i. 81 Organic Chemistry. The Influence of the Structure of Organic Compounds on their Sulpho-chromic Oxidation. L. J. SIMON (Corn@. rend. 1922 175 1070-1072; cf. A. 1922 ii 867 868).-Among the substances which are completely oxidised by the sulpho-chromic mixture are certain straight-chain compounds phenolic derivatives and ring-substituted aromatic acids. I n other cases molecular structure influences the extent of oxidation e.g. methyl benzoate is completely whilst toluic or phenylacetic acid is only partly oxidised. From a study of a considerable number of such examples it is seen that this method of oxidation indicates structural differ- ences and may be used to investigate such problems as tautomerism or intramolecular change. The author has devised a formula based on the number of carbon atoms present in the molecule and the number which escape oxidation and regards the '' oxid- ation deficit" which i t furnishes as a first step towards a new method of structural investigation.P. A. LEVENE and F. A. TAYLOR ( J . Biol. Chem. 1922 54 351-362).-Further examples are given of the preparation of tertiary hydrocarbons by the malonic ester synthesis (cf. Levene and Cretcher A. 1918 i 250). The various stages of the syntheses are represented by the following series of compounds. Revised constants are given for certain compounds which have previously been described. Ethyl ethylbutylmalonate [previously prepared but not char- acterised by Raper (T. 1907 91 1837)] b. p. 128-129"/7 mm. df 0.9646 n$ 1.4284. Ethylbutylmalonic acid C,H,,O m. p.115". a-Ethylhexoic acid b. p. 228-229" (Raper gave 225"); ethyl- a-ethylhemute C,oH,oO b. p. 189-191" di0 0.8628 ng 1.4128. p-Ethylhexyl alcohol C,H,,O b. p. 181-183" df 0.8328 nZ1.4328 ; p-ethylhexyl iodide CsH1,I b. p. 89-90"/11 mm. d? 1.3365. y-Methylheptane b. p. 120-122" (cf. Clarke A. 1909 i 349) dp 0-7069 ?a 1.3980. Ethyl cc-methylheptylmalonate Cl5H2,O4 b. p. 157-158"/10 mm. 0-9496 ng 1.4324. p-Methylnonoic acid CloHzoOz b. p. 147- l48"/12 mm. & 0.9012 niy 1-4342 ; ethyl P-methylnonoate Cl2HZ4O2 b. p. 115"/13 mm. di0 0.8653 nz 1.4240. y-Wethylnonyl alcohol C10H220 b. p. 103-103*5"/9 mm. di0 0.8342 nz 1.4361 ; y-methyl- nonyl iodide Cl0H2,I b. p. 115"/10 mm. djo 1-2515. y-Methyl- nonane C10H22 b. p. 165*5-166.5" d;O 0.7354 nz 1.4126. c-Methylnonane (p-butylhexane of Levene and Cretcher loc.cit.) c-Propylnonane C,,H2 b. p. 804-305" dy 0.7559 nz 1.4228 was obtained by reduction of 6-butyloctyl iodide (cf. Levene and Cretcher). Ethyl butylmalonate b. p. 122"/12 mm. (cf. Adams and Marvel VOL. CXXIV. i. e H. J. E. Oxidation of Tertiary Hydrocarbons. b. p. 164-166" d;' 0.7319 ?i$ 1.4116.i. 82 ABSTRACTS OF CHEMICAL PAPERS. A. 1920 i 283) dy 0-9745 .ze 14222. Ethyl butylheptylmlonate Cl8H,O4 b. p. 177-178"/12 mm. d;O 0.9318 nz 1.4366; but$ heptylmlonic acid C14H26O4 needles m. p. 117". cc-Butylnonoic acid C,,H,,O b. p. 179"/13 mm. di0 0.8860 n; 1.4403; ethyl a-butylnonoate C15H3002 b. p. 115"/1 mm. d? 0.8560 ng 1.4290. p-Butylnonyl alcohol C,,H,,O b. p. 112-114"/0*5 mm. dy 0.8359 nz 1.4430 ; p-butylnonyl iodide C,H,,I b. p.121-123"/0.5 mm. 6-Methyldodecane C13H28 b. p. 22r6-227" djo 0.7576 nz 1-4244. When r-methylnonane was oxidised with alkaline permanganate acetic and butyric acids were identified amongst the products of the reaction; the presence of some formic acid was also detected. Apparently the oxidation proceeds mainly according to the following scheme CH2Me*[CH2]2*CHMe*[CH2]2*C€€2Me -+ CH,Me*[CH,],*C O*CH,+ CH,Me*CH,*CO,H. CH,Me*[CH,],*CO*CH -+ CH,Me*CH,*CO,H+ CH,*CO,H. E. S. The Formation of Hydrocarbons during the Action of Potassium on Ethyl Acetate. HELMUTH SCHEIBLER HEIN- RICH ZIEGNER and EMIL PEFFER (Ber. 1922 55 [B] 3921- 3931) .-The action of potassium on an ethereal solution of ethyl acetate has been shown to lead to the formation of ethyl potassio- acetate CH,:C(OK)*OEt (cf.Scheibler and Voss A. 1920 i 366; Scheibler and Ziegner A. 1922 i 426). In addition acidic sub- stances are produced which will be described subsequently and neutral compounds which are the subject of the present com- munication. Potassium and ethyl acetate in varied proportions are allowed to react in the presence of ether and the products are decomposed either by sulphuric acid (30%) or carbon dioxide. The neutral portions are freed from admixed ester and ketones by successive treatment with concentrated potassium hydroxide solution a t 80" and sodium hydrogen sulphite and are subsequently distilled whereby a series of fractions boiling over the range 120-260" are isolated. The carbon and hydrogen content of these increases with increasing boiling point at the expense of the oxygen content.The analytical composition of the fraction b. p. 210-240"/atmospheric pressure agrees with that required by the formula C12H2 whereas that of the fraction b. p. above 250" harmonises with the formula Cl2HZ4. The constitution of the substances has not been elucidated. The insolubility of the bulk of the product in concentrated sulphuric acid indicates the absence of olefines and the low hydrogen content makes it unlikely that paraffins are present. It is most probable that the hydrocarbons are homologues of benzene or mono- or poly-cycloparaffins . Only traces of hydrogen are evolved in the gaseous state during the action of potassium on ethyl acetate. It appears that the liberated hydrogen reduces a portion of the ethyl potassioacetate extensively and that in the course of the change intermediateORGANIC CHEMISTRY. i.83 substances are produced which are more readily hydrogenated than ethyl potassioacetate a considerable proportion of which remains after complete solution of the potassium. The fractions boiling below 200" consist mainly of hydrocarbons but appear to contain a certain proportion of ethers; their instability towards alkaline permanganate indicates the unsaturated nature of one or both components. The formation of hydrocarbons is not observed when potassium is replaced by sodium"under otherwise identical conditions. H. W. The Chlorination of Methane. ARTHUR SCHLEEDE and CURT LUCKOW (Ber. 1922 55 [B] 3710-3726).-The chlorination of methane has been studied by passing mixtures of the gases through electrically heated quartz or glass tubes.Inflammation occurs when the gases are used in molar proportions but the flame gradually dies out; it is permanent when an excess of chlorine is employed. If on the other hand an excess of methane is used ignition is not observed ; under these conditions chlorine is only quantitatively utilised if the temperature is not below a certain minimum depend- ing on the precise composition and rate of flow of the mixture. The phenomena are not greatly affected by the presence of catalysts or by an increase of temperature of 100" above the minimum. The products of the reaction are freed from hydrogen chloride by passage through warm concentrated potassium hydroxide solution (water is unsuitable since it dissolves appreciable amounts of methyl chloride) and are subsequently condensed at a temperature not exceeding -110".(The vapour tensions of methyl chloride below -30" of methylene chloride and chloroform below 0" and of carbon tetrachloride below -20" have been measued.) Methyl chloride can be conveniently separated from the condensate by Stock's method of fractional distillation in a high vacuum but the process is inapplicable to the separation of methylene chloride chloroform and carbon tetrachloride which are therefore estimated approximately by the technical distillation method. When the velocity of passage of the gaseous mixture per unit of surface exceeds a certain value and a large increase of the heating (catalysing- surface is not also provided the reaction tends more and more towards a steady ignition and increased formation of higher pro- ducts is observed which takes place particularly at the expense of the chloroform.The greater the velocity of the gas the greater must be the catalysing surface. If the rate of flow is below the ignition value (or a suitable increase of the heating surface is pro- vided) higher products are not formed in appreciable amount but the relative proportions of the chlorinated methanes are not those which would be expected from a step-wise reaction. When the rate of flow is below the ignition value marked changes resulting in the production of carbon tetrachloride and higher products are only observed when the catalysing surface is enormously increased (for example by the use of activated charcoal).This result is The results may be summarised as follows. e 2f . 84 ABSTRACTS OF CHEMICAL PAPERS not appreciably modified when the minimum temperature is ex- ceeded by as much as loo" when the heating zone is increased or diminished or when catalysts such as ferric chloride molybdenum pentachloride or antimony pentachloride are present. On the other hand the results obtained are not in harmony with the values calculated from Martin's formula (2. Elektrochem. 1921 27 150). If the rate of flow is considerably below the ignition value an increased production of chloroform and carbon tetrachloride or substances of similar boiling point at the expense of methyl and methylene chlorides is observed. The causes which are operative in producing results differing so greatly from those of a step-wise reaction are discussed in detail.The most probable are the dissociation of methane and subsequent changes such as those represented by the equations CH3C1+ CH4= C2H6 + HCl CH,CI +CH4= CH2C1*CH3+ HC1 CH3C1 + CH,C1=CH,C1*CH3+HC1 and CH,C1,+CH3C1=CH,Cl*CH,C1 (or CH,*CHCl,) +HC1. The preparation of methyl chloride from methane and carbonyl chloride (cf. Hochstetter A 1916 i 625) has been re-examined. The action appears to be due to greatly diluted chlorine. R. DE FORCRAND (Compt. rend. 1923 176 20-23).-Thallous ethoxide was prepared by the method of Lamy (Ann. Chim. Phys. 1863 67 395; 1864 3 373) and was isolated as an oily liquid d 3-55. It reacts with other alcohols to give the corresponding alkyloxides.I n this way the author has prepared the monothallium derivatives of ethylene glycol OH*C,H,*OTl and of glycerol C,H,( OH),*OTl both of which separate as yellow solids and thallous phenoxide which is white and micro- crystalline. Thallous hydroxide may be similarly prepared. For the preparation of thallous acetate the ethoxide is the best starting point and for the neutral sulphate the hydroxide is best used. H. W. The Thallous APkyloxides. W. G . The Alkylglycerols. Preparation of Alkylvinylcarbinols [Alkylallyl Alcohols]. RAYMOND DELABY (Compt. rend. 1922 175 967-970; cf. Lespieau A. 1911 i 347).-Homologues of glycerol were obtained by the action of magnesium alkyl compounds on acraldehyde yielding unsaturated secondary alcohols.These were brominated and the resulting bromohydrins converted into diacetins by means of potassium acetate ; hydrolysis under pressure or alcoholysis by methyl alcohol of the diacetins then yielded the alkylglycerols. The alkylallyl alcohols obtained in the first stage have the following properties. Aa-Buten-y-ol (methylvinylcar- binol) b. p. 94-96' ; d 0.854 ; d:* 0.835 ; r2$ 1.4087 (cf. Wohl and Losanitsch A. 1908 i 934). Aa-Pentene-y-ol (ethylvinylcarbinol) b. p. 37'/20 mm. ; d f 0-839 ; ng 1.4182 (cf. Wagner A. 1885,370). ha-Hexeue-y-ol (propylvinylcarbinol) b. p. 133.5-134" ; d! 0.851 ; df2 0.834 ; n'," 1.4215. A-Heptene-7-01 (butylvinylcarbinol) b. p. 153.5-154"; d 0.852; diz 0.835; n$ 1.4275. These alcohols yielded crystalline allophanates m. p. 151-152" 152-153" 139.5-140" and 156.5-157" respectively.ha-Heptene-y-ol wasORGANIC CHEMISTRY. i. 85 resolved into its optical isomerides by Pickard and Kenyon's method (T. 1911,99 45). Alkylglycerols Transformation of Alkylvinylcarbinols into Alkylglycerols. RAYMOND DELABY (Compt. rend. 1022 175 1152-1 154 ; cf. preceding abstract).-Experimental details of the transformation of brominated alkylvinylcarbinols into diacetins and of the hydrolysis of the latter into homologues of glycerol are described. Propylglycerol m. p. 60-62" b. p. 167-6- 16S0/14 mm. and butylglycerol m. p. 52-54' b. p. 175-175.5"/17 mm. are hygroscopic crystalline substances ; their triacetutes have b. p. 157-159"/15 mm. and b. p. 174'121 mm. respectively. Methylglycerol has b. p. 162-5-163*5"/15 mm.(cf. Lieben and Zeisel A. 1881 710); ethylglycerol has b. p. 165-166"/15 mm. (cf. Wagner A. 1889,231). The method gives yields corresponding with 60% of the secondary alcohol used. SVIGEL POSTERNAK (Swiss Pat. 91727; from Chem. Zentr. 1922 iv 837-838; cf. A 1921 i 225).-A solution of inositol is heated with excess of orthophosphoric acid in the presence of sufficient excess of phos- phoric oxide to combine with the water produced in esterification. The resulting products are dissolved in dilute sodium hydroxide solution and the sodium metaphosphate is changed into sodium pyrophosphate by heating at 100". The latter salt is fractionally crystallised out from the viscid solution of the sodium salt of inositol polyphosphate. Other metallic inositol polyphosphates may be obtained from the sodium salt by double decomposition.Calcium magnesium and ferric salts are mentioned and a crystalline calcium sodium salt having the comnosition H. J. E. H. J. E. Preparation of Inositol Polyphosphate. c 6H602~P6N~8Ca,,3H@ is also obtained ; it probably occurs in seeds. G. W. R. Cyclic Derivatives of Mannitol. P. VAN ROMBURGH and J . H. N. VAN DER BURG (Proc. K . Akad. Wetensch. Amsterdam 1923 25 335-340).-The unsaturated oxide C6H,01 produced by heating the hexaformate of mannitol is shown to he identical with 2-vinyrdihydrofuran and its structural formula ~~$~>CH*CH:CH confirmed. The reduced oxide C6H,,0 prepared by reduction of C,H80 with hydrogen under a pressure of two atmospheres in the presence of palladium sol is shown to be identical with %ethyl- tetrahydrofuran. The structural formula 1-0-1 ?H CH,*~H*~H*CH*CH*CH,*OH OH OH proposed by van Romburgh and van Maanen (Diss.Utrecht 1909) for mannitan is confirmed and that similarly proposed for isomannide YH,*CH*CH*YH*YH*CH2 shown probably to be p T O P l 0- OHi. 86 ABSTRACTS OF CHEMICAL PAPERS. correct. isoMannide increases the electrical conductivity of an aqueous solution of boric acid only very slightly. The increase in the case of mannitol and mannitan respectively is considerable (cf. Boeseken A. 1921 i 843). J. S. G. T. Investigations on the Dependence of Rotatory Power on Chemical Constitution. XIV. The Normal Aliphatic Ethers of d-p-Octanol. JOSEPH KENYON and REGINALD ARTHUR MCNICOL (T. 1923 123 14-22). The Ability of ay-Glycols to Form Acetone [isoPropylidene] Compounds.J. BOESEKEN and P. H. HERMANS (Ber. 1922 55 [B] 3758-3760).-Trimethylene glycol unites with acetone to form an isopropylidene ether a mobile liquid with a camphoraceous odour b. p. 123-125" difi5 0.9587 nz5 1.4252 the isolation of which is rendered difficult by the unfavourable position of the equilibrium of the reaction glycol+acetone isopropylidene + water and by the considerable volatility of the compound with the vapours of acetone. The relatively difficult formation of an isopropylidene compound from trimethylene glycol is due not only to the presence of the hydroxyl groups in the ay-position and the consequent necessity of forming a six- instead of a five-mem- bered ring but also and chiefly to the unfavourable steric position of these groups.Instances in which the hydroxyl groups are more f avourably placed are found in anhydroenneaheptitol (Mannich and Brose A. 1922 i 1118) and pentaerythritol which yields mono- and di-isopropylidene compounds m. p. 116' and 135' respectively. The place of hydrogen chloride in the condensations may be taken by other acids which are soluble in acetone. Good results are obtained with $-&% of concentrated sulphuric acid which is subsequently neutralised with a considerable excess of recently- ignited finely-divided lime. H. W. FAR- BENFABRIKEN VORM. FRIEDR. BAYER & CO. ( D . R . - P . 358125 ; from Chem. Zentr. 1922 iv 888).-The alcohol is converted by the usual methods into the corresponding carbamate. For example tri- chloroethyl alcohol dissolved in anhydrous ether is treated with carbamide hydrochloride or the alcohol dissolved in benzene in the presence of quinoline is treated with carbonyl chloride and the trichloroethyl chloroformate thus formed treated with ammonia.Trichloroethyl carbarnate forms white needles m. p. 64-65'. It is a soporific. G. W. R. A New Phosphoric Ester Produced by the Action of Yeast Juice onHexoses. ROBERT ROBISON (Biochem. J. 1922,16,809- 823).-The new ester is prepared in the following way. Lawdose or dextrose is fermented with yeast-juice and constant additions of disodium hydrogen phosphate. When the inorganic phosphate is no more utilised barium acetate is added to the fermentation mixture which is then neutralised and precipitated with an equal volume Preparation of an Ester of Trichloroethyl Alcohol.ORGANIC CHEMISTRY.i. 87 of alcohol. The precipitate is then extracted with 10 parts of cold water reprecipitated several times with basic lead acetate purified by treatment with mercuric acetate and finally precipitated with alcohol as the barium salt. Hexosemonophosphoric acid has [a]? +25.0" in water. The metallic salts with the exception of the basic salts of the heavy metals are all readily soluble in water and are amorphous. A crystalline brucine salt was obtained. The phenylhydrazine salt of the osazone of hexosemonophosphoric acid is not identical with the isomeric compound obtained from the hexosediphosphoric acid. On hydrolysis by acids or by emulsin the hexosemonophosphoric acid yields free phosphoric acid and a dextrorotatory substance from which glucosazone is obtained.The rotatory power of this product of degradation is however less than that of dextrose. Preparation of Thiohydrins. FARBWERKE VORM. MEISTER LUCIUS & BRUNMG (Brit. Pat. 185403) .-Propylene thiohydrin or mixtures of this with ethylene thiohydrin are obtained by heating propylene chlorohydrin or mixtures of this with ethylene chloro- hydrin such as are obtained by the action of hypochlorous acid on olefine gas mixtures with aqueous sodium sulphide solutions the product being isolated by evaporating the water in a vacuum separating the sodium chloride by pressing and distilling the crude thiohydrin under reduced pressure. Propylene thiohydrin boils a t 120°/4 mm. and probably has the constitution S(CH,-CHMe-OH),.The thiohydrins are useful as solvents in dye printing. Varying Valency of Platinum with Respect to Mercaptanic Radicles. SIR PRAFULLA CHANDRA RAY (T. 1923 123 133- 141). Stability of Sodium Formate Acetate and Oxalate towards Oxidation under Pressure. HANS SCHRADER (Ges. Abh. Kenntnis Kohle 1920 5 193-199; from Chem. Zentr. 1922 iii 1154).-At 160" no oxidation of these salts takes place in three hours. Oxidation takes place freely at 210" and rapidly a t 260". Sodium formate and sodium oxalate are oxidised equally quickly whilst sodium acetate is more slowly attacked. The presence of sodium carbonate or sodium hydroxide has no marked effect. Intermediate stages between the three organic acids and carbon dioxide were not observed. No oxalate was formed from formate neither was oxalate or formate formed from acetate.s. s. z. G. F. M. G. W. R. Investigations on the Dependence of Rotatory Power on Chemical Constitution. XIII. The Spatial Configuration of the Unbranched Aliphatic Chain. ROBERT HOWSON PICKARD JOSEPH KENYON and HAROLD HUNTER (T. 1923,123 1-14). The Anodic Preparation of Pure Lead Tetra-acetate Tetra- propionate and Silver Diacetate. C. SCHALL and W. MELZER (2. Elektrochem. 1922 28 47-77) .-The specific conductivities and the corresponding temperature coefficients of solutions ofi. 8s ABSTRACTS OF CHEMICAL PAPERS. anhydrous and crystalline lead diacetate in acetic acid have been measured and the products obtained on electrolysis of these solu- tions examined. When the water content of the mixture is less than 0.6y0 very little change is observable at the anode and when greater than 2y0 the anodic product is mainly lead dioxide.At intermediate concentrations of water pure lead tetra-acetate may be obtained in good yield if the apparatus is surrounded by ice. The pure salt may be prepared also from a solution of the diacetate in 99.4% acetic acid in the presence of dry sodium acetate. Small amounts of lead tetrapropionate and silver diacetate may be similarly obtained. W. E. G. Relationship between the Iodine Values and Refractive Indices of some Hardened Vegetable Oils. J. J. SUDBOROUGH H. E. WATSON and I). Y . ATHAWALE ( J . Ind. Inst. Xci. 1922 5 v 47-69).-Samples of cotton-seed linseed a,rachis mohua (Bassia latifolia) sesamk sardine castor hongay (Pongamia glabra) and coconut oils were refined and hydrogenated a t 180" using nickel catalysts and the relationship between iodine value and refractive index was determined.Except in the case of castor and hongay oils the relationship between the two constants is independent of the time and of the type of catalyst used. In the case of the f i s t six oils mentioned the curves representing the relationship between iodine value and refractive index lie very close together and may be represented by the equation nr= 1 -4468 + 1 -03 x lo4+ ( I . v.) + 7.3 X 1W8(I. v.)2 to an accuracy of about 0-0005. The refractive indices a t 60" of the above six oils when completely hardened are practically identical a t the value 1.4468. The refractive indices of hardened coconut oil are much lower than those of other oils with the same iodine value.I n the case of castor oil the relationship between the iodine value and the refractive index is not independent of the type of catalyst or of the time of hardening owing to the varying extent to which the hydroxyl groups are reduced. It is possible that the case of hongay oil is similar although its acetyl value is only 24. H. C. R. The Catalytic Decomposition of Castor Oil. A. MAILHE (Compt. rend. 1923 176 37-39).-VYhen passed over aluminium and copper turnings castor oil undergoes decomposition the products varying with the temperature. The products are hydro- carbons and heptaldehyde. Below 600" the hydrocarbons formed are almost entirely homologues of methane but above 600" aromatic hydrocarbons such as benzene toluene and m-xylene are also formed.W. G. The Transition from the Colloidal to the Crystalloidal State. Solutions of Potassium Oleate. LOUIS LEIGHTON BIRCUMSHAW (T. 1923 123 91-97). Dissociation Constants of Sulphoacetic and a-Sulphopro- pionic Acids. H. J. BACKER (ROC. K . A k d . Wetemch. Amster- dam 1923 25 359-363).-Values of the respective molecular conductivities at 25" of aqueous solutions of sulphoacetic acid,ORGANIC CHEMISTRY. i. 89 mono- and di-sodium sulphoacetates sulphopropionic acid mono- and di-sodium sulphopropionates propionanilide-a-sulphonic acid and sodium propionanilide- a-sulphonate in dilutions corresponding with 1 g. mol. per 16 32 64 128 256 512 and 1024 litres have been determined.The mean values of the respective dissociation constants of. sulphoacetic and a-sulphopropionic acids calculated therefrom are 8-9 x J. S. G. T. Optical Resolu- tion of the cis-Potassium Salt. M~RCEL DEL~PINE (Cmpt. rend. 1922 175 1408-141 1 ).-Potassium iridodichloro-oxalate K,[IrC12(C20,)2] as prepared by Vdzes and Duffour (cf. A. 1909 i 762) was found to occur in the cis- and trans-forms and the cis- form was resolved into its two optical isomerides by means of its strychnine salt. The I-salt is the less soluble and has [a],-23*8". The active salts are more soluble than the racemic mixture. The trans-salt is not resolved by means of its strychnine salt. The cis- and trans-isomerides are capable of being transformed into one another under suitable temperature conditions.Investigations on the Dependence of Rotatory Power on Chemical Constitution. XVI. The Di-d-P-octyl Esters of the Saturated Dicarboxylic Acids. LESLIE HALL (T. 1923 123 3 2 4 ) . The Syntheses of y-Hydroxy- and a-Amino-y-hydroxg- pimelic Acids. HERMANN LEUCHS and WERNER NAGEL (Ber. 1922 55 [B] 3950-3960).-The compounds are prepared from the lactone ester ?0-o>CH*CH2*CH(C02Et)2 described by Leuchs and Mobis (A. 1909 i 361). Attempts to halogenate the lactone ester or the corresponding acid in the malonic residue and subsequently to eliminate the carbethoxy- or carboxy-group did not lead t o satisfactory results since only impure non-crystalline products could be obtained which even after re-esterification could not be distilled without decomposition.The lactone ester is transformed by aniline a t 160-170" into the mono-anilide C1,H,,O,N lustrous leaflets m. p. 79-80". Hydrolysis of the lactone ester and subsequent removal of carbon dioxide from the product leads to the formation of the lactone of 7-hydroxypimelic acid ?o-o>CHfCH2],*C02H four- or six-sided plates m. p. 80-82-5". It is converted by methyl alcohol and hydrogen chloride into methyl y-chloropimelate which could not be isolated in a homogeneous condition but is characterised bv its conversion by and 6.0 x lO.-5 The cis- and trans-Iridodichloro-oxalates. W. G. CH,*CH2 CH,*CH2 ammda into P-pyrrolidoneprop~o~~~de (?~~~~!>CW*[CH2-J2*CO*NH2 -~ colourless needles m. p. 144-145" (decoinp.) when slowly heated. Hydroxypimelolactone is brominated by Volhard's method and the product is converted by methyl alcohol into the methyl ester e*i.90 ABSTRACTS OF CHEMICAL PAPERS. of or- bl.omo-y-iiydroxypimelokr;ctone which could not be distilled without decomposition ; the brominated ester is hydrolysed with boiling hydrobromic acid and the resulting mid which could not be caused to crystallise is aminated with aqueous ammonia. The expected products could only be isolatcd in the form of the copper salts one of which (+H,O) sky-blue crystals is the normal salt CH,*CH2-C0 /"' of y-hydroxy- a-aminopimelic acid OH*CH< whereas the remaining two greyish-blue crystals (+iH,O) and paler greyish-blue needles (+ $H,O) are considered to be derived from the two racemic forms of a-aminopimclolactone CH,*CH(NH2)*C02 \ ?H2*CH2> CH*CH,*CH (kH,)*CO,H.CO-0 The lactone ester is converted by ethyl bromide in the presence of alcoholic sodium ethoxide into the corresponding a-ethyl deriv- ative (?o-o>CH*CH2*CEt(C0,Et) an odourless liquid b. p. CH9*CH9 206-208"jl4 &m. which is hydrolysed to a non-crystalline acid and is converted by methyl alcoholic ammonia into the di-amide C,,Hl,04N2 colourless prisms or oblique plates m. p. 170-171". The acid is transformed by loss of carbon dioxide into or-ethyl- y-hydroxypimeloluactone ni. Q. 89-91' to which the constitution _ - o>CH*[CH2]2*C0,H is assigned. The latter substance is c 0.- bHEt*CH treated wkh bromine and a trace of phosphorus at 100" and the product after being purified through the methyl ester is treated successively with hydrobromic acid and ammonia.The product is isolated as a homogeneous copper salt (CpH140,N),Cu,~H20 although the constitution of the parent acid indicates the existence of four racemic isomerides; the corresponding free acid could only be obtained as a non-crystalline very hygroscopic mass. H. W. Keto-enolic Tautomerism. I. Desmotropy-isomerism of Ethyl Diacetylsuccinate. H. P. KAUFMANN (Annalen 1922 429 247-283) .-A general account of this work has already appeared (A. 1922 i 985). The investigation shows that in 0-1N-alcoholic solution a t 30" ethyl diacetylsuccinate becomes equilibrated to an approximation of 1 or 2% in twenty-four hours and that the mixture then contains 10% of the y-ester (diketo) 30% of the p-ester (diketo) 16% of the cc,p-ester (mono-enol) and 44% of the or#-ester (mono-enol).ERICH MULLER (2. angew. Chem. 1922 35 689-692 69&700).-In a lecture delivered before the Dresden Chemical Society the author gives an extended general account of his work on the electrolysis and catalytic decomposition of solutions of formaldehyde and related compounds and outlines his views on the hydration of formaldehyde to CH,(OH) and the amphoteric nature of the hydrate. The observations are extended to other substances C. K. I. The Amphoteric Nature of the Carbonyl Group.ORGANIC CHEWISTRY. i. 91 containing the carbonyl group such as ketones carboxylic acids and carbon monoxide. The assumption of the formation of ions the existence of which cannot be demonstrated or the quantity measured is justified by analogy with inorganic compounds such as the complex cyanides and by the fact that the very involved behaviour of formaldehyde cannot be so clearly and uniformly EMU; BAUR and A.REB- MANN (Helv. Chim. Acta 1922 5 828-832).-Moore and Webster are stated to have obtained formaldehyde from carbon dioxide solutions by the action of sunlight in presence of colloidal oxides of uranium or ferric iron (A. 1913 i 1303). The present authors have repeated these experiments and have not succeeded in de- tecting the formation of oxalic acid glyoxylic acid formic acid or formaldehyde. Failure to confirm the results obtained by the above investigators may be due to their having omitted to describe with sufficient exactitude their experimental conditions. explained by any other hypothesis.H. w. The Photolysis of Carbonic Acid. E. H. R. Acraldehyde Transformations and Antioxygenisers. CHARLES MOUREU and CHARLES DUFRAISSE (Bull. Soc. chim. 1922 [iv] 31 1152-1176; cf. Moureu and Dufraisse A. 1919 i 574; Moureu and Lepape A. 1919 i 574 ; 1920 i 10 ; Moureu Dufraisse and Robin A. 1920 i 143 ; Moureu Dufraisse Robin and Pougnet A. 1920 i 144; Moureu Dufraisse Lepape Robin Pougnet Boutaric and Boismenu A. 1921 i 395; Moureu and Dufraisse A. 1922 i 250 824).-&4 review of previous publications showing the trend of the work leading to the stabilisation of acraldehyde by addition of pyrocatechol and also by other phenols is followed by development of the subject in the direction of reviewing the chief substances capable of undergoing autoxidation and also of de- tecting the antioxygenising function in substances other than phenols.The oxidation of the acraldehyde appears to be a neces- sary preliminary to the formation of disacryl and the addition of a phenol inhibits the change. But on submitting pure acraldehyde to the action of oxygen some disacryl is always formed the quantity appearing to vary for no definite reason although the transform- ation into disacryl was not observed to take place to any considerable extent in the presence of excess of oxygen. These facts appearing to be somewhat contradictory the action of light on acraldehyde was investigated. The results showed that light effects the con- densation into disacryl in absence of oxygen and conversely that acraldehyde undergoes no condensation when prepared and kept in the dark and free from contact with oxygen.In absence of light extremely small quantities of oxygen bring about the con- densation and the authors regard the phenomenon as one of catalysis suggesting that a peroxide oi acraldehyde is the catalyst. The theory put forward to account for the lack of condensation by excess of oxygen is based on the supposed existence of two types of acraldehyde molecule one of which is activated. These molecules combine with oxygen but when oxygen is absent they e* 2i. 92 ABSTRACTS OF CHEMICAL PAPERS. combine with each other. This is shown by the fact that acraldehyde in contact with oxygen over mercury remains clear the mercury meanwhile rising in the tube but when the mercury ceases to rise oxidation is no longer occurring and the acraldehyde becomes opaque.The time at which the latter change takes place may be predicted by extrapolating the curve obtained by plotting rise of mercury against time. Reasons are given for continued use of the term " antioxygeniser )' (cf. Seyewetz and Sisley A. 1922 ii 628). H. J. E. The Hydrogenation of Aldehydes and Ketones in Presence of Pure and Impure Platinum Black. FAILLEBIN (Cmpt. rend. 1922 175 1077-1079).-1n the reduction of aldehydes and ketones to alcohols in presence of platinum black considerable experimental difficulties are experienced. The action is slow the activity of the catalyst is in many cases extremely limited and the yields obtained are poor owing to the formation of hydro- carbons. I f however the catalyst is prepared by the reduction of chloroplatinic acid containing 5% of its weight of ferric chloride good yields are obtained especially in the case of aromatic aldehydes and the speed of the reaction is increased.I f the ferric chloride is replaced by iridium chloride similar advantages are obtained but to a lesser extent. In hydrogenating an ethyl acetate solution of 4-piperonyl-2-butanone in presence of '' ferric platinum " to the corresponding secondary alcohol hitherto unknown a quantitative yield was obtained. The alcohol i s not described. H. J. E. Syntheses by means of Sodamide. A. HALLER (Bull. Xoc. chim. 1922 [iv] 31 1073-1 144).-A lecture delivered before the Soci6t6 Chimique de France in which the use of sodamide as a reagent is discussed mainly with regard to the work of the author and his collaborators.The work surveyed comprises substitution reactions of ketones and nitriles the action of the alkyl derivatives so obtained on various cyclic compounds and also the decomposition and condensation reactions which may be effected by the use of sodamide. The chief papers to which reference is made are those of von Auwers and Krollpfeiffer A. 1915 i 818; Bodroux and Taboury A. 1910 i 257; Cornubert A. 1921 i 730; Haller A. 1904 i 600; 1905 i 214 602; 1913 i 629 984 1357; 1914 i 418 ; Haller and Bauer A. 1908 i 987 ; 1909 i 108 654; 1910 i 219 300; 1911 i 299 726; 1913 i 488 829; 1914 i 418 549 724; 1915 i 411; 1918 i 24 428; 1922 i 258; Haller and Benoist A. 1912 i 570; 1922 i 350; Haller and Cornubert A.1914 i 291 842 968; 1920 i 390 441 ; Haller and Louvrier A. 1918 i 397; Haller and Rarnart-Lucas A. 1914 i 1072; 1917 i 665. Diacetone-glucose [Diisopropylidene-glucose]. P. A. LEVENE and G. M. MEYER ( J . Biol. Chem. 1922 54 805-807).- When oxidised with nitric acid the monomethyl glucose obtained from diisopropylidene-glucose yields a methyl sacchccrohctone H. J. E.ORGANIC CHEMISTRY. i. 93 C,HlOO m. p. 206-207" (after sintering and darkening at 190") [a] +15" (cf. Irvine and Hogg T. 1914 103 1386). The pro- duction of this compound indicates that the isopropylidene radicles in diisopropylidene-glucose are attached to the aP and c[ carbon atoms. If diisopropylidene-glucose has the structure of a y-glucose (Irvine and Patterson T. 1922,121 2146) the methyl group in the saccharolactone will be attached to the y-carbon atom ; otherwise it will be in the p-position.E. S. Glucosides. XI. The Glucosides of Glycerol. P. KARRER and 0. HURWITZ (Helv. Chim. Actu 1022 5 86+869).-With the object of synthesising a glucoside of glycerol of which the constitution could not be open to doubt the action of acetobromo- glucose on isopropylidene-glycerol in presence of silver carbonate was studied. The reaction proceeded smoothly giving a good yield of a-tetra-acetyl-d-gluwsidoglycerol iso-propylidene ether (annexed formula) white crystals m. p. 132" [a] -20.77". By careful hydrolysis of $JH-O yH2'o> CMe cH,*O*~,H,O,A~ this compound with dilute sulphuric acid the acetone residue was removed leaving a-tetra-ucetyl-d-glucosi~o-glycerol an amorphous substance which can be readily reconverted into the isopropylidene derivative.By the action of acetic anhydride on the amorphous substance the glycerol residue is acetylated with formation of a- (tetra-acetyl-d-glucosido)- glycerol py-diacetute a well-crystallised compound m. p. 98" [a]:' -30.96". Alkaline hydrolysis of the last gave 1-p-d-glucosidoglycerol OH*CH,*CH( OH)~CH,*O*C,HI105 [ a]g - 27.72". The p-configura- tion of the glucoside is proved by its ready hydrolysis by emulsin. It appears to be identical with the glyceryl glucoside synthesised by Bourquelot Bridel and Aubry by means of emulsin from glycerol and dextrose (A. 1917 i 379). P. KARRER and J. TERKUILE (Helv. Chim. Acta 1922 5 870-876).-1t was shown by Karrer and Smirnov (A.1921 i 766) that when tetra-acetyl- glucosidotrimethylammonium bromide is submitted to alkaline hydrolysis l-glucosan is formed. It is now found that by acid hydrolysis only the acetyl groups are removed and d-glucosido-a- trimethylammonium bromide is formed (using hydrobromic acid for the hydrolysis) as hygroscopic crystals m. p. 161-162" ; [a]? +5*0". The hydroxide C6HIl0,*NMe,*O~ was prepared in solution by the action of silver hydroxide on the bromide or chloride; it is a strong base but decomposes with evolution of trimethylamine when its aqueous solution is warmed. The chloride forms very hygroscopic crystals ; the iodide transparent slightly hygroscopic crystals m. p. 162-163" ; the chlorophtinate orange-brown crystals ; chloro- aurate; picrate beautiful yellow needles m.p. 141". Tetra- acet yl-d -glucosido -a- trimethylammonium bromide f oms rhombic crystals (a b c=0.4520 1 0.3443) ; the hydroxide resembles the parent compound in preparation and properties ; the chloride forms colourless hygroscopic crystals m. p. 173" ; [a]r +6.26" ; the perchbrute forms microscopic needles m. p. 190"; the picrate E. H. R. Glucosido-trimethylammonium Salts.i. 94 ABSTRACTS OF CHEMICAL PAPERS. crystallises in fine yellow needles m. p. 133"; the chloroplatinate in h e orange needles m. p. 209-210" (decomp.) and the chloro- aurate yellow needles which decompose when heated. MARC CRAMER and EDWARD H. Cox (Helv. Chim. Actu 1922 5 88&887).-The structure proposed by Pietet and Castan (A. 1920 i 594) for glucosan has been confirmed; the objections to this formula raised by Bergmann (A.1921 i 645) and Irvine and Oldham (T. 1921,119,1744) are therefore invalid. The attempt was first made to obtain an osazone from tribenzoylglucosan but this was unsuccessful owing to the daculty of hydrating the substance without hydrolysing the benzoyl groups at the same time. By the action of methyl sulphate on glucosan in presence of sodium hydroxide a t 35-40' a trimethylglucosan was obtained b. p. 210-212"/9 mm. This gives a reducing sugar when boiled with water and forms a yellow crystalline osazone m. p. 163-164" (decomp.). This can only be derived from a trimethylglucose of the formula E. H. R. The Constitution of Glucosan. OH*yH-yH*OMe OH*CH* O *CH*CH( OMe) *CH,*OMe ' which coiifirms the glucosan structure under discussion.A further proof depends on the observation that when glucosan is heated in a sealed tube with methyl iodide at 125-130" an additive compound is formed which when reduced with sodium amalgam gives the p-deoxymethylglucoside described by Pischer Bergmann and Schotte (A. 1920 i 422). This reaction is readily explained by the proposed formula for glucosan thus o<CH*O*CH*CH (OH) *CH,*OH'OMe.CH*O*CH*CH (OH) *CH,*OH FH-YH*OH I*yH-$!H*OH FH,-YH*OH OMe*CH*O*CH*CH( OH)*CH,*OH' -+ E. H. R. Cellulose Nitrate. EUGENE C. BINCHAM and WILLuML. HYDEN ( J . Franklin Inst. 1922,194 731-740).-The mobility of solutions of cellulose nitrate (12*11y0 nitrogen) in acetone was determined by the viscometer method at various temperatures and concentrations and with varying shearing stresses in order to investigate whether the fluidity is a linear function of the concentration only or whether as seemed probable it is influenced by the shearing stress.It was found that there was a steady increase in the mobility with the pressure and as already known to be the case with many plastic materials the efflux in ml. per second was a linear function of the shearing stress. When the shearing stress is extrapolated to its value when the efflux is zero the friction or " yield value " is obtained which may be defined as the shearing stress a t the wall of the tube necessary t o start the flow. The "yield value '' increased with increasing concentration and with decreasing tem- perature. On plotting the " yield value '' against temperature for a 7-708 :& nitrocellulose mixture a sharp transition temperatureORGANIC CHEMISTRY. i.95 was indicated at 43" where the " yield value " became zero and the material loses its plastic character and becomes a viscous liquid. It seems probable that this transition temperature is independent of the concentration. The mobility of cellulose nitrate dispersions is characterised by the great depression produced by very small percentages of the solid a 1% solution having a mobility only 20% of that of the pure acetone whilst that of an 8% solution is 0.1% of that of the solvent. This is one of the most noteworthy distinctions between the polar and the nonpolar type of colloid. The mobility increases with the temperature in a nearly linear manner.G. F. M. Cellulose Acetate. EMII; KNOEVENAGEL and KARL KONIG (Cellulosechemie 1922 3 113-122).-Gelatinisation of solutions of cellulose acetate takes place when kept in the presence of small quantities of suitable catalysts such as sulphuric acid or sulpho- acetic acid the gelatinised product becoming insoluble in the usual solvents. If the system is not completely anhydrous as for instance solutions of cellulose acetate in acetone or in glacial acetic acid this gelatinisation may be accounted for by a large decrease in the acetyl value owing to acid hydrolysis of the ester and it takes place more rapidly as the concentration of the catalyst is increased. If fhe system (solution of cellulose acetate in glacial acetic acid) be rendered anhydrous by the addition of acetic anhydride another set of conditions arises and gelatinisation may be delayed for three months when an increase of acetyl value and a large increase in copper value indicative of acetolysis are recorded.When however moisture is totally excluded from the first a different type of gelatinisation is characterised which may be described as the result of condensation or polymerisation of the cellulose ester. The purified gelatinised product is then insoluble in the usual media with the exception of tetrachloroethane the acetyl value is only slightly changed either slightly increased or lowered and the copper value is slightly decreased. This type of gelatinisation may be produced by completely drying 3 g. of the cellulose acetate a t 105" in a glass tube dissolving this in lOOyo acetic acid sufficient to give a 15% solution and keeping this solution in a desiccator over sulphuric acid.Next day a small quantity of catalyst e.g. 47 mg. of sulphuric acid or 96 mg. of sulphoacetic acid is stirred into the viscous solution 0.476 g. of acetic anhydride is added to react with any moisture which may have been absorbed during the operations and the tube is sealed up. Gelatinisation takes place after two to three days but if the quantity of catalysk is increased it is further delayed. With 115 mg. or more of sulphuric acid the solution does not gelatinise but considerable acetolysis takes place. J. F. B. Hydrocellulose [Preparation of Dimethylhydrocellulose 1. E m HEUSER and WALTER VON NEUENSTEIN (Cellulosechemie 1922 3 101-107).-Chopped viscose fibre partly dried and con- taining &lo% of moisture was placed in a bottle with a paraffinedi.96 ABSTRACTS OF CHEMICAL PAPERS. cork and the air was displaced by passing dry carbon dioxide through for two hours. A current of dry hydrogen chloride was passed very slowly so that fumes appeared a t the outlet after two to three hours. The bottle was then tightly closed and allowed to remain for several hours until a test showed that the hydrocellulose was completely soluble in 10% sodium hydroxide. The fibre was neutralised and washed. For methylation 5 g. of the hydro- cellulose (1 mol.) was dissolved in 50 C.C. of 10% sodium hydroxide solution (4 mols.) and after some hours 10 g. of methyl sulphate (3 mols.) were added in small quantities at a time with continuous agitation.The temperature rose to 60" after ten minutes. Subse- quent operations were performed without separating the methylated product using the same proportions of the reagents but taking the sodium hydroxide in the form of a 25% solution instead of 10%. After the fifth operation the greater portion of the methylated cellulose remained insoluble in the saline liquid. An excess of alkali was added and while hot the cellulose ether was collected. The precipitate was dissolved in cold water the solution filtered to remove traces of under-methylated residue and the ether which was precipitated on heating collected while hot and washed with boiling water. When completely methylated this ether retained its solubility in cold water even after drying.Theoretical yields were obtained when the operations were performed with mechanical stirring and the maximum degree of methylation corresponded with a methoxyl content of 33.76. (See also this vol. i 17.) J. F. B. Complex Magmesim Salts. 11. G. SPACU (BuZ. Foc. Ytiinfe Cluj 1922 1 247-266; from Chem. Zentr. 1922 iii 1045-1046 ; cf. A. 1922 i 859).-Magnesium tetrapyridine chZoride Mg(C,NH5)4Cl prepared from magnesium chloride and anhydrous pyridine a t the ordinary temperature is a white crystal- line powder; it is very hygroscopic with separation of pyridine. This salt differs from the corresponding bromide and iodide in that it is not co-ordinately saturated. Magnesium triethylenediamine chloride Mg en3 CI forms colourless crystals which cannot be resolved into their optically active components.It is highly hygroscopic. The chloride reaction is given with silver nitrate. Magnesium diaquodiethylenedinmine iodide Mg en,(H20),12 forms crystals ; it is unstable in air. Magnesium triethylenediamine sulphate Mg en3 SO,,aq is a hygroscopic white powder. Mag- nesium tetrapyridine thiocyanate Mg(C5N€€5),(CNS)2 from magnesium thiocyanate and anhydrous pyridine forms small colourless highly refractive crystals ; it is very hygroscopic. Magnesium hexa- pyridine thiocyanate forms large colourless highly refractive crystals. Magnesium hexammine thiocyanate Mg(NH3),( CNS),,. pre- pared by the action of ammonia on solid magnesium tetrappdme thiocyanate is a colourless powder. Magnesium tetra-aquodipyridine chloride dipotassium chloride Mg(C5NH,),(H,0),C1,,2KC1 is obtained by prolonged shaking of powdered carnallite with anhydrous pyridine.It is a white crystalline powder. G. W. R.ORGANIC CIIEMISTRY. i. 87 Preparation of Chloro-substitution Products of Hexa- methylenetetramine. ROMOLO BURATTI (Swiss Pat. 90703 ; from Chem. Zentr. 1922 iv S9l).-An aqueous solution of hexa- methylenetetramine is mixed with a solution of a hypochlorite previously neutralised with an organic or weak mineral acid and the mixture concentrated to the point of crystallisation. Using neutral sodium hypochlorite containing 10 yo of active chlorine the reaction is as follows C,~~12N4+4HC10=C,H,~4cl~+4H20. The tetrachlorohexamethylenetetramine thereby formed separates on con- centration of the solution as a colourless mass; it forms salts with acids.G. W. R. A Hydrolysis of Glycine. EMIL BAUR (Nelv. Chim. Acta 1922 5 825-828).-When a freshly prepared solution of glycine in air-free water is shaken for a long time in absence of air with animal charcoal gradual decomposition takes place. After 240 hours a t 40° using 10 g. of charcoal to 1 g. of glycine 17.6% of the nitrogen present is found in solution as ammonia. Since the final solution is always neutral the reaction is probably a simple hydrolysis according to the equation NH2*CH2*C02H+H20= OH*CH2*C02*NH4. Attempts to separate calcium glycollate from the solution were unsuccessful but the salt was recognised micro- scopic ally. E. I€. R. LESLIE JULIUS HARRIS (Biochem. J. 1922,16 739-746).-Cysteine forms metallic derivatives much more readily than cystine.The former compound was found to give coloured metallic derivatives with the following ions Fe'" Mn"' MnVI0 CU" Co Ni Cr Bi. Cuprous salts give a characteristic white derivative insoluble in neutral solutions. Tin also gives a compound with it. No oxidation takes place on addition of stannic chloride and it is suggested that owing to the strong affinity of tin for sulphur it protects that atom in cysteine from oxidation. The mercuric ion but not the mercurous acts as precipitant of cysteine. The colorimetric ammonia test for cysteine is only effective in the presence of traces of a metallic compound. I n alkaline solution and in the presence of oxygen the reduced metal formed an oxygen acceptor whilst in the absence of oxygen the oxidised metal can act as an oxygen-donator the cysteine acting as the oxidisable substance.The system is there- fore reversible so long as unoxidised cysteine is present. A. LANGHANS (2. ges. Schiess- u. Sprengstoflw. 1922,17 122-126 131-133 141- 143 150-153 159-162).-The properties of the brown product produced by drying mercury fulminate for prolonged periods (A. 1922 i 328) and the action of dilute and concentrated nitric acid aqua regia sulphuric acid chlorosulphonic acid and hydro- fluoric acid on mercury fulminate are described. Mercury fulminate dissolves readily in boiling sodium sulphite solution. Two forms of crystals are obtained from the solution both having the formula HgSO,,Na,SO,,€I,O. If enough fulminate is added to produce a marked foaming the salt Na2S0,,2HgS03,H20 A Series of Metallo-cysteine Derivatives.S. S. Z. The Decompositions of Mercury Fulminate.i. 98 ABSTRACTS OF CHEMICAL PAPERS. is obtained. When treated with concentrated sodium sulphide solution mercury fulminate gives off a little carbon dioxide and is coloured black owing to formation of mercuric sulphide which dissolves in excess of the sodium sulphide. With yellow ammonium sulphide the black precipitate a t first produced turns red and cinnabak is produced. Wet mercury fulminate is converted into mercuric sulphide ammonium thiocyanate and carbon dioxide by treatment with hydrogen sulphide. Grey mercury fulminate dissolves in sodium thiosulphate giving an almost clear solution but the solution of white fulminate is always opalescent and a greyish- black mass is gradually deposited which however shows no trace of inercury globules under the microscope.The decrease of alkalinity on keeping of a thiosulphate solution in which mercury fulminate has been dissolved is ascribed to the liberation of acid owing to the oxidising action of fulminat'e on the thiosulphate the reaction being analogous to that with mercuric chloride (Na2S,03~2HgC1,+H20=Na2S0,-tHg,Cl,fS+ 2HC1). The change of thiosulphate into sulphate was followed quantitatively by precipitating the latter with barium chloride after various intervals of time had elapsed and the amount of sulphate formed mas found to be directly proportional t o the amount of fulminate added and inversely proportional to the concentration of the thiosulphate.I n using Brownsdon's titration method for estimating mercury fulminate 23 % of the thiosulphate was found to be converted into sulphate ten minutes after the addition of the fulminate. A thiosulphate solution stronger than N/10 is recommended in this determination. The electrolytic determination of mercury in mercury fulminate was successfully carried out using the following solutions potassium cyanide ammonia pyridine warm 10 yo sodium chloride potassium bromide yellow sodium sulphide (10 yo) 2 yo hydrogen peroxide 5% potassium chlorate sodium hypochlorite hypobromite and hypoiodite and ammonium oxalate. The electrolytic deposition of mercury was not quantitative using sodium thiosulphate potass- ium iodide ammonium thiocyanate or sodium picrate.H. C. R. The Salt-like Nature of Sodium Ethyl. Indirect Electro- lysis of Zinc Ethyl. FRANZ HEIN (2. Electrochem. 1922 28 469471).-A solution of sodium ethyl in zinc ethyl is a good conductor of electricity and can be readily electrolysed giving metallic zinc a t the cathode and a niisture of hydrocarbons a t the anode. The gaseous products consist of 82:/ of ethane and ethylene in equal proportions and the remainder contains hydrogen and butane. Their production can be represented by the equation 2C2H5+2 @ - 2C,H - C,H,+C,H,. These experiments prove the salt-llke nature of sodium ethyl and support the view that the solution contains Na' and either C,H,' or Zn(C,H,),' ions. The specific conductivity of a solution of 2 rnols. NaC,H in 3 mols.Zn(C,H,) is 0.01082 mho at 25". Zinc ethyl on the other hand does not conduct the electric current appreciably . W. E. G.ORGANIC CHEMISTRY. i . 99 Is Kekul6’s Benzene Theory Tenable? C. W. A. LELY (Chern. WeekbM 1922 19 593-598).-A triangular formula is put forward for benzene in which the six carbon atoms lying all in one plane are differentiated into three primary and three secondary. The three primary form a chain of three each having two valencies saturated by the other two primaries and two saturated by neighbouring secondaries. The three secondaries have each two valencies saturated by primaries and two by hydrogen atoms the latter lying in planes perpendicular to the plane of the six carbon atoms ; an additional hypothesis of synchronous rotation of the hydrogen atoms or their substituents accounts for the occur- rence of only three isomeric disubstituted benzenes and the absence of optical isomerism.Numerous reactions are cited to support the new formula and the idea is extended to put forward formulze for naphthalene Willstiitter’s cydooctatetraene anthracene etc. s. I. 1,. Specimens of Cymene and Ethylbenzene of Different Origin. K. VON ATJWERS and H. KOLLIGS (Ber. 1922 55 [B] 3872-3879).-1n a previous communication (A. 1922 ii 174) it has been pointed out that the physical constants of hydrocarbons of the benzene series and in particular the refractive indices are frequently dependent on their mode of production. Since par- ticularly marked differences in the various specimens are noticeable in the case of cymene the hydrocarbon has been re-examined and the observations have been extended to ethylbenzene as a simple representative of the series.It is found that the “ cymene ” prepared from toluene isopropyl bromide and aluminium chloride is contrary to the general rule a meta-derivative since it gives isophthalic acid when oxidised. Specimens of p-cymene obtained from p - p-iodoisopropyltoluene cuminol terpinene a-terpineol and p-aa-dichloroisopropyltoluene and oil of Ajowan have closely similar physical constants if the preparation first mentioned (the uniformity of which is somewhat doubtful) is not taken into account. It is remarkable however that p-cymene derived from camphor has a particularly low refrac- tive index; even lower values have been observed by Wheeler (Schimmel and Co.Rep. 1921 105) for a preparation from spruce oil of turpentine. Specimens of ethylbenzene have been examined which are obtained by reduction of freshly prepared styrene by sodium and alcohol of acetophenone by amalgamated zinc and hydrochloric acid by Fittig’s method from bromobenzene and ethyl bromide by Friedel and Crafts’ process from thiophen-free benzene and ethyl bromide and by the action of concentrated hydrochloric acid at 130” on p-ethylbenzenesulphonamide m. p. Pogo repectively. Although all the preparations are to be regarded as “ pure ” in the generally accepted sense of the term they exhibit differences in their physical constants similar to those observed with cymene. The products obtained from styrene or by Clemmemen’s method have higher densities and refractive indices whereas the constantsi.100 ABSTRACTS OF CHEMICAL PAPERS. of all the other specimens agree well among themselves and with the values recorded previously. Since there can be no question of steric influences in the case of ethylbenzene it appears to be established that the differences are due to traces of impurity which cling obstinately to the various specimens. These cannot be detected by elementary analysis. Attempts to use the determination of the heat of combustion as a criterion of purity do not appear to be successful. H. W. Investigations on the Dependence of Rotatory Power on Chemical Constitution. XVII. A New Type of Walden Inversion. HENRY PHILLIPS (T. 1923 123 44-59). Nitration of 3-Chloroacenaphthene.GLADYS FARNELL (T 1923 123 60-61). Studies in the n-Butyl Series. 11. The Four Stereo- isomeric Py-Di-p-tolylamino-n-butanes. GILBERT T. MORGAN and WILFRED JOHN HICKINBOTTOM (T. 1923 123 97-105). The Reduction of Nitriles with Hydrogen in Presence of Nickel. HANS RUPE and KARL GLENZ (Helv. Chim. Acta 1922 5 937-942) .-When optically active valeronitrile was reduced with hydrogen in presence of nickel instead of the expected hexyl- amine the secondary amine dihexylamine was obtained. Reduc- tion of the nitrile with sodium and alcohol readily gave optically active hexylamine. The reaction appears to be general since phenylacetonitrile and P-phenylpropionitrile likewise gave almost entirely secondary amine when reduced with hydrogen and nickel. a-Cyanocamphor did not give the corresponding secondary amine but a base which has not been identified.When o-tolylcarbylamine was used it acted as a poison towards the nickel catalyst and a poor yield of a base was obtained probably methyl-o-toluidine. The dihexylamine obtained from ordinary amyl alcohol through the cyanide had b. p. 109-115"/12 mm.; its hydrochloride C,,H,,NCl forms lustrous leaves; the nitrosoamine is an oil with an aromatic odour giving Liebermann's nitroso-reaction. Di- p- phenylethylamine (C,H,*CH,-CH,),NH from phenylacetonitrile is a viscous oil ; its hydrochloride C,,H,,NCl forms white lustrous leaflets; the nitrosoamine forms white leaflets m. p. 53-54'. Di-7-phenyZprop ylumine NH( CH,*CH,*CH,Ph) gives a hydro- chloride C1,H,NC1 white silky leaflets ; the nitrosoamine crystallises in white felted needles m.p. 55-56'. E. H. R. Phenyltrimethylammonium Perhalides . HAMILTON Mc- COMBIE and THOMAS HAROLD READE (T. 1923 123 141-153). The Thiocarbimide Reaction. SHINTARO KODAMA (Japan. J . Chem. 1922,1,83-93).-An English version of the paper previously published in Japanese (cf. A 1921 i 237). K. K.ORGANIC CHEMISTRY. i. 101 Preparation of Complex Silver Compounds of Aromatic Thiocarbamides. 3'. HOFFMANN-LA ROCHE & Co. (Swiss Pats. 90808 91780 and 91781 ; from Chem. Zentr. 1922 iv 944-945).- Aryl thiocarbamides of the composition Aryl*NH*CS*NE€ the aryl group being substituted or unsubstituted in excess are treated with silver salts such as silver chloride or silver nitrate. The complex silver compounds of the p-hydroxyphenylthiocarbamide of p-thiocarbamidosalicylic acid OH*C,H,*( CO,H)*NH*CS*NH m.p . 221" are thus prepared. The latter compound is prepared from 4-aminosalicylic acid hydrochloride and ammoniuin thiocyanate. The complex silver compound of 6-amino-3-thiocarbumido-10- methylacridine (from 3 6-diamino- 10-methylacridinium chloride) is also mentioned. L. MASCARELLI (Atti R. Accud. Lincei 1922 [v] 31 ii 116--118).-Since the molecule of o-methylcyclohexanol contains two asymmetric carbon atoms having unequal rotatory values four optically active and two racemic forms of this compound should exist. The compounds obtained by Godchot and Bddos (A. 1922 i 334) and by Sabatier and Mailhe (A. 1905 i 275) probably represent the two racemides. The cycloheptylcycloheptanol prepared by Godchot and Brun (A. 1922 i 350) should exhibit similar optical isomerism to o-methyl- cyclohexanol. An analogous case is presented by the two modi- fications of decahydro-p-naphthol (cf.Mascarelli A. 1911 i 964 ; Mascarelli and Recusani A. 1912 i 761). The Oxide of Methyl-A3-cycZohexene and the Dimethyl- cyclohexanols. MARCEL GODCHOT and PIERRE BBDOS (Compt. rend. 1922 175 1411-1414).-1t has previously been shown (A. 1922 i 334) that the action of organomagnesium complexes on the oxide of cyclohevene gives ortho-substituted homologues of cyclo- hexanol and that the latter compounds are stereoisomerides of the secondary alcohols obtained by the catalytic hydrogenation of the corresponding phenols. It is now shown that in a similar manner methyl-A3-cyclohexene gives a dimethylcyclohexanol different from those obtained by the catalytic hydrogenation of the xylenols. Methyl-A3-cyclohexene oxide b.p 141-142' (corr.) d14 0.949 ng 1.4518 is obtained from methyl-A3-cyclohexene either by the action of iodine and yellow mercuric oxide followed by the action of solid potassium hydroxide on the product in ethereal solution or by oxidation with perbenzoic acid. When heated with water at 130" for six hours it is hydrated giving 1-methykyclohexan-3 4-dio1 in. p. 60° b. p. 112-115" which gives a diphenylurethane m. p. 176-177". The oxide described above reacts with magnesium methyl iodide to yield an alcohol b. p. 173-174" (corr.) which is probably a 1 4-dimethylcyclohexan-2-01. It has dls 0.9106 ng 1.452.It does not give a phenylurethane and on oxidation yields a dimethyl- cyclohexanone b. p. 171" (corr.) &lo 0.9044 ni; 1.4427 which gives a semicarbaxone m. p. 122". These physical constants are different from those of similar compounds obtained from 1 3 4- and G. W. R. The T w o Forms of o-Methylcyczohexanol. T. H. P.i. 102 ABSTRACTS OF CHEMICAL PAPERS. 1 4 2-xylenol by catalytic hydrogenation and subsequent oxid- ation. W. G. Catdytie Hydrogenation of Liquids by means of the Com- mon Metds. VII. Phenols. A N D R ~ BROCHET (Bull. SOC. chim. 1922 [iv] 31 1270-1280).-The effect of temperature and pressure on the catalytic hydrogenation of phenols in presence of a reduced nickel catalyst was studied. With pheno1 itself the absorption of hydrogen commences a t about 50" a t atmospheric pressure and the velocity of the reaction increases rapidly as the temperature is raised to 150".To obtain complete hydrogenation to cyclohexanol within a reasonable time it is necessary to use pressure however and a t 20 atmospheres and temperatures of 200-230" the absorption of 6 atoms of hydrogen per mol. was complete within an hour. For the preparation of pure cyclo- hexanol pressures of 10-20 atmospheres and a temperature of about 150" constitute good practical working conditions and if an apparatus is employed in which the catalyst can be filtered from the hydrogenised product and used again with fresh phenol it is possible completely to reduce 150 g. of the latter per 1 g. of catalyst without any appreciable loss in activity.The cyclo- hexanol obtained boiled a t 160*5"/760 mm. d:5=0.950. The cresols are not hydrogenated appreciably below loo" but a t 150" absorption proceeds with the same facility as with phenol and by using pressures of 10-20 atmospheres reduction to the correspond- ing methyl cyclohexanols was readily achieved. o-Methylcyclo- hexanol boils a t 163-164" m-inethylcyclohexanol a t 170-171" and p-methylcyclohexanol a t 170-171". G. 3'. M. Catalytic Hydrogenation under Pressure in the Presence of Nickel Salts. 111. Phenetidine. JULIUS VON BRATJN and ERICH HAEX (Ber. 1922 55 [B] 3770-3779).-Under the con- ditions adopted by the authors phenetidine can be hydrogenated without loss of the ethoxyl groups; the main product appears to be a mixture of stereoisomeric 4 4'-diethoxydicyclohexylamines OEt.C,H,*NH*C sH,o* OEt . The hydrogenation of phenetidine is effected a t 210-230" in the presence of tetrahydronaphthalene ; the absorption of the gas occurs somewhat slowly and the relative amounts of the various products appear to be considerably affected by unknown catalytic influences. About a quarter of the phenetidine used is converted into a product b. p. 208-216"/16 mni. which can be separated by means of light petroleum into two isomeric 4-ethoxyphenyl-4'- ethoxycyclohexylamines m. p. 78-79" and 37-38" respectively. The isomeride of higher melting point is produced in relatively very small amount so that it has not been possible to make a complete investigation of it. It yields a non-crystalline hydro- chloride picra,te and picrolonate a crystalline quaternary rnethiodide m.p. 156" and an acetyl derivative colourless crystals m. p. 84". The variety of lower melting point yields a non-crystalline hydro- chloride picrate and picrolonate an acetgl compound m. p. 4 2 4 3 " ,ORGANIC CHEMISTRY. i. 103 b. p. %35-240"/18 mm. a nitroso-derivative in. p. 'is" and a very hygroscopic quaternary methiodide (the corresponding methochloride is hygroscopic but yields a well-defined chloroplatinate a rnicro- crystalline orange-red powder). Fission of the quaternary am- monium hydroxide is effected very readily and gives 4-ethoxy-Al- cyclohexene a mobile liquid b. p. 158-160" (which is converted by fuming hydrobromic acid into truns-1 4-dibromocyclohexane m. p. 113") dimethylphenetidine m.p. 35-36" (picrate m. p. l42') and the tertiary base C1,H,,O,N ni. p. 40" which gives a non- crystalline hydrochloride picrate and picrolonatc. 4- Ethoxyphenyl-4'-ethoxycyclohexyZamine is readily hydrolysed by concentrated hydrochloric acid a t the temperature of boiling water the ethoxyl group attached to the cyclohexyl nucleus being first affected . 0H.C ,HIo*NH*C,H4*OEt crystallises in colourless leaflets m. p. 84-85' (di-p-nitrobemoyl derivative m. p. 192"); it loses water when distilled under diminished pressure giving p-ethoxyphenylcyclohexenyhmine OEtbC6H4*NH-C6H? a viscous pale yellow liquid which could not be caused to solidify and has not been investigated further. 4-Hydroxyphen yl-4'- h ydroxycyclohexylamine has m. p 136-1 3'7 O. 4 - E t hox yphen y 1 - 4'-hgd rox y c y clo hex y lme thy lamin e 0Et.C ,H,*NMe*C ,H OH 112.p. 72" yields a p-nitrobenxoyl compound a pale yellow crystal- line powder m. p. 189". The residues left after the distillation of 4 4'-diethox-yphenyl- 4- Ethoxyphenyl-4'-hydroxycyclo hexy Zarnine cyclohexylamine yield a substance C20H2802N2 colourless "&ystds m. p. 169-170"; it yields a di- OEt OEt OEt OEt hgdrochloride decomp. 290" a di- /\ /\ ' \ nitroso-derivative C,,H2,0,N pale [I )-I 11 I 1' ( 11 yellow crystals m. p. 190° and \N/HE>iEt'\<i>dHEt a di-acetyl compound m. p. 240". ,... . .- . One of the annexed formula is (1.1 (11.) suggested. H. W. Investigations on the Dependence of Rotatory Power on Chemical Constitution. XVIII. The Di-Z-menthyl Esters of the Saturated Dicarboxylic Acids.LESLIE HALL (T. 1923 123 105-113). Catalytic Hydrogenation under Pressure in the Presence of Nickel Salts. 11. Hexahydrodiphenylene Oxide from 2 2'-Dihydroxydiphenyl. JULIUS VON BRAUN (Ber. 1922 55 [B] 3761-3770).-The catalytic reduction of 2 2'-dihydroxy- diphenyl follows a somewhat unexpected course which leads to the complete hydrogenation of one nucleus and the formation of the oxide ring. The latter is extraordinarily stable towards further reduction but readily undergoes fission when treated with oxidising agents. Hexuhydrodiphenylene oxide ?GH4->0 a colourless liquid b. p. C6HlOi. 104 ABSTRACTS OF CHEMICAL PAPEFUS. 138-141"/100 mm. is conveniently prepared by treating 2 2'- dihydroxydiphenyl at 230" with four atomic proportions of hydrogen and subsequent fractional distillation of the product.It is in- merent towards sodium and alcohol zinc and acetic acid or magnesium phenyl bromide. It reacts slowly with bromine but does not yield a crystalline product. Warm concentrated sulphuric acid converts it into a monosulphonic acid the sodium salt of which is described. It is converted by cautious nitration with nitric acid (d 1.4) and glacial acetic acid a t -15" 'N into nitrohexahydrodiphenylene oxide (annexed formula) ) m. p. 126" which is reduced by stannous chloride \/\ and concentrated hydrochloric acid to aminohexa- I hydrodiphenylene oxide pale yellow leaflets m. p. 56" /\/ (hydrochloride colourless needles decomp. 250" after m. p. 186"; acetyl derivative m. p. 123"). , I I darkening at 225" and softening a t 235"; picrate Hexah ydrodiphenylene oxide readily becomes resirdied under the conditions of the griedel-Crafts' synthesis but under definite conditions it can be converted into the corresponding methyl ketone (annexed formula) colourless crystals,m. p.67-68". The oxime of the latter m. p. 167' is reduced by sodium and alcohol to aminoethylhexahydro- \ /\ diphenylene oxide a colourless liquid which rapidly I (I absorbs atmospheric carbon dioxide b. p. 210- /'/ 213"/16 mm. (hydrochloride m. p. 267-268" after I 1 darkening a t 264"; chloroplatinate decomp. 200" \/ after gradual darkening ; picrate decomp. 250" after darkening at 240"). It is remarkable that the oxide ring is not ruptured during the reduction. With oxalyl chloride hexahydrodiphenylene oxide gives the acid C02H*%H3->~ colour- less crystals m.p. 250" after softening a t 240" in very poor yield. Hexahydrodiphenylene oxide is readily oxidised by chromic acid in glacial acetic acid to 6-o-hydroxybenzoylvaleric acid OH*C6H,*CO* [ CH2],*C02H long colourless needles m. p. 94" b. p. 240-242"/12 mm. (oxime m. p. 128"; semimrbaxone m. p. 186"; phenylhydrazone yellow leaflets m. p. 173'; benzoyl derivative m. p. 82"). The acid is slowly converted by methyl iodide and alkali in methyl-alcoholic solution into the methoxy-acid OMe*C6H,*CO*[CH,],*C02H m. p. 82" (semicurbaxme m. p. 175-176" ; methyl ester prismatic needles m. p. 28'). 6-o-Hydroxybenzoylvaleric acid is smoothly converted by isatin into 2-o-hydroxyphenyl-3-propylpuinoline-y 4-dimrboxylic acid C,H4< c ~ c 0 ~ H ) ~ ~ * ~ c H ~ ~ 3 ' c 0 ~ H m.p. 295" (decomp.). Molten N=C *C sH,* OH sodium hydroxide rapidly transforms 6-o-hydroxybenzoylvaleric acid into phenol and adipic acid. Nitrohexahydrodiphenylene oxide is oxidised by chromic acid to 8-m-nitro-o-hydroxybenxoylvaleric acid small yellow needles m. p. 116" which does not appear to be readily obtainable by mtration of the parent acid. coYe~>' C6H10 H. W.ORGANIC CHEMISTRY. i. 105 Elimination of the Amino-group of Tertiary Amino- alcohols. I. ALEX. MCKENZIE and ANGUS CAMPBELL RICHARD- SON (T. 1923 123 79-91). Catalytic Hydrogenation of Liquids by means of the Com- mon Metals. VIII. Naphthols. A. BROCHET and R. CORNU- BERT (Bull. Xoc. chim. 1922 [iv] 31 1280-1285).-By the hydro- genation of a- and p-naphthols under pressure using a reduced nickel catalyst both the ac- and ar-tetrahydronaphthols are obtained in each case.The hydrogenation of a-naphthol a t 130" gives a mixture of 85% of ac-tetrahydro-a-naphthol and 15% ar-tetra- hydro-a-nuphthol the former boiling at 139-140"/17 mm. (corr.) and having d;' 1.0896 ng 1.5671 R 44.37 showing an exaltation+0*27. It is a thick liquid becoming brown on prolonged contact with air. The phenylurethane melts a t 121". ar-Tetrahydro-a-naphthol is a solid m. p. 68" (corr.). Hydrogenation of fi-naphthol a t 150" gives 75% of the alicylic tetrahydro-derivative and 25% of the phenolic derivative. ac-Tetrahydro- p-naphthol is a viscous liquid which darkens on contact with air and on keeping. It boils a t 1445- 145*6"/20 mm.(corr.) G7 1.0715 mg 1.5523 R 44-16 showing exaltation+0.06. The phenylurethune melts at 99". ar-Tetrahydro- /%naphthol is a solid m. p. 57.5". Benzo-polymethylene Compounds. 11. Hydroxy-bases and P-Ketones of the Tetrahydronaphthalene and Hydrindene Series. JULIUS VON BRAUN OTTO BRAUKSDORF and GEORG KIRSCHBAUM (Bey. 1922 55 [B] 36484663) .-The bromine atom of ac-2-bromo - 1 -hydroxy (alkyloxy) tetrahydronaphthalenes (A 1921 i 407) can be replaced readily by basic residues. The hydroxy-bases formed in this manner differ greatly from ac-p-amino- tetrahydronaphthalene and its alkyl derivatives since they are relatively non-toxic and have a more or less marked antipyretic character whereas the corresponding alkyloxy-compounds are strongly poisonous. The constitution of the latter substances is established by the mode of production but in the case of the hydroxy-compounds the possibility that the hydroxyl group may occupy position 2 is not defiiritely excluded (cf.Straus and Rohr- backer A. 1921 i 171); comparative experiments on the fission of piperidino-derivatives by cyanogen bromide indicate that this is not the case and that the striking difference in physiological properties is therefore attributable solely to the replacement of the hydroxyl by the alkyloxyl group. The following alkyloxy-bases are described 2-Dimethyhmino- CH(OEt)*yH*NMe b. p. 1 - et hox yt etrahydromp hthalene C 6H4<CH CH 152"/13 mm. and its oily hydrochloride picrate m. p. 199" and methiodide m. p. 166" ; p-piperidino- 1 -ethxytetrahydronuphthalene an almost colourless liquid b.p. 170-172"/10 mm. ; 2-diethylamino- 1 -all yloxyt etrah ydronuphthulene b . p . 1 65 " / 10 mm . ac- 1 -Aminotetrahydronaphthalene is readily prepared by the reduction of 1 -oximinotetrahydronaphthalene with sodium and G. F. M.i. 106 ABSTRACTS OF CHEMICAL PAPERS. alcohol and is converted by ac-dibromopentane in boiling alcoholic solution into ac-1-piperidinotetrahydronaphthlene a colourless liquid b. p. 1'74-176"/17 mm. The hydrochloride hydrobromide and methiodide could not be caused to crystallise; the chloro- platinate is amorphous whereas the picrate crystallises in dark yellow needles m. p. 145-146" after softening a t 140". ac-2- Piperidinotetrahydronaphthlene is a colourless odourless liquid b.p. 186-187"/16 mm. (hydrochloride m. p. 230-231"; hydro- Iwonzide m. p. 233-234"; methiodide m. p. 209"; picrate m. p. 203-204"). Both piperidino-compounds suffer fission under the influence of cyanogen bromide but the production of bromotetra- hydronaphthalene is much more marked with the 1- than with the 2-derivative. 2-Piperidino- 1 -hydroxytetrahydronaphthalene (cf. Straus and Rohrbacker loc. cit.) has b. p. 192-194"/14 mm. m. p. 75-76' (hydrochloride m. p. 185-186"; picrate m. p. i 52" ; chloroplatinate m. p. 192"). 2-Piperidino-1-benzoyloxy- tetrahydronaphthalene is acted on by cyanogen bromide giving cyanopiperidine in a yield which points to the presence of the basic group in the 2- rather than in the 1-position. ac-2 - Amino- 1 -hydroxytetrahydrompht~lene m .p . 109 " b . p . l6Oo/ll mm. is prepared by the action of concentrated ammonia solution on the corresponding bromo-compound. The hydro- chloride m. p. 227"; chloropkltinute m. p. 215"; picrate yellow leaflets m. p. 192" ; phenylthiocarbamide m. p. 134" ; benxylidene derivative m. p. 115"; salicylidene derivative m. p. 108"; acetyl derivative m. p. 203" ; p-nitrobenxoyl deriva'tive m. .p. 228" ; p-aminobenxoyl derivative m. p. 208" and the hydrochlorzde of the latter m. p. 239" are described. When the base is warmed with an equivalent quantity of ethylene oxide in chloroform solution it is converted into 2 - p - hydroxyethylamino- 1 - hydroxytetrahydronuph- thlene C6H4<CH,-CH a very viscous liquid b. p. 190-196"/vacuum ; the corresponding hydrochlcy-ide has m.p. 155". 2-Benxylamino-1-hydroxytetrahydronaphthalene is a colourless viscous liquid b. p. 243-246"/13 mm. (picrate m. p. 192" ; hydrochloride m. p. 237"). 2-Nortropyl-1-hydroxytetrahydro- naphthulene b. p. 210-212"/14 mm. yields a picrate m. p. 174" and an extremely hygroscopic hydrochloride. 2-Dimethylamino- 1-hydroxytetrahydronaphthalene m. p. 40" (cf. Straus and Rohr- backer loc. cit.) yields a picrate m. p. 138" and a methiodide m. p. 138-139"; the benzoate and acetate and its salts show little tendency towards crystallisation. The p-nitrobenxoate and p-arnino- benzoate have m. p. 112" and 137" respectively. 2-Methylamino- 1-hydroxytetrahydronaphthalene picrate has m. p. 1'72". 2-Diethyl- amino-1 -1iydroxytetrahydronaphthalene gives a crystalline chZoro- platinate m.p. 192" and a liquid acetate. If the methiodide of 2 - dimethylamino- 1 - hydroxytetrahydro - naphthalene is treated with silver oxide and the solution thus obtained is heated tetrahydronaphthalene oxide is produced. When the iodide is heated alone at a temperature very slightly above its melting point i t is decomposed into trimethylarnine CH( OH)*yH*NH*CH,*CH,*OHORGANIC CHEMISTRY. i. 107 hydrochloride and 2-ketotetrahydronaphthalene b. p. 140"/20 mm. 130"/10 mm. di6'9 1-1055 (semicurbaxone m.. p. 190-191") the yield being 80-85% of that theoretically possible. The behaviour of the ketone does not appear to be well expressed in the formula C6H4<CH,.CH,. which does not explain the formation of an intense blue colour under the influence of air and alkali the difficulty with which it forms additive compounds and the apparent absence of an activated methylene group.The application of hard's reaction to tetrahydronaphthalene gives a mixture of the 1- and 2 -ketones. CH2-70 ac-2 - Dimet hyhmino- 1 - hydroxyhydrinden e C H < ~ ~ ~ ~ 1 > O H * N M e 2 a colourless crystalline subsiance which rapidly darkens when exposed to air b. p. 153-156"/9 mm. m. p. 62" gives a hydro- chbride m. p. 183-184" a picrate m. p. 145" and a methiodide m. p. 161-162". The latter substance is decomposed when dis- tilled under diminished pressure into trimethylamine hydriodide and P-ketohydrindene; the conditions of the change are more drastic and the yield of ketone is smaller than with the corresponding tetrahydronaphthalene derivative.H. W. Benzo-polyrnethylene Compounds. VII. Pharmaco- logical Analogues of ac-2-Aminotetrahydronaphthalene. JULIUS VON BRAUN HEINRICH GRUBER and GEORG KIRSCHBAUM (Ber. 1922 55 [B] 36364-3674).-Substances which contain an aliphatic amino-group in the P-position with respect to an aromatic nucleus have the property of causing an increase in the blood pressure. In addition to this property ac-2-aminotetrahydro- naphthalene has also a mydriatic action and causes a marked increase in the body temperature. The effects obviously depend on the simultaneous presence of an aromatic and a hydroaromatic ring. The examination of a number of analogous compounds has shown that similar physiological properties are exhibited by sub- stances which contain an aliphatic amino-group in the P-position to the aromatic portion of the tetrahydronaphthalene complex ; if this condition is fulfilled the further mode of attachment is immaterial.The catalytic reduction of o-hydroxydiphenyl with hydrogen under pressure a t 210-220" in the presence of nickel salts to 2-phenylcycZohexano1 is attended with considerable experimental difficulties since the completion of the reaction is not marked by any abrupt change in the rate of absorption of the gas. The products obtained when the action is continued to the almost complete cessation of absorption are dicyclohexyZ b. p. 103- 105'/12 mm. and 2-cycZohexylcycZohexano1 b. p. 134-135"/12 mm. which is probably a mixture of cis- and trans-isomerides (cf. Wallach A. 1911 i 473). The latter substance is oxidised by chromic acid to 2-cyclohexyZcyclohexano~ze b.p. 128-130"/10. mm. (benzylidene compound m. p. 100"; semicarbazone rn. p.i. 108 ABSTRACTS OF CHEMICAL PAPERS. above 200"). Less complete reduction of o-hydroxydiphenyl gives a mixture of phenylcyclohexane dicyclohexyl 2-phenylcycZohexano1 and 2-cyclohexylcycbhexanol. The alcohols are nearly exclusively produced when only six atomic proportions of hydrogen are used but they cannot be separated conveniently from one another. Oxidation of the mixture with chromic acid gives the corresponding ketones from which 2-phenylcycZohexanone is isolated partly in substance and partly as the oxime slender needles m. p. 1 7 6 1 7 5 " ; 2-phenylcyclohexanonesemicarbaxone has m. p. 193". 2-Phenyl- cyclohexanol prepared by the reduction of the pure ketone with sodium and alcohol has b.p. 143-144"/11 mm. m. p. 5 6 - 5 5 " and yields a phenylzrrethane m. p. 138-139'. 2-PhenylcycZo- hexanoiieoxime is reduced smoothly by sodium and alcohol to 2-phenylcyclohexyZamine b. p. 133-134"/12 mm. m p. 59-60'. The hydrochloride m. p. 253" chloroplatinate reddish-yellow needles decornp. 222" ncetyl derivative m. p. 130" phenylthiocarbamide m. p. 185" and methiodide m. p. 235" are described. The base causes an increase in the blood pressure; but is not otherwise similar to ac-2-aminotetrahydronaphthaleneY probably for the reason that the association of the aromatic and hydroaromatic nucleus is not sufficiently close. ac- l-Aminomethyltetrahydronaphthalene is obtained in 20% yield by -the reduction- of a-naphthonitrile ; the hydrochloride m.p. 230" picrate m. p. 170" and benzoyl deriv- ative m. p. 125-126" are described. The following process does not give the base in better yield. a-Ketotetrahydronaphthalene is condensed with zinc and ethyl bromoacetate to form the unsatur- ated ester b. p. 183-184"/16 mm. which appears to have the double bond in the nucleus as shown bv the formula It is readily hydrolysed t o the ckesponding a&d colourless needles rn. p. lOO" which is slowly hydrogenated in the presence of palladium chloride to 1-tetrahydronaphthylacetic acid m. p. 35-36'. The latter is convertible through the amide but in small yield into ac- 1 -aminomethyltetrahydronaphthaIene. The physiological properties of the base are similar to but weakerthan those of ac-2-aminotetrahydronaphthalene.ar- 1 -BenzoylaminomethyltetrahydronaphthaIene is converted by phosphorus pentachloride into ar- 1 -chZoromethyltetrahgdronaphthal- ene b. p. 144-145"/13 mm. m. p. 50-51". (ar-2-ChZoromethyZ- tetrahydronaphthalene has b. p. 141-142"/12 mm. but does not solidify when cooled.) It is converted smoothly by potassium cyanide in aqueous alcoholic solution into ar- 1 -cyanomethyltetra- hydronaphthlene colourless needles m. p. '\/\ 69-70' b. p. 168-169"/10 mm. The latter 1 is reduced by sodium and alcohol to ar- >'\/ l- P-aminoethyZtetrahydronaphthulene (annexed NH,*CH,*CH formula) a colourless somewhat viscous ItORGANTC CHEMISTRY. i. 109 liquid b. p. 146-149"/16 mm. The corresponding hydrochloride m.p. 245" (decomp.) picrate small leaflets m. p. 231" and benxoyl derivative m. p. 123" are described. Physiologically the base closely resembles ac-2-aminotetrahydronaphthalene. H. W. P. FRIED- UNDER and A. SIMON (Ber. 1922 55 [B] 3969-39SO).-An- thracene is readily converted by sulphur chloride into 9-anthryl dithiochloride C,,H,*S,*Cl yellowish-red lustrous crystals m. p. 117-118" (decomp.) when rapidly heated; it reacts readily with ammonia and aromatic bases but only gives resinous products. With dimethylamine it gives the compound C,4H,=S,-NMe yellow prisms m. p. 70-71" ; the corresponding piperidide is described. A solution of anthryl dithiochloride in benzene is converted by a concentrated solution of sodium sulphite a t the atmospheric tem- perature into sodium 9-anthryl thiosulphate C,,H,*S*SO,Na pale yellow leaflets ; the corresponding barium salt is intensely yellow but appears to yield a colourless hydrate.9-Anthryl hydrogen thiosulphute C,,H,*S*SO,H crystallises in slender colourless needles; it decomposes slowly in warm aqueous solution rapidly in the presence of hydrochloric acid into sulphuric acid and 9-anthryl disulphide lustrous orange-yellow octahedra m. p. 223" (see later) ; this compound (in addition to thiosulphate) is also produced when a solution of the sodium salt is warmed with sodium hydr- oxide. 9-ThioZanthrucene orange-yellow octahedra m. p. 90-91" is most conveniently prepared by the action of anthryl dithiochloride on a solution of hydrated sodium sulphide in methyl alcohol; the corresponding sodium salt forms orange-yellow leaflets and the methyl ether needles m.p. 153". The thiol is the sulphur analogue of anthranol (Meyer A 1911 i 193) but unlike this compound it exhibits little tendency to pass into a substance of the anthrone type. 9-Anthryl disulphide (see above) is obtained quantitatively by the oxidation of an alkaline solution of the thiol with potassium ferricyanide. 9-Anthrylthiohcetic acid Ci,HS*S*CH,*CO,H slender very pale yellow needles m. p. 164" is prepared by the addition of the sodium compound of 9-thiolanthracene to an alkaline solution of sodium chloroacetate ; the sodium salt lustrous leaflets am- monium salt and the methyl ester small yellow needles m. p. 67" are described. The acid is converted by phos- CH phorus pentachloride in the presence of light petrol- /\ eum (but less conveniently by thionyl chloride 7 90 which also causes chlorination) into 9-anthrylthiol- ,/\ ,/\/\ acetyl chloride compact yellow needles from which I I I I the corresponding amide colourless needles which \/\/\/ soften but do not melt a t 197" is prepared.The chloride is converted by aluminium chloride in the presence of light petroleum at 30-40" into 3-keto-peri-anthra- cenopenthiophen (annexed formula) sma11 tile-red crystals m. p. 150-152". The substance is transformed by hot nitrobenzene into The Action of Sulphur Chloride on Anthracene. bis-peri-anthracenope?zth~op~e?z C,,H8<&>C:C<&>C,,H darki . 110 ABSTRACTS OF CHEMICAL PAPERS. green needles and condenses with a-isatinanilide in the presence of yyridine to give 3'-indoxyl-(2 2')-peri-anthracenopenthiophen C,,R,<~,>C:C<~~>C,H dark coloured needles.Investigations on the Dependence of Rotatory Power on Chemical Constitution. XV. Some n-Alkyl Ethers of d- Benzylmethylcarbinol. HENRY PHILLIPS (T. 1923 123 22- 31). X. spim-Compounds Derived from cycloHeptane. JOHN WILLIAM BAKER and CHRISTOPHER KELK INGOLD (T. 1923 123 Preparation of Esters. LABORATOIRE-USIRE (F.P. 531960 ; from Chern. Zentr. 1922 ivy 943).-Alkyl carboxylates are heated with aluminium derivatives of alcohols. The reaction takes place according to the equabion 3R*CO2R'+A1( OR"),=3R*C0,Rf'+ &(OR') where R" has a higher molecular weight than R'. The aluminium alkyloxides are prepared by heating the corresponding alcohols with aluminium amalgam or other easily decomposed aluminium alloys.For example aluminium amyl oxide and ethyl acetate give amyl acetate. Amy1 phenylacetate a-phenylethyl acetate p-phenylethyl phen ylacetute and linalyl acetate are similarly prepared. G. W. R. L. G. RADCLIFFE and W. H. BRINDLEY (Perf. Essent. Oil Rec. 1922 13 414415).-Anisic acid was obtained in almost theoretical yield by the oxidation of anis- aldehyde with alkaline permanganate. Cannizzaro's method- treatment of the aldehyde with potassium hydroxide-proved unsatisfactory large amounts of unchanged material being recovered. The following esters were prepared by saturating solutions in the respective alcohols with dry hydrogen chloride. Methyl ester m. p. 48" b. p. 256".Ethyl ester m. p. 7-So b. p. 263" d:66 1.106 n1u9'5 1.5245. n-Propyl ester b. p. 176"/15 mm. d:6'.i 1.09 nl,s'5 1-5149. n-Butyl ester b. p. 183"/40 mm. di6'5 1.054 ~2;'~ 1.5141. isoButyl ester b. p. 170"/46 mm. di6'5 1,052 1.5072. isoAmyl ester b. p. 188"/30 mm. CZ:'~ 1.040. Phenylpropylester b. p. 256"/35 mm. d:";" 1.111 nK5 1.5623. Only the lower members of the aliphatic series have pronounced odours the higher members are practically odourless. G. F. M. Preparation of Alkyl Dihydroxynaphthoylbenzoates. 91106 and 91107; from Chem. Zentr. 1922 iv 890).-1 5- or 1 6-Dihydroxynaphthoyl-o-benzoic acid is esterified in the usual way with aliphatic alcohols. Ally1 1 6-dihydroxynaphthoyl-o- benzoate forms white prismatic needles ni. p. 128". Ethyl 1 ti- dihydroxymphthoyl-o- benzoate forms slightly yellow crystals in.p. 156". Ethyl 1 5-dihydroxynaphthoyl-o-benzoate forms white needles m. p. 146". G . W. R. H. W. The Formation and Stability of spiw-Compounds. 122-133). Some Esters of Anisic Acid. SOCIETY FOR CHEMICAL INDUSTRY IN BASLE (swiss Pats. 90806,ORGANIC CHEMISTRY. i. 111 The Degradation of Hydroaromatic Acids of the Glutaric Acid Series. A. WINDAUS F. KLANHARDT and G. REVEREY (Ber. 1922 55 [B] 3981-3987).-1n a previous communication (Windaus and Klanhardt A. 1921 i 392) it has been shown that the silver salts of the aliphatic glutaric acids react with iodine in accordance with the schemes C0,Ag*[CH,]3*C0,Ag+12=2AgI+ -OGO*[CH,],*CO*O- and -O*CO-[CH,],-CO=O- -+ (a) ~O*[CH,],-CO+O or ( b ) ~O*LCH,],-CH,+CO or (c) CH,:CHMe+ BCO,.The observations have now been extended to acids which may bo regarded as glutaric acids of which one or more carbon atoms are members of a hydroaromatic ring. The action is found to occur normally in the case of the cis-acids unless the glutaric acid substitutes a five-membered ring in the 1 3-position. Silver cyclohexanediacetate is converted by iodine into the _- - 0- -1 - o--' Zuctone CH,< CH2*CH2>~<CH2Y it colourless viscous liquid CH,*CH CH,*COy b. p. 273"/759 Gm. (slight de:omp.) 154"/16 mm. di!:! 1.0755 nz 1.48386 n 1.48668 n; 1.48631. The barium and silver salts of the corresponding hydroxy-acid have been prepared. The lactone is oxidised by alkaline permanganate or by potassium dichromate and sulphuric acid to cyclohexane-1 -acetic-1 -carboxylic v acid CH,<CH,.CH CH2'CH2>C(C02H)*CH2*C02€€ colourless prisms m.p. 131" (uncorr.) [Gf. NOOrris and Thorpe T. 1921,119 12061. Silver cis-hexahydrohomophthalate is readily converted by iodine into hexahydrophthulide FH2*CH2*YH*CH~ 0 a colourless liquid b. p. 134-138"/25 mm. the constitution of which is estab- lished by its oxidation with potassium dichromate and sulphuric acid to cis-hexahydrophthalic acid; the same lactone is obtained in smaller yield and with greater difficulty from silver truns- hexahydrohomophthalate. isoPhthalic acid is hydrogenated in glacial acetic acid solution to a mixture of cis- and truns-hexahydroisophthalic acids in which the former preponderates the more considerably when the reduction proceeds rapidly; the cis-acid however is not converted into the trans-variety if subsequently agitated with glacial acetic acid and platinum black.Silver cis-hexahydroisophthalate and iodine give the lactone of cis-cyclohexanol-3-carboxylic acid b. p. 127- 135O/2l mm. (cf. Perkin and Tattersall T. 1907 91 488) the identity of which is established by converting it into cis-cyclo- hexanol-3-carboxylic acid m. p. 132-133" and cis-3-bromocycb- hexanecarboxylic acid m. p. 61-63". Silver camphorate is transformed by iodine into camphorio anhydride; a neutral lactone does not appear to be produced. CH,*CH,*CH*CO H. w. The- Chemistry - of the Glutaconic Acids. XIII. The JOCELYN FIELD THORPE Isomerism due to Retarded Mobility. and ARTHUR SAMUEL WOOD (T. 1923 123 62-64).i. 112 ABSTRACTS OF CHEMICAL PAPERS.a-Benzyl-P-methylglutaconic Acids. PRANZ PEIST and ED. RATJTERBERG (Ber. 1922 55 [B] 3697-3705).-In extension of previous investigations (A. 1922 i 521 522 553) the authors have examined the ozonisation of a substituted glutaconic acid which might be expected to yield stable fission products and for this purpose have selected a-benzyl- @-methylglutaconic acid (cf. Bland and Thorpe T. 1912 101 1740). It is shown that the (normal) acid of higher melting point and its esters and also the esters of the labile acid are constituted in accordance with the annexed formula (I) whereas the labile acid appears to be a mixture of the two trans-forms (I1 and 111). The complex nature of the CH2Ph*R* C0,H CH,Ph*E*CO,H CH,Ph*CH (CO,H)*$Me Me*C* CH *CO ,H CO,H* CH,*C*Me H*C*CO,H (1.1 (11.1 (I11 .) latter product explains its low melting point as compared with that of the homogeneous cis-acid.cis- a-Benzyl- P-methylglutaconic acid gives an oxonide which is hydrolysed to phenylpyruvic acid m. p. 153-154" [oxime m. p. 159" phenylhydraxone m. p. 187-188" (decomp.)] acetone and carbon dioxide. The ozonide of its ethyl ester similarly yields ethyl benzylglyoxylate and ethyl acetoacetate. trans-a-Benzyl- @-methyl- glutaconic acid is converted successively into its ozonide and acetic acid phenylethyl methyl ketone and benzoic acid ; the formation of oxalic and glyoxylic acids could not be established. The oxonide of ethyl trans-a-benzyl- @-methylglutaconate is hydrolysed to acetic acid ethyl acetoacetate ( ?) ethyl phenylpyruvate and benzoic acid.Ethyl phenylpyruvate is converted into a diphenylhydraxone CH,Ph*C( :N*NPh,)*CO,Et small yellow crystals m. p. 105" and a p-nitrophenylhydrazone pale yellow crystals m. p. 181". Ring-chain Tautomerism. IV. The Effect of the Methyl Ethyl Grouping on the Carbon Tetrahedral Angle. BALBIR SINGH and JOCELYN FIELD THORPE (T. 1923 123 113-122). M. GOMBERG and D. L. TABERN ( J . Id. Eng. Chem. 1922 14 1115-1117).-Pure tetraiodo- fluorescein was prepared by direct halogenation of fluorescein in hot dilute acetic acid with excess of iodine and subsequent puri- fication by washing with dilute sulphuric acid and alcohol and precipitation from solution in dilute sodium hydroxide. This product did not contain free iodine but was amorphous and the iodine content was low owing to the presence of about 6% of di- iodofluorescein from which it was purified by conversion by boiling with acetic anhydride into the diacetate which after three recrystallisations from bromobenzene alternated with acetone was obtained analytically pure and melting a t 293-294".Hydrolysis of the acetate gave pure crystalline tetraiodofluorescein. This was converted into erythrosin by neutralisation with sodium carbonate in suspension in absolute alcohol. The salt was deposited in red H. W. Composition of E+ythrosin.ORGANIC CHEMISTRY. i. 113 crystals after concentration of the solution and addition of ether. It contained both water and alcohol of crystallisation and the former was not completely expelled even by long d r p g at 160- 170".The fully hydrated salt contains 4H20 and the conclusion seems justified that the true composition of erythrosin is C&,O5I4Na2,H20 and that it is almost impossible to dehydrate it beyond this point without risk of decomposition. G. F. M. The Semi-pinacolinic Transformation of Alkylhydro- benzoins Influence of the Alkyl Groups. MARC TIFFENEAU and ALEX. OR~KHOFF (Compt. rend. 1922 175 964-967; cf. Meerwein A 1920 i 2 ; Orekhoff and Tiffeneau A 1922 i 458).- The transformation of alkylhydrobenzoins under the influence of concentrated sulphuric acid may take place in two different ways yielding either a phenyl a-alkylbenzyl ketone or a benzhydryl alkyl ketone and the authors have investigated the influence of the nature of the alkyl group on the relative proportions of the resulting products.The conclusion is drawn that alkyl groups such as methyl isobutyl and phenyl which have a strong " satur- ation capacity" (Meerwein loc. cit.) yield the former type of product and conversely. The groups of intermediate " satur- ation capacity "-ethyl butyl propyl isoamyl-form hydro- benzoins which yield on dehydration mixtures of the two possible products. The results show that the influence of the alkyl group may render the secondary less stable than the tertiary hydroxyl group. This property is evident only in presence of concentrated sulphuric acid and may result from a temporary linking of the acid with the secondary group. H. J. E. The Action of Alcohols on Phenyl a-Bromostyryl Ketone. Formation of Various Saturated and Ethylenic Compounds.CH. DUFRAISSE and P. G~RALD (Bull. Xoc. chim. 1922 [iv] 31 1285-1304) .-Phenyl a- bromostyryl ketone readily unites with alcohols in presence of small quantities of the corresponding sodium alkoxide giving a- bromo- P-alkoxybenzylacetophenones of the general formula COPh*CHBr*CHPh*OR which can be isolated if the reaction is conducted within suitable narrow limits of tem- Derature. but which when warmed with the reaction mixture. lose hydrogen bromide giving phenyl p-alkoxystyryl ketones COPh*CH:CPh*OR. The course of the reaction and the nature of the product initially obtained by Wislicenus (A. 1900 i 37) by the action of alcoholic potassium hydroxide on phenyl p-bromostyryl ketone is thus elucidated. The phenyl P-alkoxystyryl ketones are obtained with even greater facility and in purer condition by the addition of alcohol to benzoylphenylacetylene in presence of sodium alkoxide. All attempts to prepare the two stereoisomerides which are theor- etically possible in the case of both the saturated and unsaturated compounds or to prepare derivatives from secondary alcohols or phenols failed.The preparation and properties of the following compounds are described. Phenyl a-bromo- p-ethoxystyryl ketone VOL. CXXTV. i. fi. 114 ABSTRACTS OF CHEnlICAL PAPERS. m. p. 60-61' b. p. 182-183'/3-4 mm. white crystals readily decomposed by alkalis and hydrolysed to dibromobenzylaceto- phenone by hydrobromic acid. a-Bromo- p-methoxybenzylacetvphen- one white crystals m. p. 76-77". a-Bromo-p-propoxybenzylaceto- phenone m.p. 95-96". u-Brmo-p-butoxybenxylacetophenone m. p. 81-82'. a-Bromo-p-isobutoxybenxylacetophenone m. p. 110-111". Phmyl p-ethoxystyryl ketone yellowish-white crystals m. p. 77- 78" b. p. 209'15 mm. readily hydrolysed to dibenzoylmethane by boiling with 15% alcoholic hydrogen chloride. Phenyl p-methoxy- styrp? ketone whitish-yellow crystals m. p. 65-66". Phenyl P-propoxystyryl ketone m. p. 59-60". Phenyl p-butoxystyryl ketone a yellow oil b. p. 204-206"/2 mm. Phenyl p-isobutoxystyryl ketone yellowish crystals m. p. 55-56'. Phenyl p-isopropoxystyryl ketone crystals m. p. 49-50" b. p. 180-183'/2-3 mm. G. F. M. Preparation of Condensation Products of ap-Unsaturated Ketones and Phenols. CHEMISCHE FABRIKEN VORM. WEILER- TER MEER (D.R.-P. 357755; from Chem. Zentr. 1922 ivy 890- 891) .-up-Unsaturated ketones in the presence of acids or their acid additive products are condensed with phenols.The reaction in the case of the unsaturated ketones is as follows CR With reactive phenols such as resorcinol or N-substituted m-amino- phenols the reaction is almost quantitative. The compound from the condensation of phenol and styryl methyl ketone is a grey powder. The compound from resorcinol and styryl methyl ketone is a yellow powder. The compound from the hydrogen chloride additive product of phenyl styryl ketone (3-chloro-1-keto-1 3- diphenylpropane) and resorcinol has a red colour ; it forms an acetyl derivative (annexed formula). Other con- densation products are formed from phenyl /\/\CH m-hydroxystyryl ketone and pyrogallol OAc I I ICHPh (brown) ; from phenyl styryl ketone and pyro- \/\/ gallol (bluish-grey) ; from phenyl p-dimethyl- aminostyryl ketone and m-dimethylamino- phenol (greyish-blue) ; from resorcinol and thiodiketobenzylidenethi- azolidine (reddish-brown) ; from styryl methyl ketone and p-cresol (light brown) ; from p-acetamidophenyl o-chlorostyryl ketone and quinol (reddish-violet).G. W. R. CPh 0 Crystallographic Study of uy-Diketohydrindene. ANGELO PICHSTW (Atti R. A c d . Lincei 1922 [v] 31 ii 143-146).- ay-Diketohydrindene d21 1.37 forms crystals belonging t o the hipyramidal tetragonal clags of the tetragonal system a c= 1 0.9414. T. H. P.OWAXIC CHEMISTRY. i. 115 o-Quinones and 1 2-Diketones. VI. JI-Benzils. II. Benzils of the Peroxide Type 2 2'-Diethoxybenzil a De- rivative of Benzil which is Colourless in Solution.A. SCH~N- BERG and W. M A L ~ O W (Ber. 1922 55 [B] 3746-3752).-1n a previous communication (Schonberg and Kraemer A. 1922 i 663) the isolation of a number of colourless solid benzils has been described which give more or less intensely coloured solutions. The superoxide structure TogR has been assigned to the colour- less and the diketonic structure to the coloured benzils. This con- ception is strengthened by the observation that the faintly coloured solutions react less readily than those which are intensely coloured with the usual reagents for 1 2-diketones. In 2 2'-diethoxybenzil colourless quadratic leaflets m. p. 157" the authors have now found a substance which not only is colourless when solid but also yields colourless solutions in cold alcohol acetic acid benzene and light petroleum (b.p. 100-llO") which become pale yellow when heated. It melts to a dark yellow liquid which re-solidifies to colourless crystals. It is unimolecular in its colourless solutions so that the absence of colour cannot be attributed to polymerisation. It is an unusually stable compound which is indifferent to con- centrated aqueous ammonia under pressure and is not affected by hydrogen peroxide in acid or alkaline solution. It does not react with naphthylenediamine hydrochloride in boiling glacial acetic acid but in the presence of boiling dimethylaniline it is slowly converted into 2 3-di-o-phenetyEu~-naphthaquinoxaline O*CR OEt*C6H4*$!:N OEt*C,H4*C:N>C1*6' colourless pointed prisms m.p. 180". Similarly 2 2'-dimethoxy- benzil is transformed by o-phenylenediamine hydrochloride in the presence of boiling dimethylaniline into 2 3-di-o-unisylpuino~line colourless quadratic prisms m. p. 183" by 3 4-diaminofoluene hydrochloride into 2 3-di-o-anisyZ-6-methyZquinomline colourless prisms m. p. 135" and by naphthalene-1 2-diamine hydrochloride into 2 3-di-o-anisyl-up-~pht~qui~~line colourless prisms m. p. 180". The method of condensation appears to be of general applicability . 2 2'-Dimethoxybenzil is reduced by amalgamated zinc and concentrated hydrochloric acid to up-di-o-anisylethane m. p. 86" (cf. Spath A 1914 i 1). 4 ; 4'-Diethoxy-~-phenylbenzoin E t O*C,H4*C( OH)Ph*CO *C6H,*OE t colourless needles m. p. l l l " is prepared by the addition of an ethereal suspension of 4 4'-diethoxybenzoin to an ethereal solution of magnesium phenyl bromide.3 3'-Dimethoxybenzoin colourless prisms m. p. 55" obtained by the action of potassium cyanide on m-methoxybenzddehyde is oxidised by Fehling's solution in boiling aqueous alcohol to 3 3'-dimethoxybenzil yellow prisms m. p. 83". The latter sub- stance is converted by o-phenylenediamine hydrochloridc in the f si. 116 ABSTRACTS OF CHEMICAL PAPERS. usual manner into 2 3-di-m-anisylquino~aZine~ colourless leaflets m. p. 110". H. W. o-Quinones and 1 2-Diketones. VII. +b-Benzils. 111. Separation of a 1 2-Diketone into its Coloured Crystalline Ketonic and its Colourless Crystalline Peroxide Form. A. SCHONBERG and W. BLEYBERG (Ber. 1922,55 [B] 3753-3758; cf.A. 1922 i 163 and preceding abstract).-4 4'-Dibenzyloxy- benzil has been prepared in coloured and colourless forms. A solution of 4 4'-dihydroxybenzil in absolute alcohol is treated with the calculated quantity of potassium ethoxide and a slight excess of benzyl bromide whereby the diketonic form of 4 4'-di- benzyloxybenzil is obtained as dark yellow prisms m. p. 126". If the hot concentrated solution of the dibenzyl ether in alcohol glacial acetic acid or light petroleum (b. p. 100-110") is suddenly cooled by immersion in ice-water the colourless peroxide variety ?*f?c6H4*0*CH2Ph m. p. about 124" after becoming distinctly O*C~C6H4*O*CH2Ph' yellow a t 121" separates. The success of the isolation depends greatly on the rapidity with which the cooling is effected so that it is only possible to work with small quantities a t a time.A further essential condition is that the mother-liquor should be removed as rapidly as possible since although the colourless form is stable when dry it rapidly passes into the yellow variety when in contact with solvents; this change occurs with such rapidity that the colourless compound appears to yield immediately coloured solutions. 4 4'-Dibenzyloxydibenzyl is rapidly converted by hydrogen peroxide in boiling alcoholic solution in the presence of sodium hydroxide into p-benzyloxybenzoic acid m. p. 189" (cf. Cohen and Dudley T. 1910 97 1732). It is converted by naphthalene- 1 2-diamine hydrochloride in the presence of boiling dimethyl- aniline into the corresponding wphthaquinoxaline C38H2802N2 prisms m.p. 156". H. W. E. GRAND- MOUGIN (Compt. rend. 1922 175 970-973).-The introduction of acyl groups into aminoanthraquinones results in the formation of a series of substances the tinctorial characters of which depend on the nature of the substituent entering the amino-group. If a hydroxyl group which acts as an auxochrome is present in addition acetylation or benzoylation of this group brightens the colour but usually diminishes the value of the substance as a dye. In the case of a-benzamidoanthraquinone the colour is changed from bright yellow to deep rose by the introduction of a hydroxyl group in the para position and to scarlet by a methoxyl group. Further addition of benzamido-groups results in a deepening of the colour.Isomerism influences the colour and in addition the tinctorial properties ; the author states that no adequate explanation of these facts has yet been put forward. The following substances all crystalline and of high m. p. were prepared; their colours and the tints they impart to vegetable fibres are noted 4-Benxamido- The Acylamidoanthraquinones as Vat Dyes.ORGANIC CHEMISTRY. i. 117 I-hydroxyanthraquinone red deep red. 4-Benxamido-l-acetoxy- anthrapuinone yellow rose 4-Acetumido-I -acetoxyanthraquin brownish-red light brown. 4-Benxamido-l-benxoyIccnthraquinone orange- yellow pale rose. 4-Benzumido-l-methxyanthraquinone orange -yellow scarlet. 4- A cetamido- 1 -methoxyanthraquin orange-yellow light salmon. 1 4-Diacetamidacnthraquinone orange-yellow light brown.1 4-Dibenzumidixznthraqukne reddish -yellow 1 5 - Di benzuvnido - 8 - h ydroxy - anthraquinone red red. 1 5-Dibenzamido-4 8-dihydroxyanthra- quinone violet-blue violet-blue. 1 5-Dianisarnido-4 8-dihydr- oxyanthraquinone violet-blue violet-blue. 4 5-Dibenmrnido- 1 8-dihydroxyanthraquinone violet-blue light violet. 1 5-Di- acetamido-4 8-dihydroxyanthraquinone orange-brown. 1 5-Di- acetumido-4 8-diacetoxyanthraquinone brownish-yellow. 4 5-Di- acetamido-1 8-diacetoxyanthraquinone reddish-brown. 4 5-Di- benmmido-4 8-diacetoxyanthrapuinoneY yellowish- brown. The tinctorial colours of the four last-named are not stated. red& h -yellow. H. J. E. The Investigation of meso-Thioanthracene Derivatives. I. Observations on the Production of Dithioanthraquinone Dithiodianthrone and Other Closely Related Derivatives. ISIDOR MORRIS HEILBRON and JOHN STANLEY HEATON (T.1923 Condensation Products of Phenylhydroxylamine With Hydroxymethylene Compounds and Carbinols. IV Methyl- enecamphorphenylhydroxylamine. H. RUPE and W. DIEHL (Helv. Chim. Acta l922,5,906-922).-The reduction of methylene- camphorphenylhydroxylamine to anilinomethylene-camphor (cf. A. 1921 i 426) can be accomplished by means of sodium hydrogen sulphite in aqueous alcoholic solution. When methylenecamphor- phenylhydroxylamine is treated with thionyl chloride in ethereal solution methylenemmphorphenylchloroamine is formed ; it crystal- lises in short yellowish-white prisms m. p. 102-103". The chlorine appears to be firmly attached to the nitrogen atom; that it has not wandered into the phenyl group was shown by con- densing hydroxymethylenecamphor with p-chloroaniline when p-chloroanilinomethylenecamphor was obtained which crystallises in small white prisms m.p. 167-169". Methylenecamphorphenyl- hydroxylamine combines with 1 mol. of hydrobromic acid to form a hydrobromide C8H14<~~*cHBr*mh*0H yellow needles m. p. 121" (decomp.). Similarly it readily combines with bromine in glacial acetic acid solution to form a dibrmide yellow needles m. p. 117" (decomp.). When the hydroxylamine compound is heated with phenylhydrazine in glacial acetic acid phenylhydroxyl- amine is removed and a derivative formed which was found to be identical with the phenylcamphopyrazole described by Bishop Claisen and Sinclair (A.1895 i 63). C8HI4<(?:">NPh C=N 123 173-185). The formulai. 118 ABSTRACTS OF CREMICAL PAPERS. is to be preferred to that given by those authors in which the phenyl group is attached to the 1-nitrogen atom. By the action of semicarbazide on methylenecamphorphenylhydroxylamine the simple semicarbazone of hydroxymethylenecamphor is fornied. Its m. p. is 205-206" not 217-218" as given by Wallach (A 1904 i 106). The oxidation product of methylenecamphorphenylhydroxyl- amine obtained by boiling it with cupric acetate (A. 1921 i 425) can be obtained in much better yield by oxidising in cold alcohol with potassium ferricyanide It crystallises in monoclinic needles or prisms a b c=0*9971 1 1.065; p=103" 50'. Its properties agree with the double formula [c,H14<c0 ?:CH'y<gh]2. Its form- &ion requires one atom of oxygen to two molecules of the hydroxgl- amine.The double molecule takes up four atoms of bromine to form a tetrabromide C34H4004N2Rr4 which decomposes when heated. When a solution of the tetrabromide in chloroform is warmed hydrogen bromide is evolved and a compound crystallises ; white lustrous aggregates of needles m. p. 205-208" (decomp.). From the chloroform solution of the tetrabromide a dibromide was also obtained C3,H3804N,Br m. p. 123-126" (decomp.). All these bromo-derivatives are very unstable. When the above oxidation product of methylenecamphorphenyl- hydroxylamine is dissolved in 75% sulphuric acid and then pre- cipitated by dilution a new cumpound C34H4406N2 is obtained. It appears to be a hydration product and the formula [c8H14<~~H)*CH:NPh(oH).1 2 cf 1 ,H,oO,NBr is suggested for it. It has no sharp melting point but sinters from 170" decomposing a t 190-195". Molecular-weight determinations in different solvents gave very discordant results. It forms a methyl derivative C,,H,,O,N m. p. 226-228". The above formula expresses the fact that the compound is a stronger acid than the parent substance dissolving readily in barium hydroxide solution. There is a possibility that in the condensation of hydroxymethyl- enecamphor with phenylhydroxylamine intramolecular change occurs to give a p-aminophenol derivative. This has been disproved by condensing p-aminophenol with hydroxymethylenecamphor. The product could not be crystallised but when methylated with methyl sulphate gave p-anisidinomethylenecamphor white leaflets m.p. 169-172". It also gave an acetyl derivative white leaflets m. p. 221-223". E. H. R. Action of Sulphur and certain Compounds of Sulphur on Terpenes. P. P. BUDNIKOV and E. A. SCHILOV (Ber. 1922 55 [B] 3848-3853; cf. A. 1922 i 944).-Pinene or mixtures of limonene and silvestrene are converted by prolonged boiling with sulphur into a viscous reddish-brown liquid from which a homo-ORGANIC CHEMISTRY. i . 119 geneous substance could not be isolated by fractional distillation ; the pwcentage of sulphur in the various fractions increases with increasing boiling point The products yield unstable precipitates when mixed with the chlorides of mercury gold or platinum lead acetate or arsenic iodide in the presence of acetone or alcohol but these are not suitable for analysis.The action of methyl iodide on the fractions gives methyl terpenesulphinium iodide CloH16S,CH31 a microcrystalhe precipitate m. p . about 121" after previous darkening (the constant is given with reserve by reason of the lack of uniformity of the original material). The substance behaves as a typical sulphinium iodide and is converted by moist silver oxide into the corresponding base. When treated with the theoretically necessary quantity of mercuric iodide in accordance with Smiles's procedure (T. 1900 77 163; 1907 91 1394) the iodide gives the compounds C,oH,6S,CH31,HgI pale yellow microscopic prisms and Cl~oH16S,CH31,2HgI lemon-yellow prisms decomp. about 100". With arsenic tri-iohde the com- pound ClOHl6S,CH3I,AsI3 dark orange-coloured crystals is pro- duced.The yields of the methiodide are not satisfactory but it was not found possible to effect any improvement in the initial action by the use of aluminium chloride mercuric chloride or mercuric iodide as catalysts. Better results are obtained when sulphur is replaced by sulphur chloride but the course of the change has not yet been elucidated. The primary action of sulphur on terpenes appears to consist in the formation of a monosulphide which probably combines with a further quantity of sulphur to give polysulphides. Evidence of the formation of thio-ozonides as suggested by Erdmann has not been obtained. H. W. Higher Terpene Compounds. VII. The Constitution of Eudalene Selinene and a-Santalene.The Carbon Frame- work of the Sesquiterpenes. L. RUZICKA and M. STOLL (Helv. Chim. Actu 1922,5 923-936; cf. A. 1922 i 560).-On theoretical grounds it seemed probable that eudalene is formed from sesqui- terpene compounds of the type (I) by dehydro- /\/\ genation and loss of a carbon atom. If this were \-I 1 I the case eudalene should be isomeric with the / \/ \ / 2 -methyl- 8-isopropylnapht haIene obtained from cadalene (A. 1922 i 1001) and should give the (1.1 same naphthalene-1 7-dicarboxylic acid as this when oxidised with dilute nitric acid. This was found to be the case. Naphthalene-1 7-dicarboxylic acid forms a nearly white amorphous precipitate sinters from 200° and melts at 265" to a dark brown liquid. There were also obtained in small quantity a nitro-2-methyl-8-isopropylnaphthulene yellow needles m.p. 112-113" and a rtitrmphthoic acid m. p. 225-227". The identity of eudalene with 1 -methyl-7-isopropylnaphthalene was confirmed by synthesis of the hydrocarbon as follows. Cuminol was condensed with ethyl bromoacetate by Reformatzky's method and from the product ethyl p-isopropylcinnamate was oht,nined. 1i. 120 ABSTRACTS OF CHEMICAL PAPERS. This was reduced by Bouveault's method giving p-isopropylphenyl- propyl alcohol C,H,*C,H,.[CH,],*CH,*oH b. p. 149'112 mm. This was converted through the bromide and cyanide into 7-p-cumyl- butyric acid C,H,°C,H,o[CH,],~~O,H b. p. 186'112 mm. crystal- lising in leaflets m. p. 31-32". The acid chloride is a mobile oil,- b. p. 156"/12 mm. When this chloride was treated with aluminium chloride 8-keto-2-isopropyl-5 6 7 8-tetrahydronaphthalene was obtained b.p. 156"/12 mm. It forms a aemicarbazone m. p. 195". This was boiled with magnesium methyl iodide and 1-methyl-7- isopropyl-3 4-dihydronaphthalene was obtained b. p. 137"/12 mm. This was dehydrogenated by heating with the theoretical proportion of sulphur and gave 1 -methyl-7-isopropyl naphthalene identical with eudalene The constitution of eudalene having been established it is shown that the two forms of the sesquiterpene selinene which readily gives eudalene when treated with sulphur probably have the following constitutions H Me H H Me B-Selinene. a-Selinene. It has now been shown that there are two types of carbon skeleton in the sesquiterpene series derived from three isoprene units.These may be called the cadinene type (11) and the eudesmol type (I). The structure of a-santalene was /\/\ determined by Semmler (A. 1910 i 574) who how- I l- ever considered two formula possible belonging to h/ what he called the camphor-type and the camphene- type. It is shown that these two "types" are but /\ (II.) different plane projections of the same three- dimensional structure. The structure is definitely related to the eudesmol type (I). It can now be concluded that all known sesquiterpenes are closely related and are derivable from a regular tri-isoprene chain of the kind present in farnesol. I I E. H. R. The Nature of Shellac. Shellolic Acid. C. HARRIES and W. NAGEL (Ber. 1922 55 [B] 3833-3848).-A preliminary account of partly completed work on shellac and sticklac.The lac is freed from wax and colouring matter by successive treatment with light petroleum and water and the residue is repeatedly agitated with ether thereby leaving an insoluble '' pure resin " which is the subject of investigation and is the vehicle of the characteristic properties of shellac. The " pure resin " is attacked by N-potassium hydroxide solution at the atmospheric temperature which causes the deposition of potassium aleuritate the amount being 22-24% of the resin taken. The filtrate from the salt is acidified with sulphuric acid and extracted with ether thus yielding a mixture of shellac resin acids; these are purifiedi. 121 ORGANIC CHEMISTRY. by agitating their ethereal solution with aqueous barium hydroxide and treating the latter with carbon dioxide when the precipitated barium carbonate adsorbs a portion of the coloured resinous matter.The filtrate contains small amounts of sparingly soluble barium aleuritate and the freely soluble barium salts of the shellac resin acids. Attempts to isolate a homogeneous material from the latter by dialysis were not completely successful but it is shown thereby that the salts are devoid of colour. More rapid but less complete purification can be effected through the zinc salts. The free acids do not crystallise readily. The most satisfactory results are obtained by taking advantage of the observation that the shellac resin acids unlike other resin acids are at any rate in part esterifiable by methyl-alcoholic hydrogen chloride (3%) a t the atmospheric temperature where by methyl shellohte C,,H,O 6 long prismatic rods m.p. 149" b. p. 284-288"/0-1 mm. (slight decomp.) [a]$ +32-61" in methyl-alcoholic solution is obtained the amount being S-lO% of the weight of the "pure resin" taken. The ester is hydrolysed by boiling aqueous N-sodium hydroxide solution to shellolic acid (shellenedioldimrbox ylic acid) colourless leaflets m. p. 199.5-201" decomp. 202-203" which gives the Liebermann cholesterol and the Salkowski-Hesse reactions. It does not reduce Fehling's solution or decolorise a solution of bromine in chloroform ; its unsaturated nature is demonstrated by its instability towards alkaline permanganate and the apparent formation of an ozonide. The sodium and barium salts are colourless amorphous and freely soluble in water; the silver copper and lead salts dissolve more sparingly.The hydrazide Cl5H,O4N crystallises in well-defined prisms m. p. 243-244" (decomp.) the non-crystalline acetyl derivative and the diphenyldiurethane C3,Ha08N2 m. p. 92-94" (decomp.) according to the rate of heating are described. The CH.C02H annexed formula for shellolic acid is suggested tentatively ; the positions of the hydroxyl and carboxyl groups can- HO*HC!,~,CH*OH not yet be definitely assigned. The position of the double bond accounts for the inactivity of the acid towards The non-crystalline residue of esters from which methyl shellolate has been separated (21.5.) has b. p. 180-210"/0*1 mm. and appears to consist of compounds of hydroxy-acids.The latter are themselves amorphous but give solid phenylurethanes. The investigation has rendered it very improbable that shellac resin is the aleuritic ester of a higher alcohol since no trace of the latter could be detected and i t must be present in considerable quantity if it were an actual constituent Since it is established that the resin does not contain a free acid it appears probable that the shellac molecule is composed of hydroxy-acids which are united in the form of lactides. For one of the simpler components the constitution Xo>C,,H,,( OH)CO*O*CIGH,,(OH)<?o 0 is sug- C02H.CH H2C'\8/)CH HC CH 1 1 bromine. H2C CH2 \/ CH2i. 122 ABSTRACTS OF CHEMICAL PAPERS. gested the composition of which is very similar to that of the pure resin." H.W. Centaureidin a Decomposition Product of Centaurein the Glucoside of Roots of Centaureu Jucea. M. BRIDEL and G. CHARAUX (Compt. rend. 1922,175 1168-1170).-Centaureidin C,,H,,O (this vol. i 50) crystallises from 50% alcohol in micro- scopic yellow needles containing water of crystallisation which is lost a t 50". The crystals melt a t 197" the anhydrous substance at 203". It is insoluble in water but dissolves in many organic solvents. With aqueous alkalis or sulphuric acid it gives a golden- yellow solution. As the result of a general study of its properties the authors suggest that it may be a flavone derivative although the evidence on behalf of such a conclusion is mainly negative. < ( H. J. E. Polysaccharides. XVII. Chitin. P. KARRER and ALEX. P. SMIRNOV (Helv.Chim. Acta 1922 5 832-852).-To determine how the glucosamine residues are combined in chitin a study was made of the decomposition products obtained when chitin was distilled with zinc dust. From 300 g. of chitin from lobster shells 37 g. of a brown oil were obtained which consisted chiefly of pyrrole compounds with a small quantity of pyridine bases among which a-picoline was identified. From the mixture of pyrrole compounds a fraction was isolated which was identified as 2-methyl-1-n-hexyl- pyrrole. This compound and 2 5-dimethyl-1-n-amylpyrrole were synthesised for comparison with the compound from chitin to which the name chitopyrrole is applied. 2 5-Dimethyl- 1-n-arnylpyrrole was prepared by heating acetonyl- acetone with n-amylamine. It is a colourless oil b.p. 225-227" with an orange-like odour and gives a cherry-red pine-shaving reaction. 2-Methyl- 1 -n-hexylpyrrole was prepared from potassium 2-methylpyrrole and n-hexyl iodide. It is a colourless oil rapidly .turning brown b. p. 200-210" smelling like old fungus and gives an intense red pine-shaving reaction. Chitopyrrole boils over a somewhat wider range than 2-methyl-1 -n-hexylpyrrole 190-220" but essentially the two appear to be identical. Both are oxidised by chromic or nitrous acid to a substituted maleinimide which when hydrolysed gives maleic acid and n-hexylamine. The last was identified by preparation of the picrolonate m. p. 188-189" which was also prepared from synthetic n-hexylamine. The formation of 2-methyl-1-n-hexylpyrrole from chitin indicates the presence in the latter of two glucosamine residues combined through a nitrogen atom as in the following formula Neither the position of the acetyl groups nor the number of glncos- amine residues present in the molecule can yet be stated.Chitosan which is formed from chitin by hydrolytic removal of the acetyl groups is converfed by nitrous acid into a reducing sugar witlhORUAHIC CHEMISTRY. i. 123 total loss of its nitrogen in the elementary form. This behaviour is quite in accordance with the above representation of chitin as an aldehyde-ammonia derivative. E. H. R. Betulin. OTTO DISCHENDORFER (Ber. 1922 55 [B] 3692- 3693) .-A preliminary account of the author's observations induced by the recent publication of Schulze and Pieroh (A. 1922 i 1045).Monobromobetulin m. p. 215" and the corresponding diacetute m. p. 193" have been prepared. Analyses 6f the latter indicate the possibility of the formulae C3,H,,O2 or C,,H,O for betulin whereas Schulze and Pieroh (Zoc. czt.) regard C32H5202 or C,H,O as probable. H. W. Strophanthin. I. Strophanthidin. WALTER A. JACOBS and M~CHAEL HEIDELBERGER ( J . Biol. Chem. 1922 54 253-261).- When dried in a vacuum a t 110" over phosphoric oxide stroph- anthidin decreases in weight by an amount corresponding with the loss of +H,O. The anhydrous substance therefore has the formula C23cH32O6. The loss of l&H,O noted by Windaus and Hermanns (A. 1915 i 704 705) was evidently due to partial decomposition. A final decision between the C formula and the C27 formula of Feist is rendered possible by the preparation of the p-bromo- benzoate C30H3507Br,H20. In the anhydrous condition this compound has m.p. 222-224' (decomp.) [a]? +42" in acetone. isostrophanthidin (strophanthidinic acid lactone of Feist iso- cyxnarigenin of Windaus and Hermans) has the same formula as strophanthidin crystallises with +H20 and forms a benzmte C30H3607 rosettes of microscopic leaflets m. p. about 270" after sintering [a]E + 38.0" in chloroform. When reduced with hydrogen in the presence of colloidal palladium strophanthidin slowly absorbs two atoms of hydrogen with the formation of dihydrostrophanthidin C23H3P06 which melts a t 190-195" when anhydrous and crystallises with one or two molecules of water according to the method of crystallisation. The dihydrate has [a]2+34*85' in methyl alcohol.Dihydrostrophanthidin forms a benzoate CNH3,O7 minute glistening prisms m. p. 225-227" (decomp.). The presence of a carbonyl group in strophanthidin is shown by the pre- paration of the oxime C,,H,O,N glistening prisms m. p. 270-275' (decomp.) [a]2,7+71.3" in pyridine the phnyZhydraxone glistening prisms m. p. 230-232" after sintering a t 175" [ar]5-5-0" in chloroform and the p-bromophenylhydraxone stout pointed prisms which soften a t 180-185" and become com- pletely molten at ZOO" and have [a15 6+ 105%" in chloroform. C2,H3805N2,2H20 9 C,,H3 70 5N,Br 1 +MeOH E. S. [Catechin.] M. NIERENSTEIN (Ber. 1922 55 [B] 3831- 3833).-A reply to Freudenberg (A. 1922 i 756). The author maintains that catechincarboxylic acid can be pre- pared according to his method and promises further details with regard to Kostanecki's catechone.With respect to the production f* 2i. 124 ABSTRACTS OF CHEMICAL PAPERS. of optically active catechin from inactive catechincarboxylic acid the inactivity of the latter is maintained but the optical activity of Gambier catechin is not considered to be established. Pure Gambier- and aca-catechins are not precipitated by solutions of elatin. Freudenberg's observation that tetramethylcatechin cannot t e demethylated by the author's method is correct so far as the derivative of Gambier catechin is concerned but is not true for that of aca-catechin. The identity of the methylated product of the reduction of catechin with pent amet hoxy - a'y -dip henylpro pane is not regarded as established.The homogeneity and optical inactivity of aca-catechin m. p. 204-205° is maintained. H. W. Tannins and Similar Compounds XII. The Tannin of the Native [German] Oak. KARL FREUDENBERG and ERICH VOLLBRECHT (Annalen 1922 429 284-317).-A more expanded account of work already published (cf. A. 1922 i 1046). C. K. I. Constitution of Thiophen. WILHELM STEINKOPF [with HAL- VARD AUQESTAD-JENSEN and HANS DONAT] (Annalen 1922 430 78-112).-The lability of the hydrogen atoms in the thiophen nucleus is well illustrated by its behaviour towards cyanogen bromide with which it reacts analogously to compounds of the type of ethyl acetoacetate and ethyl malonate which are known to contain labile hydrogen. Thus ethyl acetoacetate and cyanogen bromide react giving ethyl y-bromoacetoacetate. Ethyl malonate yields ethyl bromo- malonate.Acetophenone yields O- bromoacetophenone and in a similar way 2 - acet o t hienone yields O- bromo - 2 - acet o t hienone. Phenol and cyanogen bromide yield p-bromophenol whilst indene and cyanogen bromide give 1 - bromo-2-hydroxyhydrindene. Thiophen reacts with cyanogen bromide giving bromothiophen b. p. 151-151.5" and dibromothiophen b. p. 195-206" and with cyanogen iodide giving 2-iodothiophen (identified as the 5-iodo-2-mercurichloride). Bromothiophen is converted by cyano- gen bromide into dibromothiophen and 2-thiotolen into bromo- thiotolen b. p. 173-177". Alkylthiophens in which both a-positions are substituted are also brominated by cyanogen bromide. 2 5-Dipropylthiophen yields 3-bromo-2 5-dipropyZthiophen b.p. 130-132~5"/10 mm. and 2-ethyl-5-isoamylthiophen yields 3-(or 4-)bromo-2-ethyl-5-isoamylthio- phen b. p. 122-127"/14 mm. The above 2 5-dialkylthiophens mere synthesised by way of the corresponding ketones. 5-Ethyl-%popiothienone b. p. 137- 138"/19 mm. prepared from 2-ethylthiophen propionyl chloride and phosphoric oxide gives a semicarbazone m. p. 195-196" and on reduction with zinc and hydrochloric acid gives 2-ethyl- 5-propylthiophenY b. p. 196-197". 5-Propyl-2-propiothienone b. p. 137-138-5"/13 mm. prepared from 2-propylthiophen propionyl chloride and either phosphorus pentoxide or aluminium chloride yields a semicurbaxone prisms m. p. 174-175" and on reductionORGANIC CHEMISTRY. i. 135 gives 2 6-dipropylthiophen b. p.213-214". 6-isoAmy1-2-aceto- thienone b. p. 149-151"/13 mm. prepared from 2-isoamylthio- phen acetyl chloride and phosphorus pentoxide gives a semi- carbaxone leaflets m. p. 212" and on reduction yields 2-ethyZ- 5-isoamylthiophen b. p. 103~5-106~5"/12 mm. None of these 2 5-dialkylthiophens gives well characterised mercury compounds on treatment with mercuric chloride. The Thiophen Series. XV. Cyclic Mercury Compounds and Experiments on the Formation of Mixed Thiophen- Mercury Compounds. WILHELM STEINKOPF WILHELM BIELEN- BERG and HALVARD AUGESTAD- JENSEN (Annalen 1922 430 40-78).-The experiments carried out with the object of pre- paring the mixed mercury compound C4H,S*HgPh were based on the reaction between magnesium phenyl bromide and phenyl mercurichloride the products of which are mercury diphenyl and magnesium chloride and bromide.Pure mercury phenyl thienyl could not be isolated from the product of the action of magnesium phenyl bromide in 2-thienylmercurichloride although evidence of its formation was obtained. Mercury diphenyl is also formed by the action of mercurous chloride on magnesium phenyl bromide and by the action of stannous chloride on magnesium phenyl chloride. Similarly mercury 2 2'-dithienyl is the product of the reduction of thienyl-2-mercurichloride by stannous chloride. A series of cyclic mercury-thiophen compounds is described of which dimercury 2 2' 5 5'-dithienyEene I >S S< is typical. This substance is obtained from 2 5-thienylenedimercuri- chloride and either sodium iodide or sodium thiocyanate in the presence of pyridine. Pyridine and mercuric chloride convert it into mercury 5 5'-dithienylene-2 2'-dimercurichloride which may also be obtained by the action of pyridine on thienylene-2 5- dimercurichloride.Dimercury 3 3'-(or 4'-)diethyl-2 2' 5 5'-di- thienylene mercury diethyl-5 5'-dithienylene-2 2'-dimercurichloride and dimercury 3 4 3' 4'-tetramethyl-2 2' 5 5I-dithienylene are also described; like the unakylated parent substances they are all exceedingly insoluble compounds which do not melt at 320". Dimercury diethyldithienylene on treatment with mercuric chloride under regulated conditions yields mercury diethyldithienylene- dimercurichloride but excess of the reagent effects complete dis- ruption of the molecule the product being 3-ethylthienylene- 2 5-dimercurichloride. A colorimetric method of estimating the rate of separation of mercuric sulphide when an organic mercury compound is treated with sodium sulphide is described in the original.Synthesis of Substituted Thianthrens. I. Thianthren and Nitrothianthren. SRI KRISHNA (T. 1923 123 156-160). Production and Reactions of 2-Dithiobenzoyl. nhRy MCKIBBEN and ERNEST WILSON MCCLELLAXD (T. 1923 123 C. K. I. CH:C-Hg-C:CH CHX-Hg-C H C. K. I. 170-173).i. 126 ABSTRACTS OF CHEMTCAL PAPERS. The Alkaloids of the Northern Aconite (Acon2tum septen- trtonak Koelle). GUNNAR WEIDEMANN (Arch. ex@. Path. Phurm. 1922 95 166-180).-Two alkaloids isolated from the northern aconite have been investigated. forms hard glass clear six-sided prisms m.p. 223" [a] +27.0' in chloroform. It is a monoacid base and contains three methoxyl p u p s . It forms a chloroplatinate C,H4,0,N,,HPtC1 and a chlorouurate C,,H,20,N,,HAuC14. On hydrolysis with alcoholic potash it yields lappaconitic acid C,H,O,N shown to be acetyl anthranilic acid and a base lappaconine C,H,50,N,2H20 m. p. 93O [a]2"+2%41°. It forms a hydrochloride C,,H,507N,HCl large colourless crystals. Septentrionaline C,,H,O,N,( OMe)4 the other alkaloid investigated is an amorphous white powder m. p. 131" [a32 '+32.71" forming a chloroplatimte C,H4,0gN,,HPtC1,. On hydrolysis with alcoholic potash there is obtained (1) an acid C,H,O,N m. p. 125-126" which when boiled with sodium hydr- oxide solution yields anthranilic acid losing CH,O and (2) a base C2,H,O7N m.p. SO" [a] 6+29.55" forming a hydrochloride CZ5H3,O ,N,HCl. W. 0. K. Lappaconitine C32H420$N2 Paniculatine the Alkaloid from Acodtum punicuZutzcm Larn. G . E. BRUNNER (Schweiz. Apoth. Ztg. 1922 60 357- 358 ; from Chem. Zentr. 1922 iii 1007).-Paniculatine C,,H,,O,N the alkaloid from Aconitum panniculatum Lam. is not identical with aconitine. It forms small rhombic prisms with m. p. 263". G. W. R. Melanins Arising from Adrenaline. PIETRO SACCARDI (Biochem. Z. 1922 132 439-442).-Melanin-like pigments ob- tained by the oxidation of adrenaline with chlorine water are described. W. 0. K. Pre aration of a Quinine Derivative. AKTIEN-GESELLSCHAFT BACHSTEZ (D.R.-P. 357753; from Chem. Zentr. 1922 iv 951).- Equimolecular amounts of quinine and 4-ethoxyphenylmalonamic acid are fused together and crystallised from hydrolysing solvents or allowed to react as such or in form of their salts in the presence of hydrolysing solvents.Quinine 4-ethoxyphenylmalonamate forms long colourless needles m. p. 72-73'. FGR L ILIN-FABRIKATION JULIUS ALTSCHUL and M~RCELL G. W. R. Quiteninone. SIGMUND FRANKEL CHARLOTTE TRITT-ZIRMING and Lrzly GOTTESMANN-GRAUER (Ber. 1922,55 [B] 3931-3935).- The action of hydrogen peroxide (30%) on a solution of quinine sulphate in dilute sulphuric acid in the presence of copper or ferrous sulphate as catalyst and a t the atmospheric temperature leads to the production of quiteninone C,gH2,,04N2~ needles m. p. 156". The reaction appears to be considerably mfluenced by external factors and to take place through a number of intermediate pro- ducts of which quinine oxide (cf.Speyer and Becker A. 1922,ORGANIC CHEMISTRY. i. 127 i 674) is one. Quiteninone is also produced by the oxidation of quitenine (cf. Nierenstein A. 1920 i 875). It gives a p'crate m. p. 140". The constitution of quiteninone is partly elucidated by the preparation of the methyl ester picrate C2,,H220,N2,2C6H30,N decomp. 270° and the methyl ester dihydrocMde m. p. 181". Quiteninone could not be caused to react with phenylhydrazine p-nitrophenylhydrazine or semicarbazide hydrochloride ; with hydroxylamine in alkaline solution it gives an oxime which is identified as the corresponding $cr& C,,H210,N3,C3H30,N m. p. 126". H. W. Esterifmation of Creatine. ARTHUR W. Dox and LESTER YODER ( J . Biol. Chem. 1922 54 671-673).-Saturation of a suspension of creatine in an alcohol with hydrogen chloride results in the formation not of creatinine but of an ester of creatine the hydrochloride of which separates on the addition of ether.By this means the author has prepared creutine methyl ester hydrochloride slender needles m. p. 139-140" ; crmtine ethyl ester hydrochloride needles m. p. 163" ; creatine n-butyt ester hydrochloride flat needles m. p. 138". Each salt melts with the evolution of gas and leaves a solid residue of creatinine hydrochloride. Ergot. A. STOLL (Schweix. Apoth. Ztg. 1922 60 341-346; from Chem. Zentr. 1922 iii 1007; cf. Spiro and Stoll A. 1922 i 47).-The specific effect of ergot is not considered to be due to the presence of simple amines such as tyramine. After addition of acid reagents such as aluminium sulphate to ergot extraction of one kg.of the material with ether and benzene removes 350- 400 g. of alkaloid-free ergot oil together with soluble acid and neutral substances such as organic acids phytosterol and colouring matters. From the acidified cell material a crystalline alkaloid was separated. Ergotarnine CsH3503N5,2COMe2,2H,0 (from acetone) forms highly refractive rhombic pmms ; it has [a]$ -155" in 0.6% chloroform solution ; the monoacid base and its compounds de- compose on heating a t 140"; a t 180" a brown mass is formed with evolution of gas. On keeping an ethyl-alcoholic solution or warm- ing a methyl-alcoholic solution of ergotamine an isomeride ergot- aminine of weaker basic character is formed.It crystallises in triangular leaflets and has [a] +381" in 0.6% chloroform solution. Ergotaminine may be reconverted into ergotamine. Both isomer- ides give a blue coloration with strong sulphuric acid. Ergotarnine is unstable in air. G. W. R. E. S. The Hydroxycodeinone Series. EDMUND SPEYER [with 8. SEUG and MARTIN HE=] (Annulen 1922,430,1-40).-A further account (see A. 1915 i 580; 1916 i 157 758; 1921 i 685) of derivatives of codeine and thebaine including the reduction of hydroxycodeinone to hydroxythebainol and the conversion of the latter into a nitrogen-free substance by exhaustive methylation All results are interpreted on the basis of Knorr's farmula for morphine.i. 128 ABSTRACTS OF CHEMICAL PAPERS. Hydroxycodeinone is converted either by electrolytic reduction or by zinc and formic acid into hydroxythebainol m.p. 234" CH,- NMe f CH NRft the formate of hydroxythebainol m. p. 227" and 7-hydroxy- codeine being obtained as by-products in the latter case. Hydroxy- thebainol yields a crystalline hydrochloride hydrobromide m. p. 252-253" [ ~ ] g -157*7" hydriodide m. p. 247" and picrate m. p. 204-5206". On treatment with benzoyl chloride it gives a benxoyl derivative needles m. p. 257"; with bromine and acetic acid a perbromide which on reduction by sulphur dioxide gives a mono- homo-derivative m. p. 230-231" reducible to hydroxythebainol and with hydrogen peroxide an N-oxide which crystallises in prisms m. p. 237". On methylation with methyl sulphate and alkali hydroxythebainol yields the methyl ether methiodide rhombs decomp. 233" which with hot alkali hydroxide gives deR-N-methyl- hydroxythebainol methyl ether OMeH\ OMe/\ OMell I OMel \//\CH \/\ 1~ ~ E k m e or IIcH CH,:CH-CH/\/ NMe,*CH,*CH,CH/\d HO-CH~ llCCH H O ~ H ! ,,!ICE CH-OH \/ CH-OH small needles m.p. 195-197". The hydriodide forms needles m. p. 255O and the methiodide obtained with the aid of methyl iodide microscopic prisms decomp. 239-240". On treating this substance successively with silver oxide and concentrated alkalis trimethyl- amine is eliminated and the nitrogen-free substance 2 3-dihydroxy- 5 6-dimethoxy-4-vinyl-2 3 4 4a-tetrahydro~henanthrene~ is ob- tained. This forms stout prisms m. p. 188-189" and is insoluble in alkali which shows that the two hydroxyl groups are in the aliphatic part of the molecule.An improved method of preparation of hydroxythebainone is described. Its dibromide obtained by the use of bromine and chloroform decomposes at 258" and on reduction by hydrogen and palladium black gives hydroxydihydrothebainone which may be obtained directly from hydroxythebainone by means of the same reducing agent. The methiodide of hydroxythebainone forms prisms m. p. 245" the acetyl derivative needles m. p. 197" andORGANIC CHERIISTRY. i. 139 the oxime of the acetyl derivative prisms m. p. 216-218". The methiodide of the acetyl derivative decomposes a t 212-213" and gives a non-crystalline deoxy-base on decomposition by alkalis. The methiodide of the acetyl derivative of 7-hydroxycodeine sinters a t 230" and decomposes a t 256".Reduction of hydroxycodeinone oxime by hydrogen and palladium under various conditions yielded hydroxydihydrocodeinone and it was not found possible to prevent the elimination of the oximino- group. Parallel experiments with styrylmethylketoxime and di- styrylketoxime yielded analogous results the products being benzylacetone and dibenzylacetone respectively. The Pilocarpine Series. 11. Pilocarpic Esters and their Derivatives. MAX POLONOVSKI AND MICHEL POLONOVSKI (Bull. SOC. chim. 1922 [iv] 31 1185-1201; cf. this vol. i 52).- Although it is generally accepted that a y-lactone group is present in pilocarpine and its isomeride few of the properties of a y-lactone have been actually shown to be characteristic of these substances. The authors find that the esterification of pilocarpine by means of methyl or ethyl alcohol and hydrogen chloride does not yield the ethyl ester of pilocarpic acid but the ethyl y-chloro-ester thus confirming the y-lactone grouping by the characteristic reaction of simultaneous esterification of acidic and alcoholic groups by alcohol and halogen respectively.Methyl y-chloropilocarpte forms small transparent prisms m. p. 4 2 4 4 " [.ID +32*6" and is a strong base; the nitrate forms lustrous plates m. p. 157" [alD 3-20". Ethyl 7-chloropilocaryate was obtained as an oil [a]D +29.2"; the nitrate forms lustrous plates m. p. 136" [.ID + 23.4". On treat- ment of the nitrate with concentrated sulphuric acid it is converted into ethyl 7-chloronitropilocarpate. The y-chloro-esters of pilo- carpine are readily relactonised with loss of the halogen atom and the alkyl group; this is accompanied by the transformation of cz portion of the alkaloid into its isomeride.isoPilocarpine behaves in a similar manner to pilocarpine on esterification. Methyl 7-chloroisopilocarpate is an oil of very alkaline reaction [.ID -7.5" (nitrate hygroscopic crystals m. p. about loo" [aJD -5.6") ; the ethyl ester is also an oil aD -5" (nitrate crystalline m. p. 95" [.ID 0). All these chloro-esters are unstable even in the solid state a t the ordinary temperature a portion of the substance being transformed into the hydrochloride of a quaternary base. The chloro- esters react readily with sodium ethoxide or methoxide yielding an oily mixture of the ethyl esters of a- and p-anhydropilocarpic acids separable by the Werence in solubility of their nitrates in water.a-Anhydropilocurpic acid C,,H,,O,N forms lustrous plates m. p. 243" [a]= -19"; the hydrochlorzde has m. p. 18'7"; the ethyl ester is an oil and yields a nitrate colourless needles m. p. 165" [.ID -19" in alcohol and +3-4" in water. @-Anhydropilocarpic acid forms prismatic crystals m. p. 186" [.ID +42" ; the hydrochloride has m. p. 142" ; the ethyl ester forms large transparent tablets m. p. 48" and gives a nitrate m. p. 95" [.ID -28" in water. C. K. I.i. 130 ABSTRACTS OF CHEMICAL PAPERS. On bromination ethyl a-anhydropilocarpate gives a bromo-deriv- ative C13H1902N2Br from which on hydrolysis the bromo-acid m. p. 138-139". is obtained. are suggested as possible for the isomeric acids.The Pilocarpine Series 111. isoPilocarpinani1 or Phenyl- isopilopyrrolidone. MAX POLONOVSKI and MICHEL POLONOVSHI (Bull. SOC. chim. 1922 [iv] 31 1201-1204; cf. preceding abstract).-Pilocarpine combines with aromatic amines yielding very stable pyrrolidone compounds. This reaction points to the existence of a lactone group in pilocarpine (cf. Emmert and Meyer A. 1921 i 268). The same substances may be obtained from the y-halogen acids derived from the alkaloid. Prolonged heating being necessary the products are derivatives of isopilocarpine. isoPiZocurpinuniZ is a hard substance slightly alkaline in reaction which gives with mineral acids crystalline salts of acid reaction. The nitrate colourless plates m. p. 162" [a]D+21*20 and the hydro- chloride white hygroscopic prisms m.p. 135" were prepared. The base yields on nitration a substance containing a nitro-group in the benzene ring which was not further investigated. The Pilocarpine Series. IV. Metapilocarpine. MAX POLONOVSK~ and MICHEL POLONOVSI~ (Bull. Soc. chim. 1922 [iv] 31 12061208 ; cf. preceding abstracts).-As substances obtained by the authors from the decomposition of 7-chloro-esters of pilo- carpine seemed to be identical with the metapilocarpine described by Pinner (A. 1905 i 658) it appeared that the latter could not be a simple stereoisomeride. An examination of its properties resulted in confirmation of Pinner's observations. The substance is inactive towards polarised light and although neutral to litmus and giving with acids salts of acid reaction does not combine with alkalis.Its reactions lead to the conclusion that in metapilocarpine the lactone grouping is absent and the acid group which replaces it is neutralised either by the basic glyoxaline ring or by a betaine linking CO-0-N-. The authors prefer the second explanation and therefore represent the transformation of pilocarpine into metapilocarpine by Et-$!H-YH-CH2-F :CH>NMe + Et-yH-yH-CH2-y :CH>NMe H. J. E. H. J. E. \I/ COO CH2-N:CH + - COO CH N:CH + - H. J. E. The Pilocarpine Series. V. Isomerism of Pilocarpine and isoPilocarpine. lMAx POLONOVSKI and MICHEL POLONOVSKI (Bull. Xoc. chim. 1922 [iv] 31 131&1330).-The authors discuss the evidence for and against the various hypotheses advancedORGANIC CHEMISTRY. i. 131 to account for the isomerism of pilocarpine and isopilocarpine and reject as untenable theories based on position isomerism in favour of stereoisomerism arising in or near the lactonic group. Their conclusions are largely based on the isomerisation of pilocarpine and its derivatives to isopilocarpine and its corresponding deriv- atives by action of small quantities of sodium ethoxide on alcoholic solutions of the alkaloid in the cold.It was found that pilo- carpine and nitropilocarpine were completely isomerised under these conditions by traces of sodium ethoxide that molecular quantities of sodium ethoxide gave an isomerised and delactonised product the de-lactonisation proceeding much more slowly than the isomerisation and requiring twenty-four to forty-eight hours for completion and that sodium pilocarpate sodium nitropilo- carpate and ethyl chloropilocarpate were not isomerised to the iso-derivatives ; in other words esterification or salt formation causes a stabilisation of the pilocarpine molecule whence it is con- cluded that for isomerisation by sodium ethoxide the presence of the unchanged lactone grouping is essential and this group is probably the seat of the isomerism.Whilst the nature pf the stereoisomerism must for the present be left an open question the authors incline to the view that it may be a case of partial racemisation of one of the two asymmetric C atoms of the lactonic group in view of the similarity between the isomerism of pilocarpine and that of hyoscyamine. G. F. M. Strychnos Alkaloids. XXXV. Ethers of Hydroxydihydro- brucinolone and the Violet Colour Reaction of the Nitro- pinones obtained therefrom.HERMANN LEUCHS JOHANNES GR~~ss and HARRY HEERING (Bey. 1922 54 [B] 3729-3738; cf. A. 1921 i 883).-The mtion of boiling methyl-alcoholic potassium hydroxide solution on brucinolone or its acetyl derivative leads to the formation of small amounts of methoxy- dihydrobrucinolone c22H2,@6~2,3H2qy coarse prisms or plates m. p. 82" [a]" -50.3" m glacial acetic acid solution and (mainly) a product m. p. 200-202" after softening at 190" which is con- verted by acetic anhydride into acetylmethoxydihydrobrucinolone C,-H,,0,N2 m. p. 258-270° [a] -92*5" in glacial acetic acid solution and acetylcryptobrucinolone. Attempts to separate the mixture m. p. 200-202" into its components by methyl-alcoholic ammonia at 100" were unsuccessful.Under these conditions cryptobrucinolone is converted into a base C2,H2,05N3 (isolated as the hydrochloride) identical with that isolated previously but in much poorer yield from crude acetylbrucinolone and ammonia (Leuchs A. 1914 i 317); its formation in the latter instance appears to depend on the presence of cryptobrucinolone or its ester in the acetylbrucinolone. It unites with phenylcarbimide to give the compound C,,H,O,N small prisms m. p. 200". Ethoxydihydrobrucinolone is converted by 6N-nif1-k acid at 0" into the corresponding quinone a yellowish-red resin which yields a semicurbame slender orange-coloured needles m. p. 240" after change at 211". The free quinone is reduced by sulphur dioxidei. 132 ABSTRACTS OF CHEMICAL P,QPERS.to a colourless amorphous product ; the corresponding acetate could not be caused to crystallise. The action of 5N-nitric acid on ethoxydihydrobrucinolone a t 50-60" leads to the formation of the nitroquinone hydrate C21H,0a3 golden-yellow leaflets which is further transformed into the semicurbaxone C22H260~ slender pale yellow needles and the crude monoethyl ester. The nitroquinone hydrate is reduced by sulphurous acid to the nitro- quinol hydrute C2,H2,0,N3 an amorphous dark violet powder m. p. (variable) about 185" (decomp.). The latter is converted by hydrogen chloride in methyl and ethyl alcohols into the methyl and ethyl esters amorphous violet substances. The triacetyl derivative of the nitroquinol has m. p. 175-180" (decomp.).The relationships of the quinonc and quinol compounds are illustrated by the scheme 0 0 .. .. /\-R NO,J'\-?*CO,H + y - N \/-NH No2011 II 1 >co -f II 11 0 G 0 OH OH NO/\-E~CO,H -+ HO*N:/\-~CO,H I I \/-N I I \/-'YH OH OH H. W. Strychnos Alkaloids. XXXVI. The Preparation of Meth- oxy- and Ethoxy-dihydrostrychninolones and of Strychninol- one-c; Oxidation of the Latter. HERMANN LEUCHS and RUDOLF NITSCHKE (Ber. 1922 55 [BJ 3738-3745; cf. A. 1921 i 883 ; Leuchs Griiss and Heering preceding abstract).-Strych- ninolone-a is converted by methyl-alcoholic potassium hydroxide solution initially into the b-form and finally into an inseparable mixture of methoxydihydrostrychninohe and strychninolone-c. The former has been isolated only as a resin which is converted by sodium acetate and acetic anhydride into a crystalline acetate C,,H,P5N2 colourless quadratic crystals m.p. 237-239" after softemng a t 230" [a] -109" when dissolved in glacial acetic acid. The course of the change is followed more readily when the methyl- is replaced by ethyl-alcoholic potassium hydroxide solu- tion since in this instance ethoxydihydrostrychninolone needles m. p. about lOO" [a] -51*9" in glacial acetic acid solution m. p. (+MeOH) 65-70' can be isolated directly. Strychninolone-c is transformed by acetic anhydride and sodium acetate into acetylstrychninolone-c C21H2004N2 domatic prisms m. p. 256-257" [a]. -229.6" in glacial acetic acid solution. The latter substance is oxidised in acetone solution by powdered potassium permanganate to an acid C2,H,0,N2,H20 colourless prisms m.p. 280-282' (decomp.) after softening a t 260" in which however the water of crystallisation appeam to be retained withORGANIC CHEMISTRY. i. 133 unusual tenacity. The acid is converted by hydrochloric-wid into acetic and oxalic acids and a non-crystalline unstable kych- chloride. The behaviour of acetylstrychninone-c is closely- ana- logous to that of cryptobrucinoIone; each probably contains the oxidisable group CHGRzr which is transformed by oxygen into CO,H*R:N*CO*CO,H. H. W. Strychnos Alkaloids. XXXVII. The Degradation of Methylcacotheline and its Violet Colour-reaction with Sodium Sulphite. HERMANN LEUCHS BERNHARD WINKLER and W. ROBERT LEUCHS (Ber. 1922 55 [B] 3936-3950).- Among the violet products which are formed from cacotheline and analogous substances of the brucine series methylcacotheline methosulphite (A.1919 i 35) occupies a peculiar position since it becomes isomerised when heated with the production of a violet isomeride which is therefore not formed in the usual maimer by the addition of two atoms of hydrogen. The isomerism cannot be regarded as definitely proved by analytical methods since the presence of two additional atoms of hydrogen in the molecule does not greatly affect the composition but it is now placed beyond doubt by the observations that the substance is produced in 50% yield by the action of one molecular proportion of sodium sulphite on two molecular proportions of methylcacotheline that sulphuric acid is not produced thereby and that the remainder of the methyl- cacotheline passes into the yellow methylbetaine (cf.A. 1920 i 179) which is convertible by further treatment with sodium hydrogen sulphite into the violet methosulphite. The reactions which take place in the production of the violet methosulphite are indicated by the scheme CH-CO’ .. 0 H 9 Hb’ P O$Il /\*$*CO,H I1 + ?yIe 02NI/J8.C02H + i NOS 0,H iN*O*SO,- \/*m 0 (1.) (11.1 OH 0 OH Me 02N-/f)i~-C0,H + Me HO*N/\*?*CO,H 1 iN-O-SO,*\/.N I 1. IN* o*s o,*\/ *NH OH OH (111.) (IV.1 The ammonium sulphite becomes added to the quinone nucleus with the formation of a phenylsulphite ester group or a sulphonic acid whereupon the quinone becomes isomerised to quinol without addition of extraneous hydrogen ; the nitro-group subsequently passes into the isonitro-form with production of a new quinonoid arrangement and development of the dark colour. The so-called violet ‘‘ methosulphite ” (formula IV) is convertedi. 134 ABSTRACTS OF GHEMICAL PAPERS.by drastic tmatment with hydrogen chloride and ethyl dcohol into a rnolzoe6hyZ ester (&~Ol&El dazk violet p h w and by acetic anhydride and anhydrom sodium acetate at 100" into an anhydride C,,H,O&S (the ammonium salt orange-coloured needles and the diacetyl derivative C,,H2,OllN3S pale yellow leaflets are described). The sulphite C,,H,,010N3S is trans- formed by methyl sulphate and alkali hydroxide mto a monomethyl derivative C23H2,0,0N3S a blackish-violet crystalline powder. Reduction of the violet sulphite or of the corresponding quinone with tin and hydrochloric acid has been shown previously (A.1919 i 36) to yield a colourless compound C,,H2,0,N3S the formation of which is now interpreted as due to the reaction of the :NH and *C02H groups to form :N*CO*. The main product formed by the action of sodium sulphite and sulphurous acid on the quinol from methylcacotheline (A. 1920 i 178) is the compound C2,H2,011N3S2 almost colourless four-sided plates or prisms in which the :N*CO* group of the substance described above has become transformed into :NH*CO,H and the NH group into *NH*S03H. A second product C,,H,08N,S needles appears to be derived from C22H2507N3S by the conversion of the amino- into the hydroxy-group. The violet methosulphite is converted by passing oxygen through its ammoniacal solution into a substance (cf.A. 1919 i 35) to which the composition C,,HBO1+S (instead of Cz1HSOl,N3S) is now assigned and for which the annexed 0 structure is suggested. The substance is i;\.a,co2H re-transformed into the violet metho- sulphite by drastic reduction with sulphur- i N*O*S02*\/*NH ous acid whereas when treated with tin and hydrochloric acid it gives the amine hydrochloride C,,H,,0,N3S,HCl,4H20 (the corresponding sulphate nitrate and hydrobromide are described) and small quantities of a cornpound C22H290&3S,HC1,2H20 thin prisms. Esterification of the compound C,,HSO1,N3S by methyl or ethyl alcohol in the presence of hydrogen chloride yields pro- ducts which contain halogen whereas when su€phuric acid is used the corresponding dimethyl compound C,H,,Ol2N3S pale green prisms and diethyl derivative C2,H,01,N,S rectangular pmms which is hydrolysed by potassium carbonate solution to the monoethyl compound C,H2,0,,N3S almost colourless short prisms are obtained.The dialkyl derivatives are produced by the esterifica- tion of the carboxy-group and the addition of alcohol consequent on the rupture of the ethylene oxide bridge with production of the group :C( OH) *C( OAlk) . The oxidation of the violet nitroquinol by air does not therefore lead to any consLderable degradation of the molecule. A more drastic change is effected when the original material methyl- cacothehe is treated with a solution of bromine in hydrobromic acid. Two products are thereby obtained the first of which produced in 30% yield has the formula CzoHz50sNzBr and crystal- lises in colourless rect'angular prisms or leaflets [aJ$ -6.75" in No2'l)0 11 I 5ORGANIC CHEMISTRY. i.135 aqueous solution (corresponding nitrate colourless needles or prisms [a]': -7.4" when dissolved in water); it appears to be analogous to the salt Cl&C,0,N2*HBr prepared by Hanssen (A. 1887 505; cf. Leuchs Millbrand and Leuchs A. 1922 i 1052) by the action of bromine water on cacotheline. The second cmpound is obtained initially in unstable union with sulphurous acid after the removal of which it has the composition Cl8H2,O,N2Br3 [a]:" -4.3" in aqueous solution; it appears to be a N-methyl derivative of a dibrominated bromide. The methyl ester of the oxime of cacotheline methochloride is converted by methyl alcoholic ammonia a t 100" into the methyl ester of the nitrosophenolmethylbetaine C,H2,07N4 apple-green prisms.H. W. Nitropyrroles. ANGELO ANGELI (Atti R. A c d . Lincei 1922 [v] 31 ii 3-5).-Fischer and Zerweck (A. 1922 i 758) state inaccurately that the nitration of alkylpyrroles may be effected smoothly by means of nitric acid (cf. Angeli A. 1911 i 397). The compounds investigated by these authors are not alkylpyrroles but esters of carboxylic acids derived from keto- formyl- etc. derivatives of pyrrole and the action on them of nitric acid con- sists not in true nitration but in replacement of acetyl aldehydo- methyl etc. groups by nitro-groups. Melanins from Pyrrole Derivatives. PIETRO SACCARDI (Biockm. Z. 1922 132 443-456).-A general review of the relations of melanin pigment to the pyrrole group.The Tetrachlorodipyridioiridiates. UCEL DELBPINE (Compt. rend. 1922 175 1075-1077; cf. A. 1922 i 859).- Details of the preparation of the red and orange isomerides of potassium tetrachlorodipyridinoiridiate by the action of pyridine on potassium iridichloride or potassium aquapentachloroiridiate are given together with an account of certain of their properties. The introduction of pyridine into the complex diminishes the number of acid functions and a t the same time renders the substance more stable. H. J. E. The Reaction between Acetylene and Aniline at High Temperatures. RIKO MAJIMA TADASHI UNNO and KASHICHI ONO (Ber. 1922 55 [B] 3854-3859).-The production of small quantities of indole by subjecting a mixture of acetylene and aniline to a red heat has been observed previously in the presence of aluminium oxide as catalyst.Experiments in which the latter was replaced by the oxides of silicon iron chromium thorium nickel cobalt manganese molybdenum tungsten vanadium or titanium or by metallic nickel iron cobalt platinum palladium osmium or copper did not lead to satisfactory results; nickel is the most powerful catalyst but speedily loses its activity. Better results are obtained by leading a mixture of aniline vapour acetyl- ene and carbon dioxide through a tube heated at 600-700" whereby under the most favourable conditions t'he yield of indole T. H. P. W. 0. K.i. 136 ABSTRACTS OF CHEMICAL PAPERS. amounts to 34% of the changed aniline. Benzene carbazole and p-naphthylamine are produced in considerably smaller quantity and still smaller amounts of pyrrole naphthalene and quinoline are formed.Glass and porcelain tubes are unsatisfactory on account of frequent breakage but excellent service is rendered by wide iron tubes which have been subjected to previous protracted heating a t 600-650" in a current of acetylene whereby the inner surface becomes coated with a thin black compact layer possibly composed of carbon. The formation of indole and @-naphthylamine is probably due to the intermediate production of vinylaniline thus NH,Ph+ CHiCH . . -+ NHPh*CH:CH 1% C6H4<gg>cH and NHPhGHzCH -+ CH,:CH*C,H4*NH2 +'aH# C,,H,*NH,. Carbazole is probably attributable to intermediate diphenylamine. It is uncertain whether carbon dioxide acts merely as a diluent or takes part in the change in accordance with the equation NH,Ph + C,H,+ CO,= C6H4<g>CH+ CO +H,O ; other gases such as ammonia hydrogen nitrogen or water vapour give less satisfactory results.Catalytic Hydrogenation under Pressure in the Presence of Nickel Salts. IV. Qwinoline Bases. JULIUS VON BRAUN ADOLF PETZOLD and JON SEEMANN (Ber. 1922 55 [B] 3779- 3792).-Under the conditions used by the authors quinoline is readily and quantitatively hydrogenated a t 210-215" into 1 2 3 4-tetrahydroq~inoline~ which can thus be prepared more readily than by the older methods. At 250" a portion of the tetrahydro-base is reduced further to decahydroquinoline and another portion is reduced to tertiary bases of pyridine character which have not been fully examined. Quinoline derivatives which contain a substituent in the benzenoid nucleus or in position 1 behave in the same manner as the parent bases whereas those which have a substituent in position 3 are also hydrogenated in the benzenoid nucleus to an extent which depends on the particular substituent present.The reductions are effected with the pure bases or with solutions of them in tetra- or deca-hydronaphthalene. The results are similar in every case and the rate of absorption of the gas is not appreciably influenced by the diluent. 6-Methylquinoline is converted at 120" into 6-methyl-1 2 3 4- tetrahydroquinoline b. p. 131-133"/9 mm. m. p. 37-38" the yield being 90% of that theoretically possible. 6-Chloro-1 2 3 4- tetrahydropinoline long colourless needles m. p. 43" b. p.155"/18 mm. is prepared a t 160"; the corresponding hydrochloride m. p. 190" picrate yellow needles m. p. 150" and nitroso-derivative yellow lustrous prisms m. p. 67" are described. [With A. SCHULTHEISS.]-~ 7-Ethylenedioxyquinoline (Sonn and Benirschke A. 1921 i 805) is very readily hydrogenated at 180-190" to 6 7-ethylenedioxy-1 2 3 4-tetrahydroquinoline H. W.ORGANIC CHEMISTRY. i. 137 b. p. 193"/11 mm. m; p. 101" (picrate m. p. 168"; hydrochloride m. p. 199" (Sonn and Benirschke give m. p. 201") ; nitroso-derivative m. p. 110"). 2-Phenyl-1 .2 3 4-tetrahydroquinoline b. p. 196-197"/12 mm. is obtained from 2-phenylquinoline a t 150". 3-Ethylpinoline an almost colourless liquid with an odour of quinoline is prepared in 80% yield by the action of n-butyraldehyde on o-aminobenzaldehyde a t 120-130" (cf.Wislicenus and Elvert A. 1909 i 420); it has b. p. 135-138"/12 mm. d? 1*0508 ng 1.6030 (hydrochloride m. p. 173" ; picrate m. p. 197" ; methiodide m. p. 191"). The base is readily hydrogenated a t 180-lW" yielding thereby a mixture of 3-ethyl-5 6 7 8-tetrahydroquimline and 3-ethyl-1 2 3 4-tetrahydroquinoline which are separated from one another with the aid of benzoyl chloride. The former is a colourless liquid b. p. 125-128"/12 mm. di0 0.99218 n1,8 1.5311 (it gives a methiodide m. p. 120" and a picrate yellow crystals m. p. 158"). The latter has b. p. 140"/12 mm. d 19041 ng 1.5625 (hydrochloride m. p. 210" ; picrate m. p. 142" ; methiodide C,,H,,,NI m. p. 205"; the benxoyl and nitroso-derivathes are non- crystalline). 3-n-AmyZquinoline prepared in almost quantitative yield from o-aminobenzaldehyde and heptaldehyde at 180" is a colourless liquid b.p. 179'116 mm. di7 1.0048 n$ 1.5715 (hydrochloride m. p. 174"; picrate m. p. 153"; methiodide m. p. 69"). It is hydrogenated at 180-190" with the formation of approximately equal amounts of 3-n-amyl-1 2 3 4-tetrahydroquimZine a colour- less liquid b. p. 159-164"/12 mm. di7 0.96625 ng 1.5339 (hydro- chloride needles m. p. 124" methiodide C,,H,,NI m. p. 145"; the picrate benzoyl compound acetyl derivative and nitroso-com- pound are non-crystalline) and 3-n-amyl-5 6 7 8-tetrahydro- quimline a colourless nearly odourless liquid b. p. 167"/12 mm. di7 0.96028 ng 1*5188. The latter substance gives a picrate long needles m.p. 135" and a methiodide m. p. 105" whereas the chloride and chloroplatinute are non-crystalline; it does not react with acetic anhydride or nitrous acid. The base is oxidised by potassium permanganate to oxalic and pyridine-2 3 5-tricarb- oxylic acids. 3-Phenylquinoline is hydrogenated with remarkable ease a t 160" to a mixture of 3-phenyl-1 2 3 4-tetrahydroquinoline and 3-phenyl- 5 6 7 8-tetrahydroquinoline from which the former is very readily separated by taking advantage of its sparing solubility in alcohol with which the latter is readily miscible. 3-Phenyl- 1 2 3 4-tetrahydroquinoline i s a colourless crystalline sub- stance m. p. 83" (hydrochloride matted needles m. p. 229" ; picrate m. p. 181"; picrolonute a yellow powder m. p. 205"; acetyl deriv- ative feathery crystals m.p. 78"; nitroso-compound m. p. 147"). 3-Phenyl-5 6 7 8-tetrahydroquinoline is a colourless liquid b. p. 211-212"/18 mm. (hydrochloride m. p. 235" after darkening at 225-230"; picrolonute a yellow powder m. p. 201"; methiodide C,,H,,NI m. p. 240-243"; the base does not react with nitrous acid or acetic anhydride). 3-Phenyl-5 6 7 8-tetrahydro-i. 138 ABSTRACTS OF CHEMICAL PAPERS. quinoline is reduced by sodium and ethyl alcohol to 3-phenyl- deuzhydroquinoline colourless crystals m. p. 98" after softening at 95" (nitroso-derivative m. p. 110" ; picrate m. p. 210-212" ; the acetyl compound could not be caused to crystallise). The Relative Stability of Cyclic Bases. VII. Substituted Tetrahydroquinoline Rings. JULIUS VON BRAUN JON SEE- MANN and ADAM SCHULTHEISS (Ber.1922 55 [B] 3803-3817).- In previous communications (von Braun and Neumann A. 1917 i 282; von Braun Heider and Neumann A. 1917 i 167) it has been pointed out that whereas the stability of the tetrahydro- quinoline ring towards scission during reduction by sodium amalgam is little affected by the introduction of the methyl group in position 2 that of the dihydroindole ring is modified profoundly by 2 or 3 methylation. The presence of a methyl group in position 3 or 4 in the tetrahydroquinoline ring is now shown not to exert a marked effect on the course of the change. On the other hand the presence of a phenyl group in position 2 causes the almost quantitative) rupture of the non-aromatic ring linking whereas when the group is in position 3 the three possible types of change are realised.H. W. CH2-FHMe 1 ; 3-Dimethyl-1 2 3 4-tetrahydroq~inoline~ "H4<Nlyre.CH. ' an almost colourless liquid b. p. 130-132"/17 mm. is obtaiGed by the reduction of 3-methylquinoline methiodide by tin and hydrochloric acid ; the very hygroscopic hydrochloride m. p. about 110" picrate m. p. 131" and methiodide m. p. 204" are described. The corresponding quaternary chloride is converted by sodium amalgam into a mixture of 1 3-dimethyl-1 2 3 4-tetrahydro- quinoline and y-phen yl- p-methylpropyldimethylamine CH,Ph*CHMe*CH,*NMe from which the former is removed by treatment with formaldehyde in hydrochloric acid solution. The latter base is a colourless liquid b. p. 100-105"/7 mm. 22l"/atmospheric pressure (hydro- chloride m.p. 90"; picrate m. p. 87"; methiodide m. p. 140"). 1 4-Dimethyl-1 2 3 4tetrahydroquinoline methiodide is con- verted into the corresponding chloride which is transformed by sodium amalgam into 1 4-dimethyl-1 2 3 4-tetrahydroquinoline and y-phenylbutyldimethyiylamine CHMePh*CH,CH,*NMe a liquid b. p. 112-115"/7 mm. (hydrochloride m. p. 100"; picrate rn. p. 98"; methiodide m. p. 125") which constitutes 60% of the mixture of bases. 2-Phenyl-1 2 3 4-tetrahydroquinoline is converted by methyl iodide and aqueous alkali mainly into 2-phenyl-1-methy,?-1 2 3 4- tetrahydroquinoline colourless crystals m. .p. 101" b. p. 188- 192"/14 mm. ; the corresponding hydrochlorzde m. p. 157" chloro- platinate m. p. 172" nitroso-compound a microscopic green powder m.p. 75" and diphenylmethane derivative C33H34N2 m. p. 60" after slight previous softening are described. The base unites with difficulty with methyl iodide to give the quaternary iodide C17H20NI rn. p. 185". The quaternary chloride (chloro-ORGANIC CHEMISTRY. i. 139 pkrtinute orange-coloured crystals m. p. 203") is almost quantit- atively converted by sodium amalgam into o-y-phenylpropjldi- methylaniline NMe,C6H4*CH,*CH2*CH2Ph b. p. 175-178"/10 mm. The picrate golden-yellow needles m. p. 110" the non-crystalline hydrochZoride and the chloroplatinate a yellowish-brown powder m. p. 170° are described. The base does not unite reaGly with methyl iodide thus proving that the dimethylamino-group is attached to the benzenoid nucleus and is sterically hindered and that the compound is not the isomeric substance CH,Ph*CH,*CHPh*NMe,.2-Phenyl-l-methyl-l 2 3 4-tetrahydroquinoline suffers fksion in accordance with the scheme C,H,< CH2-(?H2 +BrCN -+ CN*N~e-.C6H,.CH2*CH,*CHPhBr to the extent of a t least 50% when treated with cyanogen bromide in a gently boiling water- bath. Since cyanoamides cannot be distilled without decom- position and seldom crystallise the product of the action is treated directly with an excess of piperidine whereby i t is converted into a portion insoluble in acid b. p. 218°/vacuum and a bromine-free amorphous base which readily unites with methyl iodide giving the substance C2,H3&1 a pale-brown microcrystalline powder which loses methyl iodide without definitely melting above 60".The hydroxide C 6 H 4 (CH2-cH,>CHPh "Me2(OH) is mainly decomposed with loss of methyl alcohol when distilled under diminished pressure giving 2-phcnyl-1-methyl-1 2 3 4-tetrahydroquinoline. In striking contrast to the 2-phenyl derivative 3-phenyl- 1 2 3 4-tetrahydroquinoline is readily converted by methyl iodide into the quaternary iodide C,,H$I m. p. 172" which loses methyl iodide when distilled under diminished pressure and gives 3-phenyl-1-methyl-1 2 3 4-tetrahydroquinoline m. p. 42" b. p. 195-202"112 mm. The corresponding picrate m. p. 17S0 the non-crystalline hydrochloride and chloroplatinate m. p. 192" the nitroso-compound m. p. about 105" and the diphenylmethane derivative C,,H,,N m. p. 92" are described. The quaternary chloride (chloroplatinate m.p. 204O) is converted by sodium amalgam into a mixture of 3-phenyl-1-methyl-1 2 3 4-tetra- hydroquinoline (47 % ) P y -diphen ylpropyldimethylarnin e CH,Ph-CHPh*CH,*NMe (45%) and 0- p - phen yl pro pyldimeth ylaniliiz e NMe,*CGH4*CH2-CHPhMe (8%). The two bases last mentioned are separated by taking advantage of the inability of the o-dimethylaniline derivative to unite with methyl iodide in ethereal solution. The quantity of 0- p -phenylpropyldimethylanihe available was insufficient to permit its isolation in a homogeneous condition and it is characterised by its picrate m. p. 166-167" non-crystalline hydrochloride and chloroplatinate m. p. 181 '. py-Diphenylpropyldimethylamine meth- iodide is exceedingly hygroscopic; it is conrerted into the corre- sponding chloride (chhoaurute yellow needles m.p. 164" ; chloro- platinate a microcrystalline powder m. p. 236"). The constitution NMe-CHPhi. 140 ABSTRACTS OF CHEMICAL PAPERS. of the iodide is established by the observation that the corre- sponding hydroxide is decomposed when distilled under diminished pressure into u-benzylstgrene CH,Ph*CPh:CH a liquid b. p. 140-143"/vacuum @ 1.0143 nz 1.5903 (corresponding dibromide colourless crystals m. p. 98"). 3-Phenyl-1-methyl-1 2 3 4- tetrahydroquinoline is converted to a slight extent by cyanogen bromide into the corresponding methobromide (identified by its conversion into the chloroplatinate m. p. 203") but mainly into 1-cyano-3-phenyl-1 2 3 4-tetrahydroquinoline m. p. 78" ; fission of the ring as in the case of the 2-phenyl compound is not observed.6 7-Ethylenedioxyquinoline methiodide m. p. 270" is reduced by tin and hydrochloric acid to 6 7-ethylenedioxy-1-methyl- 1 2 3 4-tetrahydroquinoline colourless leaflets m. p. 53" b. p. 165-170"/8 mm.; the hydrochloride m. p. 203" picrate m. p. 170" and the methiodide m. p. 220° are described. The latter compound is prepared more advantageously by exhaustive methyl- ation of 6 7-ethylenedioxy-1 2 3 4-tetrahydroquinoline. It is converted into the corresponding chloride which is transformed by sodium amalgam into a mixture of 6 7-ethylenedioxy- 1-methyl- 1 2 3 4-tetrahydroquinoline (40%) and y-3 4-ethylenedioxy- phenylpropyldimethylamine C2H4O2:C6H,*[CH2J,~NMe2 b. p. 162- 164"/7 mm. which are separated from one another by careful fractional distillation ; the methiodide m.p. 165" picrate m. p. 167" and hydrochloride m. p. 169" of the latter base are described. H. W. The Mechanism of Syntheses of isoQuinoline from Benzyl- amine Derivatives. PAUL STAUB (Helv. Chim. Acta 1922 5 888-894).-Many attempts to synthesise isoquinoline derivatives by ring closure in appropriate benzylamine derivatives are recorded in the literature but few were successful. In an attempt to deter- mine the conditions requisite for success the following new com- pounds were prepared for ring closure experiments. u-Benzylidineaminoiso~o~l alcohol CHPh:N*CH,-CHMe*OH was obtained by reducing nitroisopropyl alcohol NO,CH,*CHMe *OH and condensing the product with benzaldehyde; it crystallises in needles m.p. 76-76.5". Benzylaminoisopropyl alcohol CH,Ph*NHCH,*CHMe*OH was obtained by the action of benzyl chloride on aminoisopropyl alcohol; b. p. l42"/11 mm. Benxenesulphonbenxyl- p-hydroxyethylamide CH,Ph*N(SO,Ph)*CH,*CH,*OH was prepared by heating benzenesulphonbenzylamide with ethylene oxide a t 170"; it formed a viscous liquid crystallising slowly and was not purified. CH,Ph*N( S02Ph)-CH2*CH2Br was formed by the dction of ethylene bromide on sodium benzene- sulphonbenzylamide in ether ; m. p. 82-5-83" Benzenesulphonbenz yl- p-bromoethylamide,ORGANIC CHEMISTRY. i. 141 Benzenesulphonbenzyl phenacylamide CH,Ph*N( S0,Ph) *CH,*COPh was prepared by the action of bromoacetophenone on sodium benzenesulphonbenzylamide in ether. The product was not crystallised.Acetobenzylphenmylarnide CH2Ph*NAc*CH,*COPh was obtained as a syrup from the action of bromoacetophenone on sodium benzyl- acetamide. In no case could any i.soquinoline derivative be obtained by the action of any of the usual ring-closing agents on any of the above six compounds. It is concluded that a benzene derivative containing the chain Ph*C*N*C*C- must if it is to be convertible into an isoquinoline base conform to two rules. In the lateral chain it must have (1) a system of conjugated double bonds actual or potential; (2) a hydroxyl or alkoxyl group in P-position to the nitrogen atom. The compound CHPh:N*CH,*CH( OEt) conforms to condition (1) by loss of alcohol and readily forms isoquinoline. Benzyl- aminoacetaldehyde CH,PhmNH*CH,*CHO which is converted by fuming sulphuric acid into isoquinoline may be supposed to undergo oxidation and by passing into the tautomeric form CHPh:N*CH:CH*OH satisfies the conditions.E. H. R. Catalytic Hydrogenation under Pressure in the Presence of Nickel Salts. V. The Carbazole Complex. JULIUS VON BRAWN and HEINRICH RITTER (Bet-. 1922 55 [B] 3792-3803; cf. Padoa and Chiaves A. 1908 i 772).-The process used by the authors does not cause the hydrogenation of pyrazole even a t 260" and under a pressure of 30 atmospheres in spite of the use of a material which has been exhaustively purified in several different manners. Nevertheless the authors are inclined to the view that pure pyrazole is not unusually resistant to hydrogenation and that the failure of their experiments is due to some unknown catalytic impurity.The N-alkylcarbazoles are readily hydro- genated with initial addition of four hydrogen atoms to one benzene nucleus. Subsequently the second benzene nucleus is attacked with addition of a further four atoms of hydrogen. 9-Methylcarbazole readily absorbs about seven atomic pro- portions of hydrogen a t 210-215" and 25 atmospheres pressure. About 40% of the initial material remains unchanged whilst a portion is perhydrogenated with production of ammonia. The liquid pro- duct of the reaction is a mixture of the octahydro- and tetrahydro- derivatives with 9-methylcarbazole from which the first-named is readily separated by treatment with hydrochloric acid (20y0) in which it alone dissolves. The remaining bases cannot be separated from one another by distillation or crystallisation ; the presence of the tetrahydro-compound is established by the formation of the strongly basic hexahydro-derivative when the mixture of bases is treated with tin and hydrochloric acid.i.142 ABSTRACTS OF CREMICBL PAPERS. c I3 C3-:l'lethyEocta~~yd~ocu~~~zoZe MeN< I 6 8 lustrous leaflets m. p. 94" b. p. 176-178"/16 mm. is somewhat unstable towards air and does not give the pine-shaving or the dimethylaminobenzalde- hyde reaction. It does not give a well-defined picrate or meth- iodide; it is readily oxidised by chromic acid potassium perman- ganate or ferric chloride but it has not yet been found possible to isolate uniform products of the action. It is conveniently reduced by tin and boiling hydrochloric acid (20%) to 9-methyZ- decahydrocarbaxole MeN< I 6 8 a colourless mobile liquid b.p. 138-139"/12 mm. which is stable towards air ; the non-crystalline hydrochloride hydrobromide m. p. 202-204" picrate m. p. 162" and methiodide m. p. 189" are described. The presence of the double bond in the base is betrayed only by its instability towards permanganate ; it yields only the hydrobromide when treated with fuming hydrobromic acid and does not unite with bromine or hydrogen even under widely varied conditions. The methiodide is transformed by successive treatment with silver oxide and dis- tillation into 2 -dimeth ybamino- A1 1'- dic yclohexen yl "-'\ (annexed formula) an almost colourless liquid b. p. ()We2 'I I 148"/15 mm. which is conveniently characterised \' as the picrate m.p. 157". The new base is some- what unstable and is readily converted by sulphuric acid (10%) into 2-A1-cyclohexenyZcyclohe~anone (annexed formula) b. p. 137- CH CH 139"/14 mm. diB 1.015 n 1.515 (semicurbaxone /\ /\ m. p. 191"). The ketone is certainly not p - 9 ~ (p2 identical with the compound obtained by Wallach (A. 1911 i 473) by the auto- %' '' CH CH2 condensation of cyclohexanone and it would \' \A2 therefore appear that the latter has the CH2 alternative constitution C,H,O:C,H,o sug- gested by Wallach. 2-Al-cycZoHexenylcyclohexanone is readily reduced by hydrogen in the presence of palladous chloride to 2-cycZohexylcycZohexanone b. p. 135"/11 mm. d:6 0.987 ny 1.4915 (cf. Wallach Zoc. cit.). The reduction of 9-ethylcarbazole is very similar to that of the methyl compound.9-EthyZoctahydrocarbuzoZe lustrous leaflets m. p. 43" b. p. 162-163"/9 mm. is unstable towards air; it does not give a methiodide or a picrate. It is readily reduced to 9-ethyldecahydrocarbuzoZe a colourless liquid which is stable towards air b. p. 140-141"/12 mm. ; the non-crystalline hydrochloride the picrate m. p. 133" and the methiodide m. p. 176-177" are described. 2 - M ethylet h ylarnino-A1 1'- d icyclohexen yl is a colourless liquid b. p. 148-150"/12 mm. which gives a non-crystalline picrate and methiodide. It is readily transformed by dilute sulphuric acid into 2-A1-cycZohexenylcycZohexanone and methyl- ethylamine. 9-EthyZhexahydrocarbuzoZe is a colourless odourless liquid b. p. 155-167"/8 mm. 292-293"/749 mm.; it gives a well-defined methiodide m.p. 174". The base is transformed by a mixture of nitric and sulphuric acids at 0" into the nitro-com- C6H8 C HORGANIC CHEMISTRY. i. 143 /\ pound (annexed formula) golden-yellow needles f) I I m. p. 142" which is reduced by stannous chloride \/\/\/ to 7-amino-Q-ethylhexahydmcarbazole an almost colourless very viscous liquid b. p. 224-226"/24 mm. which shows all the cdour reactions characteristic of the simpler meta-diamines of the aromatic series. It has not been found possible to effect the smooth dehydro- genation of 9-methyl- or 9-ethyl-hexahydrocarbazole ; if the substances are passed over lead oxide and pumice the alkyl residues are lost and carbazole is produced. Benzo-polymethylene Compounds. VIII. Cyclic Analogues of Atophan. JULIUS VON BRAUN and PAUL WOLFF (Ber.1922 55 [B] 36763688) .-1 -Ketotetrahydronaphthalene condenses readily with isatin to give 5 6-dihydro-a-naphthacridine-7-carb- oxylic acid (annexed formula) which readily undergoes further change. Constitutionally the new compound (for f2\ which the name tetrophan is proposed) is allied I somewhat closely to 2-phenyl-4-cinchoninic acid &hAh/\/ 120 (atophan) from which however it differs entirely 'I 16a 61 in its physiological action having a characteristic p/x($/ effect on the spinal marrow somewhat resembling that of strychnine. The physiological properties of a number of its derivatives have been examined. Substitution in the benzenoid nuclei of the isatin and tetrahydronaphthalene complexes does not affect the qualitative nature of the action.The presence of the basic nitrogen atom and of the carboxyl group appears essential. A new ring produced by the insertion of more than one carbon atom into the atophan complex must be present which must not be eccentrically united. The group -CH,*CH,- need not be so fully hydrogenated. Since tetrahydroatophan shows close physiological resemblance to tetrophan it is to be expected that the action of the latter would be considerably enhanced by transforming it into its tetrahydro- derivative; the reverse is however found to be the case. 5 6-Dihydro-a-naphthacridine-7-carboxylic acid slender pale yellow needles m. p. 252" (decomp.) is prepared smoothly by heating a mixture of isatin and 1-ketotetrahydronaphthalene with aqueous- alcoholic potassium hydroxide solution.It reacts incompletely with mineral acids yielding salts which are hydrolysed by water. The sodium lead basic copper and silver salts are described; the ethyl ester long coarse needles m. p. 80" is prepared from the latter. The acid is decomposed when heated somewhat above its melting point into 5 6-dihydro-u-naphthucridine b. p. 237- 238"/10 mm. m. p. 60"; the corresponding hydrochloride long needles m. p. 226" picrate m. p. 206" and methiodide a reddish- yellow crystalline powder decomp. 191" are described. The base is readily dehydrogenated by lead oxide at 300-320" with production of a-naphthacridine m. p. 108". It is oxidised by chromic acid in glacial acetic acid solution to 5 6-diketo-u-nuphth- acridine an orange-coloured powder m.p. 242" (a salt with chromic acid is described). NEt H. W.i. 144 ABSTRACTS OF CHEMICAL PAPERS. 5 6-Dihydro-ct-naphthacridine-7-carboxylic acid is converted by bromine and glacial acetic acid a t 120-130" into u-naphth- acrid~ne-7-mrboxylic acid a yellow powder m. p. 261" ; the cobalt copper mercuric and silver salts are described. The methyl ester has m. p. 83". When heated above its melting point the acid is converted into a-naphthacridine m. p 108". 5 6-Dihydro-a-naphthacridine-7-carboxylic acid /\ is reduced by sodium and ethyl or amyl alcohol to 5 6 6a 7 12 12a-hexahydro-a-naphthmridine- /\/\/\/ 7 -carboxylic acid (t etrahydrot etrophun) (annexed I I I formula) m. p. 190" (decomp.) after softening \/\/\/ at 180" ; the corresponding nitroso-compound after softening a t 244" are described.6 6-Dihydro-ll-methyl-a-naphthucridine-7-carboxylic acid colour- less crystals m. p. 188" (decomp.) is prepared from l-ketotetra- hydronaphthalene and o-methylisatin ; its metallic salts closely resemble those of the parent acid. 5 6-Dihydro-11-methyl- u-naphthacridine has m. p. 93-94" b. p. 248-253"/16 mm.; it yields a picrate m. p. 141" and a hydrochloride long pale yellow needles m. p. 173" but could not be caused to react with methyl iodide. 11-Methyl-a-naphthucridine m. p. 107" gives a picrate m. p. 155" and a hydrochloride m. p. 136" but does not react with methyl iodide. 5 6- Dihydro-9-methyl- a-naphthctcridine-7-carboxylic acid has m. p. 293" (decomp.); the sodium salt dissolves very sparingly in water.9-Clhloro-5 6-dihydro-a-naphthacridine-7-carboxylic acid m. p. 283" (decomp.) is converted when heated above its melting point into 9-chloro-5 6-dihydro-a-naphtFridine m. p. 102" b. p. 257- 26Zo/14 mm. The hydrochloride of the latter has H2 m. p. 240". The base loses the chlorine atom H2/\H2 when it is treated with lead oxide. I 9-Bromo-5 6 - dihydro-u-napht7Lcccridine- 7 -curb - H k / N oxylic acid has m. p. 265"; the silver salt and the It ,) methyl ester yellow leaflets m. p. 128" are de- /\/\/\ /' scribed. 9-Bromo-5 6-dihydro-a-naphthacridine 1 (I I IH crystallises in pale yellow needles m. p. 168" \/\/\/ (hydrochloride m. p. 211-212"). 5 6-Dihydro-1 2 - tetramethylene - a - naphthacr - idine-7-carboxylic acid (annexed formula) is pre- pared from isatin and a-keto-octahydrophenanthrene ; it has m.p. 210" (decomp.) after darkening at 180". 'OZH 2-Acenaphthylquinoline-4-carboxylic acid pre- pared from 4-acetylacenaphthene and isatin has pared from ar-u-aminotetrahydronaphthylamine pyruvic acid and benzaldehyde forms yellow crystals m. p. 260" after softening at 255"; its NH I eo,H I m. p. 149" and acetyl derivative m. p. 247" I" H2 '\/\ 'I 1 IPh m. p. 234" after previous softening. T'etramethyleneafophan (annexed formula) pre- H2 physiological action resembles that of atophan. H. W.ORGANIC CHEMISTRY. i. 145 The Relationship between Fluorescence and Chemical Constitution in the Case of Derivatives of Benzoxazole. 11. F. HENRICH [with H. SUNTHEIMER and c. STEINMANN] (Ber. 1922 55 [B] 391 1-3921).-1n a previous communication (A.1921 i 886) it has been shown that the development of fluor- escence in alkaline solutions of hydroxybenzoxazole derivatives an aromatic nucleus directly attached by one of its carbon atoms to the p carbon atom and when the hydroxyl group is in the para- position to the nitrogen atom. This regularity is now shown to be true of certain derivatives of 2 4-dihydroxytoluene. 2 4-Dihydroxytoluene m. p. 106-107" is prepared from 2 4-diaminotoluene through the compounds 2-amino-4-acetamido- toluene 4-acetamido-2 - hydroxytoluene and 4- amino-2 - hydroxy- toluene ; the necessary conditions for each change are described fully in the original. It is converted by amyl nitrite and potassium hydroxide in absolute alcoholic solution into 5-nitroso-2 4-di- hydroxytoluene decomp.175-180" according to the rate of heating after darkening and softening at 146"; the potassium salt and the dibenzoyl derivative m. p. 146-153" are described. 2 4-Dihydroxytoluene is converted by fuming nitric acid in the presence of ether into a mixture of 3-nitro-2 4-dihydroxy- toluene bright red needles m. p. l l l " and 5-nitro-2 4-dihydroxy- toluene orange-yellow needles m. p. 118-119" which can be separated from one another by takiag advantage of the volatility of the former with steam with which the latter does not volatilise. 5-Amino-2 4-dihydroxgtoluene hydrochloride long colourless needles is prepared by the reduction of the corresponding nitroso- or nitro-compound with stannous chloride and concentrated hydro- chloric acid.The free amine is obtained by the addition of sodium hydroxide to an aqueous solution of the hydrochloride; it dis- solves to a dark blue solution in an excess of the reagent and readily absorbs oxygen without however becoming converted into dyes of the type of litmus. The hydrochloride is transformed by acetic anhydride into 6-acetoxy-2-methyl-5-methylbenzomzole (an- nexed formula) colourless needles m. p. 94" which is converted by alcoholic potassium hydroxide solution into Me/\/N\ 6-hydroxy-2 5-dimethylbenxoxaxole colourless crystals m. p. 221"; the latter substance does ~ c ~ ( / j ~ / not fluoresce in aqueous alkaline solution. 6-Benzoxy-2 -phenyl- 5-methylbenxoxaxole colour - less crystals m. p. 164-165" after softening a t 163" is prepared from 5-amino-2 4-dihydroxytoluene hydrochloride and benzoyl chloride and is hydrolysed by alcoholic potassium hydroxide solu- tion to 6-hydroxy-2-phenyl-5-methylbenzoxazole colourless crystals m.p. 242" after softening a t 236" which has a green fluorescence in aqueous sodium hydroxide solution. 3-Nitro-2 4-dihydroxytoluene is reduced by stannous chloride and concentrated hydrochloric acid to 3-amino-2 4-dihydroxy- toluene hydrochloride colourless needles which is converted by HO*C,H,<O>C*R(p) N is observed only when R in position p is CMe VOL. CXXIV. i. Bi. 146 ABSTRACTS OF CHEMICAL PAPERS. successive treatment with benzoyl chloride and alcoholic potassium hydroxide solution into 4-hydroxy- 2 -phen yl-5-meth ylbenxoxazole colourless crystals which do not exhibit fluorescence in alkaline solution.H. W. The Relative Stability of Cyclic Bases. VIII. The Phen- morpholine and Homotetrahydroquinoline Rings. JULIUS VON BRAUN and JON SEEMANN (Ber. 1922 55 [B] 3818-3825).- It is shown that the stability of the morpholine ring is increased to an unexpected extent by its association with an aromatic ring so that it is at least as stable towards cyanogen bromide as 1 2 3 4- tetrahydroquinoline and even more stable towards sodium amalgam. Homo-1 2 3 4-tetrahydroquinoline does not suffer fission under the action of cyanogen bromide. The nine rings (pyrrolidine piperidine morpholine dihydro- indole dihydroisoindole 1 2 3 4-tetrahydroquinolineY 1 2 3 4- tetrahydroisoquinoline phenmorpholine and homo- 1 2 3 4-tetra- hydroquinoline) exhibit uniformity in their behaviour towards cyanogen bromide sodium amalgam and the Hof mann degradation if the dihydroindole and the phenmorpholine systems are not considered.The peculiarity of the latter rings appears to be caused by the presence in them of the group -Ar-T-CH,-CH,-. N - p-Hydroxyethyl-o-anisidine is conveniently prepared in 75- 80% yield by heating o-anisidine with a considerable excess of ethylene chlorohydrin on a water-bath; the picrate m. p. 140" and the hydrochloride m. p. 134" are described. The base is con- verted by successive treatment with concentrated hydrochloric acid at 160-180" and with dilute aqueous alkali into Dhenmor- I I pholine C&,< '-YH2 b. p. 127-128"/12 mm. The latter is NH*CH transformed quantitatiGly by methyl iodide and sodium hydroxide into l-methylphenmorpholinium methiodide which is decomposed when distilled under diminished pressure into 1 -methylphenmorph- oline b.p. 124"/12-5 mm. (picrate m. p. 144"). 1 -Methylphenmorpholine reacts slowly with cyanogen bromide at the temperature of boiling water giving unchanged material 4-methylphenmrpholine methobromide in. p. 213" and 4-cyano- phenmrpholine C,H,< '-YH2 aliquid b. p. 115-118"/1 mm. ; the'latter could not be obtained in a perfectly homogeneous con- dition and is characterised by converting it into the guanidine derivative C6H4< YH2 small colourless needles m. p. 152" the non-crystalline hydrochloride and chloroplatinate m. p. 144" of which are described. 4-Methylphenmorpholine methochloride is converted by sodium amalgam almost entirely into 4-methylphenmorpholine which is characterised by its con- version into the diphenylmethane derivative C,,H,20,N2 a very viscous liquid b.p. about 260"/2 mm. (decomposition occurs before completion of the distillation) (dimethiodide m. p. 157"). N( CN) *CH O-- N[C(:NH)*NHPh]*CH,'ORGANIC CHEMISTRY. i. 147 The difficulty of methylating homo-1 2 3 4-tetrahydroquinol- h e c6H4<NH2%H2>CH2 CH *CH has been indicated previously (von Braun and Bartsch A 1913 i 197) ; it has now been found possible to isolate the products of the action in a homogeneous form. The quaternary iodide C,,H,,NI has m. p. 155". 1-Methylhomo- 1 2 3 4-tetrahydroquinoline has b. p. 108-110"/10 mm. and gives a picrate m.p. 139" a non-crystalline hydrochloride and a chbroplatinate m p. 186"; it combines very slowly with methyl iodide. It is obtained as the sole basic product when the correspond- ing quaternary hydroxide is distilled in a vacuum It is converted by cyanogen bromide into 1-methylhomo-1 2 3 4-tetrahydro- quinoline methobromide (identified as the corresponding chloro- platinate m. p. 197") and 1-cyanohomo-1 2 3 4-tetrahydro- quinoline C,H,<F$*> b. p. 178-182"/13 mrn. which could not be obtained completely free from bromine and is identified bv transformation into the guanidine derivative 2 Condensation of Benzidine with Formaldehyde. HEISA- BURO KONDO and SUEZO ISHIDA ( J . Pharm. SOC. Japan 1922 979-985; cf. H. Schiff A. 1892 1223).-By adding 30 C.C.of 35% formaldehyde solution to benzidine (10 g . ) dissolved in abso- lute alcohol (200 c.c.) a greyish-white light amorphous condens- ation product dimethanolbenxidine C1,H,(NH*CH2*OH) is pre- cipitated; it has m. p. 271-272" after sintering a t 260" and is sparingly soluble in alcohol or water but readily soluljle in hydro- chloric acid. It is not identical with dimethylenebenzidine m. p. 140-141" obtained by Schiff by the same method. When dissolved in dilute hydrochloric acid the odour of formaldehyde is perceptible and from the solution methanolbenxidine hydrochloride NH,*C,H4*C6H4*NH*CH2*OH,2HC1 colourless needles of high melting point was isolated by adding concentrated hydrochloric acid. When kept for a long time the solution in hydrochloric acid gave a reddish-violet product identical with the compound C,,H,,N obtained by Schiff.K. K. The Behaviour of certain Dibenzamidoethylene Derivatives Prepared from Iminazoles towards Acid Anhydrides. A. WINDATJS and W. LANGENBECK (Ber. 1922 55 [B] 3706-3709; cf. Windaus Doxies and Jensen A. 1922 i 61).-Dibenzamido- ethylene derivatives are converted by the anhydrides of fatty acids into glyoxalines which contain the alkyl group of the fatty acid in position 2. ap-Dibenzamido-Aa-propylene is converted by acetic anhydride a t 180" into 2 4(2 5)-dimethylglyoxaline identified as the picrate m. p. 142". The action of propionic anhydride under similar con- ditions leads to the formation of 4( 5 ) -methyl-2-ethyZglyoxaZine hygroscopic crystals m. p. 45" (oxalate colourless leaflets m.p. Y 2i. 148 ABSTRACTS OF CHEMICAL PAPERS. 145" ; picrate yellow prisms m. p. 131" ; hydrochloride hygroscopic leaflets m. p. 132"; nitrate m. p. 129"). 4(5)-Ethylglyoxaline (cf. Kolshorn A. 1904 i 675) is converted by benzoyl chloride and sodium hydroxide into ccp-dibenxamido- A"-butylene NHBz*CH:CEt*NHBz needles m. p. 146" which is transformed by acetic anhydride a t 140" into 2-methyl-4(5)-ethyl- glyoxaline (picrate yellow needles m. p. 90-91" ; oxahte m. p. 141"). The glyoxaline obtained by Windaus and Ullrich (A. 1914 i 662) by the action of ammonia on rhamnose is identified as 4( 5) -methyl- 5( 4) -ethylglyoxaline. BRITISH DYE- STUFFS CORPORATION LTD. JAMES BADDILEY and ERNEST HARRY RODD (Brit. Pat. 189295).-New dyes of the triarylmethane series are obtained by condensing a 4 4'-dialkyldiamino-3 3'-dimethyl- benzophenone with a suitable secondary or tertiary amine by warming with phosphoryl chloride with the addition of a neutral diluent if desired.These new basic products dye cotton mordanted with tannin-antimony valuable shades of bright reddish- to bluish- violet. Those containing phenyl benzyl or naphthyl residues substituted in the amino-groups can be sulphonated by warming with 20% fuming sulphuric acid giving new acid dyes which dye wool level shades of violet. The ketones required for the synthesis may be obtained by boiling the corresponding thio-ketones with concentrated hydrochloric acid and the latter are produced by the condensation of monoalkyl-o- toluidines with formaldehyde and converting the resulting diphenylmethane derivatives into thio- ketones by the process of Brit.Pat. 20615/14. 4 4'-DiethyZdi- amino-3 3'-dimethylbenxophenone m. p. 165" when condensed with ethyl-o-toluidine gives the triphenylmethane dye in the form of a bronze-coloured paste which dyes mordanted cotton a reddish- violet. Bluer shades are obtained by condensing the ketone with benzylethylaniline or dibenzylaniline. CHEMISCHE FABRIK AUF ACTIEN (VORM. E. SCHERING) (D.R.-P. 358397 ; from Chem. Zentr. 1922 iv 950).-Ammonia is allowed to act on pyridine or its homologues in the presence of alkali metals. For example anhydrous pyridine a t 80" is added to a suspension of finely divided sodium in toluene and anhydrous ammonia is passed in. After the sodium is used up the temperature is raised to 130" and so maintained until no more ammonia is absorbed.From the pro- ducts of the reaction by extracting with ether distilling off the ether and fractionation of the residue 2-aminopyridine b. p. 103- 1lOo/20 mm. 4 4'-dipyridyl b. p. 173-180"/20 mm. and 4-amino- pyridine are obtained. 6-Amino-2-methylpyridine a yellow oil b. p. 120-130"/20 mm. is similarly prepared from 2-methylpyridine (ac-picoline). GI. W. R. Preparation of Amino-alcohols of the Quinoline Series. SOCIETY FOR CHEMICAL INDUSTRY IN BASLE (Swiss Pats. 92001 92607 92608 and 92609; from Chem. Zentr. 1922 iv 950- 951) .-2-Pheny1-4-quinolyl methyl ketone is halogenated on the H. W. Preparation of Triarylmethane Dyestuffs. G. F. M. Preparation of Aminopyridines.ORGANIC CHEMISTRY.i. 149 methyl group and the halogen derivative allowed to react with primary or secondary amines. The N-alkyl-2-phenylquinolyl 4- aminomethyl ketone thus formed is reduced to the amino-alcohol. 2-Phenyl-4-quinolyl methyl ketone yellow crystals m. p. 75" is prepared by condensation of ethyl 2-phenylquinoline-4-carboxylate with ethyl acetate and elimination of carbon dioxide from the intermediate ethyl 2-phenylquinoline-4-acetute m. p. 52-54" or from 4-cyano-2-phenylquinoline by Grignard's reaction. By the action of bromine on the ketone 2-phenyl-4-quinolyl b r m e t h y l ketone hydrobromide is obtained; it forms yellow crystals m. p. about 225" (decomp.). The free bromomethyl ketone forms light yellow crystals m p. 91". With dimethylamine it gives 2-phenyl-4- quinolyl4-damethylaminomethyl ketone which forms a hydrochloride light yellow crystals m.p. 208" (decomp.) and a hydrobromide m. p. about 206". Reduction of this compound gives p-dimethyl- amino- a-2-phenylquinolylethanol (annexed CH(oH)*CH2'NMe2 formula) a white plastic mass. The /\/\ dihydrochloride forms crystals m. p. 175" I I IPh (decomp .) . 2- Phen yl-4-quinolyl diethyl- \/\/ aminomethyl ketone is similarly prepared ; the hydrobromide forms yellow felted needles m. p. about 188" (decomp.). 2-Phenyl-4-quinolyl piper- idinomethyl ketone gives a hydrochloride m. p. 235" and a hydrobromide m. p. about 241" (decomp.). 2-Phenyl-4-quinolyl anilinomethyl ketone forms yellow crystals m. p. 123-125". These ketones may be reduced to the corresponding amino-alcohols.~-Diethylamino-cc-2-phenyl-4-quinolylethanol gives a dihydrochloride which forms crystals m. p. about 185" with darkening. p-Piperidino- ~-2-phenyl-4-quinolylethanol gives a dihydrochloride m. p. about 199" (decomp. ) . (3-Anilino- a-2-phen ylqzcinolylethnnol C,NH,Ph*CH (OH)*CH,*NHPh has m. p. 146". N G . W. R. Preparation of a Primary Amino-alcohol of the Quinoline Series. SOCIETY FOR CHEMICAL INDUSTRY IN BASLE (Swiss Pat. 92301 ; from Chem. Zentr. 1922 iv 951 ; cf. preceding abstract).- 2-Phenyl-4-quinolyl methyl ketone is changed into its oximino- derivative and this is submitted to reduction. 2-Phenyl-4-quinolyl oximinomethyl ketone forms yellow crystals m. p. 182" (decomp.). By reduction p -amino- a -2 - phen yl- 4- quinol yl ethan 01 CgNH ,Ph*CH (OH) *CH,*NH is obtained. It forms a dihydrochloride colourless crystals m.p. about 190" after softening a t about 145". OTTO DIMROTH and FRITZ FRISTER (Bey. 1922 55 [B] 3693-3697).-The formation of 1 1'-diacetyldihydro-4 4'-dipyridyl by the action of acetic anhydride on the dark violet solution obtained by the reduction of 4 4'-dipyridyl with nascent hydrogen has been described previously (Dimroth and Frister A. 1922 i 678). It is now shown that the violet solution contains a semiquinonoid compound G. W. R. Reduction of 4 4'-Dipyridyl.i. 150 ABSTRACTS OF CHEMICAL PAPERS. of molar proportions of 4 4'-dipyridyl and 1 l'-dihydro-4 4'- dipyridyl which has been isolated as the dihydrochloride for which the name dipyridyl-violet chloride is proposed. The dye differs from other semiquinonoid substances in that the reduced half is quinonoid and the oxidised half has the aromatic constitution.4 4'-Dipyridyl is obtained conveniently and in 90-95% yield by the action of oxygen on a suspension of 1 1'-diacetyltetra- hydro-4 4'-dipyridyl in glacial acetic acid and subsequent addition of sodium hydroxide to the solution. Titration of an aqueous solution of the substance with chromous chloride solution in the absence of air shows that the dye is produced by the action of one equivalent of hydrogen on a molecule of dipyridyl and that further reduction is not caused by chromous chloride. The dye is isolated as a dark green crystalline powder C,,H,,N,Cl? by the action of chromous chloride on a solution of dipyndyl in the presence of calcium chloride ; the precipitated product is washed successively with water and alcohol in the absence of air towards which it is unusually sensitive.H. W. Syntheses in the Indole Group. 11. The Influence of the Solvent on the Grignard Reaction. RIKO MAJIMA and MUNIO KOTAKE (Ber. 1922 55 [B] 3865-3872).-The unexpected differences observed in the production of indole-3-aldehyde from magnesium indolyl iodide and formic ester according as the reaction is effected in the presence of aliphatic or arylaliphatic ethers (Majima and Kotake this vol. i 156) has led the authors to examine possible further cases of this kind. Better yields of the products of the action of magnesium indolyl iodide on carbon dioxide acetone or benzaldehyde are obtained in the presence of anisole than in that of ethyl ether whereas the reverse is the case when ethyl chloroformate acetyl chloride or chloroacetyl chloride is used.Indole-3-carboxylic acid m. p. 218-ZZO" is obtained by the action of carbon dioxide on magnesium indolyl iodide in anisole or ethyl ether. The reaction has been examined previously by Qddo (A. 1911 i 486) who has thus isolated indole-1-carboxylic acid m. p. 108"; the cause of the discrepancy has not been elucid- ated. Magnesium indolyl iodide and acetone give di-3-indolyl- dimethylmethane m. p. 163-165" (cf. Scholtz A. 1913 i 520). Di-3-indoZyZphenyZnaethune CHPh CH<g&>NH) m. p. 149- 152" (+0*5C&6) rn. p. 120-121" is prepared from magnesium indolyl iodide and benzaldehyde; i t is oxidised by ferric chloride in alcoholic solution to the red dye NH<!$%>CH CPh :C<(&>N<g m.p. about 245-248". Ethyl chloroformate and magnesium indolyl iodide give ethyl indole-3-carboxylate colourless crystals m. p. 118-119"; Oddo's observation (Zoc. cit.) that ethyl indole- 2-carboxylate m. p. 107" is produced could not be confirmed. Acetyl chloride and magnesium indolyl iodide yield 3-indolyl ( 2ORGANIC CHEMISTRY. i. 151 methyl ketone NH<z&>C*COMe m. p. 188-189" (oxime m. p. 143-146") identical with the products described by Odd0 (loc. cit.). 3-lndolyl chlorornethyl ketone forms small rhombic crystals m. p. 212-214". H. W. K. VON AUWERS and H. BROCHE (Ber. 1922 55 [B] 3880-3911).-An examination has been made of the Dossibilitv of the existence of IsoYmeric Relationships in the Pyrazole Series.isomerides of the types RN HcH'(?H and iN<x?l$ in the 'N=CH pyrazole series analogous to those observed with the closely related indazoles C,H4<NR>N and C,H4<h->NR. The alkyl and phenyl derivatives of 3-methyl- and 3 5-dimethyl-pyrazole have been investigated but evidence of existence of isomerides of the expected type has not been obtained. The observations which have been made throw doubt on the identity of 3- and 5- derivatives of pyrazole as assumed by Knorr and other investigators. 3-Methylpyrazole is converted by an equivalent amount of methyl iodide in ethereal solution at 120" into 1 3-dimethyl- pyrazole (picrate m. p. 1726'; cf. Jowett and Potter T. 1903 83 467) a portion of the base remaining however unchanged.The same substances are obtained by boiling 3-methylpyrazole with methyl iodide and sodium methoxide in methyl alcoholic solution. 1 3-Dimethylpyrazole b. p. 141-142' is readily isolated in the homogeneous condition by the dry distillation of its meth- iodide m. p. 255-256". In a similar manner the ethylation of 3-methylpyrazole with ethyl bromide at 100" or in the presence of alkali gives 3-methyl- 1 -ethyZpyraxole a colourless liquid b. p. 152"/atmospheric pressure (picrate slender yellow needles m. p. 141"); the formation of quaternary salts in this case is much less pronounced than when methyl iodide is used. The products of the benzylation of 3-methylpyrazole are readily separated by taking advantage of the insolubility of the benzylated compounds in water in which the original material and the pyrazolium salt are soluble; experiments in the presence or absence of alkali yielded l-benzyl-3-rnethylpyrazole a colourless liquid with an odour of hyacinths b.p. 140-141"/14 mm. (picrate slender yellow needles m. p. 112.5-113.5"). In contrast to the indazoles 3-methylpyrazole when alkylated by different methods gives only one definite substitution product. The proof that this is a 1 3-(and not a 1 &)derivative is deduced in the following manner. The benzoic ester of hydroxymethyleneacetone CH,*CO*CH:CH*OBz (colourless crystals m p. 91-92' prepared by the action of benzoyl chloride on a suspension of the sodium compound of hydroxy- methyleneacetone in pyridine and anhydrous ether) is condensed with methylhydrazine and benzylhydrazine whereby dimethyl- and benzylmethyl-pyrazoles are obtained identical with the products described above ; unfortunately it was not found possible to isolate the intermediately formed benzoates but the mode of formation CH CH 9*i.152 ABSTRACTS OF CHEMICAL PAPERS. leaves practically no doubt that 1 3-derivatives are formed. Under similar conditions phenylhydrazine gives 1 -phenyl-3-methyl- pyrazole m. p. 37". The condensation of hydroxymethyleneacetone with methylhydrazine leads to the production of 1 3-dimethyl- pyrazole without any isomeric substance and differs therefore from the action with phenylhydrazine which yields a mixture of 1 3- and 1 5-compounds. The observations have been extended to 3 5-dimethylpyrazole I A " CMe'CH NH<N-:bMe the symmetrical structure of which excludes - the formation of structurally isomeric 3- and 5-derivatives ; any case of isomerism must therefore be due to difference in a ring structure since the production of 4-derivatives is excluded.It is converted by methyl iodide in the presence of alkali into 1 3 5- trimethylpyrazole the picrate of which has m. p. 144-145" whereas Knorr gives 131-133". Similarly 3 5-dimethylpyrazole and benzyl chloride a t 170" give exclusively l-benxyl-3 5-dimethyl- pyrazole a colourless liquid b. p. 144-146/10 mm. (picrate lustrous greenish-yellow needles m. p. 126.5-127-5"). The latter compound is also produced by the action of acetylacetone on benzylhydrazine although in this case the primary production of a three-membered ring might be expected.Finally l-phenyl- 3 5-dimethylpyrazole bas been prepared by the action of phenyl- hydrazine on acetylacetone and methyl a-bromopropenyl ketone CHMeXBrCOMe respectively ; the compound obtained in either case gives a picrate m. p. lolo and there is no evidence of the production of an isomeric compound. fiom the foregoing it appears therefore that cases of isomerism due to the different structure of the ring are not found in the cases of alkylated and arylated pyrazoles. It is however still more remarkable that the number of position isomeric alkylpyrazoles is smaller than would be expected from theoretical considerations since 1 3-derivatives are frequently formed exclusively when mixtures of them with 1 5-compounds or when the latter alone would be normally expected.In the hope that the reactions in the pyrazole series would resemble those in the indazole group an examination of the effect of heat on pyrazolium salts has been made. 1-Phenyl-3-methylpyrazole methiodide m. p. 144" and l-phenyl-5-methylpyrazole methiodide m. p. 282" lose methyl iodide when subjected to dry distillation and re-form 1 -phenyl- 3-methylpyrazole and l-phenyl-5-methylpyrazole respectively. Contrary to expectation however pure 1 3-dimethylpyrazole and 3-methyl-l-ethylpyrazole are obtained from l-benxyl-3-methyl- pyrazole methiodide m. p. 153-154" and l-benxyl-3-methylpyraxole ethiodide colourless prisms m. p. 149.5-150*5". The formation of 1 3- instead of 1 5-compounds discloses a surprising atomic .A v CH-CMe migration in accordance with the scheme zHeN(C,H,) CH-CMe >N.The irn-ORGANIC CHEMISTRY. i. 153 portance of the observation has caused special attention to be directed to the constitution of the pyrazolium salts. The possibility that contrary to rule the addition of alkyl iodide occurs a t the nitrogen atom which is already attached to an allryl group is excluded by the observation that 1-benzyl-3-methylpyrazole methiodide differs from 1 3-dimethylpyraxole benzyliodide colourless leaflets m. p. 167' (corresponding picrate flat yellow needles m. p. 126"). The further possibility that the difference in the two iodides is due to the different spatial arrangement of the sub- stituents around the same nitrogen atom and not to structural differences is improbable and is more defmitely excluded by the apparent impossibility of effecting their mutual interconversion.The wandering of alkyl groups during the fission of certain pyrazol- ium salts must therefore be regarded as cstablished. Nevertheless. it does not follow that 1 5-dialkylpyrazoles are incapable of existence since the high temperature involved in the fission can greatly facilitate the transformation of the primary products of the change into the 1 3-isomerides. The synthesis of 1 5-dialkylpyrazoles has therefore been attempted under conditions which are less likely to involve trans- formation. The action of phenylhydrazine on a-bromocroton- aldehyde leads to the formation of 1 -phenyl-5-methylpyrazone (cf. Vignier A. 1913 i 444) but under precisely similar conditions methylhydrazine gives 1 5-dimethylpyrazole in good yield.A possible explanation of the difference in the reaction lies in the assumption that the alkylhydrazine unlike the aryl compound becomes primarily added a t the double bond thus (?Br:CHMe + NH,-NHMe-+ CHBr-CHMe>NH I + CHO CHO NHMe yHBr*CMe>N -f CH-CMe>N 1 1 CH,-NMe CHeNMe * This hypothesis is shown to be improbable in the following manner. P-Keto-n-butyl alcohol condenses with methylhydrazine sulphate in aqueous solution in the presence of acetic acid and sodium - ... acetate to yield 1 3-climethylpyrazoline (?H2'rH2>NMe a colour- CMe-N less liquid b. p. 127-129" (picrate yellow needles m. p. 131-5- 132.5") ( I-phenyl-3-methylpyrazoline m. p. 71-72' is prepared similarly from phenylhydrazine) .If crotonaldehyde which may be assumed to react in the same manner as its bromo-derivative is similarly condensed with methylhydrazine it must yield 1 3- dimethylpyrazoline if the hydrazine is added initially a t the double bond and 1 5-dimethylpyrazoline if the initial reaction occurs a t the aldehyde group. It is found that the compound which is produced is not identical with that derived from p-keto-n-butyl alcohol and hence must be regarded as 1 5-dimethyZpyraxoliney it is a colourless very hygroscopic liquid ?H,*CHMe CH-N > m e ; b. p. 124-125" which gives a picrate thin golden-yellow leaflets ff 2i. 154 ABSTRACTS OF CHEMICAL PAPERS. m. p. 113-114”. It must therefore be considered that the production of 1 3-dimethylpyrazole from methylhydrazine and bromocrotonaldyde instead of the expected 1 5-derivative is actually due to atomic wandering. Attempts to convert 1 3- and 1 5-dimethylpyrazolines into the corresponding pyrazoles by cautious oxidation with bromine led to remarkable results. The 1 3-compound is transformed smoothly into 1 3-dimethylpyrazole ; the 1 5-derivative is largely resinified but is partly converted into 1 3-dimethylpyrazole. It therefore appears that under usual conditions 1 5-dimethyl- pyrazole (and presumably other 1 5-dialkylpyrazoles) are incapable of existence whereas 1 -phenyl-3-methylpyrazole and 1 -phenyl- 5-methylpyrazole are stable substances. Knorr’s conception that the pyrazole molecule is symmetrically constituted owing to the oscillation of the double bonds and of ” the hydrogen atom united to the nitrogen RHoNH>N e CH-CH YH’”>NH can scarcely be extended to the alliyl derivatives CH.CH of pyrazole since a continuous oscillation of an alkyl group between two nitrogen atoms is contrary to all experience of tautomerism and the firmness of attachment of akyl groups.The authors do not regard the production of 1 3-dialkylpyrazoles in place of the expected 1 5-derivatives as due to the identity of the two series of compounds in consequence of ‘‘ mobile ” double bonds and oscillatory atomic displacements; i t is caused by the great instability of the 1 5-derivatives which immediately after formation become isomerised in accordance with the scheme H. W. The Polymorphism of Antipyrine Vanillin and the Ery- thritols.PAUL GAUBERT (Compt. rend. 1922 175 1414- 1417).-In general a substance which can be superfused or better can pass into the amorphous state will occur in several crystalline forms depending on the temperature. This appearance may be favoured by the presence of some impurity or sometimes by the more or less prolonged heating of the substance a t a temperature above its melting point. I n this manner it is shown that antipyrine melted on a glass plate may occur in three crystalline modifications vanillin may give four crystalline modifications and the erythritols two crystalline forms. The conditions for obtaining these modifi- cations and the forms obtained are detailed. W. G. Preparation of a Derivative of Barbituric Acid Soluble in Water.HERMANN STAUDINGER (Swiss Pat. 91561; from Chem. Zentr. 1922 ivy 840) .-Diallylbarbituric acid or ethylallylbarbituric acid is treated with ozone and the ozonide is changed into the corre- sponding carboxylic acid. Diallylbarbituric acid gives a crystalline ozonide which is decomposed by heating over a water-bath withORGANIC CHEMISTRY. i. 156 formation of a dicurboxylic acid (I) colourless crystals m. p. 280" (decomp.) . Ethylallylbarbituric acid gives similarly a monocurb- oxylic mid (11) which forms colourless crystals m. p. 280-281" (I.) 7 O-NH-$!O The compounds have therapeutic uses and serve as intermediates (11.) F O-NH-YO NH-CO-C( CH,*CO,H) NH-CO-CEt*CH,*C02H' for the preparation of other barbituric acid derivatives. G. W. R. Preparation of a Compound of 5-isoPropy1-5-allylbarbi- turic Acid.F. HOFFMANN LA ROCHE & Co. (Brit. Pat. 188251). - 5-isoPropyl-5-allylbarbituric acid when fused in ap- proximately molecular proportions with 4-dimethylamino- 1-phenyl-2 3-dimethyl-5-pyrazolone form as double compound which is appa,rently homogeneous and melts sharply a t 92-93'. It is yellow in colour and dissolves in hydrocarbons to a deep yellow solution. Water and other solvents containing hydroxyl give only slightly yellow solutions indicating that in these solvents the compound is resolved into its colourless components to a con- siderable extent. The compound is valuable therapeutically as it combines the soporific action of the barbituric acid with the analgesic action of the pyrazolone and can therefore partly replace the opiates.G. F. M. Complex Metallic Compounds of Indigotin. I. K. KUNZ (Ber. 1922 55 [B] 3688-3691).-During the course of experi- ments on the production of 1 1-diphenylindigotin (Friedlander and Kunz A. 1922 i 765) it was observed that boiling solutions of indigotin in o-nitrochlorobenzene are rapidly decolorised by cuprous chloride and sodium acetate or copper powder. A similar change is now shown t o occur in other solvents of high boiling point. It leads to the formation of the compound C3,H1804N,~ which is rapidly decomposed by mineral acids and gives a normal indigo vat with hyposulphite. In analogy with Willstatter's formulation of chlorophyll the C,H~<-N>C:C<NH> C ,H annexed constitution is suggested. The action of zinc foil on a solution C6H4<&>C:c<co->C,H4 of indigotin in tar oil (b.p. 240- 250") methylnaphthalene or o- nitrochlorobenzene leads t o the formation of a zinc compound C3,Hl8O4N4zn black lustrous prisms which unlike the copper compound is rapidly hydrolysed by warm water with production of indigotin. A crystalline nickel derivative has also been prepared by the use of nickel sulphate and sodium acetate but it has not yet been found possible to prepare a magnesium compound. co CO- N> c u ':::.NH H. W. Benzbisthiazoles. 11. STEPHEN RATHBONE HOLDEN EDGE (T. 1923 123 153-156).i. 156 ABSTRACTS OF CHEMICAL PAPERS. Syntheses in the Indole Group. I. A New Synthesis of r-Tryptophan. RIKO MAJIMA and M ~ I O KOTAKE (Ber. 1922 55 [B] 3859-3865) .-Indole- 3 -aldehyde C,H,<$gz!>CH m.p. 193-195" is conveniently prepared in 40% yield by the successive action of indole and formic ester on a solution of mag- nesium ethyl iodide in anisole which is cooled in a mixture of ice and salt; in ethereal solution only traces of the aldehyde are formed (cf. Alessandri and Florence A. 1915 i 452). Phenetole may replace anisole and ethoxymethyleneaniline may take the place of formic ester but the yields are thereby somewhat reduced; amyl ether is as unsuitable as ethyl ether. The aldehyde is heated with hydantoin anhydrous sodium acetate and acetic anhydride a t 106-108" and the product is treated with aqueous sodium hydroxide solution where by 3 - indol yl id eneh ydantoin; /C H CO-YH NH*CO' NH\-&$>C*CH :C< m.p. about 320" decomp. 325" is formed the yield being 46.6% of that theoretically possible. 1 -Acetylindole-3-aldehyde is obtained as a by-product of the action; it therefore appears probable that it is this derivative which actually undergoes con- densation and that acetyl-3-indolylidenehydantoin is hydrolysed by the treatment with sodium hydroxide; the inefficiency of acetic acid and sodium acetate as condensing agents thus receives an explanation. 3-Indolylidenehydantoin is reduced by sodium amalgam in alkaline solution to o-hydantylscutole NH<-@wcH,*cH<~.~ C H CO-NH * colourless needles m. p. 220-221". The latter is hydrolysed by aqueous barium hydroxide solution a t 108" to r-tryptophan NH<~&>C*CHz*CH(NHz)*COz€€ hexagonal leaflets m.p. 283- 285' after becoming discoloured at 250". As by-product a sub- stance Cl,H130,N3 leaflets in. p. 207" (decomp.) is isolated. H. W. Preparation of a Condensation Product from p-Iminazolyl- ethylamine [4-~-Aminoethylglyoxaline]. SOCIETY FOR CHEMI- CAL INDUSTRY IN BASLE (Swiss Pat. 92297; from Chem. Zentr. 1922 iv 890).-When formaldehyde or prefer- ably inethylal is added slowly to a suspension of 4- p-aminoethylglyoxaline in fuming hydrochloric EH acid at 100" and the mixture heated for several HzC C-N hours in a reflux apparatus p-imidaxolylisopiper- \/ idine [tetrahydro-l 3 6-benxtriazolel (annexed CH2 formula) is formed. The hydrochloride is crystal- line and has m. p. 249". Determination of the Constitution of Colouring Matters from their Absorption Spectra.IV. F. KEHRMANN and M SANDOZ (Helv. Chim. Acta 1922 5 895-905; cf. A. 1921 i 276). -In continuation of previous v70rk the absorption spectra of CH NH '"' Hr G. W. R.ORGANIC CHEMISTRY. i. 157 salts of &amino-derivatives of methyl- and phenyl-phenazine have been examined. The introduction of a second symmetrical amino-group into the monoaminomethylphenazine previously described (Zoc. cit.) lightens the colour of the mono-acid salt from cherry-red to orange-red. The absorption maxima are at X 530 and X 483. Introduction of a phenyl group into each amino-group deepens the shade to bluish-violet and there is now only one absorption maximum at X 572. The corresponding ditolyl deriv- ative has an absorption maximum a t X 580. The monoacid salts of the diaminophenylphenazies chosen for examination were the perchlorates on account of their great stability.The mono-acid salt of 2 7-diaminophenylphenazine (annexed formula) is blue and to it may be ascribed the same para-quinonoid structure as to aposafran- ine which is violet-red. The di-acid salt /\N\/\NH2 is red corresponding very closely with I I 1 I aposafranine from which it is inferred NHzxk\/\'\/ that the second amino-group has been neutralised but that no change in the disposition of the double bonds has occurred. When the dizimino- phenylphenazonium perchlorate is dissolved in concentrated sul- phuric acid the colour of the solution is green. The colour is due to the tri-acid salt (annexed formula) and is similar to that of the di-acid salt of aposafranine.Again there is no change of structure and it is in- "/\'\NHsX teresting to note the colour-deepening effect of the fixation of an equivalent of acid by a doubly-linked atom form- ing part of the chromogen. Finally in fuming sulphuric acid the colour is reddish-brown resembling the di-acid salt of phenylphenazine. This salt must have an ortho- quinonoid structure. The perchlorate of 1 3-diaminophenylphenazine is green in water yellowish-green in alcohol approaching a yellow of the second order. Although the amino-group in position 1 has a considerable colour-deepening effect the structure is still probably paraquinonoid. The di-acid salt of this compound resembles aposafranine as in the case of the 2 7- isomeride but the tri-acid salt obtained by direct solution in concentrated sulphuric /\@\/\ acid is not green but yellow resembling the 1 I I 1 mono-acid salt of phenylphenazine.The NH3X'\/&$>/ structure of this salt must therefore be orthoquinonoid (annexed formula). When the perchlorate is dissolved in fuming sulphuric acid the reddish- brown colour of the tetra-acid salt is obtained. The perchlorate of the third isomeride phenosafranine is red with an ofange fluorescence in solution. Here again as with methylphenazine the second symmetrical amino-group brightens the shade whilst the colour intensity is increased. The di-acid N NPh H X \/ N I I I I NH,XH\/\ /\/ NPh m3x I Ni. 158 ABSTRACTS OF CHEMICdL PAPERS. salt is blue indicating that the second acid equivalent is combined with chromogenic nitrogen (annexed formula).The tri-acid salt is green resembling as expected t,he di-acid salt of uposafranine. The tetra- XNH /\@\/\ acid salt is the normal reddish-brown. I I I INH It is concluded that all the observed m2xH\/\/\/ facts can be explained if it is granted that two states ortho- and para- quinonoid are possible in this series. It is further shown briefly that acetylation of an amino-group in this series has a similar effect on its auxochromic properties to neutralisation of the group with acid. GEORG BARKAN (2. BioZ. 1922 76 267-266).-The solubility of freshly pre- cipitated amorphous sodium urate (C,H,O,N,Na,H,O) a t 18" is 2.03 g. or 9.8 x On keeping or more particularly on stirring in contact with its solution the solubility decreases and approaches that of the crystalline form.NPh E. H. R. The Solubility of the Salts of Uric Acid. grain-molecules per litre. W. 0. K. Oxidation of Benzeneazoresorcinol. DINO BIGIAVI and GIULIO GIANNMI (Atti R. Accad. Lincei 1922 [v] 31 ii 109- 116).-Of the two possible diacetyl derivatives (a) C6H,'NO:N*C6H3( OAC) and (p) C,H,*N:NO*CGH3(OAc) obtainable by the oxidation of the diacetyl compound of benzeneazoresorcinol by means of hydrogen peroxide only the latter is actually obtained the a-form probably undergoing further oxidation. OH - p-4-Benxeneaxoxyresorcinol C,H,*N:NO-( >OH crystallkes \-/ in red granules m. p. 144" giving a chestnut-yellow powder. The dibenzoyl derivative forms pale yellow needles m. p. 113" and the diacetyl derivative long pale yellow silky needles in.p. 102". 4-Benzeneuxoxy-2 6-dibrornoresorcinoZ C,H5*N:NO*C6HBr,( OH) crystallises in transparent greenish-yellow cubes m. p. 153" and yields aniline when reduced by means of zinc and acetic acid. 4-op- Di bromo benzeneaxox y-2 6-di brornoresorcinol forms long red needles m. p. 229" (decornp.) and gives 2 4-di- bromoaniline when reduced by means of tin and hydrochloric acid. 2 6-Dibenxeneuxo-4- benxeneaxoxyresorcci?zol C ,H 5*N20*C ,H ( OH) 2( *N,Ph ) is a red compound m. p. 220". When treated with alcoholic potassium hydroxide the diacetyl derivatixe of p- benzeneazoxyresorcinol yields a dark red compound m. p. about 190" which has not yet been characterised. C12HG03N2Br4 T H. P.ORGANIC CHEMISTRY. i. 159 Oxidation of Benzeneazonaphthols. I. DINO BIGIAVI and RENATO CERCHIAI (Atti R.A&. Lincei 1922 [v] 31 ii 27- 30).-Owing to the divergent behaviour towards alkali shown by the three isomeric benzeneazonaphthols a phenylhydrazonic constitution is attributed to 1 - benzeneazo-2-naphthol whilst 4-benzeneazo-1-naphthol and 2-benzeneazo-1-naphthol are con- sidered 'to possess the ordinary hydroxylic formuls (Goldschmidt and Brubacher A. 1891 1209; Willstatter and Parnas A. 1907 i 425 1056). Marked differences are found also in the behaviour of these compounds towards 30 yo hydrogen peroxide solution and acetic acid. Thus under these conditions 1 -benzeneazo-2-naphthol and also its acetyl-derivative undergo slow oxidation in the cold to carboxy- cinnamic acid CO,H*C,H,*CH:CH*CO,H m. p. 200" whilst in the hot they yield the isomeric dihydroisocoumarincarbosylic acid C,J34<CH,.CH,Co2H m.p. 153" which is also obtained when carboxycinnamic acid is fused for a few moments. Similar oxidation of P-naphthol yields the products obtained by Ehrlich by the action of permanganate (A. 1889 1001) namely (1) Carboxycinnainic acid for which however this author gave m. p. 183" instead of 200"; (2) an acid C,H,,O m. p. 281" which although regarded by Ehrlich as dibasic gives only mono- silver and monoethyl derivatives and has probably the structure f)co2H ?o'o-I ) 1 This acid reduces permanganate but \/-CHIC--\/\/ not so rapidly as does carboxycinnamic acid. Reduction of Ethyl Diazobenzylidenegluconate. P. A. LEVENE ( J . Biol. Chern. 1922 54 809-813).-When reduced with aluminium amalgam in isopropyl-alcoholic solution ethyl diazo- benzylidenegluconate yields a mixture of ethyl benzylidenedeoxy- gluconate and ethyl benzylidene chitosamate.The former product is identical with that previously obtained (A. 1922 i 1028) by reduction of ethyl benzylidene-ccp-anhydromannonate ; the latter was not isolated but was identified by hydrolysis into chitosamic acid. The reduction apparently takes place asymmetrically for no trace of an epimeride of chitosamic acid can be detected. That the deoxy-compound is not produced through the intermediate formation of an unsaturated compound is shown by the fact that the amide of benzylidene- ap-anhydromannonic acid is unchanged when submitted to similar treatment. co-p / /\ T. H. P. E. S. The Formation of Aromatic Thiocyanates by the Diazo- reaction. A.KOECZYNSKI [with J. KNIATOWNA and F. KANIN- SKI] (Bull. Xoc. chim. 1922 [iv] 31 1179-1185).-1n the trans- formation of o-nitrodiazobenzene thiocyanate into o-nitrophenyl thiocyanate under the catalytic influence of a metallic thiocyanate the yield obtained depends on the metal of which a salt is used and on the temperature being greater in the majority of casesi. 160 ABSTRACTS OF CHEMICAL PAPERS. studied at 15-20" than a t 60-70". The most effective catalysts of the substances used are the thiocyanates of iron and tungsten. The Tryptophan Content of some Proteins. CLARENCE E. MAY and EMBREE R. ROSE ( J . Biol. Chem. 1922 54. 313- 216) .-The colour which tryptophan yields with Ehrlich's reagent has been utilised for the estimation.The protein (0.05 to 1 g.) is added to a mixture of concentrated hydrochloric acid (50 c.c.) water (50 c.c.) and a 5% solution of p-dimethylaminobenzaldehyde in 10% sulphuric acid (1 c.c.). It is then incubated a t 35" for twenty-four hours left a t the ordinary temperature for at least twenty-four hours and the colour then compared with a casein standard. It is assumed that casein yields 1.5% of tryptophan. Application of the method to a number of proteins yielded the following values lactalbumin 2.4 gliadin 1.05 glutenin 1.80 edestin 1.5 glycinin 1.65 ovovitellin 1.74 egg-albumin 1.11 phaseolin 0.80 maize gluten 1.08 legumin (vetch) 1*05y0. No colour developed in the case of zein and gelatin. The Isoelectric Point of Globin. SEIUNGO OSATO (Biochem. Z. 1922 132 485-487).-The isoelectric point of globin from blood-corpuscles is found by the precipitation method to be at P H 8.1.W. 0. K. I. The Preparation of Nucleic Acid from the Nucleoprotein of Tubercle Bacilli (Tuberculinic Acid). 11. The Pyrim- idines contained in Tuberculinic Acid the Nucleic Acid of Tubercle Bacilli. TREAT B. J o i m s o ~ and ELMER B. BROWN ( J . Bio2. Chem. 1922 54 721-730 731-737).-I. A method is described for the preparation of tuberculinic acid from tubercle bacilli (cf. also Ruppel A. 1899 ii 237; Levene A. 1901 i 299). The protein which remains after removal of this nucleic acid has been analysed for nitrogen distribution with the following results amide-N 11.83 humin-N 4.1 1 cystine-N 1-26 arginine-N 10.63 histidine-N 11.48 lysine-N 3.69 monoamino-N 47.39 non-amino-N 9.34%. Tryptophan was also present. 11. Cytosine and thymine have been isolated from the products of hydrolysis of tuberculinic acid; the presence of uracil could not be detected. E. S. The Results and Aims of General Enzyme Chemistry. H. VON EULER (Ber. 1922 55 [B] 3583-3600).-A lecture delivered at the centenary of the Versammlung Deutscher Natur- forscher und Aerzte. K. w. The Isolation of Enzymes. RICI~ARD WILLST~TTER (Bey. 1922 55 [B] 3601-3623).-A lecture delivered a t the centenary of the Versammlung Deutscher Naturforscher und Aerzte. H. J. E. E. S. H. W. Saccharase. E. CANALS (BUZZ. SOC. chim. 1922 [iv] 31 1333-1341; cf. A. 1922 i 1075).-The author traces a relation- ship between the magnesium and phosphate content of saccharases of various origins and their respective diastatic powers whichORGANIC CHEMISTRY. i. 161 seems to some extent to support the opinion of Salkowski that invertase is the magnesium salt of a nitrogenous phosphated acid. A direct relationship cannot be traced however between the total magnesium and phosphoric acid content and the diastatic power since if the preparation is purified by dialysis the diastatic power is enhanced whilst there is a loss of about 40% of Mg and 58% of P,O and all the magnesium and phosphoric acid of the saccharase is therefore not essential for the hydrolysing activity Purification by precipitation with alcohol results in a similar loss in magnesium and an even greater loss in phosphoric acid and in this case the diastatic power is greatly diminished owing probably to a profound modification of the colloidal system of the enzyme by this treatment. Alcoholic precipitation is therefore not suitable for the purification of saccharase. It is observed that the magnes- ium of the saccharase dialyses less easily and is less readily separated by alcoholic precipitation than the phosphate and it would there- fore seem to enter to a greater extent into the constitution of the inicelle. G. F. M. The Wuence of Neutral Alkali-metal Salts on Diastatic Ferment. V. AMANDUS HAHN and HUGO MEYER (2. BioZ. 1922 76 227-246) .-Purified diastase from commercial pan- creatin ‘‘ Rhenania ” is influenced as regards its activity by the presence of neutral salts of the alkali metals similarly to saliva diastase and to unpurified pancreatin. The optimum PH depends on the buffer solution used being 7.2 with a phosphate mixture and 5-6 with an acetate mixture resembling saliva which has corresponding optimum at PH 6.6 and PH 5.6. The influence of neutral salts is much more marked with acetate buffer mixtures than with phosphate buffer mixtures. In both cases the effect is to increase the optimum Pll. Arseno-metallic Compounds. 11. Isomeric Silver Sal- varsans. A. BINZ and W. H. LUDWIG (Ber. 1922 55 [B] 3826--3831).-The silver complex of silver salvarsan has been considered by Ehrlich and Karrer (A. 1916 i 95) to be co-ordin- atively attached to the arsenic atom and by Binz Bauer and Hallstein (A. 1920 i 401) as united by the residual affinity of the nitrogen atom. The isolation of two isomeric silver salvarsans apparently justifies both formula At present the allotment of the structure to the isomeride is somewhat arbitrary and is due to the fact that one form is lighter in colour than the other and resembles t o this extent the complex compounds containing silver and nitrogen but not arsenic. Silver salvarsan I NH,*C,H3( OH)*&:As*C,H,( OH)*NH,. . . AgOH is obtained by adding sodium carbonate solution to salvarsan and silver nitrate dissolved in water ; it is insoluble in sodium carbonate and stable towards reduction with hypophosphorous acid. It dissolves in sodium hydroxide solution with the dark brown colour of technical silver salvarsan and from the solution carbon dioxide precipitates the isomeride NH,*C,H,(OH)*As:As( ... AgOH)*CGH3(OH)*NH W. 0. K.i. 162 ABSTRACTS OF CHEIIICAL PAPERS. a dark brown substance which dissolves in sodium carbonate solution and is reduced by hypophosphorous acid. Silver salvarsan I is rapidly converted by hydrochloric acid into the cmpound NH,*C,H,( OH)*As:As*C,H,( QH)*NH,. . . AgC1 a pale yellow powder. The dark brown isomende II appears to be unaffected by similar treatment after several hours but gradually becomes converted into the chloride just described the formation of which from the chloride NH,*C,H,( QH)*As:As(. . .AgC1)*C,HB(OK)*NH takes place very slowly in comparison with the change in the reverse direction in alkaline solution. The same change occurs when solutions of technical silver salvarsan are treated with hydrochloric acid the brown colour in this case persisting for a considerable time and gradually giving place to a pale yellow precipitate. H. W.