only at a somewhat elevated temperature. With increasing temperature derivatives of malic acid have their specific rotation displaced towards the left, this behaviour being characteristic of a-hydroxy-acids of the l-series. With aspartic acid, the influence of temperature is irregular and frequently unusually pronounced, probably owing to association. For the same reason, the method of obtaining the specific rotation of the homogeneous substance by extrapolation from the specific rotation of its solutions of gradually increasing concentration is only partly successful with derivatives of aspartic acid. In general, the authors are of the opinion that comparisons of the specific rotations of substances can be made only with dissolved or supercooled materials if they exhibit the closest chemical similarity. Clough's observation (J.C.S., 1918, 113, 526) that ethyl esters of a-hydroxy-acids belonging to the d-series are more highly dextrorotatory than the methyl esters cannot be immediately extended to a-amino-acids.

The biological significance of the doctrines of configuration is discussed at length. The direct production of l-malic acid by oxidation of dextrose is inconceivable, since d-malic acid must result if the configuration remains intact. The probable precursor of malic acid is oxalacetic acid, from which it is derived by asymmetric hydrogenation. The suggested course of the change is dextrose saccharic acid ketipinic acid (cf. Franzen and Schmitt, 'A., 1925, i, 231), and thence by benzilic acid transformation to citric acid, and by acid fission and reduction to malic acid, whereby the general simultaneous occurrence of the two acids is explained. A second argument in favour of the view that oxalacetic acid is the precursor of malic and aspartic acids is based on the configurational similarity of lactic acid formed by hydrogenation of pyruvic acid in the organism with alanine produced from the same substance in the presence of ammonia. The steric similarity of both pairs renders it probable that they are formed in a similar manner from ketonic substances.

The following new compounds are incidentally described ethyl cinnamoyl-1-malate, b. p. 195°/2 mm.; ethyl formyl-1-malate, b. p. 120-121°/2 mm.; ethyl B-phenylpropionyl-l-malate, b. p. 185-186°/3 mm.; ethyl ethanesulphonyl-l-malate, b. p. 154-155°/0-5 mm.; methyl ethanesulphonyl-l-malate, b. p. 167168/1 mm.; ethyl p-toluenesulphonyl-1-malate, b. p. 197-198°/1 mm.; acetyl-l-maldi-methylanilide, m. p. 116°; acetyl-1-maldianilide, m. p. 177°; ethyl cinnamoyll-aspartate, m. p. 72°; ethyl formyl-l-aspartate, b. p. 145-146°/1 mm.; ethyl B-phenylpropionyl-l-aspartate, b. p. 203°/2 mm., m. p. 34°; ethyl acetyl-1-aspartate, b. p. 183°/20 mm., m. p. 31°; methyl acetyl-1-aspartate, b. p. 154-155°/15 mm., m. p. 65°; ethyl n-heptoyl1-aspartate, p. 29°; ethyl ethanesulphonyl-laspartate, m. p. 50°; ethyl p-toluenesulphonyl-1aspartate, m. p. 79°; p-toluenesulphonyl-l-aspartic acid, m. p. 139-140°; p-toluenesulphonyl-1-aspartyl chloride, m. p. 96-97°. The specific rotations of these compounds in the liquid state and in solution. are recorded for the wave-lengths 637, 578, 546, and 436 at about 20° and 100°.

m.

The paper concludes with a reply to the recent criticisms of Kuhn and Ebel (A., 1925, i, 1378). H. WREN.

Catalytic decomposition of amides. A. MAILHE (Bull. Soc. chim., 1925, [iv], 37, 1394-1397).Amides undergo catalytic decomposition when passed over finely divided nickel at 400-440°, yielding carbon monoxide and an amine containing one carbon atom less than the original amide. In the case of aromatic amides, such as benzamide, the amine is stable and aniline is the principal liquid produced. Traces of benzene are also formed, whilst the gaseous products in this case consist of carbon dioxide 30%, carbon monoxide 25%, and hydrogen 35%. With the aliphatic amides, the amine primarily produced undergoes further catalytic decomposition on the nickel (cf. A., 1924, i, 623), yielding a nitrile, the next lower homologue to that produced in catalytic dehydration (Boehner and Ward, A., 1917, i, 11). Thus valeramide at 430° yields isobutyronitrile and a gaseous mixture containing carbon dioxide 12%. carbon monoxide 31%, hydrogen 45%, and 12% of olefines, together with some ammonia, the latter resulting from the decomposition of the amine. Propionamide at 430° similarly yields a gas containing ammonia, carbon dioxide 9%, carbon monoxide 31%, hydrogen 51%, and 9% of olefines, whilst the liquid product consists of acetonitrile. Acetamide is similarly decomposed into ammonia, hydrogen cyanide, carbon dioxide (5% at 440°), carbon monoxide (15%), and hydrogen with deposition of carbon.

R. BRIGHTMAN.

Reaction between sodium hypobromite and carbamide. M. B. DONALD (J.Č.S., 1925, 127, 2255-2259).—The reaction between carbamide and

hypobromite solutions, prepared by bromination of sodium hydroxide in varying degrees up to 100%, has been studied both by measurement of the evolved gas and by titration of the excess hypobromite with thiosulphate. The nitrogen deficit becomes greater as the percentage bromination increases until the hydroxide is 75% brominated. For brominations above 75%, the nitrogen deficit still increases, but the

carbamide taken.

thiosulphate figure indicates that the hypobromite is being used up in a slow side oxidation, so that for 80% hypobromite and upwards the thiosulphate required after 30 min. is about equivalent to the With completely brominated sodium hydroxide, however, there is no gas evolution and no hypobromite is used up. The secondary oxidation is attributed to formation of sodium cyanate (cf. Fenton, J.C.S., 1878, 33, 300; Lescœur, A., 1920, ii, 201; Margosches and Rose, A., 1923, ii, 590). This is increasingly formed as the ratio of bromination is reached, when the cyanate is slowly hypobromite to hydroxide increases until 75% oxidised, an action which, however, does not take place in absence of carbamide. The use of 10Nsodium hydroxide solutions for the determination of carbamide gasometrically leads to inaccurate results.

A. DAVIDSON.

New hydrolysis product from elastin. R. ENGELAND (Biochem. J., 1925, 19, 850-852).Hydrolysed elastin is precipitated with 50% phospho

16

tungstic acid, and the precipitate treated with lead hydroxide and then decomposed with barium hydroxide. After removing traces of lead and barium and evaporating in a vacuum to a syrup, the preparation is boiled with copper hydroxide and filtered. The filtrate is then concentrated and methyl alcohol is added. The copper salt, C6H22O5N,Cu, separates and, when decomposed with hydrogen sulphide, yields hyphasmine, C16H24O5N2. This compound gives a pink colour with Millon's reagent. It is discoloured at 200° and softens at about 220°. S. S. ZILVA. Double sulphates and chromates of guanidine and bi- and ter-valent metals. G. CANNERI (Gazzetta, 1925, 55, 611-615).-Double guanidine sulphates of the type (CHN), H2SO,,XSO4,6H2O have been prepared, in which X = Mg, Zn, Cd, Fe, Ni, Co, Mn, Cu, UO2; mixed crystals of the zinc and nickel compounds were also prepared. Magnesium guanidine chromate, (CHN)2,H2CrO4, MgCrO4,6H2O, is isomorphous with the corresponding sulphate. Chromium guanidine alum,

(CHN)2,H2SO4, Cr2(SO4)3,12H2O, forms mixed crystals with the corresponding aluminium guanidine alum; ferric guanidine alum and vanadium guanidine alum are described.

T. H. POPE.

Action of nitriles on organo-magnesium compounds. Chloroacetonitrile. L. MATHUS (Bull. Soc. chim. Belg., 1925, 34, 285-289).a-Chlorobutyronitrile reacts with magnesium ethyl bromide, forming an imine (De Booseré, A., 1923, i, 311), but with chloroacetonitrile, the reaction is complicated by the formation of polymerides, from which definite compounds could not be isolated. Polymerisation and hydrocarbon formation are more marked in the case of magnesium alkyl halides than with magnesium phenyl bromide. From the reaction product of 11 mols. of the latter and 1 mol. of chloroacetonitrile, the only products isolated were diphenyl, m. p. 70-71°, and chloroacetophenone, m. p. 58°, b. p. 140-143°/15 mm. (yield 14%), the normal product of the Blaise reaction. The latter compound, when distilled in a current of steam, yielded benzoyl carbinol, m. p. 72-73°; semicarbazone, m. p. 185°. The formation of diphenyl is explained on the assumption of the intermediate formation of the dinitrile of succinic acid, which could not be isolated on account of polymerisation.

J. S. H. DAVIES.

Reduction of some aliphatic cyano-compounds. Reduction of cyanoacetylcarbamide. H. RUPE, A. METZGER, and H. VOGLER (Helv. Chim. Acta, 1925, 8, 848-852).-Catalytic reduction of cyanoacetylcarbamide in aqueous solution at 60-70° yields uracil, which results from an internal condensation of the aldehyde produced by hydrolysis of the aldimine, NH2 CO-NH CH, CH:NH, the first product of the reduction. B-Keto-a-phenylbutyronitrile is reduced to benzyl methyl ketone, b. p. 98-100°/10 mm., and a little y-keto-3-phenylbutaldimine, m. p. 96°. Propionitrile resists reduction and is recovered unchanged (cf. A., 1923, i, 1088). n-Valeronitrile gives a good yield of di-n-amylamine, b. p. 95-97°/12 mm. (nitrosoamine, hydrochloride,

decomp. 275°, and oxalate, m. p. 210-211°, are described). G. M. BENNETT.

minium ethyl and diethyl iodide. V. GRIGMixed organo-aluminium compounds. AluNARD and R. L. JENKINS (Bull. Soc. chim., 1925, [iv], moisture, finely divided aluminium reacts readily 37, 1376-1385).-In the absence of oxygen and with ethyl iodide. Solvents such as benzene, ether, or chloroform give less satisfactory results. Small quantities of ethane and ethylene are produced, but the main product is a mobile liquid, which decomposes below 300° under ordinary pressure, but on fractional distillation under reduced pressure yields aluminium diethyl iodide, AlEt,I, b. p. 118-120°/4 mm., d1-6091, and aluminium ethyl di-iodide, AlEt12, m. p. 35-37°, b. p. 158-160°/4 mm. Both compounds are spontaneously inflammable in air, and react violently with water, yielding ethane. Mol. wt. determinations indicate a double molecule. They dissolve in ether with liberation of heat, yielding the etherates, AlEt21,0.75Et2O, and AlEtI,Et2O (cf. Krause and Wendt, A., 1923, i, 289), which appear to be stable in dry air but react violently with water and are decomposed by alcohol. Neither compound reacts with carbon dioxide; with acetone, they yield mesityl oxide and more complex products, which were not examined. Benzaldehyde similarly gives complex condensation products of high mol. wt. Both compounds reduce mercuric chloride.

Distillation of the reaction product of aluminium and ethyl iodide at 285-300°/760 mm., yields a mixture of ethylene and ethane, the proportion of ethylene increasing as the decomposition proceeds, whilst in the final stages the saturated hydrocarbon produced differs from ethane. On similar distillation, the etherates yield a little ether, ethyl iodide, and at 90-110°, a colourless, inflammable liquid, containing iodine, which reacts vigorously with water. With ethyl bromide, aluminium similarly yields a colourless liquid, separated into two fractions, b. p. 118-120°/ 15-16 mm., and 121.5-124.5°/15-16 mm., the compositions of which agree only approximately with AlEt,Br and AlEtBr2, respectively, possibly through the presence of AlEt. R. BRIGHTMAN.

in

G. H.

Applications of thallium compounds organic chemistry. II. Titrations. CHRISTIE and R. C. MENZIES (J.C.S., 1925, 127, 2369-2373). Thallous salts of organic acids have been prepared by titration of the acids with standard thallous hydroxide solution (cf. Menzies and Wilkins, A., 1924, i, 704). With tartaric acid, using four equivalents of thallous hydroxide, a tetrathallium tartrate is formed in which two of the thallium atoms are hydrolysable by carbon dioxide in aqueous solution. Attempts to prepare the corresponding tetrapotassium tartrate failed. B-Diketones and phenols also yield well-crystallised thallium salts, easily prepared pure by addition of a concentrated aqueous solution of thallous hydroxide to an alcoholic solution of the phenol etc. The thallium in these salts is hydrolysed by water and may be accurately determined by titration. The usefulness of both classes of thallium salts for synthetic purposes is indicated. The following salts are described: Thallous fumarate, m. p.

268° (decomp.); thallous maleate, m. p. 164—166°; thallous succinate, m. p. 246-248°; thallous phthalate, m. p. 268-270°; thallous stearate, m. p. 119°, palmitate, m. p. 115°, and oleate, m. p. 78-82° (these three compounds show double m. p. and appear to be anisotropic liquids between these points); thallous salts of ethyl acetoacetate, m. p. 91-92°, benzoylacetone, m. p. 103-105°, phenol, m. p. 231-235°, m-cresol, m. p. 187°, resorcinol monomethyl ether, m. p. 146-148°, guaiacol, m. p. 160-161°, a-naphthol, m. p. 180-190°, and vanillin, m. p. 193-201°.

A. DAVIDSON.

two different acids are obtained of the formula

CH2 ¢HH CO,H
CH,CHCHÍCHÍCH, CO,H

(III).

If the symmetrical formula for dicyclopentadiene is to be retained, these acids must be represented as cisand trans-isomerides. Attempts at interconversion of the two acids or their esters met with little success, and the exact spatial representation is still in doubt. Ketodihydrodicyclopentadiene, b. p. 102°/12 mm., exists in two forms, (a) m. p. 83-84°, and (b) a liquid, which are readily interconvertible. As no colour is obtained with ferric chloride, the possibility of ketoenolic tautomerism is rejected and two alternative explanations are given. The semicarbazone, m. p. 211°, dibromide, m. p. 142°, and tribromo-derivative, m. p. 186°, of this ketone are described. Reduction with palladium and hydrogen yields ketotetrahydrodicyclopentadiene (I), b. p. 100°/12 mm., which on oxidation with nitric acid yields the (?) trans-acid (III), m. p. 200°.

Dihydrodicyclopentadiene dibromide, m. p. 62.5°, yields the series: acetate of the bromohydrin, m. p. 83.5°; glycoldiacetate, b. p. 182-184°/20 mm.; cisglycol (II), m. p. 80°; and the (?) cis-acid (III), m. p. 133.5° (dimethyl ester, b. p. 134-135°/13 mm.). Dihydrodicyclopentadiene oxide,

CH2- -CH-CH- -CHCH2·CH2·CH·CH·CH2·CH ́

2

m. p. 91.5°, is hydrolysed to the trans-glycol (II), m. p. 118°, which yields the (?) cis-acid (III) on oxidation. Other substances described are: dicyclopentadiene oxide, m. p. 64°; dicyclopentadiene glycol, m. p. 120°; dicyclopentadiene dioxide, m. p. 165°; the oxime of diethylaminoketodihydrodicyclopentadiene, m. p. 96-97°; and a substance,

m.

CH[CH CH>CH-|NH,

p. 201-202°.

CH·C(N⋅OH)

12

R. W. WEST.

Action of sulphuryl azide on benzene. K. F. SCHMIDT (Ber., 1925, 58, [B], 2409-2412; cf. Curtius and Schmidt, A., 1922, i, 776).-The "pseudoaniline formed by the action of sulphuryl azide on a large excess of benzene at 140° is identified as pyridine. H. WREN.

Syntheses in the p-cymene series from isopropyl alcohol. I. propyl alcohol. I. Syntheses of p-cymene. L. BERT (Bull. Soc. chim., 1925, [iv], 37, 1252—1270). -To confirm Widman's work (A., 1891, 686) on the constitution of p-cymene the author has attempted the synthesis using Grignard reagents. p-Bromotoluene was obtained in 12% yield by the action of methyl sulphate on magnesium p-bromophenyl bromide, and in 30% yield from p-toluidine by Sandmeyer's method, the Gattermann process giving a 27% yield. Magisopropyl sulphate gives p-cymene in 10% yield. nesium p-tolyl bromide obtained from this, with Similarly, cumene, obtained in 75% yield by the action of isopropyl bromide on benzene in the presence of aluminium chloride, is converted by Jacobsen's method (Ber., 1879, 12, 430) into p-bromocumene (yield 83%), b. p. 111°/28 mm., 216-217°/729 mm., da 1.289, na 1-539, and the magnesium cumyl bromide obtained from this gives with methyl sulphate a 12% yield of p-cymene. A third synthesis was effected by treating cumene with trioxymethylene and hydrogen chloride in the presence of zinc chloride (cf. Blanc, A., 1923, i, 549), when cuminyl chloride, CH,CICH CHMe2, b. p. 228°, d 1.020, n 1.523, is obtained in 75% yield, which on treatment with magnesium in ether and decomposition of the Grignard reagent with water yields 74% of the theoretical amount of p-cymene, together with some dicuminyl. The p-cymene obtained by these last two syntheses gives identical constants in close agreement with those of p-cymene obtained from thymol by converting this into 3-bromo-p-cymene with phosphorus pentabromide, and decomposing the Grignard compound obtained from this with water. Pure p-cymene accordingly has b. p. 175-176°/735 mm., d 0.858, n 1.493. R. BRIGHTMAN.

Syntheses in the p-cymene series starting from isopropyl alcohol. II. Syntheses with magnesium p-isopropylphenyl bromide. L. BERT (Bull. Soc. chim., 1925 [iv], 37, 1397-1410; cf. preceding abstract).-Magnesium p-isopropylphenyl bromide is much more readily prepared than the corresponding compound from p-bromotoluene, the reaction being catalysed initially by the addition of a few drops of bromine. Decomposition with water after 1 hrs. showed that 80% of the p-bromoabout 15% being converted into pp'-diisopropylcumene had been converted into the Grignard reagent, bromide, the following substances have been prepared diphenyl. From magnesium p-isopropylphenyl by the usual Grignard reaction methods: p-cumenol, (methyl ether, b. p. 207-5-208°/728 mm., 100°/22 b. p. 119°/19 mm., 226-227°/729 mm. (yield 18%) mm., d 0.955, n 1-513; ethyl ether, b. p. 222°/728 mm., 113-114° /22.5 mm., d5 0.938, n5 1·506); p-cumenic acid (yield 40%), m. p. 116°, p-cumyl alcohol, b. p. 144-145°/31-5 mm., 244-245°/735 mm.,

8.5

d2 0.978, n 1-522 (in 35% yield from trioxymethylene), and p-cumaldehyde, b. p. 110°/14 mm. (yield 33%, on hydrolysis of the acetal first formed from ethyl orthoformate). The acetal of p-cumaldehyde, b. p. 142-143°/18.5 mm., 263°/724 mm. (slight decomposition), d 0-944, ni 1-484, is new. Glycol monochlorohydrin gives a 90-100% yield of B-p-isopropylphenylethyl alcohol, b. p. 155°/29 mm., di 0-969, ng 1-521 (acetate, b. p. 143-146°/17 mm., d 0-986, n 1-499). p-Cumenic acid is better prepared by the action of ethyloxalyl chloride on cumene in the presence of aluminium chloride and carbon disulphide; ethyl cumenylglyoxylate, CH Pro-CO.CO2Et, b. p. 177°/17-5 mm., 180/20 mm., d2 1.047, n 1.516, is thereby obtained in 75% yield, hydrolysed with 10% sodium hydroxide, and the p-cumenylglyoxylic acid on heating with concentrated sulphuric acid is converted quantitatively into p-cumenic acid. Since an improved procedure enables ethyloxalyl chloride to be obtained by Bouveault's process (A., 1896, i, 551) in consistent yields of 80%, this process affords a technical method of synthesising p-cumenic acid. The phosphorus pentachloride is prepared in situ from the trichloride and the mixture with ethyl oxalate heated at 135140° until phosphoryl chloride ceases to distil, when the temperature is raised to 165-170°. The following esters of p-cumenic acid are described: methyl, b. p. 126°/14 mm., de 1-018, n 1.515; ethyl, b. p. 134-5 13.5 mm., d 1.000, n 1.508; propyl, b. p. 148°/14 mm., d 0.981, n 1-503; isopropyl, b. p. 138°/14 mm., d 0.978, n 1.500; butyl, b. p. 161-1620/14 mm., dis 0-970, n 1-501; isobutyl, b. p. 155°/14 mm., di 0-966, n 1.497; isoamyl, b. p. 173°/16.5 mm., dis 0.961, n 1.497; isooctyl, b. p. 199°/16.5 mm., da 0.938, n 1-491; benzyl, b. p. 218°/16.5 mm., d 1.059, n 1.554; cumyl, b. p. 243°/16.5 mm., da 1.023, n 1.545. p-Cumenylglyoxylic acid reacts readily with boiling aniline, yielding first an acid imide, CH Pr CO CO2H+PhNH2 →→ H2O+ CH ̧Pr3·C(:NPh)•CO2H, which subsequently decomposes into carbon dioxide and p-cumenylimide, CHMe C&H CH:NPh; this on hydrolysis with 25% sulphuric acid gives p-cumaldehyde in nearly theoretical yield. The process affords a technical means of synthesisng the aldehydes.

4

19

19

195

R. BRIGHTMAN.

Reactions of compounds of triphenyl and triphenylsilicyl in liquid ammonia. C. A. KRAUS and R. ROSEN (J. Amer. Chem. Soc., 1925, 47, 27392748). Sodium and potassium triphenylmethyls, prepared by the action of the respective metals on triphenylmethane in liquid ammonia solution, have equivalent conductances, in 0.05 and 0.08N-solution in liquid ammonia, of 64 and 32, respectively. They are regarded as true salts which undergo normal ionisation in the above solvent (cf. Schlenk and Marcus, A., 1914, i, 823). In liquid ammonia solution, the salts absorb two atoms of oxygen per molecule of salt, a product being precipitated which appears to contain the metal in combination with the organic residue, but which, after warming to the ordinary temperature, consists, in the case of the sodium salt, of sodium peroxide and an ether-soluble oxide of triphenylmethyl. In the case of the potassium

salt, the product initially precipitated, after being kept for some time at the ordinary temperature, absorbs a further molecule of oxygen when redissolved in liquid ammonia, with formation of potassium tetroxide. When potassamide is treated in liquid ammonia solution with triphenylmethane, immediate and practically complete formation of potassium triphenylmethyl takes place. Sodamide behaves similarly, but whilst the potassium salt is stable when isolated by evaporation of the solvent, the sodium salt changes rapidly as the temperature is raised, and the corresponding calcium salt is unstable even at low temperatures (cf. Schlenk and Ochs, A., 1916, i, 379; Kraus and Kawamura, A., 1924, i, 276). These phenomena indicate that triphenylmethane behaves, in liquid ammonia, as an acid of such strength that the salt of the strong base, potassium, is only slightly hydrolysed ("ammonolysed "), whilst salts of the weaker bases, sodium and calcium, are decomposed in this way to a considerable extent. Potassium triphenylmethyl crystallises from liquid ammonia free from solvent, whilst the sodium salt prepared similarly contains 1 mol. of ammonia of crystallisation. Triphenylmethane forms a sensitive indicator in liquid ammonia solution, giving intense red colorations in presence of metallic amides (bases), and colourless solutions in the pure solvent or in presence of ammonium salts (acids).

Triphenylmethyl chloride combines with pyridine (1 mol.) with formation of an additive compound, m. p. 173-174°, and with aniline (1 mol.), yielding an additive compound, m. p. 189-190°. When treated with liquid ammonia, or with ammonia in ethereal solution, triphenylmethyl chloride forms a complex compound, which decomposes when heated into triphenylmethylamine and ammonium chloride. When triphenylmethyl chloride is dissolved in liquid ammonia, it is hydrolysed to a slight extent, reversibility of the hydrolysis being indicated by the formation of triphenylmethyl when ammonium chloride is added to a solution of triphenylmethylamine in liquid ammonia. Triphenylmethylamine reacts with sodium and potassium in liquid ammonia solution, with formation of the metallic amide and salt of triphenylmethyl.

Triphenylsilicyl chloride undergoes rupture at the phenyl-silicon linking when treated with metallic sodium in liquid ammonia, in which the chloride is only slightly soluble. Definite compounds could not be isolated from the reaction product. Triphenylsilicyl chloride combines with 2 mols. of ammonia, the product decomposing into triphenylsilicylamine, m. p. 55-56°. F. G. WILLSON.

"Tervalent " carbon. III. Pentaphenylcyclopentadienyl. K. ZIEGLER and B. SCHNELL (Annalen, 1925, 445, 266-282; cf. A., 1924, i, 850).2:3:4: 5-Tetraphenyl-A2:4-cyclopentadiene (cf. Wislicenus and Carpenter, A., 1899, i, 60), of which an improved method of preparation is described, condenses with p-nitrosodimethylaniline with formation of 2:3:4: 5-tetraphenyl-A2:4-cyclopentadienone-p-dimethylaminoanil, violet-black, m. p. 224-226°. This is hydrolysed by boiling hydrochloric acid to 2:3:4:5-tetraphenyl- 42:4-cyclopentadien - 1 - one,

blackish-violet, m. p. 217-218°, which is converted by the action of excess of magnesium phenyl bromide into 1:2:34:5-pentaphenyl-A2:4-cyclopentadien-1-ol, colourless, m. p. 175-176°. This affords 1-chloro1:2:3:4:5-pentaphenyl-42:4-cyclopentadiene, yellow, m. p. 167°, when treated with hydrogen chloride in boiling acetic acid. The corresponding bromo-derivative, yellow, m. p. 188-189°, is obtained similarly; it is much less readily hydrolysed than the corresponding triphenylmethyl bromide. Both the chloroand bromo-derivatives yield pentaphenyl-A2: 4-cyclopentadiene, m. p. 244-246°, when treated with zinc dust in glacial acetic acid solution. The mixture of benzamarones produced by Knoevenagel's method (A., 1893, i, 352) is slowly reduced by zinc dust in glacial acetic acid solution to 3: 4-dihydroxy-1: 2:3:4:5pentaphenylcyclopentane, m. p. 241° (cf. Wislicenus, loc. cit.). This is very resistant to dehydrating agents, but is converted by concentrated sulphuric acid, in boiling acetic acid solution, into the above pentaphenylcyclopentadiene, m. p. 250°. Treatment of the bromopentaphenylcyclopentadiene with molecular silver in benzene affords 1:2:3:4:5-pentaphenylA2:4-cyclopentadienyl, reddish-violet, m. p. 260°. The free radical is only slightly less stable in air than the tetra-arylallyl radicals (cf. A., 1924, i, 308). It absorbs oxygen in nitrobenzene solution, but the resulting peroxide is itself further oxidised.

F. G. WILLSON.

Polarity theories and four-membered rings. Non-existence of 2:3: 3-triphenylmethylene1:2-oxaimine. G. N. BURKHARDT, A. LAPWORTH, and J. WALKDEN (J.C.S., 1925, 127, 2458-2461). The interaction of nitrosobenzene and as-diphenylethylene yields a compound to which Ingold and Weaver (A., 1924, i, 1116) gave an oxaimine formula; this is in reality the N-phenyldiphenylnitrone, CPh2:NPh:O, a methylene group having been lost in the reaction. Thus the facts are found to be in accord with the theory of alternate polarities, as in the other cases investigated (cf. A., 1925, ii, 937). B. W. ANDERSON.

Influence of nitro-groups on the reactivity of substituents in the benzene nucleus. VIII. 2:3- and 2:5-Dinitro-p-chlorotoluenes. J. KENNER, C. W. TOD, and E. WITHAM (J.C.S., 1925, 127, 2343—2349). The action of methyl-alcoholic ammonia at 150° on 4-chloro-2: 3-dinitrotoluene, m. p. 106.5° (prepared from 2: 3-dinitro-p-toluidine), gave a 70% yield of 4-chloro-2-nitro-m-toluidine, m. p. 53°; acetyl-derivative, m. p. 210-212°. 4-Chloro-2: 5-dinitrotoluene was similarly converted into 20% and 30%, respectively, of 4-chloro-6-nitro-m-tolyl methyl ether, m. p. 122°, and 4-chloro-6-nitro-m-toluidine, m. p. 121; acetyl-derivative, m. p. 135°. To verify completely the constitution of these compounds they were synthesised from 4-chloroaceto-m-toluidide, and the alternative bases, 4-chloro-3-nitro-o-toluidine (m. p. 60-62.5°; acetyl-derivative, m. p. 195°) and 4-chloro-5-nitroo-toluidine (m. p. 164°; acetyl-derivative, m. p. 182°) from 4-chloroaceto-o-toluidide. 4-Chloro-5-nitro-otoluidine yielded on reduction 5-chloro-2-methylp-phenylenediamine, m. p. 146°. Thus in each case the influence of the methyl group on the mobility of

the m-nitro-group is seen to be stronger than that of the chlorine atom, in contrast with the results obtained by the study of direct substitution. B. W. ANDERSON.

con

Preparation of isonuclear bromonitronaphthalenes by dehydrogenation of the corresponding derivatives of tetralin (ar-tetrahydronaphthalene). V. VESELÝ and L. K. CHUDOŽILOV (Bull. Soc. chim., 1925, [iv], 37, 1436-1444).-ar-Bromonitroand bromodinitro-tetrahydronaphthalenes are verted into the corresponding bromonitronaphthalenes by the method previously described (A., 1923, i, 550; cf. Braun, Hahn, and Seemann, A., 1922, i, naphthalene, although an ac-dibromo-derivative is 728). The reaction fails with ar-trinitrotetrahydrofrom this compound, however, takes place only very formed at 100°. Elimination of hydrogen bromide slowly on raising the temperature and the product remains liquid. In contrast to 1-chloro-2 : 4-dinitronaphthalene (Ullmann and Bruck, A., 1909, i, 21) the halogen atom in ar-1-bromo-2: 4-dinitrotetrahydronaphthalene does not react with ethyl sodiohydronaphthalene and ar-2-bromo-1 : 3-dinitrotetramalonate, and only very slow reaction occurs with 2-bromo-1 : 3-dinitronaphthalene. The following compounds are described: ar-2-Bromo-1-nitrotetrahydronaphthalene, m. p. 101-102°, b. p. 162–185°/ 13-15 mm., from ar-1-nitro-2-aminotetrahydronaphthalene (Schroeter, A., 1922, i, 123) by the Sandmeyer reaction. ar-2-Nitro-1-aminotetrahydronaphthalene, obtained by the action of alcoholic ammonia on the 1: 2-dinitro-compound, similarly yields ar-1-bromo-2-nitrotetrahydronaphthalene, m. p. 50-51.5°; ar-2-bromo-3-nitrotetrahydronaphthalene, m. p. 50-51° (yielding 2-bromo-3-nitronaphthalene, m. p. 82-83° on dehydrogenation); ar-1-bromo-3nitrotetrahydronaphthalene, m. p. 106-106.5°; ar-3bromo-1-nitrotetrahydronaphthalene, m. p. 76-76-5° (yielding on dehydrogenation 3-bromo-1-nitronaphthalene, m. p. 97-98°, also obtained from 4-nitro-ẞnaphthylamine), and ar-4-bromo-1-nitrotetrahydronaphthalene, m. p. 68-69°, were similarly obtained from the corresponding nitroaminotetralins described by Schroeter (loc. cit.). Aceto-2 : 4-dinitro-ar-tetrahydro-a-naphthalide after hydrolysis to the amine, m. p. 184°, with sulphuric acid, is similarly converted (yield 60%) into 1-bromo-2: 4-dinitro-ar-tetrahydronaphthalene, m. p. 94° (cf. Morgan, Micklethwait, and Winfield, J.C.S., 1905, 87, 747), yielding on 'dehydrogenation 1-bromo-2 4-dinitronaphthalene, m. p. 151–152°. ar-2-Bromo-1 : 3-dinitrotetrahydrofrom the acetodinitrotetrahydronaphthalide and by naphthalene, m. p. 135-136°, was similarly obtained lene. The product, m. p. 105-106°, obtained by nitration of 2-bromo-1-nitro-ar-tetrahydronaphthaMorgan (loc. cit.) from the nitration of a mixture of 1- and 2-bromo-ar-tetrahydronaphthalene is impure, repeated crystallisation raising its m. p. to 133—134°. 2-Bromo-1 3-dinitronaphthalene, m. p. 183-184°, is obtained on dehydrogenation. R. BRIGHTMAN.

:

Reactivity of the methyl-hydrogen atoms in 2: 4-dinitro-1-methylnaphthalene. V. VESELÝ and I. PASTAK (Bull. Soc. chim., 1925, [iv], 37, 1444-1451).-Nitration of a-methylnaphthalene