. A. J. G.

Synthesis of Phenaceturic Acid. By E. HOTTER (Ber., 20, 81-85). When glycocine, dissolved in as little water as possible, is

treated with sodium hydroxide and phenylacetic chloride and then with hydrochloric acid, a compound separates which, after drying and extraction with ether, dissolves in alcohol or water, and crystallises in the groups of white laminæ characteristic of phenaceturic acid. The yield is small. Ethyl phenaceturate, CHINO,Et, crystallises in long broad prisms, melts at 79°, and is readily soluble in hot alcohol, less soluble in warm ether and benzene, insoluble in carbon bisulphide.

Copper phenylacetate, (C,H,O2),Cu, is soluble in hot alcohol, and crystallises in small nodular masses consisting of radiating deepgreen needles. W. P. W.

Phenylamidopropionic Acid obtained from the Decomposition of Proteïds. By E. SCHULZE and E. NÄGELI (Zeit. physiol. Chem., 11, 201-206). One of the authors has previously shown (Abstr., 1883, 1122) that one of the products of the decomposition of proteïds by means of hydrochloric acid is a substance with the properties and composition of phenylamidopropionic acid, and that the same substance, probably again the result of the decomposition of proteïds, can be obtained from the germinating seeds of certain plants (Abstr., 1882, 189).

This substance is very similar to the amido-acid prepared synthetically by Erlenmeyer and Lipp (Abstr., 1882, 971), and called by them phenyl-a-amidopropionic acid or phenylalanine. This opinion is held because by dry distillation both acids yield the same decomposition products, namely, carbonic anhydride, phenyllactimide, and phenylethylamine. Still there are certain differences between them. The acid obtained from proteïds and plants crystallises from warm concentrated aqueous solutions in shining plates, from dilute solutions in slender, white needles, containing water of crystallisation; whilst phenylalanine crystallises from hot water in prisms free from water, and from aqueous alcohol in shining plates. The melting point of the former acid is 275-280°, of the latter 263-265°. The copper salt of the latter contains 2 mols. H2O, that of the former is free from water. Some of these differences might be explained on the supposition that the acid obtained from proteïds and plants is not so pure as that prepared synthetically. This would not, however, account for the difference of melting point. The optical examination of these substances led to the elucidation of this point. Phenylalanine prepared synthetically is optically inactive, whilst that obtained from proteïds and plants is active. In many pairs of isomerides, one of which is optically active, the other inactive, it is found that other differences, for instance, of solubilities and melting points, occur (see Abstr., 1886, 373). It is therefore concluded that the acid obtained from proteïds and plants is an optically active modification of phenyl-a-amidopropionic acid or phenylalanine. This conclusion is supported by the fact that tyrosin giving its typical microscopic appearances, and Piria's and Hoffmann's reactions, can be obtained from both acids by the two methods described by Erlenmeyer and Lipp. W. D. H.

Reduction of the Phthalic Acids. Constitution of Benzene. By A. BAEYER (Ber., 19, 1797-1810).-An account of the theoretical portion of this paper will be found in the March number of the Transactions (A. K. Miller, Trans., 1887, 208-215).

When tere- and iso-phthalic acids are reduced with sodium amalgam with the aid of heat, they both yield tetrahydro-acids, no intermediate dihydro-acid being formed; no reduction takes place in the cold. Phthalic acid, on the other hand, is reduced by sodium amalgam, slowly in the cold, more readily when warmed, to dihydrophthalic acid. The author explains this difference of behaviour on the assumption that only those double affinities (taking Kekulé's formula) can be reduced which belong to carbon-atoms in combination with carboxyl.

The

It is remarkable that the partly reduced acids behave towards bromine like unsaturated compounds; dihydrophthalic and tetrahydroisophthalic acids each take up two atoms of bromine. dibromide of tetrahydroterephthalic acid yields a hydroxy-acid, from which by the action of bromine, a compound identical with tetrabromocatechol was obtained. Tetrahydroterephthalic acid has therefore the constitution

[H·COOH : H : H : H·COOH : H2 : H2 = 1 : 2 : 3:4: 5: 6].

Tetrahydroterephthalic acid is prepared by boiling for 20 hours 5 grams of terephthalic acid, dissolved in the smallest amount of aqueous soda, and gradually adding 4 per cent. sodium amalgam (500 grams). It dissolves in 120 parts of boiling water, and separates on cooling in small prisms. It melts above 300° and sublimes. The silver salt is amorphous. The methyl salt is obtained as an oil which has the odour of fennel, and solidifies to very large prisms melting at 39°; its solutions have a blue fluorescence. When the ethereal solution of the salt is treated with sodium ethoxide, a rose-coloured transient precipitate is formed, similar to that obtained from ethyl succinosuccinate. When tetrahydroterephthalic acid is heated with hydriodic acid (b. p. 127°) for six hours at 240°, it is converted into hexahydroterephthalic acid, C.H2O. This is less soluble than the tetrahydro-acid, from which it is also distinguished by its stability towards potassium permanganate. It melts at about 295°, and sublimes. The methyl salt melts at 58°; its solution has no fluorescence. Tetrahydroisophthalic acid is prepared by boiling 2 grams of isophthalic acid for two to three days with sodium amalgam. It dissolves readily in hot water, from which it crystallises in needles melting at 199°. The silver salt is amorphous; the methyl salt is an oil.

The author confirms the statement of Graebe and Born (Annalen, 142, 345), that dihydrophthalic acid cannot be further reduced by sodium amalgam. It is best prepared by boiling phthalic acid with sodium amalgam; the yield is quantitative.

Dibromohexahydroterephthalic acid, CH10Br2O, + H2O, is formed when 10 grams of tetrahydroterephthalic acid is shaken with 100 grams of a 5 per cent. ethereal solution of bromine, left for some time, and the ether poured off. This is repeated with fresh bromine solution until almost all the acid has dissolved. The united solutions

are decolorised with sulphurous acid and extracted with soda. The methyl salt crystallises from ether in large prisms, melting at 73°. The acid has about the same solubility in water as tetrahydroterephthalic acid; it is not decomposed when the solution is boiled for a short time. Warm soda solution decomposes it with formation of a dihydroterephthalic acid, CH,O,. When the dibromo-acid is treated with silver oxide, an acid is obtained, probably a dihydroxyhexahydroterephthalic acid.

According to Kekulé's formula, the two bromine-atoms and hence also the hydroxyl-groups, would both have the ortho-position:[H.COOH : H2 : H2: H·COOH : HBr : HBr = 1:2:3:4: 5: 6].

This position of the hydroxyl was proved by treating the dihydroxyacid with bromine; tetrabromocatechol was obtained, which gave an intense blue coloration with ferric chloride.

Dry hydrophthalic acid combines readily with bromine (1 mol.), yielding dibromotetrahydrophthalic acid. The acid is subjected in small portions to the action of bromine-vapour, the product freed from excess of bromine, dissolved in soda, filtered, and acidified. The oil is exhausted with ether, and distilled in a vacuum; it solidifies to well-formed rhombohedra (compare Graebe and Born, loc. cit.).

N. H. M.

Action of Potassium Hydroxide on Mixed Alkyl Bisulphides. By R. Orro and A. RöSSING (Ber., 20, 189—191).—When an alcoholic solution of ethyl phenyl bisulphide is warmed with small quantities of alcoholic potash until the reaction is complete, the following compounds are obtained :-Ethylsulphinic and benzenesulphinic acids, ethyl mercaptan and phenyl bisulphide; the latter is formed by the oxidation of the potassium phenylmercaptide first formed. W. P. W.

Action of Sulphurous Anhydride on Benzene. By C. E. COLBY and C. S. McLOUGHLIN (Ber., 20, 195-198).-When 100 grams of benzene and 35 grams of aluminium chloride, placed in a reflux apparatus, are treated with sulphurous anhydride until absorption of the gas is complete, then heated until hydrogen chloride ceases to be evolved, and the product, after cooling, poured into water, diphenylsulphoxide, Ph2SO, is obtained, which crystallises from xylene in small, transparent crystals, which seem to be triclinic. It melts at 70-71°, and is readily soluble in alcohol, ether, acetic acid, and benzene, sparingly soluble in cold light petroleum. When oxidised with potassium permanganate, it yields diphenylsulphone, and when reduced with sodium, diphenyl sulphide is obtained. The same compound can be prepared by the action of thionyl chloride on benzene. For this purpose, 50 grams of benzene and 16 grams of thionyl chloride, placed in a reflux apparatus, are treated in the cold with successive quantities of aluminium chloride until hydrogen chloride ceases to be evolved; the product, after heating for half an hour, is poured into water, and the diphenylsulphoxide purified by crystallisation.

When equivalent quantities of diphenylsulphoxide and sodium nitrate are dissolved in sulphuric acid, and the two solutions mixed and heated at 100°, three nitro-compounds appear to be formed:(1) Gericke's dinitrosulphobenzide, melting at 163° (Annalen, 100, 211); (2) an oil not yet further examined; and (3) dinitrodiphenylsulphoxide, (C,H,NO2)2SO, which forms yellow, indistinct, microscopic crystals, melts at 116°, and is soluble in alcohol, readily soluble in ether, benzene, acetic acid, and carbon bisulphide.

W. P. W.

Action of Potassium Hydroxide on Phenylenemetadiphenylsulphone. By R. OTTо and A. RöSSING (Ber., 20, 185—189).— Since ethylidenediethylsulphone and ethylidenediphenylsulphone are not attacked by alcoholic potash at 140° (Ber., 19, 2814), whilst ethylenediphenylsulphone readily yields benzenesulphinic acid and. phenylsulphonethyl alcohol (Abstr., 1885, 261), the authors have examined the behaviour under like conditions of a disulphone of a third class, phenylenemetadiphenylsulphone, and find that the products of the action are benzenesulphinic acid and diphenylsulphonephenyl ether, 2C,H,(SO2Ph), +2KOH= 2PhSO2K +(C‚É‚SØ2Ph),O + H2O.

Diphenylsulphonephenyl ether, obtained by heating phenylenemetadiphenylsulphone with alcoholic potash at 160-170°, crystallises in small white feebly lustrous needles, melts at 69-70°, and is readily soluble in alcohol, ether, and benzene, insoluble in water. It can be volatilised unchanged only when carefully heated in small quantities, and distils above 200°. When heated with concentrated aqueous ammonia, no change occurs. Sodium amalgam in alkaline solution reduces it to benzene, benzenesulphinic acid, and a crystalline compound, which yields a bromo-derivative crystallising in large rhombic tables melting at 54-55°.

Ethylenediphenylsulphone when treated with an excess of concentrated aqueous potash, yields benzenesulphinic acid, and a small quantity of a compound possibly polymeric with diphenylsulphonethyl ether are obtained. W. P. W.

Synthesis of Carbazole. By A. GOSKE (Ber., 20, 532-534).— The author has succeeded in effecting the synthesis of carbazole from thiodiphenylamine. When this base is heated with an excess of metallic copper for two hours in an atmosphere of coal-gas, and the product afterwards distilled, a distillate is obtained which consists of carbazole; the yield amounted to 60 per cent. of the weight of the thiodiphenylamine employed. The synthetical carbazole melts at 238°, and gives the characteristic compound with trinitrophenol. Bearing in mind Bernthsen's synthesis of thiodiphenylamine from catechol and orthamidophenylmercaptan (this vol., p. 245), the constitution of carbazole must be represented by the formula