129 Organic Chemistry. Preparation O f Isoprene. BADISCHE ANILIN- 8 SODA-FABRIK (D. R.-P. 268100 268101).-Isoprene is obtained by applying the process described in the chief patent (255519 A 1913 i 438) which consists in heating the substances there mentioned to a high tempera- ture under diminished pressure in the presence of catalysts to methyl- butenyl acetate (1st patent) or methylbutenyl ether (2nd patent). J. C. C. Preparation of Isoprene. BADISCHE ANILIN- & SODA-FABRIK (D.R.-P. 268102).-Instend of as-dimethylallene as described in the chief patent (A. 1913 i l) isopropylacetylene is dropped on to strongly heated aluminium oxide under a presgure of 40-50 mm. J. C. C. Preparation of Erythrene. FARBENFABRIKEN VORM. FBIEDR. BAYER & Co. (D.R.P. 262553 2628S4).-Erythrene can be obtained by submitting cyclopentane or 1-methylcyclopentane-3-01 or other homologues and derivatives of cyclopentane to pyrogenic decomposition. The second patent states that cycZohexano1 can also be used.J. C. C Preparation of E r y t h r e n e a n d Other Unsaturated Gom- pounds ALBERT GERLACR and RUDOLF KOETSCHAU (D.R.-P. 267079 267080).-The pyrogenic decomposition of oleic acid or oils fats and waxes derived from the acids of the oleic series is advantageously carried out by means of an electrically heated metallic spiral placed inside the reaction vessel the latter being externally heated to the boiling point of the contents. The second patent mentions that such substances as wool fat cholesterol etc. may be used in this process for obtaining erythrene and the other unsaturated hydrocarbons which are also produced.J. C. 0. Catalytic Decomposition of Alkylidenehydrazines. Investi- gation of the Hydrocarbons from Gitral and Citronellal. Compare A 1913 i 1161).-The hydrocarbon b. p. 164-165'/ 755 mm. obtained from citralhydrazone (A. 1911 i 1027) has been prepared in larger quantity from unpurified citral and subjected t o further investigation. It is found to be a mixture of two isomerides the one present in larger quantity (119 grams) being (1) the limonene modification CH2:CMe*CH2*CH2*CH,-CMe:CHMe b. p. 162-163'/ 760 mm. DiO 0.7677 n 1.4422. When this hydrocarbon is converted by means of fuming hydrobromic acid into the dibromo-derivative C,,H,,Br and the latter either distilled with aniline or boiled in a reflux apparatus with alcoholic potassium hydroxide the linaloolene obtained by Semmler by the action of sodium on linalool ('g Die N.KISHNER (J. RU58. PhyS. Chem. SOC. 1913 45 1779-1787. VOL. CVI. i. 1;i. 130 ABSTRACTS OF CHEMICAL PAPERS. Aetheriachen Ole," I 527) is formed ; when alkali is used for decomposing the dibromo-compound the unsaturated alcohol C,,H,,*OH b. p. 190-195O is also obtained. Reduction of the limonene modification by Sabatier and Senderens' method yields inactive /3[-dimethyl- octane CHMe,*CH,*CH,*CH,*CHMeEt b. p. 158-159*5"/776 mm. (157*4-158'/748 mm.) DF 0.7287 (0.7281) n 1.4100 (1.4097). Similar reduction of the linaloolene obtained above does not yield this same dimethyloctane but an isomeride b. p. 161-162'/772 mm. Dfo 0-7399 m 1.4146.CMe, (YH*CH,*CH,*CMe CHMe obtained in smaller quantity (25 grams) has b. p. 163-165'/760 mm. DF 0.7699 n 1.4439 and yields acetone and other products on oxidation with permanganate. An optically active /I[-dimethyloctane was previously obtained (A 1911 i 1087) by reducing with hydriodic acid in a sealed tube the hydrocarbon C,,H, formed when citronellaldehydehydrazone is distilled with fused potassium hydroxide. As this method of reduction of ten leads to partial racemisation Sabatier and Senderens' method was employed the resulting hydrocarbon having the proper- ties b. p. 159.5'/753 mm. Di0 0.7308 n 1.4106 [a] +6*43' which are changed after purification with fuming nitric acid t o b. p. 160-161'17'72 mm Di0 0.7301 nD 1.4109 [a]D + 6-21'. Combination of the hydrocarbon C10H2 with hydrogen bromide and distillation of the bromo-derivative C10H27Br with quinoline yields a hydrocarbon CI0H2 which has slightly different physical constants and on reduc- tion by Sabatier and Senderens' method gives the hydrocarbon C10H2 b.p. 159*5-160'/760 mm. DF 0.7293 nD 1.4115 [a] +6*97'. For the dihydromyrcene obtained by Semmler (A. 1901 i 732) the author suggests the formula CH,:C~e*CH,*CH,*CH,*CEt:CH,. (2) The terpinolene modification T. H. P. Simultaneous Oxidation of Saturated and Unsaturated Hydrocarbons by means of Potassium Permanganate. N. KISHNER (J. Russ. Phys. Chem. Xoc. 1913 45 1788-1792).-The results of the author's previous experiments on the purification of saturated hydrocarbons by treatment with potassium permangaaate raise the question Does admixture of an unsaturated hydrocarbon influence the stability of a saturated hydrocarbon as regards its oxidisability? In other words is the oxidation of a mixture of a saturated and an unsaturated hydrocarbon to be regarded as a coupled or induced reaction ? Various experiments have been made with P[-dimethyloctane the unsaturated hydrocarbon C10H22 b.p. 161-163'/772 mm. (preceding abstract) the hydrocarbon C,,H (Zoc. cit.) and d-P[-dimethyloctane (Zoc. cit.) in order to ascertain whether they are oxidised more readily in presence of menthene than alone. The results show that the unsaturated hydrocarbon does not occasion considerable increase in the rapidity with which the saturated hydrocarbon oxidises. This reaction shows complete analogy to the coupled or induced oxidations investigated by Schilov (A.1903 ii 276). The oxidation of different saturated hydrocarbons is influenced inORGANIC CHEMISTRY. i. 131 the above manner t o different extents. PS-Dimethyloctane contains two tertiary hydrogen atoms and these undergo oxidation far more readily than those of other hydrocarbon radicles such as the methyl or methylene group ; this circumstance explains the results obtained by Nametkin (A. 1913 i 1285) with mixtures of propane or cyclopropane with an unsaturated hydrocarbon and it is probsble that a mixture of isobutane with propylene would exhibit different behaviour. When the products of oxidation of a mixture of P[-dimethyloctane with menthene are distilled a distinct odour of tertiary alcohols is per- ceptible indicating that the initial stage of the oxidation of the dimethyloctane consists in hydroxylation of the tertiary hydrogen.T. H. P. Influence of Constitution on the Rotatwy Power of Optically Active Substances. VII. Optically Active Hydro- carbons from Citronellel. H. RUPE and ALPEO~US JBGE~ (AnnaZun 1914 402 149-186. Compare A 1913 i 884).-Continuing the investigation of the influence of certain radicles on the rotatory power of optically active substances the authors have selected ci tronellaldehyde as a very suitable material for their purpose because it coctsins only one centre of optical activity and by the introduction of alkyl and aryl groups by the Grignard reaction is readily converted into secondary alcohols from which unsaturated hydrocarbons are obtained by the loss of water.Semmler's hypothesis that citronella1 reacts sometimes according to the " limonene " formula CH,:CMe*[CH2],*CHM~*CH0 at other times according to the '( terpinolene " formula CMe,:CH*[CH,],*CHMe*CH,*CHO is even if true without weight in so far as the authors' purpose is concerned because the double linking in both formulz is so widely removed from the centre of optical activity as to be without appre- ciable influence on the rotatory power. Quite different is the case of the double linking produced in the unsaturated hydrocarbons when these are obtained from the secondary alcohols by the (indirect) elimination of water. This double linking can occupy different positions (in suitable substances) and since i t is within the sphere of influence of the asymmetric carbon the determination of its exact position is a matter of prime importance. /3[-Dirnethyl-A"-nonen-B-ol (Rupe and Splittgerber A 1907 i 7 1 l ) P[-dlmethyl-Af-decen-O-ol (Austerwell and Cochin A 1910 i 572) pi- dirnethy Z-AF-undecem#-ol CMe, CH*CH,*CH,*CHMe*CH,*CHPra*OH b.p. 124'19 mm. and 1u-ctimethyl-AaL-undecadien-6-01 CISle,:CH*CH,*CH,*CHMe-CH2*CH(OH)~CH2-CH :CH b. p. 123-124O/lO mm. are obtained in the usual manner from citronella1 and magnesium methyl ethyl propyl and ally1 bromides respectively. These carbinols are converted by phosphorus tri bromide at 100-105° into tho corresponding bromides from which hydrogau bromide is then eliminated by boiling with pyridine or better alcoholic sodium ethoxide.pt- Rimethgl- AFV-nonndiene CMe,:CH*CHI,*CH2*CHRlu*CH:C KRle k 2i. 132 ABSTRACTS OF CHEMICAL PAPERS. bas b. p. 61-62'/9 mm. Di0 0.7730 [a]:' -8*12' [a] - 10*37' [a] - 12*29' and [a] - 16.17'. By treatment in glacial acetic acid with 6-8% unpurified ozone and subsequent decomposition by water on the water- bath the dimethylnonadiene yields 6-acetyl-a-methyt- oaleric acid CH,Ac*CH,*CH,*CHMe*CO,H (isolated as the semi- carbazone C,HI7O,N m. p. 129-1 30° colourless needles) and a-methylglutaric acid (and possibly also a little P-methyladipic acid). The formation of these two acids not only proves that the new double linking in the dimethyloonadiene is as near as possible to the asymmetric carbon atom but also is evidence in favour of Semmler's hypothesis that citronellaldehyde reacts in two forms (v.supra). P1-Dimetnyl-APe-decadiene CMe,:CH*C H,gCH,*CHMe*CH,*CH:CHl\le has b. p. 81-82'/9 mm. Dt0 0.7813 and [a] - 6.64'. The position of the new double linking is proved by the decomposition of the ozonide by hot water whereby c-cccetyl-p-methylhexoic acid (isolated as the semicarbazone C,,H,,O,N m. p. 135-1 36') and P-methyladipic acid are obtained (again evidence of Semmler's hypothesis). &- Dimethyl-APe-undecadiene CMe,:CH*CH,*CH;CHMe* CH,* CH:C KEt has b. p. 9@-91'/8 mm. Dao 0.7873 [a]:' -5*22' [a] -6-6So [a] - 7-98' [a]? - 10*55' and n 1,44614 n 1.44903 np 1,45602 ny 1.46215. The position of the new double linking is determined by the value of [a] which is practically identical with that of the dimethyldecadiene. The abnormally high value of [a] of the dimethyl- nonadiene must be due to the presence of the double linking in the immediate neighbourhood of the asymmetric carbon atom.The molecular refractions of the three hydrocarbons nre very nearly normal. &-D~rnethyl-AaYL whdecatrians C Me, CH * CH CH,. C HMe* CH,* C H CH C H CH bas b. p. 94-95'18 mm. D?," 0.8005 [a]Zo - 7-87" - 10-12' - 12.34' - 16.92' for the C D E and F lines and 2' 1.46887,1*47261 1.48216 1.49040 for the a D p and y lines. The presence of the conjugated double linking is indicated by the pronounced exaltations of the specific arid the molecular refractions. The molecular rotation is considerably greater t ban that of the dimethylundecadiene ; also it is noteworthy that a comparison of the specific rotatory powers of the dimethyl- nonadiene and the dimethylundecatriene shows that the double linking in the immediate neighbourhood of the asymmetric carbon atom has the same effect 011 the rotatory power as the conjugated double linking at a greater distance The interaction of citronellaldehyde and a n excess of magnesium cyclohexyl bromide in ether leads t o the forma tion of O-cyclohexyl-P[-di- methyl-AP-octen-8-ol CMe,:CH*CH2*CH2*CHMe*CH2*CH(OH)~C6Hl b.p. 166-167'/10 mm. ; by-products are isopulegol and probably cyclohexane and decahydrodiphenyl (isopulegol is a1 so obtained as a by-product in the preparation of ~:-ddmrnthy~-AaL-undecadien-6-o1 unless a large excess of ally1 bromide is employed). The brornido obtained from the carbinol is converted by boiling pyridine into yy-dimethyl- Aac-octadienyZcyclohexane CMe,:CH*CH,*CH,*CHMe* CH C H*C,H 11ORGANIC CHEMISTRY.i. 133 2>CH,) b e p. (or CMe,:CH-CH,*CH,-CHDle.CH2-CH:C<cH2,cH2 CX,*CH 142-143"/9 mm. D:00.8468 [a] - 5-03' [a] - 6*40' and [a] - 7-65'. The alternative formula is necessary because the decomposition of the ozonide by water yields adipic and cyclohexanecarboxylic acids. The adipic acid is produced from the intermediately formed cyclohexanone ; the authors show that by treatment with ozone i n carbon tetrachloride and subsequent decomposition by water cyclohexanone is converted into adipic acid. The specific rotatory power of the dimethylocta- dienylcyclohexane despite the position of the double linking in the immediate neighbourhood of the asymmetric carbon atom is almost the same as those of the dimethyldecadiene and dimethylundecadiene. Apparently therefore the influence of a saturated hydroaromatic group on the rotatory power is smaller than t h a t of a n alkyl group.I n comparison with the preceding dimethyloctadienylcyclohexane yq-dimethyl-Aa~-octadienylbenzene (Klages and Sautter A. 1906 i 489) has -a very high rotatory power [a] -63.24'. This is evidently due to the conjugated double linking i n the immediate neighbourhood of the asymmetric carbon atom because its product of reduction CMe,:CH-CH,=CH,-CHMe*CH2*CH,Ph has [a]. - 7*26' and moreover when the conjugated double linking is situated further away from th6 centre of optical activity the rotatory power is quite small.CMe,:CH*CH,*CH,*CHMe*CH,*CH,Ph b. p. 17S-179'/9 mm. obtained in 20% yield from citronellaldehyde and magnesium benzyl chloride (the main product SO% yield is a-phen y l- O-beizz y l- %dime th y Inonan- p-ol CH,Ph*CMe2* CH,*CH,*CH,* CHMe*CH,*CH( OH)*CH,Ph b. p 243-244"/9 mm) is converted through the bromide into 68-dameth yl- Aa?-nonctdienylbenxene CMe,:CH*CH,*CH,*CHNe*CH,*CHPh b. p. 159-160'/9 mm. Uy 0.8894 [a] -3*33' and n 1.51360 nD 1.51770 np 1.52960 and ny 1.53964 Thus l-phe~~~Z-P~-dimet~~yl- A p-nonen- 8-01 K - Phenyl-pc dirnethy LAP-decen-9-0 I? CMe,:CH*CH,-CH,-CHMe-CH,*CH(OH)*CH,-CH,Ph b. p. 188-189"/9 mm. obtained from citronellaldehyde and magnesium P-phenylethgl bromide is converted through the bromide into €1-dimethy E - h@-decudienyl benzene CMe, CH*CH,*CH,*CHMe CH,*CH:CH*CH,Ph b.p. 163-164"/9 mm. DSo 0-8852 - 4-54' - 5-76' - 6*84O - 8.90' for the C D E and F lines and n 1.50375 n 1.50766 nzp 1.51768 and lyly 1.52629. c. s. Preparation of Ally1 Alcohol. A. KOEHLER (Bull. Sod. chim. 1913 [iv] 13 1103-1105. Compare Chattaway T. 1013 105 15l).-The author has increased the yield of ally1 alcohol from 20% to 32% by replacing the oxalic acid with formic acid in its preparation. One hundred grams of glycerol are heated with 80 grams of formic acid for one hour on a water-bath and the product fractionally distilled and collected in three fractions (1) up to 200' (2) 200-260° (3) residue. The saponification index of fraction (2) is determined,i. 134 ABSTRACTS OF CHEMICAL PAPERS. and it is then carefully poured on t o twice the calculated quantity of Eolid potassium hydroxide.The whole is boiled for one hour allowed to cool and the top layer docanted and dried over anhydrous potassium carbonate. The intermediate product is monoformin. W. G . Preparation of Chlorohydrins. HENRY V. WALKER (D.R.-P. 267205).-Chlorohydrins are obtained by treating gas-naphtha (which contains about 40% of olefines and 60% of paraffins) with R mixture of a1 kali hydrogen carbonate and hypochlorite solutions J. C. C. Preparation of Monochlorobutylene Glycol Ether. FARBEN- FABRIKEN VORM. FRJEDR. BAYER & Co. (11.R.-P. 262832).-ChZoro- tutylene glycol ether [y-chlorobutyl y-hyh-ozybusyl ether] CH,*CHCI *CH,*CH,*O*CH,* CH,*CH( OH')* CH3 R colourless liquid b. p. 82-8G0/18 mm. is prepared by treating butylene ay-glycol with concentrated hydrochloric acid.It may be used for the preparation of erythrene. J. C. C. Preparation of True Acetylenic Derivatives s t a r t i n g with Dipropargyl. LESPJEAU (Gompt. Tend. 1913 157 1439-1440).- J n the preparation of the dimethyl ether of the glycol (Aac-octadi- inene-a0-diol) OH*CH,*CiC*CH,*CH,*CiC*CH,*OH from the di- magnesium derivative of dipropargyl (compare A 1910 i 535) there is always produced a t the same time an appreciable quantity of rl-metl~oxy-AaE-he;nlndi-inene CH i C* CH,*CH,* Ci C*CH,*OM e b. p. 67*5O/ 11 mrn. D1"' 0.9156 n1,5" 1,50125. The fact that this is a true acetylenic derivative is shown by the following properties. With aqueous silver nitrate folution i t gives a precipitate soluble in warm water crystallising out in slender needles on cooling. With cuprous chloride i t gives a yellow precipitate which with iodine yields aapE~-~~enIniodo-rl-methoxy- AcGeptadiene CI,:CL*CH,*CH,*CI:CI*CH2*OMe m.p. 95". Finally the ether reacts with magnesium ethyl bromide giving a magnesium derivative which with chloromethyl ether yields the dimetbyl ether of AP~-octadi-inene-aO-diol. This * magnesium derivative also reacts with carbon dioxide giving y-metAoxy-Aac-lieptadi- inene-a-carboxyZic acid CO,H*CiC*CH,*CH,*CiC*CH,*OMe m. p. 46-47". W. G. Ar-Acetylenic Glycols and the Ketohydrofurans Derived from them. GEORGES DUPONT (Ann China. Phys. 1913 [viii] 30 4S5-587).-A rdeum6 of work already published (ccmpare A. 1909 i 545; 1910 i 85 379 456; 1911 i 173 554 804; 1912 i 290 483; 1913 j 696 and Iotsitch J.Russ. Phyn Chern. SOC. 1902 34 239 242 ; 1903 35 430 1269 ; 1906 38 252 656). The following new ccmpounds are described 817-Dip?.o~yZ-b~-dec;nene-Gr]-diol OH*CPr,*CiC*CPr,*OH m. p. 1204 which on hydrogenation yields sr]-d.zj?rop?lldecan-677-dio2 m. p. 82-84' and aas6-telrccivhenyZ-A.8-butinene-a8-diol OH*CPh,*CiC*CPh,.OH m. p. 193". C€T,.CH:CMe*CiC.CMe:CH~CH y [ - D imetlr yloctn- APC- diene- A8-inene,ORGANIC CHEMISTRY. i. 135 a liquid with an agreeable odour b. p. 170' m. p. - 45" nD 1.4977 D22 0,807 1. 8rj-Dipropyldeca- Am-diene- Ae-inene CH,*CH2*CH :CPr* Ci C*CPr:CH*CH2*CH a yeliow liquid b. p. 125-127'118 mm. IZ 1,4890 D19 0.81313 on hydrogenation yielding 8y-dipropyldecane b. p. 125-1 27'118 mm. Dipropyl Ae-decinene-6q-diol does not give the corresponding hydro- ?ZD 1,4450 D 1.7887. furan with mercuric sulphate but an internal anAydride C*CPr C-CPr I l l >o b.p. 137"/18 mm. n 1,4747 D23 0.8404. On hydrogenation i t yields 3-kelo-2 2 5 5-tetrapropylhydrofuran) b. p. 132'118 mm. n 1.4498 D 0.8203. Similarly acetylenedicrotonaldehyde (compare A 191 1 i 804 b. p. 105'/20 mm. n 1.5802 C-CH*CH:CHMe C*CH* CH C El Me gives the compound 111 >O D20 0.9548. W. G. Products Obtained by the Action of Bromine on Ether by Schutzenberger McIntosh and the Author. V. V. TSCHELINCEV ( J . Russ. Phys. Chem. Soc. 1913 45 1845-1860).- Criticism of Arbuzov's results and conclusions (compare A. 1913 i 815). T. H. P. Dimethyl Phosphates of the Rare Earths. J. C. MORGAN and C. JAMES (J. Amer.Chem. Soc. 1914 36 10-16; Chem. News 1914 109 13-15).-Tn order to determine the value of dimethylphosphoric acid for the separation of the rare earths a study has been made of the following salts. The solubilities are expressed as parts of the anhydrous salt per 100 parts of water. Yttrium dimethyl phosphate long white needles solubility 2.80 a t 25' and about 0.55 a t 95". La2( Me2P0,)6,4H,0 white hexagonal crystals solubility 103.7 a t 25'. Cerous dimethyl p?iosphate Ce,(Me2P0,)o,H,0 white hexagonal plates solubility 79.6 a t 25" and about 65 a t 95". Praseodymium dimethyl phosphate green hexagonal crystals solubility 64.1 a t 25'. Neodymium dimethy2 phosphate Nd2( Me,PO,) lilac-coloured hexagonal plates solu- bility 56.1 at 25' and 22.3 a t 95'. Sumarium dimethyl phosphate cream-coloured hexagonal prisms solubility 35.2 at 25" and about 10.8 at 95".Gadolinium dimethyl phosphate long white needles solubility 23.0 a t 25' and 6.7 a t 95'. Erbium dimethyl phosphate long needles solubility 1.78 a t 25". Ytterbium dimethyl phosphate long white needles solubility 1.2 a t 25" and 0.25 at 95". Fractionation experiments were made by preparing a solution of the rare earths in dimethylphosphoric acid gradually raising the tempera- ture and collecting the precipitates ; additional fractions were obtained by evaporation of the mother liquor. I n the case of a material Lanthanum dimethyl phosphate,i. 136 ABSTRACTS OF CHEMICAL PAPERS. containing gadolinium with just sufficient terbium to colour the oxide orange-brown the gadolinium collected in the most soluble fraction and the terbium in the least soluble. Similar experiments with other mixtures of rare earths showed that the rate of separation by this method is much greater than by other methods.Lanthanum cerium praseodymium and neodymium are left in the mother liquor. Samarium europium and gadolinium are much less soluble than these but more soluble than terbium dysprosium and holmium. Erbium thulium yttrium and ytterbium collect in the least soluble f ractione. Some difficulty is encountered in the application of this method owing t o the fact that the dimethyl phosphates gradually decompose with formation of a gelatinous precipitate which renders filtration troublesome. E. G. Preparation of Methylbutenyl Esters. BADISCHE ANILIN- & SODA-FA BRIK (D.R.-P.26 755 3) .-When @y- or /3&dibrorno-@-methyl- butane is boiled with acetic acid and sodium acetate mixtures of the corresponding methylbutenyt acetates are obtained. 3 . C. 0. Decomposition of Hydroxy-acids by Soda-Lime. LEONARD CARPENTER (Chem. News 1914 109 5)-When sodium lactate intimately mixed with soda-lime is destructively distilled in an iron tube under reduced pressure at a temperature just below redness a product is obtained which consists of an aqueous solution of acetone and a brown oily layer composed mainly of mesityl oxide. It appears t h a t ethyl alcohol if formed is immediately oxidieed probably to acetic acid which then loses water and carbou dioxide giving acetone. The same product was also obtained by the action of soda-lime on citric acid.J.t is t o be noted also that very little charring occurs during heating. H. W. Influence of the Position of the Oxygen Atom in t h e Chain on the Velocity of Hydrolysis of Esters. M. H. PALOMAA (Chem. Zentr. 1913 ii 1956-1960; from Ann. Acud. Sci. Fennicae 1913 [A] 4 No. 2 1-104. Compare A . 1913 i 6).-The rate of hydrolysis of esters of hydroxy- and alkosy-acids of mono- esters and ether-esters of glycols of esters of ketonic acids and keto- alcohols by 0.1 N- and 0*2N-h~drochloric acid at 25O has been determined in comparison with the rates of hydrolysis of esters of fatty acids. The constants are calculated from the formula for a reaction of the first order k= 1 / 0 - 4 3 4 3 t . log U / ( U - z) The results show that a reaction minimum occurs at a definite median position of the oxygen atom in the chain which since it occurs when the oxygen atom is in the @-position to the ester group suggests the possibility of intramolecular ring formation substances being thereby produced which have a more saturated character than open chain compounds and thus yield with greater difficulty the additive products which are initially formed in the hydrolysis of esters byORGANIC CHEMISTRY.i. 137 acids. indicated by the dotted lines The following formulae are suggested partial valencies being H25+-3O H,Y-O n 2 c 7 ,yo Hz?-7? \,(,,,' \ ;/' \,!,/' \\ p' H,C /OR' H,C ' h O R ' Meh / O R MeC /COR OK 6 R 0 0 Esters of Ether esters of Est.crs of Ester of alkoxy-acids. ethylene glycol. acetoacetic acid.acetone alcohol. The following substances are described Ethyl glycollate b. p. 15S0 DF 1.0869 ; methyl methoxyacetate b. p. 131-131~4' Di5 1.0578; methyl ethoxyacetate b. p. 144-5-145' Di5 1.0112 ; ethyl ethoxy- acetate b. p. 158'/752 mm. DY 0.9754 ; methyl n-propoxyacetate b. p. 16 1 *5-162O Di5 0,9867 ; ethyl n-propoxyacetate b. p. 174-1 74*5'/ 764 mm. Di5 0.9551 ; methyl n-butoxyacetate b. p. 180-180*5°/ 765 mm. Di5 0.9675 ; methyl isohutoxyacetate b. p. 1715-172.5'/ 767 mm. DY 0.3605; methyl a-methoxypropionate b. p. 129*5'/752 mm. Di5 1.0108 ; methyE y-methoxybutyrate (from y-methoxybutyric acid and methyl alcohol in the presence of phosphoric acid) b. p. 162*5-163'5"/ 767 mm. Di5 0.9879 ; methyl yethoxybutyrate b. p. 175*5'/760 mm. Di5 0,9622 ; methyl 6-methoxyvalei.ate b.p. 184*5-185*5O/767 mm. Di5 0.9741 ; n-propoxymethyl acetate CH;CO,*CH,OPr (from chloro- methylpropyl ether and potassium acetate) b. p. 150-151*5'/759 mm. Di5 0,952 1 ; methoxymethyl propionate C,H;CO,*CH,*OMe b. p. 134.7-135.5@/752 mm. Di5 1.0024 ; ethoxymetliyl propionate b. p. 149-5-150 5@/75.5 mm. DY 0.9709 ; n-propoxyrnethyl propionate b p. 166-1 68' Dy 0.9484 ; P-hydi*oxyethyl formate H*CO,( CH,),*OH (from ethylene glycol and 95% formic acid) b. p. 179-150.5'/766 mm. Di5 1.1989 ; P-methoxyethyl formate (from methoxyetbylene glycol and formic acid) b. p. 131-131*5'/767 mm. Di5 1.0484 ; P-ethoxyethyl formate b. p. 141-0-14 1*7'/748 mm. Di5 1.0013 ; ,B-n-popoxyethyZ formate b. p. 157-159'/766 mm. D:" 0.9762 ; P-ethoxysthyl acetate (from ethylene glycol monomethyl ether and acetyl chloride in the presence of pyridine) b.p. 156-1565O/761 mm. Di5 0.9810; P-ethoxyethyl n-butyrate b. p. 188' Di5 0.9438 ; y-hydroxypropgl formate (trimethylene glycol foymate) (from trimethylene glycol and formic acid) b. p. 195-197'/757 mm. D'4" 1.1405 ; y-rnetlioxypropyZ formate b. p. 146-1 47@/767 mm. Uy 1.0057 ;. y-ethoxypopyl formate b. p. 157.5-159'/742 mm. Di5 0.9731 ; tramethylene glycol mono- n-propyl ether b. p. 170-1 72" Ui5 0,9076 ; y-propoxypropyl formate (from previous compound and formic acid) b. p. 174-5-1 76' ; y-hydroxy- propyl acetate (from trimethylene glycol and glacial acetic acid) b. p. 202*5-204" ; y-met?ioxypropyl acetate (from trimethylene glycol mono- methyl ether and acetyl chloride in the presence of pyridine) b.p. 162-163*5O/762 mm. DF 0.9803 ; y-ethoxypropyl acetate b. p. 174.5-1 75*5' D:" 0.9567 ; cccetylpropyl acetate CH,-CO,*CIT,*CH,* CO-CH (fiwm acetylpropgl alcohol and acetic anhydride) lo. p. 2 1 I -213' 11:" 1.0217. H. W.i. 135 ABSTRACTS OF CHEMICAL PAPERS. A c t i o n of Dimethylamine on tbe Two Chlorohydroxyiso- butyric Acids and their Derivatives. E. FOURNEAU and M. TIFFENEAU (BuZl. SOC. cliim. 1914 Liv] 15 19-26).-1t is known that the two isomeric iodohydrins derived from styrene namely OH*CHPh*CH,Il and CHIPh*CH,*OH (Tiffeneau and Fourneau A. 1913 i 1337)) as also the two chlorolactic acids react with dimethyl- amine yielding the same product ; for example in the former case OH*CHPh*CH,*NMe,. In the present investigation i t is shown that even where one of the positions affected is a tertiary one a similar behaviour is observed.When ethyl p-chloro-a-hydroxyisobutyrate is heated wi t h fused sodium acetate at 1 90-200' ethyl a-hydroxy-p-acetoxyisobutyrata OAc*CH,*CMe(OH)*CO,Et D! 1.135 b. p. 226-2229' is obtained. The following compounds can be prepared in R similar manner Propyl a-h ydvox y-P-acetox yisobutyrate Dj 1.001 b. p. 244-245'1770 mm. Propyl u-hydroxy-P-valeroxyiso- butyrate b. p. 272'. Ethyl a-h3droxy-P-benzoyloxyisobutyrate m. p. 35' b. p. 175-1 7 6'11 8 mni. Propyl a-hydroxy-P-benzoyloxyisobutyrate Di 1,1457 b. p. 205-208O/29 mm. Ethyl a-hydroxy-P-salicyloxgiso- butyrate rectangular prisms m. p. 52' b. p. 197'/16 mm. Yropyl a-hydrotzy-P-salicyloxyisobutyrate b. p. 200°/1 7 mm.By the gradual addition of thionyl chloride to a mixture of the substance with pyridine at 0') propyl a-hydroxy -P-acetoxyisobutyrate can be converted in to prop91 a-chloro-~-acetoxyisobutyrate OA c*CH,*CClMe*CO,*C,H b. p. 147-150'/50 mm. Bthyl a-chloro-P-acetoxyisobutyrate D8 1.1686 b. p. 2 16-2 17'/760 mm. and propyl a-chloro-P-benxoyloxyisobutyrute b. p. 198-200' are obtainable i n a similar manner. These immediately preceding esters are hydrolysable by boiling with five times their weight of 10% hydrochloric acid with formation of a-c~~Zoro-P-hydroxyisobutyric acid hygroscopic rectangular prisms m. p. 77" soluble in most solvents; its ethyl ester obtained by the action of alcohol containing a little sulphuric acid has b. p. 201-202'. When heated at looo for ten hours with dimethylamine in benzene solution ethyl a-chloro-P-hydroxyisobutyrate is converted into the ethyl ester b.p. 192-1 94') of P-dimethylamino-a-bydroxyisobutyric acid which is produced by hydrolysis of the ester with water ; this acid m. p. 173-174" is identical with that previously described (Fourneau A. 1909 i 210); the ester gives a benzoyl derivative m. p. 140' (compare Fournesu Zoc. cit.) of which the hyds.ochloride and picrate form hygroscopic prismatic needles m. p. 140-141' and elongated lamellae m. p. 127O respectively. Ethyl a-chloro-p-acetoxyisobutyrate and propyl a-chloro-P-benzoyl- oxyisobutyrate also slowly react with dimethylamine in benzene solution the former at loo' and the latter at 125-130'. The product is again the dimethylamino-ester just described but in the experiment with the former of the two esters the main product was accompanied by a small quantity of a n acid substance m.p. 130° and possibly ~-dimethylarnino-a-acetox~isobutyric acid,ORGANIC CHEMISTRY. i. 139 Thus a-chloro-P-hydroxy- and ~~chloro-~-hydroxy-isobutyric acid derivatives yield the same product with dimethylamine. D. F. T. Double Aluminium Oxalates. W. STORTENBEKER (Rec. trav. Chim. 1913 32 226-243. C'ompare Wyrouboff and Rosenheim Zeitsch. anorg. Chem. 1909 63 121).-The author has examined and described a number of double alumino-oxalates of sodium ammonium and rubidium together with mixed crystals of these substances. It is shown that mixed crystals and compounds of the same substances can be obtained and also that chemical compounds may be accompanied by mixed crystals which have the same crystalline form.The most probable interpretation of the last fact is that mixtures are formed of the compound and the simple salt which crystallise with the same amount of water of crystallisation and in the same form. J. F. S. Stereochemistry of the Halogenosuccinic Acids. 11. BROR HOLMBERG (J. p r . Chem. 1913 [iiJ 88 553-603).-1n the previous paper (A 1913 i 824) the author has shown that measurements of the rate of formation of bromine ions from I-bromosuccinic acid in neutral solution did not give concordant values for a unimolecular reaction and pointed out that this was probably due to the reconver- sion of the P-propiolactonecarboxylic acid under the influence of the bromide formed in the reaction into the bromo-acid.This view has been confirmed in the case of the decomposition of sodium I-bromo- succinate by removing the bromide almost as fast as i t is formed by the gradual addition of silver nitrate when concordant values for a reaction of the first order were obtained. According t o Senter ('I!. 1910 97 346; 1911 99 95 1049) the presence of silver bromide exerts an accelerating influence on the hydrolysis of the bromo-derivatives of fatty acids. The behaviour of I-bromosuccinic acid in the presence of silver ralts has been investi- gated by the author but no accelerating influence could be observed. Further the presence of silver salts during the decomposition of sodium I-bromosuccinate has no influence on the configuration of the resulting malic acid the sign of rotation of the latter being determined by the reaction of the solution in which the hydrolysis of the inter- mediately-formed P-propiolactonecarboxplic acid takes place ; in acid solution the lactone is hydrolysed to I-malic acid and in alkaline solution to the d-acid.Attempts have also been made to isolate the lactonic acid formed by the decomposition of sodium E-bromosuccinate in neutral solution but hitherto the compound has been obtained only as a syrup. An aqueous solution of the sodium salt of the lactonic acid is readily obtained by removing the malate and unchanged bromosuccinate from the solution resulting from the decomposition of the sodium salt of I - bromo-acid by precipitation with silver nitrate.The sodium salt reacts with sodium chloride bromide and iodide in aqueous solution to form the salts of the corresponding I-halogeno-i. 140 ABSTRACTS OF CHEMICAL PAPERS. succinic acids ; with ammonia i t yields almost pure d-P-malamic acid which is hydrolysed by hydrochloric acid to d-malic acid. In neutral solution the lactonic acid is moderately stable but is readily hydrolysed in acid or alkaline solution t o I- and d-malic acids respectively ; the salts with the common metals are soluble in water. The transformation of I-bromosuccinic acid into propiolactone- carboxylic acid and reconversion of the latter into the Lbromo-acid is accompanied by very little racemisation and the author therefore assumes that these reactions take place without change of corifiguration ; I-bromosuccinic acid must therefore have the same configuration as the d-lactouic acid t o which it gives rise Further since change of configuration is unlikely to occur when substitution is not effected at the asymmetric atom the conclusion is drawn that d-/3-malamic acid obtained from the d-lactone has the same configuration as the I-bromo-acid. The author considers that d-malic acid corresponds with the d-lactone and that the hydrolysis of the latter is accompanied by configurative change only in acid solution.The racemisation of I-bromosuccinic acid by sodium bromide and hydrobromic acid has also been studied. It is found that racemisation takes place readily in acid solution but not in neutral solution. From the results so far obtained the conclusion is drawn that the racemisa- tion is almost entirely restricted t o the non-ionised acid and is deter- mined by the presence of the bromine ion.The halogen in the I-bromo-acid is readily replaced by iodine by the action of potassium iodide in neutral solution ; the iodosuccinic acid thus formed is lzevorotatory but could not be isolated in a pure con- dition. The pure I-acid can however be obtained by the action of potassium iodide on a solution of sodium salt of d-propiolactonecarboxylic acid prepared in the manner described above. It has m. p. 150- 152' (decomp.) [a]=. - 74.1' in ethyl acetate solution. The interaction of sodium and potassium xanthate with the sodium potassium barium strontium and calcium salts of kbromosuccinic acid under various conditions has also been investigated and it is found that the sign of rotation of the resulting xanthosuccinic ucids CO,H*CH(S* CS*OEt)*CH,* CO,H depends not only on the particular salt of the bromo acid but also on the concentration of the solution and the amount of xanthate employed.The purest specimen of E-xanthosuccinic acid hitherto obtained has m. p. 131-132" [UlD -92.6' in etbyl acetate solution. I-Bromosuccinamic acid reacts with the potassium and sodium xanthates yielding d-xunthosuccinarrvic acid CO,H* CH( S* CS*OEt) *CH,* C0.N H which forms flat obliquely-cut colourless prisms m. p. 137-138' (decomp.),.[,llD + 64O in etbyl acetate and is hydrolysed to d-xantho- siiccinic acid. F. B. Crystallography of Symmetric Dibasic Aliphatic Acids. Malonic acid (D" 1,6305 ; triclinic a b c = 0.7440 1 0.4573 ; a = 102'42' /3 = 100"44' y=.63'48').JULIEN DRUGMAN (&?ilScA. Kryst. Lwh,. 1913 53 240-259).-ORGANIC CHEMISTRY. i. 141 Dimethylmalonic acid (D17* 1.357 ; tetragonal-trapezohedral cc c = Potassium hydrogen dimethylmalonate + 2H,O ( D1lZ 1.540 ; /3 = 10Ool4g' 1 14830). triclinic a b :c = 0,6299 1 1,2682 ; a = 89"14' y = 94'39'). Calcium dimethylmalonate (rhombic (6 b c = 0.6 17 1 0,937). Diethylmalonic acid (triclinic a b c = 0.6925 1 0.5334 ; a = 89'35' Potassium hydrogen diethylmalonate + H,O '1 (tetragonal pseudo- Glutaric acid (D14'5 1.429 ; monoclinic a b c = 2.0448 1 2,6734 ; P/3-Dimethylglutaric acid (T. 1911 99 434 ; monoclinic a 6 c = 0,4909 1 1.1171 ; p=91'55'). ay-Dihydroxy-ay-dimethylglutaric acid (triclinic a b c = 1.6589 1 0.9910; a=91'54' p=9s019fr' y=95'46').ay-Dihydroxg-ay-dimethylglutaric acid-monolactone (rhombic a b c = 0.6186 1 1.4939 Prendel 1891). n-Pimelic acid,a-modification (m.p.(?) 101-102c D1"'j 1.329 ; monoclinic a b c = 3.697 1 1.2058; /3 = 105"40' von Lzng 1893) ; P-modificit- tion (m. p. 101-102' DI5 1.282 ; monoclinic a b c = 2.15 1 :?; p = 136'). y-Methylpimelic acid (D" 1,641 ; monoclinic cc b c-1.696 1 C r y s t a l l o g r a p h y of the I s o m e r i c Citraconic and Mesaconic JULIEN DRUGMAN (Zeitsch. Kryst. Min. 1913 53 (triclinic a b c = 1.1525 1 0.7550 ; p = 93'37' y = 88'29'). cubic a c = 1 1 .OOO ; optically uniaxial and positive). p = 97"Y). 1.648 /3 = 103'22'). Acids (C,H6O4>. 260--262).-C!itraconic acid L.J. S. a = 98'22' p = 10S048&' r = 88'3 1'). Mesaconic acid (rhombic a b c = 0.8536 1 0*8421). Syntheses by means of Organometallic Derivatives of Zinc. Preparation of a-Ketonic Acids. E. E. BLAISE (Compt. rend. 1913 157 1440-1443. Compare A. 1912 i 236 410).- a-Ketonic acids can be prepared indirectly by aid of zinc alkyl iodide through the formation of the corresponding mixed cycloacetals. Ethyl chloroglyoxylate condenses with a-hydroxgisobutyric acid giving the acid OEt*CO*CO,*CMe,*CO,H m. p. 83' which on warming with thionyl chloride gives the chloride b. p. 120.5"/12 mm. and yields an anilide m. p. 95'. The acid chloride reacts with zinc propyl iodide L. J. S. $!Me2*0 co-0 t o give the cycloacetaZ >CPr*CO,Et b. p.123-124'/11 mm. which on heating on a water-bath with alcoholic hydrochloric acid yields ethyl butyrylformate b. p. 70-5'/11 mm. and ethyl hydroxyiso- butyrate. At the same time a small amount of the corresponding diethylacctal b. p. 96'/11 mm. is formed this compound being also obtained by the action of ethyl orthoformate on the a-ketonic ester in the presence of one drop of sulphuric acid. Ethyl butyrglformate is readily hydrolysed by boiling with 5% aqueous oxalic acid giving the corresponding acid b. p. 79'/12 mm. which furnishes a p-nitro- phenylhydraxone m. p. 205' and a semicarbaxone m. p. 220'. W. G.i. 142 ABSTRACTS OF CHEMICAL PAPERS. Action of Lime on Reducing Sugars and the Products formed thereby. A. SCHWEIZER (Chem. Zentr. 1913 ii 1791 ; from Arch.8uikerind. N e d . - l d e 27 reprint 12 pp.).-The action of alkali on dextrose solutions leads to the formation of products which are more or less deeply coloured according to the temperature of reaction etc. The dextrose is thereby decomposed with formation of salts of lactic and saccharic acids etc. Addition of acids renders the solutions colourless. The author is led to the conclusion that this phenomenon is t o be attributed to the alcoholic function of the decomposition products and that tho metallic atoms of these compounds are displaced by acids in accordance with the following scheme I HC(OH)*CH(OM)*CO,M -+ HC-CH*CO,H -+ bH,-CO*CO,H \/ 0 where M represents the atom of a univalent metal. This supposition receives support from the deep colorations imparted by the addition of ferric salts.From carbonation experiments made a t 2s' with an excess of lime the conclusion is drawn that the colourless salts of lactic acid and the coloured salts of saccharic glyceric acids etc. are simultaneously formed at all temperatures in contrast to the assumption that the former are alone produced by the action of alkali on dextrose below 55O whilst the latter are formed above this temperature. Por the investigation of the properties of the product of decom- position solutions of dextrose decomposition product (prepared by the action of lime on invert sugar) were treated with lime and then carbonated. The following results were obtained On boiling the colour of the solution deepens and an additional quantity of calcium salts is formed.When carbonated a portion of the coloured salt is removed (it is shown that the primary products of the action of alkali on dextrose are here involved) whilst the portion of the salts which remains undergoes further decomposition whereby fresh quantities of lime are dissolved. The removal of coloured products occurs not only in alkaline but also in neutral solution but whether in the latter case additive products are formed from calcium carbonate and the salts of organic acids has not been determined. The alteration in colour of the calcium salts of these compounds with increasing alkalinity or acidity has also been investigated. The colour is lightest in acid media and increases in intensity until the neutral point is reached; a sudden marked darkening then occurs which diminishes with increasing alkalinity.H. W. Action of Cyanides of the Alkali and of the Alkaline Earth Metals on Dextrose. E. XUPP and A. HOLZLE (Arch. Pharm. 1913,25 l,553-556).-Equal molecular quantities of dextrose and potassium cyanide in aqueous solution react according to the equation OH-CH,*[CH-UH],*CHO + KCN + 2H20 The quantity of ammonia evolved and of potassium carbonate obtained = OH*CH,*[CH*OH],*CO,K + NH,.ORGANIC CHEMISTRY. i. 143 by the incineration of the dried residue correspond very closely with the theoretical amounts. The acid has been isolated in the form of its anhydride which exhibits all the properties of a-glucoheptoic an- hydride. The reaction between dextrose and barium cyanide follows a similar course. c. s.Some Acyl Derivatives of Dextrose and Mannose. EMIL FISCHER and RUDOLF OETEER (Ber. 1913 46 4029-4040).-Follow- ing the discovery of the galloyldextroses and of pentabenzoyldextrose (A. 1912 i 888) the authors found that cinnamoyl chloride condensed even more readily than did benzoyl chloride when shaken with dextrose and quinoline in chloroform solution The study of this reaction has been extended to substituted cinnamoyl chlorides and to mannose and mannitol. The derivatives of mannose so obtained were all Zaevo- rotatory and consequently the configuration of the sugar may be retained in them. Hexacinnamoylmannitol C6H,06(C,H70) from very finely powdered dry mannitol and cinnamoyl chloride forms tufted prisms or needles from ethyl acetate m. p. 99-loo' [a] +12.96' [a] +13.15' (in chloroform).a- Pentacinnamoyldex t rose and /3-pentacinnamoyl dextrose have already been described (ibid.). Pentucinnamoylrnannose crystal- lises with 1 mol. benzene in mycelium-like masses m. p. 108-112' [a] - 91.3' - 91-66' (benzene) [a] calculated for the benzene-free compound - 99*9' - 100.3O. Pentubenzoylmannose c6H7o6Bz5 forms long radiating needles from alcohol m. p. 161-161*5' (corr.) [a] - 80*44' - 80.7' (chloroform). Penta-acetyhnannose C6H7O6AC5 was obtained by shaking together mannose pyridine and acetic a D - hydride at 0' ; it has m. p. 127.5' (corr.) [a] - 24*8O [a] - 24.9' (chloroform) and a very bitter taste. Similar derivatives were prepared from caffeic acid which was applied in the form of the dimethylcarbonato-derivative. Methyl chlorocarbonate was gradually added to a dilute solution of the sodium salt at - 5 O in a hydrogen atmosphere when the solution was acidified and the 3 4-dimethyZcarbonatocinnan.~ic acid C6H3(O*C02Me~*CH:CH* Co2H precipitated.It formed curved needles from aqueous acetone m. p. 145-146' (corr.). The chloride C13H,,07Cl was obtained by the action of phosphorus pentachloride in glistening needles m. p. 108.5-109ti' (corr.). From it the esters were prepared; the ethyl ester forms flat needles or prisms m. p. 98' (corr.) and the methyt ester crystallises in tufted spikes m. p. 95-96.5O (corr.). The chloride was condensed with a-dextrose forming penta-3 4-dimethylcarbon- atocinnamoyZ&extrose C,H706(C13H1107)5 as a colourless amorphous mass with indefinite constants.The easily hydrolysed carbonato-gronpe were removed by the action of dilute sodium hydroxide below ZOO in an atmosphere of hydrogen. On acidifying a resinous mass was formed which was fractionally precipitated by chloroform from a solution in ethyl acetate. The pale yellow amorphous mass so obtained was probably the expected penta-3 4-dihydroxycinnamoyl- dextrose but the constants and analytical data were inconclusive. It is very sparingly soluble in hot water and gives a deep greeni. 144 ABSTRACTS OF CHEMICAL PAPERS. coloration with alcoholic ferric chloride and a pale yellow precipitate with alcoholic potassium acetate. J. c. w. Influence of Alcoholic C o n t e n t on the Biochemical Synthesis of a-Methylglucoside by a-Glucosidase. A AUBRY ( J .Pharm. Chim. 1914 [vii] 9 19-23).-The biozhemical synthesis of a number of alkylglucosides has been effected by Bourquelot HPlrissey and Bride1 (A. 1913 i 323 428 747) by the action of a- glucosidase on solutions of dextrose in water containing the respective alcohols. I n every case excessive concentration of alcohol destroys the ferment. The present investigation was undertaken to determine the concentration of alcohol most favourable to the synthesis and also the point at which the ferment is immediately destroyed. Experiments were performed by adding a constant amount of a maceration of bottom yeast to solutions of dextrose in water made progressively richer in methyl alcohol. The course of the reaction was followed polarimetrically and finally the dextrose determined by the method of Mohr-Bertrand.Small amounts of toluene (which appears to have a slightly toxic effect on the ferment) were also added. The synthesising action of the ferment is very obvious even in the most dilute alcoholic solutions and increases until the solution contains 16 grams of methyl alcohol in 100 C.C. At greater concentrations of alcohol the toxic effect becomes more marked occurring immediately in solutions containing 34-36 grams of methyl alcohol in 100 C.C. a t the ordinary temperature. At higher temperatures the toxic action becomes more pronounced ; thus at 40° immediate destruction of the ferment takes place in a solution containing 16 grams of methyl alcohol in 100 C.C. H. W. Influence of the Strength of the Alcohol on the Biochemical Synthesis of a-E thylglucoside and a-Propylglucoside.lh. BOURQUELOT and A. AUBRY (Compt. T e n d 1914 158 70-72; J. Pharrn. Chirn. 1914 [vii] 9 62-66. Compare preceding ab- stract).-The optimum concentration of alcohol for the synthesis of aglucosides by a-glucosidase is for ethyl and prop91 alcohols respectively 20% and 16%. Above these concentrations the synthesis is masked by the more rapid decomposition of the glucoside by the same ferment. W. G. Optimum Experimental Conditions for the Preparation of P-Ethylglucoside. J. COIRRE (J. Piiarm. Chim. 1913 [vii] 8 653-559).-The conditions under which the maximum amount of /3-ethylglucoside is formed by the interaction of emulsin dextrose and alcohol are discussed theoretically and a process for the preparation and purification of this compound under these conditions is described.T. A. H. Characterisation of the Organ Pentose as &Ribose. J. VON BRAUN (Bey. 1913,46 3949-3951).-4 4'-Bismethylhydrazino- diphenylmethane (von Braun A. 1908 i 700 737; 1910 i 524) proves a very convenient reagent for distinguishing between theORGANIC CHEMISTRY i. 145 various pentoses. It is already known that whereas xylose fails to react with this subtance active arabinose readily gives an amorphous sparingly soluble hydrazone m. p. 180'. &-Ribose is now found to yield a microcrystalline dihydraxone (C,H,,O,:N*~~~e*C,H,),CH m. p. 141-1 42' whilst d-lyxose gives an analogous but much more sparingly soluble dihydrrmone m. p. 156'. The behaviour of a specimen of the pentose obtainable from inosic acid towards this reagent proved the identity of the pentose with &-ribose. D.F. T. Starch and its Hydrolytic Degradation Products. OSCAR VON FRIEDRICHS (Arkiv. Kenz. Mia. Geol. 1913 5 No. 2 1-40)- It has been shown already that starch can be converted completely into maltose. Since maltose is a dextrose-a-glucoside it follows that the amylase which converts starch into maltose but cannot hydrolyse it further t o dextrose is a @-enzyme and consequently every other linking between the glucoside-forming dextrose residues in starch corresponds with the P-configuration in P-dextrose. Starch may be represented therefore thus Dextrose-6-glucoside. Maltose. P P P . . . C,,H2~Og-O-C12H2009.. .O-C,~H,,O,O O P The dicarbonyl binding represented as -0- may occupy any other possible position t o that just given but both linkings must have the @configuration otherwise dextrose would be liberated during the hydrolysis of starch by diastase.That starch is not hydrolysed by emulsin is not remarkable for i t is known that gentianose which contains a P-glucoside linking is not hydrolysed by emulsin. If the amylase which hydrolyses starch is a @-enzyme i t might be expected to hydrolyse @-methylglucoside ; this however was not found to be the case; on the other hand neither does it hydrolgee a-methylglucoside. Assuming the above formula for starch to be correct then starch when hydrolysed by acids should yield dextrose maltose and a dextrose-P-glucoside. i t has been found possible to isolate a P-glucosidodextrose from the products of the hydrolysis of starch by acids which appears to be identical with Fischer's isomaltose but this point could not be established definitely.The disaccharide isolated had La]D +59*1' was readily hydrolysed by emulsin and yielded an osazone m. p. 146'. The products formed by heating starch with an aqueous solution of oxalic acid under pressure have been isolated as far as possible by means of repeated fractional extraction with aqueous alcohol. The properties of many oE these compounds are given below also the molecular weights determined by the diihsion method of Oholm (A. 1905 ii 147 ; 1910 ii 273). VOL. CVI. i. Ii. 146 ABSTRACTS OF CHEMICAL PAPERS. Amylodextrin probably a mixture [a] + 196' ; mol. wt. 13630 ; reducing power 0-6-0*9% of that maltose.Erythrodextrin I [a]:" + 196'; mol. wt. 3290; reducing power 3.6% of that of maltose. Erythrodextrin 11 [a] + 193.1'; mol. wt. 2080; reducing power 10%of that of maltose. Achroodextrin [a]? + 179' probably identical with the malto- dextrin a of Ling and Baker; mol. wt. 1060 ; reducing power 30% of t h a t of maltose. Achroodextrin [a] + 172*8' probably identical with the malto- dextrin p of Ling and Baker and the achroodextrin I11 of Prior; mol. wt. 1020; reducing power 42% of that of maltose. Achroodextrin [a] + 162.6"; mol. wt. 680 ; reducing power 58% of that of maltose. The molecular weight values obtained indicate that the last described achroodextrin is formed by the condensation of four dextrose mole- cules whilst the molecule of amylodextrin is built up of four molecules of erythrodextrin I which is probably formed by the condensation of twenty molecules of dextrose.An investigation of the rate of formation of fermentable sugar from erythrodextrin I and erythrodextrin I1 by the action of oxalic acid shows that at first very little if any is formed although a decided diminution in optical rotatory power takes place from the very com- mencement of the interaction of the reagents; fermentable sugar is formed in rapidly increasing quantity only after the lapse of a relatively long period which under the conditions of the experiment extended t o about 280 minutes in the case of erythrodextrin I and about 120 minutes in the case of erythrodextrin 11. The conclusion is drawn from this that maltose and dextrose are not formed in the early stages of the hydrolysis of starch or the higher dextrins by acids but only after these have been broken down into simpler compounds.W. H. G. Behaviour of Starch Dextrins towards Certain Yeasts. OSCAR VON FRIEDRICES (Arkir. Kern. Hin. GeoZ. 1913 5 No. 3 1-14).-The behaviour of the dextrins formed by the action of acids on starch (compare preceding abstract) towards certain yeasts has been studied and the rate a t which they are fermented by some of them measured. The achroodextrins [a] + 162.6' + 172%' and + 1 7 9 O and the erythrodextrins [a]D + 193.1' and 196' are not fermented by Saccharom yces frayilis 8. JIarxianus 8 ellipsoideus 11 8. exiguus or 8. Pastorianus I although the two last mentioned are possibly able t o ferment slowly the achroodextrins [a] + 162.6" and + 172.P.The three achroodextrins are slowly fermented by Saccharomyces thermantiton but the two erythrodextrins are not attacked by it. The yeast Sinner 11 appears to be unable to ferment the two erythrodextrins or the achroodextrin [a]. + 179O although the achroodextrins 172.8' and + 162.6' are destroyed by it the latter more readily than the former.ORGANIC CHEMISTRY. i. 147 ~chixosaccharornyces Pornbe attacks energetically all the dextrins investigated the erythrodextrin [.ID + 196" least readily and the achroodextrins most readily. The achroodextrins are also fermented very readily by Sacchsia suaveolens and Mucor Rouxii ; the erythrodcxtrins are also attacked with ease by the latter. Generally speaking the dextrins with the higher molecular weights are not so readily attacked as those with the lower molecular weights whilst Schizosacchnromyces Pombe alone is capable of fermenting the achroodextrins as rapidly as i t does dextrose The yeasts which are found to split up a-methylglucoside are capable of fermenting the dextrins only to a very limited extent whilst Sacchsia SuaveoZens which is known to be capable OF fermenting /3-methylgIucoside is found to ferment the ncht oodextrina very energetically.The results support the view that P-glucoside linkings are present in the starch and dextrin molecules. W. H. G. Synthetic isoMaltose. OSCAR YON FRIEDE~ICHS ( Arkiv. Kern. Min. Geol. 1913 5 No. 4 1-13).-The product obtained by the action of four parts of concentrated hydrochloric acid on one part of dextrose a t 10' for twenty-four hours is composed approximately of 68% dextrose 18% isomaltose 8% maltose and 6% unknown polysaccbarides.The assumption has been made that isomaltose is a /3-dexhrosegluco- side since it is hydrolysed by emulsin and not by maltase (compare E. F. Armstrong A. 1906 i 127); but several trials have shown definitely that the optical rotatory power of a solution of isomaltose which has been partly hydrolysed by emulsin is lowered on the addition of alkali a behaviour which it is difficult to explain except on the assumption that isomaltose is a n a-dextroseglucoside. It mas also observed that the optical rotatory power of the solution instead of decreasing as the isomaltose was hydrolysed into dextrose increased as the interaction of the isomaltose and emulsin proceeded the measure- ments being made in each case after the addition of alkali; i t has not been found possible to account for thie.isoMaltose is hydrolysed by a n extract of Aspeygillus niger but not by an extract of Kephir. It undergoes fermentation when treated with Aspergillus niger Sacchclromgces frap'lis 8. exiguus and 8acchsic6 suaveolens but is not affected by Xaccharcm pces ILp3r 8. fifarxianus S. ce~evisice Saaz or S. cerevisiae Frohberg. W. H. G. Theory of the Dry Distillation of Wood. PETER KLASON (Arkiv. lienz. Min. Geol. 1913 5 No. 7 1-42. Compare A . 1908 i 717 955).-The author has distilled dry beech ~ o o d under tho following conditions in a cathode-light vacuum under 5 mru. pressure and at ordinary pressure with various velocities in which the distillations between the temperatures 250° and 400' lasted respec- tively three eight and sixteen hours and four teen day,..The percent- ages of charcoal pitch acetic acid formic acid wood spirit mtthgl alcohol acetone and formaldehyde were determined in each case and 1 2i. 148 ABSTRACTS OF CHEMICAL PAPERS. also the percentage composition of the charcoal and pitch. The gases obtained were also analysed and their calorific values determined. The conclusion is arrived at that the dry distillation takes place in two stages. The first stage which is practically the only one when a cathode-light vacuum is used takes place essentially according t o the equation 2C,,H,,O,,( wood) = 3CIoH,O(primary charcoal) + 19H20 + 3c02 + 3CO + 2*5CH;CO,H + H*CO,H + CH,*OH + C,,H,,O,,(pitch) + C,H,O(primary t a r oils).The second stage which takes place cam- pletely when the heating is very slow since no pitch is then obtained is represented by the equation C,,H,,O,,( pitch) = C,oH200,(secondary charcoal) + 9H20 + 2C0 + C,H,,(secondary t a r oils). l'he slower the rise in temperature the more the yield of pitch diminishes and the yields of charcoal water carbon dioxiode and t a r oils increase. The yield of acetic acid is about 6.5% under ordinary pressures and is independent of the rate of heating ; under diminished pressure the yield is increased only about 0.5%. The yield of formic acid (2.4%) i s greatest under diminished pressure ; a t ordinary pressures it varies from 0.71 to 0.33% as the time of distillation is increased.Contrary to what has been stated previously by Norlin the yield (1.5%) of methyl alcohol is independent of the rate oi heating whilst the quantity of water and carbon dioxide formed increases with the time of distillation. Acetone is essentially a secondbry product produced from the acetic acid and consequently its yield decreases as the time of distillation iucrerses. The formaldehyde in the wood spirit forms about 1% of the dry wood taken. The heat of reaction calculated a t Oo during the dry distillation which is taken as the difference between the heats of combustion of the wood and its products of distillation is positive both for the primary and secondary reactions. The heat developed during the actual dry distillation between 2.50'- and 400' is negative for vacuum distillations and positive for distillations at ordinary pressure.The ratio between the acetic acid and formic acid percentages is a measure of the rate of carbonisation of the quantities of pitch and charcoal and of the character of the charcoal. The ratio is about 32 for vacuum distillations and 11 and 5 respectively for quick and slow distillations at ordinary pressures. If it sinks below 5 it denotes that the reaction products have been strongly overheated. T. S. P. Action of Chlorine on Pinewood. ENIL HEUSER and RUDOLF SIEBER (Zeitsch. angew. Chem. 1913 26 801-806).-The action of chlorine on pinewood is very energetic at first and quickly reaches a point when very litt.le change occurs; thus treatment of the wood for thirty minutes one hour and two hours leads to the production of hydrogen chloride to the extent of ZO% 27% and 32% respectively of the weight of dry wood taken and the treated wood loses about 20% 36% and 42% respectively of its original weight when extracted with an aqueous solution of sodium sulphite ; the interaction proceeds very slowly after two hours for only about 35% of hydrogen chloride is formed in twenty-two hours and only about 45% of the wood is removed by a solution of sodium sulphite.During the first two hours,ORGANIC CHEMISTRY. i. 149 that is so long as lignin is present the chlorine acts to only a very slight extent on the cellulose but when all the lignin has been converted into chlorinated derivatives the cellulose slowly undergoes oxidation being converted entirely into oxycellulose in twenty-two hours.As stated already about 32% of hydrogen chloride is formed in two hours but only about 9.5% of chlorine is found combined with the wood a t the end of this period ; similarly only about 7% of a chlori- nated lignin compound can be extracted from the treated wood by means of ethyl alcohol ; it is probable therefore t h a t the greater part of the hydrogen chloride formed during the reaction owes itas origin to the oxidation of the '' wood-lignin " by the chlorine. It was not found possible t o obtain a derivative of pyrogallol from the chlorinated " wood-lignin " compound and not a trace of furfuraldehyde was obtained by treating it with 12/ 0 h y drochloric acid. W.H. G. Ethylamine Compounds of Mercuric Chloride. RAGNAR WJDMAN ( A ~ k i v . Kern. Min. Geol. 1913 5 No. 1 1-36).-The author has investigated the various equilibria which exist between the com- ponents mercuric chloride ethylamine and hydrochloric acid in aqueous solutions at ordinary temperatures (compare Strom holm A 1906 i 935) and finds that the following compounds each of which is white in colour are capable of existing as solid phases HgCI,,NH,Et NHEt*Hg,Cl HgCl,,BN K,Et (NEt) Hg,Cl (NEt),.Hg5C16. The two first are obtained from solutions containing an excess of mercuric chloride and can be transformed one into the other according to the followin'g scheme 2HgCl,,NH,Et z NHEt*Hg,CI + NH,EtCI. The third and fourth compounds are prepared from solutions con- taining an excess of ethylamine and are connected by the equation 3(HgC1,,2NH2Et) (NEt),Hg,C12 + 4NH,EtCl.The compound (NEt),Hg,CI is obtained among other ways by shaking or boiling the compound NHEt*Hg,CI with a strong solution of mercuric chloride. The compound (N Et),Hg,Cl18 described by Stromholm does not exist. T S. P. Decomposition of Betaine by Alkali. FR. ALBERS (Chem. Zeit. 1913 37 1533-1534 1545-1547).-0n heating betaine with potassium hydroxide a t 200-220° about one-third of the nitrogen is eliminated as trimethylamine the only other volatile compound formed being carbon dioxide. A compound stable in alkaline solution at 220" is formed giving a hydrochloride C,H90,N,HCI m. p. 187-189O. The plcctinicldorzde forms a yellow matted crystalline mass m.p. 120- 1 2 lo. When the decomposition is effected at 500-540° rather less trimethylltniine is found as well as methylamine and ammonia the last being formed by secondary decomposition from the amines. About 15% less than the theoretical quantity of amines is formed. Methane and hydrogen are liberated as well as a little carbon monoxide and perhaps nitrogen ; no hydrogen cyanide is formed.i. 150 ABSTRACTS OF CHEMICAL PAPERS. The treatment thus fails t o liberate more than 85% at most of the betaine nitrogen as ammonia or amines. E. F. A. Preparation of Dimethylaminomethyl Alcohol. FARBEN- FABRIICEN VORM. FRIEDR. BAYER cb Go. (D.R.-P. 268012. Compare this vol. i 20).-Dimethylaminomethyl alcohol can be prepared by treating trimethylamine or its salts with a halogen (such as chlorine) or hypochlorous or hypobromous acid in the presence of water thus avoiding the preparation of the halogen additive compounds of trimethylamine as described in the chief patent (Zoc.cit.). J. C. C. Combinations of Carbamide with Acids. D. F. DU TOIT (Proc. K. Akad. Ketensch. Amstei*dam 1913 16 555-556).-The author has investigated the nature of the solid phases which occur in contact with solution in the system carbamide-acid-water for a number of different acids. I n the case of oxalic acid the compound 2CO(NH,),,H,C,04 was found at 20' and 30°. but no evidence of the compound CO(NH,),,H,C,O,,H,O mas obtained at these temperatures. With acetic acid a t 16*5O,-3Oo and 32G the compound CO( NH,) 2CH3* C0,H was obtained. It melts a t about 35" is very soluble in but not dissociated by water.In tho case of hydrochloric acid the compounds ZCO(NH,),,HCl and CO(NH,),,HCl were found at ZOO whilst with nitric acid CO(NH,),,HNO was the only salt formed at this temperature The solubility of the nitrate decreases a s the concentration of nitric acid in the solution increases attaining a minimum at about 70%. At ZOO sulphuric acid was found to give rise t o the compounds 2CO(NH,)2,dH2S0 and CO(NH,),,H,SO but no evidence was obtained of the existence of CO(NH2),,2H,SO which has been described by Hsntzscb. H. M. D. I n t e r n a l l y Complex Salts. X. Salt and Complex Salt Formation with Imido-compounds. H. LEY and I?. WERNER ( B e y . 1913 46 4040-4050).-The formation of complex salts of compounds in which the alkyl or aryl radicles and the oxygen atoms of the acid-imides have been successively replaced by imino- or amino- groups has already been studied in a number of cases (A 1907 i 301 730).The remaining possible imidocompounds namely the acid-imides acjlcarbamides biuret and the imino-acid-imides are now reviewed. 1. Acid-imides.-The copper and magnesium salts are very readily hydrolysed. The dialysis of the copper oxide sol obtained by hydrolysing a 1-2% solution of copper succinimide has been quantitatively followed succinimide being estimated by distilling with concentrated alkali and titrating the ammonia evolved. Even after twenty-one days the solution still contained about 1% of succinimide and was with difficulty prevented from coagulating. Magnesium succinimide was obtained in the form of an alkaline solution by the addition of magnesium filings to mercury succinimide or by the action of mag-ORGANIC CHEMISTRY.i. 151 nesium powder on aqueous succinimide. From such a solution silver nitrate deposited silver succinimide C,H,O,NAg,$H,O in slender columns. On evaporating the solution of msgnesium succinimide magnesium succinamflte Mg(C,R,O,N) 6 H,O crystallised in large hexagonal tablets m. p. 86-88'. The salt is extremely soluble and is considerably dissociatod in solution. Certain indications were obtained of the existence in concentrated solutions of complex salts of the type [(RN),Mg]Na interesting from their connexion with the grouping N,Mg*-* *N in chlorophyll. Ferric salts of succinimide could not be obtained.Compounds containing the above complex anion were obtained in the case of cyanuric acid. Copper acetate and the sodium salt in concentxated solution gave sodium CupricyanuratS as a hetivy violet crystalline precipitate. Claus and Putensen (A. 1889 30) assigned the formula Cu[NH,(C,HO,N,)] to this compound from which the conclusion might be drawn that copper dimethyl cyanurate would not combine with ammonia. A n ammoniacal solution of dimethyl hydrogen cyanurate gives with copper sulphate however a deep blue coloration followed by precipita- tion of the reddish-violet diammine Cu(C,H603N,),.2NH3 which becomes dark blue in a n atmosphere of ammonia probably forming a tetrammine. Claus and Putensen's comporind is therefore a diammine. I I and 111. Acplcarbamicles and Biwet.-No metallic derivatives of acylcarbamides (benzoylcarbnmide and benzoy lphenylcar bamide) could be obtained.Owing to the similarity in colour between the alkali-copper acid-imides of the type (CuS,)M and the alkali-copper biurets (compare Tschug,iev A 1907 i 595) attempts were made to prepare similarly constituted compounds of biuret. Only the known derivative K,CU(C,H~O,N,)~ and a basic salt KCu(C2H,02N3),,3H,0 well-defined reddish-violet column8 could be obtained. rhey are internally complex salt$ akin to those prepared from amino- and hydroxyacethydroxarnic acids ( A . 1913 i 346). IV. Imino-acid-imides. -Sodium dibenzamide and copper acetate gave the normal bluish-green easily hydrolysed coppel. dibenxamide. Iminodibenzamide however gave a lustrous pale greyish-brown copper salt CU(C,,H,,ON~)~.I t dissolves in pyridiue with deep blue colour. Similarly nickel forms a psle yellow very stable s d t . These abnormally-coloured salts are most probably represented by the [ cu (C3H203N3) 4 1 2H,0 Copper cyanurate combines with two molecules of ammonia. ,CP h L - N H formula N/ NCPh*O*i!I ' J. C. W. Some Double and Complex Cyanates. PAUL PASCAL (Bull. Soc. chim. 1914 [iv] 15 11-19).-When a fresh solution of potassium cyanate in 70% alcohol is added to a neutral solution of uranyl nitrate a voluminous yellow precipitate possessing a greenish fluorescence of potassium uranylcyanats [UO,( CNO),]K is produced. This compound is very soluble in water and owing to hydrolysis the solution gives all the tests for uranium.After a few minutes thei. 152 ABSTRACTS OF CHEMICAL PAPERS. aqueous solution deposits a yellowish-orange insoluble compound having the composition [(U02)2( CNO),]K. If t o a 25% solution of uranyl nitrate a tenth of its weight of potassium uranylcyanate is added and then sufficient alcohol to produce a turbidity urccnyl cyanate UO,(CNO) is obtained as a golden-yellow precipitate. On the other hand a solution of potassium uranylcyanate containing 30-50% of potassium cyanate deposits spontaneously the yellow compound [ UO,(CNO),]K. The cryoscopic behaviour of solutions of urnnyl cyanate containing various proportions of potassium cyanate indicates that the compound [UO,(CNO),]K is present in solutions containing a n excess of potassium cyanate. Solutions containing cobaltous cyanate and potassinm cyanate behave similarly indicating that potassium cobaltous cyanate is present ; moreover the molecular magnetic susceptibility is much less than that for ordinary cobaltous salts and in migration experiments the cobalt wanders t o the anode.T. 5. P. Polymerisation of Cyanamide to Dicyanodiamide in Aqueous Solution. G. GRUBE and J. KRUGER (Zeitsch. physikul. Chem. 1913 86 65- 1O5).-The polymerisation of cyanamide in aqueous solution to dicyanodiamide CN*NH2 -+ (CN-NH,) has been investigated by the usual kinetic method and by means of potential measurements I n the latter method the concentration of the hydrogen ion has been determined under different conditions. It is shown that the presence of alkali accelerates the reaction in a marked degree.Ammonia accelerates the velocity of polymerisation to a n increasing amount as its concentration increases; in the case of sodium hydroxide and calcium hydroxide a maximum velocity is reached at a definite concen- tration and at greater or smaller concentrations than this value the velocity becomes smaller The dissociation constant of cyanamide is of the order of magnitude 10-11 and that of dicyanodiamide From these figures it follows that a molecular solution of monosodium cyanamide is dissociated to the extent of 3% and one of mono- ammonium cyanamide to the extent of 79-89%. It is shown that the polymerisation occurs according to the echeme This is proved by the fact that the react.ion has a maximum velocity when the condition (CNNH‘) = (CNNH,) is fulfilled.The method of preparation of dicyanodiamide from calcium nitride is discussed and a method proposed for its preparation without the addition of any further alkali. J. F. S. CN*NH + CN*NH’ = C?N,N,H,’. Chloro- and Bromo-oximinoacetic Acids. J. HOUBEN and H. KAUFFMANN (Bey. 1913 46 4001-4010).-1t was hoped from a n examination of t,he behaviour of chloro- and bromo-oximinoacetic acids to decide whether the easy decomposition of oxalic acid bromide (and chloride) [bromo- and chloro-glyoxylic acids] occurs in the stages COBr*C02H -+ (CO),O + HBr -+ CO + CO + HBr (Staudinger and Anthes A. 1913 i 604) or COBr*CO,H - + HCOBr + CO -+ CO + HBr + CO,.ORGANIC CHEMISTRY. i. 153 The present results however fail to give a decisive answer t o this question.Chloro-oximinoacetic acid (compare Houben and Kauff mann A . 1913 i 1159) when neutralised with sodium hydroxide in aqueous solution gives a green precipitate with coprer acetate a deep red coloration with ferrous sulphate a yellow precipitate with silver nitrate and a yellow turbidity with lead acetate; ammonium salt colourless precipitate from ether. The molecular weight of the acid in acetic acid solution by cryoscopic measurement agrees with that of an acetyl derivative. When heated with acetyl chloride for half an hour the acid is converted into chloroacetoximinoacetic acid OAc*N:CCl*CO,H stellar aggregates of needles m. p. 73-74' (decornp.) which is con- verted by phosphorus pentachloride under light petroleum into chloro- acetoximinoacetyl chloride OAc*N:CCI*COCI a colourleas oil b.p. 93'/13 mm. Ethyl bromo-oximinoacetate (compare Jovitschitsch A. 1906 i 1161) is obtained conveniently by shaking a mixture of the nitrolic acid of ethyl acetate NO,*C( :NOH)*CO,Et with aqueous hydro- bromic acid and ether for two hours when the nitro-group becomes replaced by bromine; the product m. p. 93O was hydrolysed by heating under reflux condenser with ether and hydrobromic acid the resulting brorno-oximinoacetic acid OH*N:CBr*CO,H being found in the ethereal solution Bromo-oximinoacetic acid is a very hygroscopic crystalline substance decornp. a t 1 lo' which unlike the chloro-oximinoacetic acid is decomposed by water with great readiness the products of decomposition being iulminic acid hydrobromic acid and carbon dioxide.If the acid is heated carefully in a vacuum (15 mm.) decomposition occurs and pale yellow crystals form in the ice-cooled receiver but these could not be further examined as they decompose rapidly with formation of gaseous products. These crystals may have been of the formula yo>C:N*OH or CHBr:N*O€€ 0- which would be expected respectively if the decomposition had occurred analogously to either of the two suggested modes of decomposition of bromoglyoxylic acid. D. F. T. Organo-metallic Ester Compounds. 11. Iodo-zinc-ester Com- pounds. BRUNO EMMERT and WILHELM ELLER (Bey. 1913 46 1508-1511. Compare A. 1911 i 846).-Zinc reacts with ethyl iodoacetate in the presence of iodine to form the compound which when rapidly heated decomposes a t 133-140'.I n the presence of ethyl ether reaction takes place more readily and after removal of the solvent a syrupy mass remains which evolves the remainder of the ether when placed in a vacuum for a protracted period. The crystalline compound desxibed above is thereby obtained which is only partly dissolved on protracted contact with ether. When cautiously decomposed by water it yields ethyl acetoacetate which is recognised by its odour and by the ferric chloride reaction whilst with I,Zn,(CH,.CO,*Et),,i. 154 ABSTRACTS OF CHEMICAL PAPERS. dilute sulphuric acid i t gives ethyl scetoacetate and ethyl iodoacetate. On distillation i t yields a few drops of a n iodine-free organometallic liquid which commences to boil at about 180° thereby becoming black and exploding.With anhydrous ethyl alcohol it yields the crystalline additive product 1,Zn (C H,* CO,E t)3 2 E t OH needles. Compounds of zinc with ethyl P-iodopropionnte and ethyl o-iodo- benzoate were also obtained by heating the requisite components; on account of their extreme sensitiveness to moisture they were not further investigated. H. W. Hydration and Certain Other Transformations of 1 1 2- Trimethylcyclopropane. N. KISHNER and C. CHONIN (J. Russ. Phys. Chem. Xoc. 1913 45 1770-1779).-Attempts to nitrate 1 1 2-tri- methylcyclopropane by means of nitric acid (D 1*075) as suggested by Konovalov gave unsatisfactory results the hydrocarbon being either unattacked or at higher temperatures converted into oxidation products. The action of nitric acid in presence of glacial acetic acid causes rupture of the trimethylene ring and formation of products oE hydration of the trimethylcyclopropane.These products consist principally of dimethylisopropylcarbinol but the other possible alcohol methy1tert.-butylcarbinol could not be detected. The action of hydrobromic acid on 1 1 2-trimethyIcycZopropane has already been referred to briefly (compare Kishner A 1912 i 245) but has now been more thoroughly investigated. The bromo-pro- ducts obtained on distillation with aniline to convert them into the corresponding unsaturated hydrocarbons yield Py-dimethyl-AP-butene and py-dimethyl-ha-butene ; the parent bromo-compound of the former of these is P-bromo-Py-dimethylbutane whilst the latter may be derived also from ,8- bromo-py-dimethylbutane or from a- bromo- Py-dimethylbutane. The action of fuming hydrochloric acid on 1 1 2-trimethylcyclo- propane in a sealed tube at 100' gives a theoretical yield of P-chloro- fly-dimethylbutane CMe,Cl*CHMe,(?) b.p. 11 2-1 12*5'/759 mm. DF 0.8724 n 1.4178. T. H. P. Stability of Cyclic Hydrocarbons in Connexion with their Configuration. The Transformation of cycZoHexene in to Benzene and cyctoHexane. J. BOESEKEN and K. H. A. SILLEVIS (Proc. R. Akad. Wetensch. Amsterdam 1913 16 499-506).-When cyclohexoce vapour mixed with carbon dioxide is passed in a slow current over finely-divided nickel heated at lSO' it is converted into benzene and cyclohexane in accordance with the equation The quantity of benzene actually obtained amounted to about 40% the remainder consisting of cyclohexane and from 2-4% of unchanged cyclohexene.Since carbon monoxide was also found to have been yroduced it is probable that the excess of benzene is due to a part of the hydrogen which is formed as an intermediate product in accordance with C,Hl0 = C,H + ZH being used up in the reduction oE the admixed carbon dioxide. The experiments show that cyclohexene at 180' is metastable with 3CfiHio = C6H + 2CfiHIpORGANIC CHEMISTRY. i. 155 respect t o a mixture of benzene and cyclohexane and also with regard to a mixture of benzene and hydrogen. Observations of other workers relative to the stability of the cyclic hydrocarbons are referred t o in connexion with thew experiments. H. M. D. Thermal and Cryoscopic I n v e s t i g a t i o n of Mixtures of Benzene and Ethyl Alcohol.F. VIALA (BuU. SOC. chim. 1914 [iv] 15 5-ll).-The freezing point curve of mixtures of alcohol and benzene has a eutectic point at a molecular concentration of 97.5% of alcohol; no compound is formed. The freezing point of alcohol is found t o be - 113.9'. The specific heat of a mixture of benzene and alcohol is greater than corresponds with the law of mixtures. From the results of the specific heat measurements the author has calculated the heat of dilution of various solutions and then applied Baud's equation (A. 1913 ii 233) in order to determine the molecular weight of alcohol i n benzene solution. It was found that in dilute solutions t h e alcohol consist,s of simple molecules but association takes place with increasing concentration until in concentrated solutions (50% and over) the association into double molecules is complete.The association does not vary with the concentration in the latter solutions and consequently the conclusion is drawn that pure alcohol consists of double molecules at all events between the temperatures 231' and 275.5O abs. T. s. P. The Supposed Benzene-potassium of H. Abelj anz. W. SCHLENE and HEINRTCH MEYER (Ber. 1913 46 4060-4061).- Abeljanz (A. 1876 i 703) described a bluish-black crystalline mass which he obtained by heating benzene with potassium at 240-250' and assumed to be a mixture of the compounds C,H,K and C,H,K,. The authors find that the substance is merely potassium coated with a film of carbon. A considerable quantity of dipheiiyl was present in tho solvent.The evolution of hydrogen under the influence of moisture and the formation of diphenyl and diphenylbenzene which Abeljanz observed are thus accounted for. J. C. W. Vacuum Coal-tar. AMF PICTET and NAURICE BOUVXER (Compt. rend. 1913 157 1436-1439. Compare A. 1913 i 1315).-A study of the t a r obtained by distilling coal under reduced pressure (15-18 mm.) at 4504 The tar contains neither phenols nor aromatic hydrocarbons but yields both on decomposition at a red heat. The tar after washing with dilute sodium hydroxide and dilute sulphuric acid was fractionated under reduced pressure. The fractions were found to contain alcohols and unsaturated hydrocarbons which were not obtain- able in sufficient quantity for identification. Tliese fractions freed from the alcohols by treatment with sodium and from the unsaturated hydrocarbons by treatment with sulphuric acid were then furkher fractionated and two of the principal fractions studied.Fraction I b. p. 172-174' D23 0,7765 r'$ 1,4196 had by analysis and molecular weight determination the composition C,,H,,.i. 156 ABSTRACTS OF CHEMICAL PAPERS. Fraction 11 b. p. 189-191' Dz 0.7838 7'E 1.4234 had the com- position Cl,H2,. These two fractions had all the properties of naphthenes and were identical in physical properties with two hydrocarbons isolated by Mabery from Canadian petroleum (compare J. Amer. Chem. sbc. 1911 33 251). A chemical study of the hydrocarbon CloH20 shows it to be 1 2 4 5-tetramethylcyclohexane and from it the authors have prepared a dibromo-derivative m.p. 202O and a dinitro-derivative m. p. 205'. They suggest that the other hydrocarbon CllH2 is the analogue of the first being pentamethylcyclohexane. W. G. A Method of Synthesis of Benzyl Chloride and its Homo- logues. MARCEL SOMMELET (Compt. rend. 191 3 157 1443-1 445). -Methyl chloromethyl ether reacts readily with benzene and its homologues in carbon disulphide or carbon tetrachloride in the presence of stannic chloride to give benzyl chloride and its homologues The chloro-ether and the hydrocarbon in solution are cooled to - lo' and the stannic chloride run in drop by drop. By this method the author has prepared benzyl chloride p-tolpl chloride and the three dimethylbenzyl chlorides. 3 4-Bimethylbenxyl chloride is a liquid b. p. 116-117'/24 mm.which on heating with potassium acetate in acetic acid yields 3 4- dimethylbertzyl acetnte C6HsMe,*OAc b. p. 146-148'/28 mm. furnishing on hydrolysis 3 4-dimethylbenzyl ulcohol m. p. 62.5-63.5'. 2 4-Dirnethylbenzyl chloride b. p. l l O o gives the acetate '0. p. 157'/50 mm. and the cclcohol b. p. 151-152'/44 mm. yielding a phenyl- uretham m. p. 78-79". These two alcohols on oxidation give the corresponding dimethylbenzoic acids. 2 5-Dimethylbenzyl chloride b. p. 120-121'/28 mm. yields an acetate b. p. 138-141°/28 mm. and an ulcokol b. p. 142-143'1 37 mm. which gives aphenylurethccrze m. p. 86'. All these chlorides are readily converted into the corresponding aldehydes by warming in aqueous alcoholic solution with hexa- methylenetetramine (compare A. 1913 i 1395 j.Friedel and Craft's Reaction. XIII. Action of Chlorides of Ethylene on Benzene. J. B~ESEKEN and M. C. BASTET (Rec. trav. chiin. 1913 32 184-209. Compare A. 1911 i 522 and earlier abstracts),-It has been previously shown that the Friedel and Crafts' reaction takes place when three molecules are present the first of which is unsaturated the second of which can be activated to such an extent that i t is decomposed during the reaction into two parts which then unite with the first molecule and the third of which is a catalyst which can activate the two molecules (compare Boeseken and Prins A. 1911 i 173). In such con- densations in the presence of benzene it has been assumed that the molecule of the latter is always supersaturated but from observations of Sieger (Diss.) the authors are led to the conclusion that this is not invariably the case.I n the present paper they describe a series of condensations of benzene with various chloro-derivatives of ethylene in W. G.ORGANIC CHEMISTRY. i. 157 which in the first stage the molecule of benzene is disrupted and forms a compound with the unsaturated molecule; this latter molecule becomes in its turn disrupted and combines with a second or third molecule of benzene which now behaves as if it were unsaturated. Vinyl chloride b. p. - 1 8 O is obtained by the decomposition of ethylene dichloride a t 600'. A small quantity of carbon is simultaneously formed and in addition to hydrogen chloride the gaseous product contains carbon dioxide carbon monoxide oxygen and unsaturated hydrocarbons in small quantity.For purposes of experiment the vinyl chloride was not condensed but after being freed from hydrogen chloride and dried was passed directly into the reaction mixture. The latter is prepared by warming a mixture of mercuric chloride and aluminium powder with benzene. A vigorous reaction occurs with the formation of a complex product containing mercurous chloride aluminium chioride and benzene which remains for some time a t 0' in supersaturated solution. I n these circumstances vinyl chloride is readily and completely absorbed with the formation of aa-diphenylethane and small amounts of 9 10-dimethyldihydroanthracene m. p. 178-179". Benzophenone is obtained by oxidation of the former the absence of benzoic acid showing that the original product does not contain dibenzyl.A possible explanation of the course of the reaction consists in the assumption of an initial formation of styrene and subsequent combination of the latter with benzene. To test this point styrene (obtained by the rapid distillation of cinnamic acid with the help of a long fractionating column) has been condensed with benzene at 0' in the presence of aluminium-mercury couple. The product appears to be a polymeride of styrene having a mean molecular weight of 250 in benzene solution ; it does not contain aa-diphenyl- ethane. When agitated with aluminium chloride in carbon disulphide solution (without benzene) styrene yields a substance of indefinite b. p. and mean molecular weight 1'72 in benzene solution. The possibility of the intermediate formation of ethylidene chloride has also been examined.Under conditions similar to those used in experiments with vinyl chloride ethylidene chloride gives aa-diphenyl- ethane but the yields are considerably lower than in the original experiments and further augmentation of the concentration of hydrogen chloride diminishes the quantity. Finally the interpretation is adopted that the first stage of the reaction consists in a disruption of the benzene molecule and combination with vinyl chloride to form a-chlorophenylethane Ph*CHCl*CH which in its turn is disrupted and combines with a further molecule of benzene to yield aa-diphenylethane (compare Schramm A. 1893 i 563). The formation of 9 10-dimethyldihydro- anthracene is explained by the assumption of the condensation of two molecules of a-chlorophenylethane.aa-Diphenylethane when prepared by Priedel and Craft's reaction and freshly distilled shows a marked fluorescence. This appears to be due to an impurity formed during the condensation. The fluorescence gradually disappears when the substance is preserved but re-appears oni. 158 ABSTRACTS OF CHEMICAL PAPERS. repeated distillation especially in an oxidieing atmosphere. It can be removed by boiling under ordinary pressure or by treatment with sodium in alcoholic solution. The condensation of the two symmetrical acetylene dichlorides (compare Chavanne A 1912 i 330) with benzene in the presence of aluminium-mercury couple a t 20° has also been investigated. The products obtained in each case are identical consisting of dibenzyl triphenylethane and tetraphenylethane.At Oo reaction is extremely slow. A t 80° reaction is more complete than a t 20° but also more complicated a larger quantity of residue being formed whilst also the amounts of dibenzyl and tetraphenylethane (which are certainly the products of an abnormal reaction) are increased relatively to the amount of triphenylethane. Trichloroethylene CCI,:CHCI scarcely reacts with benzene in the presence of the aluminium-mercury couple a t the ordinary temperature. A t higher temperatures the main products are aa-dipheuylethane and tetraphenylethane and the coiirse of the reaction is represented by the equations CHC'l:CCI + C,H = C,H,*CKC1*CHC12 and It has been previously shown that tetrachloroethylene does not react to a perceptible extent with benzene in the presence of powdered aluminium chloride at 80° (A.1912 i 65). The authors now show that very little action occurs between the pure substances a t 80' in the presence of aluminium-mercury couple. The reaction observed with commercial tetrachloroethylene is attributed t o impurities. The authors are led to the conclusions that the intensity of the action diminishes with an increase in the number of atoms of chlorine and that reaction commences by an addition of benzene a t the double bond. The chlorine atoms do not become active even in tetrachloroethylene but are activated in the additive products. With increase in the number of chlorine atoms the second stage of the reaction becomes more complicated.H. W. C,H,*CHCi*CHCl + 3C,H6 = CH(C,H,),*CH(C,H,) + 3HCI. Quantitative Investigations on the Nitration of the Chlorotoluenes. J. P. WIBAUT (Rec. trav. chim. 1913 32 244-320).-The author has prepared the four nitro-o-chlorotoluenes in a state of purity and has studied the six curves of solidification of binary mixtures prepared from them. He has thus been enabled to investigate the products of nitration of o-chlorotoluene at Oo and has shown by chemical means that all four possible isomerides are formed whilst from thermal analyses the relative proportions of the isomerides are deduced. A similar series of experiments has been effected with rn-chlorotoluene. The results are in accord with the theoretical predictions of Holleman. o-Chlorotoluene has h. p. 159-5O/759.5 mm.(corr.) n69'4 1.4977. 2-Chloro-4-nitrotoluene is prepared by the nitration of o-toluidine dissolved in concentrated sulphuric acid and successive treatment of the nitrotoluidine with nitrous acid and cuprous chloride. It has v P 4 1.5470 m. p. 62*3* instead of 65' and 6s' recorded in the literature. Reduction with 'iron powder in the presence of a small quantity of sulphuric acid converts it into o-chloro-p-toluidine,ORGANIC CHEMISTRY. i. 159 m. p. 23*1°. The acetyl derivative m. p. 105' and the benxoyl derivative needles m. p. 122" have been prepared. 2-Chloro-6-nitrotoluene 7269.4 1.5377 m. p. 35.3" (Green and Lawson T. 1891 59 1017 give 37") is prepared by reductions of 2 6-dinitro~ toluene by hydrogen sulphide in ammoniacal solution and replacement of the amino-group by chlorine in the usual manner.6-Chloro-o- toluidine has m. p. 2%" and rapidly darkens on exposure to air. The acetyl and benzoyl derivatives have m. p. 156' and 170' respectively. For the preparation of 2-chloro-5-nitrotoluene finely powdered acet- o-toluide is added to a mixture of nitric acids (D 1-52 and 1.40 respectively) at 30'. The temperature must not rise above 35-37'. After remaining for twenty-four hours a t the ordinary temperature the product is poured on to ice. The mixture of acetyl compounds is hydrolysed with boiling hydrochloric acid and 3-nitro-o-toluidine removed by passing steam through the mixture accompanied by a smaller quantity of 5-nitro-o-toluidine. Crystallisation from alcohol enables the two isomerides to be obtained in the pure state m.p. 95' and 130' respectively. The yields are unsatisfactory but no improvement could be effected by nitrating in sulphuric acid or acetic acid solution. 2-Chloro-5-nitrotoluene has m. p. 42*9" n69-4 1.551 1. 6-Chloro-m- toluidine has m. p. 83" (acetyl derivative m. p. 92" ; benxoyl derivative needles m. p. 119.5'). 2-Chloro-3-nitrotoluene prepared from 3-nitro-o-toluidine described above has m. p. 22*1° T ' ~ ~ ' ~ 15327 ; the corresponding amino-com- pound (compare Wynne and Greeves T. 1895 67 1548) solidifies at 6.6' (acetyl derivative m. p. 133'; benzoyl derivative needles m. p. 125'). o-Nitrotoluene is nitrated by gradually adding to it four times its weight of nitric acid (D 1.52) a t - 1 to + 1". After treatment with ice and extraction with benzene the product is distilled under diminished pressure.Measurement of the index of refraction of the various fractions proves the absence of unchanged o-chlorotoluene or dinitro-derivatives. Attempts to separate the isomerides by distilla- tion under diminished pressure or by crystallisation at low temperatures were unsuccessful Better results were obtained by reduction and acetylation of the mixture. Crystallisation from benzene then yields aceto-6-chloro-o-toluidide whilst on concentrating the mother liquors aceto-2-chloro-m-toluidide is obtained. 6-Chloro-m-toluidine is isolated during the distillation of the primary reduction product with steam. Neither 2-chloro-4-nitrotoluene nor any of its derivatives could be isolated but the presence of the former is deduced from the fact that by precipitation of the acetylamino-derivatives (remaining after crystal- lisation of the greater bulk of the products) from benzene solution by addition of light petroleum fractions are obtained which show a rise in m.p. on admixture with aceto-2-chloro-p-toluidide. The solidification curves of the six binary mixtures of the o-chloro- nitrotoluenes have been determined. The curve has the ordinary form in the cases of the following pairs of isomerides (in which the methyl group and chlorine atom are in the positions 1 and 2 respectively) and the eutectic point is appended in brackets 1 2 5 and 1 2 6 (7.2') ; 1 2 ; 5 a n d 1 2 3 ( + 1 ° ) ; 1 2 6 and 1 2 4 (17.2"); 1 2 4 and 1 2 3 (8.2') ; in the case of the isomerides 1 2 5 aod 1 2 4 thei.160 ABSTRACTS OF CHEMICAL PAPERS. curve rises to a maximum indicating the existence of a compound formed by a molecule of each component whilst with isomerides 1 2 6 and 1 2 3 a maximum is also found corresponding with the formation of a compound from two molecules of the latter and one of the former The application of Valeton's method (Acccd. Xci. Amsterdam 1910 754) to synthetic mixtures of the four o-chloronitrotoluenes has shown that the proportions of the isomerides 1 2 4- 1 2 5- and 1 2 6- can be determined with sufficient accuracy; estimation of the 1 2 3-isomeride is not directly possible since the formation of the compound 1 2 3- and 1 2 6- causes too great deviations. The mixture of isomerides obtained by the action of nitric acid (D 1.52) on o-chlorotoluene at 0' during one and a-quarter hours is thus shown to contain 43.4% of 1 2 5- 17.0% of 1 2 4- 20.7% of 1 2 6- and (by difference) lS*S% of the 1 2 3-isomeride.3-Chloro-6-nitrotoluene is obtained by the following method which differs in several respects from t h a t described by Cohen and Hodsman (T 1907 91 974) ; aceto-m-toluidide is dissolved in cold concentrated sulphuric acid and nitrated a t - 5Oto - 3 O with a mixture of nitric acid (D 1.52) and Concentrated sulphuric acid. After elimination of the acetyl group 6-nitro-m-toluidine m. p. 135' (instead of 138' recorded in the literature) is obtained which is converted into 3-chloro-6-nitro- toluene in the usual manner. The latter has n68'6 1.5495 and appears t o exist in a stable modification m.p. 24*9O and a metastable form m. p. 24.2O. 3-Chloro-2-nitrotoluene is obtained by the mathod recommended by Brand and Zoller (A 1907 i 755). 1.5204 temperature of solidification 23.4O. Certain observations lead the author to the conclusion that it may also exist in a metastable state. For the preparation of 3-chEoro-4-nitrotoluene n68'5 1.5428 m. p. 24*2O acet-o-toluide has been converted into 5-chloro-4-nitro-o-toluidine m. p. 1 2 4 O according to the method of Claus and Stapelberg (A. 1893 i 580) who give 128O as m. p. Removal of the amino-group of this compound by the usual methods leads to the desired product the structure of which is confirmed by transforming it into aceto-3-chloro- m-toluidide.3-Chloro-5-nitrotoluene has n68'5 15404 m. p. 5S.4O whereas Hanig (A. 1887 1034) gives 55O. m-Chlorotoluene has been nitrated under the same conditions as the o-isomeride. Measurement of the index of refraction of the various fractions proves the absence of unchanged m-chlorotoluene or of its di- nitro-derivatives. The author has not attempted to isolate directly the various isomerides ; but since on theoretical grounds the formation of 3-chloro-5-nitrotoluene is considered improbable he has applied Valeton's method to the determination of the proportions in which the three other isomerides are formed when m-chlorotoluene is nitrated a t Oo ; he thus finds 58.9% 3-chloro-6-nitrotoluene 32.3% 3-chloro-4-nitro- toluene and 8e8yA 3-chloro-2-nitrotoluene. Investigation of the curves of solidification of binary mixtures of these three substances yields normal results in the cases of 3-chloro-4-nitrotoluene and 3-chloro- 2-nitrotoluene (eutectic temperature about - 4.5') and of 3-chloro- It hasORGANIC CHERI1STK.Y.i. 161 6-nitrotoluene and 3-chloro-2-nitrotoluene (eutectic temperature about - 5') ; in the case of 3-chloro-6-nitrotoluene and 3-chloro-i-nitrotoluene a compound appears to be formed from one molecule of each constituent which however is mainly decomposed in the liquid state. By somewhat modifying the theories of Holleman and Huisingiz a mathematical expression has been devised according to which the relative quantities of isomerides obtained by nitrating the chloro- toluenes can be su6ciently accurately calculated.Tri-arylmethyls. XI. Tetra-arylquinodimethanes [Tetra- aryldimetbylenecycZohexadienes]. W. SCHLENK and MAX BKAUNS (Rer. 1913 46 4061-4066).-1t was previously found that the introduction of diphenyl or naphthyl groups in the hexa- phenylethane molecule considerably enhanced the dissociation into free tri-arylmethyl radicles. The conclusion was therefore drawn that the above groups make larger demands on the affinity of the carbon atom than does the phenyl group. I t wag expected that the introduction of these groups into Thiele's tetraphenylquinodimethane [ 1 4-bisdiphenylmethylene-42 ' '-cyclohexadiene] (I) (A. 1904 i 49 1) would similarly loosen tbe fourth valency of the methane-carbon atom and that compounds of the type (11) would be obtained. H.W. (1.1 (11.) The diphenylbidiphenylyl- and bidiphenylyldi-a-naphthylquinodi- methanes (111 and IV) however proved t o be indifferent towards additive agents like Thiele's hydrocarbon. (111.) UV.1 The above conclusion was however justified since it was found that the tetra-aryl-p-xylylene hsloids which contained diphenyl or naphthyl groups dissociated into the quinonoid hydrocarbons and the free halogen erren more readily than did the tetraphenyl compound. Terephthaloyl chloride was condensed with diphenyl by means of aluminium chloride to form p-phmylene bidiphenylyl diketone CGH4[ 00 *C,H,Ph],. The not quite pure compound formed colourless leaflets m. p. 280-285'. I t was converted into diphenylbidiphcnylylxylylene glycol C,H,[CPh(OH)*CoH,Ph] which formed a microcrystalline powder m.p. 1 0 5 O by the action of magnesium phenyl bromide and into di-a- m~piithylbidipiLenyl?/l~~l~Zene glycol C,,H,,O by the action ol magnesium a-naphthyl bromide. Both cornpounds gave deep blue solutions in sulphuric acid. Hydrogen chloride converted them into the correspond- i ng xylylene dicldorides C,,H,,CI and C,,H,,CI,. The former gave white crystals m. p. 254' (decomp.) and the latter crystallised with 2C,H6 and they both gave coloured solutions on boiling in benzene or xylene owing to dissociation into the quinodimethanes. These compounds were prepared by boiling benzene solutions with copper powder in carbon dioxide atmospheres. Diphenylbidiphenylylquino- VOL. CVI. i. mi. 162 ABSTRACTS OF CHEMICAL PAPERS dimethane [ 1 4-bisphenyldiphenytyl~ethylene-h2 K-cyclohexadiene] (111) forms bright red slender needles decomp.about 200' and di-a- naphthylbidiphenytytquinodimethalne [ 1 4 - big - a - naphthyldiphenytyl- methylene-h2 5-cgclohexadiene] (IV) forms orange needles m. p. above 290'. Their solutions are strongly fluorescent but soon become colourless in the light. They do not absorb iodine and are scarcely affected in the air. J. C. W. Combinations of Aniline with Hydrochloric Acid. J. C. THONUS (Proc. K. Aknd. wet end^ Amsterdam 19 13 16 553-555)- The equilibrium conditions characteristic of the system anilice-hydro- chloric acid-water have been investigated. The following compounds occur as solid phases at 0' GNH,Ph,HCl ; 5NH,Pb,3HCl,H20 ; NH,Ph,HCl ; lONH,Ph,llKCl. Of the above four compounds the first and second cease to exist a t 25O whilst experiments a t 35Oshow that solid phases occur of the composition 4NH2Ph,5HCI ; lONH,Ph,liHCl ; NH,Ph,HCl; SNH,Ph,HCl ; 3NH2Ph,HC1.solid phases and solution a t 0'. A diagram is given showing the relationship between the various H. M. D. P r e p a r a t i o n of N-Monoalkyl Derivatives of Homopiper- onylamine. HEHMAN DECKER (D.R.-P. 267700). - N-Monoalkyl derivatives of homopiperonylamine are prepared by treating alkylidene derivatives of homopiperonylamine with alkylating agents in absence of water and decomposing the resulting quaternary ammouium compounds. Benzylidenehomopiperonylamine when treated with methyl iodide furnishes ultimately N-methylhomopiper- onylamine b p. 156-158'/24 mm. (corr.).This forms a hydrochloride m. p. 178-180° a nitrate m. p. 166-167' (corr,) a carbonate m. p. 72-75' and a hydviodide m. p. 135-1 36' (corr.). N-Ethythomopiper- onyZamine similarly prepared by means of ethyl iodide has m. p. 1 26 - 1 28'. J. C. 0. Preparation of 1 6-Dibromo-2-naphthylttmine. HARTWIG FRANZEN and ADOLF EIDIS (J. pr. Chem. 1913 [ii] 88 755-764). -The preparation of 1 6-dibromo-2-naphthylamine (compare Claus and Jack A 1898 j 324; also Claus and Philipson A 1891 461) is best accomplished by the bromination of benzylidene-/3-naphthyl- amine in the manner indicated below; the isolation of the inter- mediate products is however unnecessary. Benzylidene-P-naphthyl- amine combines with bromine in chloroform solution to form a dibromide CIoH,*NBr*CHPhBr which separates as a light yellow crystalline powder and is converted by boiling in alcoholic solution into 1- bromo-2-naphthylamine.This crystallises in colourless needles m. p. 63-64' and condenses with benzaldehyde in alcoholic solution yielding zt benxytidene derivative (slender yellow needles m. p. 94') which unites with bromine in chloroform solution to form a dibyornide CloH,Br*NBr*CHPhBr m. p. 220° with previous sintering and darkening at 210'. When boiled in alcoholic solution the dibromide yields 1 6-dibromo-2-naphthylamine.ORGANIC CHEMISTRY. i. 163 Bromination of aceto-/3-naphthylamide in glacial acetic acid gives rise to acsto - 1 - bromo-P-naphthg2arnk-h hydrobromide C,?H,,ONBr a colourless crystalline powder m p. 180-190' (compare Cosiner A 1881 605).On treatment with ammonia or when crystallised from alcohol the hydrobromide is converted into the free amide which is hydrolysed by alcoholic hydrogen chloride to l-brom0-2-naphthylamine identical with that obtained by the method described above but different from the compouud m. p. 76-79' described under the same name by Morawski and Gltiser (A. 1888 1096). F. €3. Acetylation of Organic Compounds. E. KNOEVENAGEL (Annalen 1913 402 lll-l48).-1t has long been known that the presence of sodium acetate sulphuric acid or zinc chloride greatly facilitates the acetylation of organic compounds by acetic anhydride but the manner in which the catalyst operates is still obscure. It does not act in virtue of its dehydrating power because not only is a very small amount of the catalyst employed but also other substances such as strong acids or metallic wlts of strong acids which are devoid of desiccating propertiee can also function as catalysts.Different catalysts produce different results and a t present generalisatione cannot be stated but this much is certain that the action of a catalyst is not dependent simply on the concentration of the hydrogen ions or on its dehydrating power Acetylatiou by means of acetic anhydride is not limited to organic compounds containing hydroxyl groups ; with a suitable catalyst oxidee especially aldehydes likewise certain unsaturated and ethereal compounds such as ethers unsaturated ketones open and cyclic anhy- drides of polyhydric alcohols amongst them poly saccharides dextrin stsrch and cellulose can unite with acetic anhydride in consequence of a more or less easy rupture of the linking between carbon and carbon cr carbon and oxygen.When the acetate formation is accompanied by a degradation of the organic molecule the phenomenon (which is quito analogous to ordinary hydrolysis) is termed acetolysis. The acetylation of hydraxylic compounds the acetate formation in the case of the compounds mentioned above and the occurrence or non- occurrence of acetoly sis are dependent apart from the temperature and the quantity of the catalyst 60 largely on the nature of the catalyst that it is possible under selected experimental conditions sim iiltarieously to acetylate hydroxyl groups to promote acetate forma- tion at a carbon-oxygen linking and with suitable Substances to degrade the molecule or sometimes to leave it undegraded.The preceding processes are being applied to elucidate the constitu- tions of the polysaccharides dextrins starch and cellulose ; the present paper however deals with their application to simpler classes of compounds. The catalysts employed are ferric chloride zinc chloride hydrated stannous chloride phosphorus trichloride sulphuric acid sulphoacetic acid hydrated ferrous sulpbate hydrated copper sul phate hydrated zinc sulphate ammonium sulphate and methylamine sulphate. The last two although active in the acetylation of hydroxyl groups are without effect in the conversion of aldehydes into diacetates. This m 2i. 164 ABSTRACTS OF CHEMICAL PAPERS. conversion is readily effected by all of the other catalysts the modus operandi being to mix the aldehyde (1 mol.) acetic anhydride (1.1 mol.) and cataljst (1-5%) a t the ordinary temperature external cooling being employed to keep the temperature below 70".Benzaldehyde gives an almost quantitative yield of benzylidene diacetate in the presence of copper sulphate or zinc chloride and also with sulphuric acid or ferric chloride when the reaction is commenced in a freezing mixture. Similar experiments have been performed with furfural- dehyde (at 0-loo) cinnamaldehyde (ferric chloride acts too vigorously but by dilution with glacial acetic acid a quantitative yield of cin- namylidene diacetate is obtained) vanillin (stannous chloride produces a nearly quantitative yield of the triacetate) anisaldehyde (successful results obtained only with stannous chloride phosphorus trichloride and copper sulphate ; anisylidene diacetate bas m.p. 67O) piperonal (quantita- tive or nearly quantitative yields with ferric chloride stannous chloride sulphoacetic acid ferrous sulphate and copper sulphate ; piperonyli- dene diacetate has m.p. SOo) and protocatechualdehyde (the tetra-acetate has m. p. 131'; no success is obtained with copper sulphate). Salicylal- dehyde yields the triacetate m. p. 1 0 3 O or disalicylaldehyde m. p. 1 2 9 O or a mixture of both; the triacetate is tho main product in the presence of ferric chloride zinc chloride or sulphuric acid whilst disalicylaldehyde is entirely or mainly produced when the catalyst is phorphorus trichloride copper sulphate or zinc sulphate ; the remain- ing catalysts produce an approximately equal molecular mixture of both substances.In all cases disalicylaldehyde is the primary reaction product and is subsequently converted into the triacetate in the presence of a suitable catalyst. The converse change OP the triacetate into disalicylaldehyde cannot be accomplished. Salicylaldehyde and acetic anhydride yield the triacetate in the presence of 2% of sulpho- acetic acid but disalicylaldehyde when about 50% of the catalyst is employed ; in the presence of acetylsulphuric acid only the triacetate is formed Paraformaldehyde and acetaldehyde are readily converted into the corresponding alkylidene diacetates by acetic anhydride and a little sulphuric acid or ferric chloride respectively. Ketones unlike aldehydes do not react additively with acetic anhydride to form diacetates even in the presence of the preceding catalysts.p-Benzoquinone and acetic anhydride however react in the presence of a little ferric chloride to give an almost quantitative yield of Thiele's hydroxyquinyl triacetate m. p. 9 6 O . [With JuNa and RUMscHI~.]-styryl methyl ketone and acetic an- hydride react in the presence of a little sublimed ferric chloride to form after three to four days 5-acetoxy-2-acetyl-3 ; 4diphenyZ-l-naethyl-A~-cyclo- pntene CMe<OH(oAc),CHPh m. p. 1 1 8 O by the hydrolysifi of which is obtained li-hydroxy- 2-acet yl-3 ; 4 -diphen yl- 1 -meth yl-A'- c yclopentene 020H2002 m. p. 95' (oxime m. p 103' [decomp.]; phenylhydraxone m. p. 9 4 O [decomp.] ; tribromo-derivative C20H,,0,Br3 m.p. 1 8 3 O [decomp.]) ; the preceding acetate forms a tribrorno-derivative C22H,10,Br m. p. 194' (decomp.). CAc-FHPhORG Ah7 C CHEMISTRY. i. 165 [With Runrsc~1~.]-Ethy1 ether is almost mattacked by acetic anhydride a t 100' in the presence of sulphriric acid potassium hydrogen sulphate pyridine hydrogen sulphate or ammonium per- sulphate. I n the presence of a little ferric chloride however acetolysis occurs t o a slight extent ethyl acetate being formed. Epichlorohydrin and acetic anhydride react at the ordinary temperature in the presence of ferric chloride (but not of copper sulphate or srilphuric acid) to give an almost quantitative yield of diacetin-a-chlorohydrin OAc*CH 2*CH( OAc) CH,Cl b. p. 116-118'/11 mm. Epichlorohydrin glacial acetic acid and a little ferric chloride react at the ordinary temperature t o form a-acetin-y-chlorohydrin OAc*CH,*CH(OH)*CH,Cl b.p. 120-1 21°/ 14 mm. in about 90% jield. [With JAKOB TRANSIER.]-cim?Ol and acetic anhydride react to form according to the temperature and the catalyst employed a terpineol acetate b. p. 104-106'/10 mm. and terpin diacetate b. p. 145'/' 14 mm. in varying amounts; the most effective catalysts for the rupture of the cineol ring are sulphuric acid and ferric chloride. The terpineol thus formed is appnren tly identical wit,h the compound syrithesised by Perkin (T. 1904 85 654) since i t has b. p. 102-103O/ 14 mm. D"," 0.9337 and lz; 1.4783 and forms a phenylurethane m. p. logo and nitrosochloride m. p. 117'. Cineole and glacial acetic acid yield dicinene by warming with a little concentrated sulphuric acid on the water-bath.The preceding experiments illastrate well the different results obtained by the action of acetic anhydride in the presence of various catalysts. The 6-ring in cineole is readily ruptured in the presence of many catalysts the 3-ring in epichlorobydrin by only a few catalysts at the ordinary temperature whilst the acetolgsis of ethyl ether is only slight even with the strongest catalyst at 100'. [With J u N a and RUjnisc~~h'.]-Ethyl furylidenemalonat,e and acetic anhydride in the presence of n little sublimed ferric chloride react ultimately on the water-bath to form a substance C,,H,,07 or C,,H,,O m. p. 70° from which an acid C15H2007 or C1H160fi m. p. 1444 is obtained by boiling with wntef or hydrochloric acid m d a n acid m.p. 192-195' (barium salt C,,H?,O6Ba,4H,0) by boiling with a n excess of aqueous barium hydroxide. The investi- Compounds of the Aminophenols with Zinc Chloride Bromide and Iodide. A. KOPPITZ (J. pr. Chem. 1913 [ii] 88 744-754).-The three isomeric aminophenols combine with zinc haloids in aqueous solution yielding additive compounds of the formula 2CGH,0N,ZnX,. On account of the tendency of the aminophenols to form basic zinc compounds the preparation of the additive compounds is best accomplished by dissolving the aminophenols in n hot concentrated solution of the zinc haloid. The compounds are decomposed by water but may be crystallised from strong solutions of the corresponding zinc haloid.The compounds of zinc chloride with 0- and p-aminophenols crystallise in reddish-violet needles m. p. 184" and 247' gstion of these substances is being continued. c. s.i. lG6 ABSTRACTS OF CHEMICAL PAPERS. resp~,ctively ; the corresponding cornpounds of zinc bromide in light brown or reddish-brown needles m. p. 170' and 234'. The compound of zinc iodide with p-aminophenol forms dark brown needles m. p. 208'. The compounds formed by m-aminopbenol with zinc chloride (lu.;trow white silky needles m. p. 235') zinc bromide (m. p. N S ' ) and zinc iodide (white needles m. p. 190') are also described. All the above compounds melt with decomposition. F. B. Nitration of Acyl DerivrttiverJ of m-Aminophenol and m-Anisidine. F R ~ D ~ R I C REVERDIN and KARL WIDMER ( B e y .1 9 13 46 4066-4076).-The knowledge of mono- and dinitro-derivntives of these bases is extended by the preparation of some new compounds and by a re-examination of several of those already known. I. A cyl Derivntires of m-Aminophenol. -m- Acet y laminophen y 1 acetate (Ikuta A 1893 i 265) was prepared by heating m-aminophenol with acetic anhydride and sodium acetate at 150-160'. It was accompanied by the unstable tyiacetyl derivative C,H,(OAc)*NAc which is less soluble in water than the diacetyl compoiind and h w m p. 75-7 7'. Toluene- p-sulphonyl-m-aminophenol 0 He C,H,*NH*S02* C6H M e (Reverdin and de Luc) has m. p. 158' and yields with acetic anhydride the acetory-derivative C,,H,,O,NS in white needles m. p. 166'. Tho di-toluene-p-sulphonyl-m-aminophenol C,oH,,0,NS2 forms white needles m.p. 110'. 11. Nitration o j the Diacyl Compounds.-By the action of nitric acid on m-acetylaminophenyl acetate Meldola obtained a mixture of 4- and 6-nitro-3-acetylaruinophenols (T 1906 89 925). This coiild not be confirmed. The addition of the compound- to fuming nitric acid below 4O resulted in the formation of 6-nitrodiacetyl-m-anainophenol [6-nitro- 3-acetylaminoplhenyl acetate] CloH1005N2 as white needles m. p. 11 3' which were hydrolysed by warm sodium carbonate to 6-nitro-3-acetyl- aminophenol (m. p. ZOO'; Meldola 221'). The latter was completely hydrolysed by means of 6% hydrochIoric acid t o 6-nitro-3-aminophenol (Meldola) the constitution of which was confirmed by conversion into 3-chloro-6-nitrophonol and 6-aitroresorciuol.Other mononitro- derivatives could not be obtained By dissolving the diacetyl compound in a mixture of fuming nitric acid and acetic anhydride 4 6-dinitro-3-acctylaminophenyl acetate mas obtained in almost colourless needles m. p. 157'. It was hydrolysed by sodium carbonate to the N-acetyl derivative and finally by con- centrated sulphuric acid to 4 6-dinitro-m-aminophenol (both described by Rleldola). Other acyl derivatives could likewise only be nitrated in the mme positions. Thus rn-benzoylaminophenyl benzoate (Ikuta loc. cit.) yielded 4 6-dinitrobeizzoyl-m-aminophenyl benzoate C,H,(NO,j,*CO~NH-C,H,*OBz as a white crystalline powder m. p. 70-72'. Similarly the above toluenesul phony1 derivative formed 4 6-dinitrototuei~e-p-sul~l3lonyl-m- ctxi&nophenyl toluene-p-sulphonate C,oH170,N,S2 in whito needles m. p.120-123'.ORGANIC CHEMISTRY. i. 167 111. Nitration of m-Acetanieididu.-By the addition of rn-acet- anisidide to nitric acid (D 1*4) 4-nitro- and 6-nitro-derivatives were obtained. The former waa extracted by light petroleum. Meldola (Zoc. cit.) obtained it by other means. It was diazotised and coupled with P-naphthol .giving a red azo-compound C17H,,0,N m. p. 2029 The insoluble 6-nztro-m-acetarcisidide crystallised from water in golden- yellow needles m. p. 165O which were further nitrated to the 4 6-dinitro-compound of proved constitution (Meldola). It W ~ E I hydrolysed to 6-nitro-rn-anis~dhedine OMe*C6.H3(NH,)*NO2 which formed dark yellow needlm m. p. 169'. By nitrating m-acetaniaidide in glacial acetic acid solution a 2-nitro-compound was also formed.It was the first to crystallise from a mixture with the 6-nitro-derivative dissolved in hot water. 2-Nitro-rn-acetanisidide is a brown powder m. p. 2 6 5 O and 2-nitro-m-anieidine forms lemon-yellow needles m. p. 143O. The only known dinitro-derivative was the 4 6-member (Meldola). Using fuming nitric acid in acetic acid solution 2 4-dinit.ro-m- acetaizisidide has also been obtained. It is less soluble in water than the 4 6-isomeride and forms pale yellow needles m. p. 202". It was hydrolysed to 2 4-dinitro-m-anisidine (Blanksma A 1909 i 150). Using acetic anhydride as the solvent 2 6-dinitro-m-acatanisidide was also obtained. It crystallises from hot water as a middle fraction between the 2 4- and 4 6-isomerides.It forms white needles m. p. 190° and is converted by nitric acid into the 4 6-compound 2 6-Dinitro-macnisidins forms dark yellow needles m. p. 146'. The constitution is assumed from the improbability of a substituent entering position 5 owing to the ortho-para orientating influence of the methoxy- and amino-groups. Other dinitro- and also trinitro-compounds could not be obtained by nitration. J. C. W. A New Synthesis of Higher Phenols. RIHO MAJIMA and IKUYA NAHAMURA (Ber. 1913 46 4089-4095).-The preparation of tetra- penta- and hexa-decylveratroles and some allied products is dexcribed. Myristyl chloride was condensed with veratrole by means of aluminium chloride giving a 25% yield of veratryl tridecyl &tom ( 3 4-dimethoxybensoyZtridecccne) C6H3(OMe)2'COoC13~,7 which formed long slender bulky needles m.p. 74-75' and was oxidised by nitric acid (D 1-12) to 3 4-dimethoxybenzoic acid. The oxime formed thick needles m. p. 54-55' whiuh were reduced by aluminium amalgam to a-veratryZtetradecyZamine the .hydroch~orids of which formed prisms m. p. 199O. The phosphate of the base was distilled under 0.8 mm. when tetradecsnylveratro~6 (a-veratryl-Aa-tatradecene) C6H,(OMe),*CH:CH0C1~HPL was obtained in thin plates m. p. 38-40°. This was reduced by platinum and hydrogen to tetradecylvemtrole (a-oeratryztetr~deca?~) C,H,(OMe),*C,,H, which formed long flat crystals m. p. 49-503 b. p. 185-190°/0*5 mm. Hubsequently the same compound was ob- tained by the direct reduction of the above ketone by means of zinc amalgam (Clemmensen A 1913 i 733).Pentadecyl chloride was also condensed with veratrole yieldingi. 168 BBSTRACTS OF CHEMICAL PAPERS. verat$ tetradecyl ketone C25H3803 b. p. 220°/0*3 mm. m. p. 64-65' which was reduced t o pentadecylveratroZe (a-veratrylpentadecane) CpH,(OMe),*Ci,H, b. p. 1 85-195"/0*5-0*8 mm. m. p. 50-51". Similarly palmit,yl chloride yielded veratryl pentadecyl ketone b. p. 230°/0*5 mm. m. p. 79-80" which was reduced to hexadecylueyatrole (cetylveratrole) C,H,(OMe),*C,,H, b. p. 1 90-200'/0*5 mm. m. p. 56-57". For the preparation of pentadecoic acid methyl myristate was re- duced by Bouveault's method t o tetradecyl alcohol which was converted into tetradecyl iodide b. p. 192-195"/17*5 mm. then into the nitrile b. p. 181-185'/23 mm. DT 0.8187 and this was hydrolysed. The method is alternative t o that devised by Le Sueur (T.1905 87 1898). J. C. W. Benzhydrol P r e p a r a t i o n of s-Tetraphenylethane. PAUL SABATIER and M. ~S~URAT (Compt rend. 1913 157 1496-1500).-A ritsum6 for the most part of the chemistry of benzhydrol. Attempts to prepare this compound by the action of water on the Grignard compound CHPh,*ONgBr gave only 3% of the desired alcohol a little benzophenone a large amount of diphenylmethane and s-tetraphenylethane the two hydrocarbons being separable by their solubilities in alcohol. Their formation is explained by the equattions 3CHPh,*OH = 2H,O + COPb + CHPb,*CHPb 2CHPh,*OH = H,O + COPh + CH,Ph and is comparable to the action of ethyl alcohol and benzhydrol (compare Schmidlin and Banus A 1913 i 34).JOSEPH SAMUEL HEPBURN (Chem. Zentr. 1913 ii 2052-2053; from J. Franklin Inst. 1913 176 405-452).-The brains of the sheep after dehydration with alcohol are extracted with three portions of ether at the ordinary temperature. Lipoids are removed from the concentrated ethereal solution by precipitation with acetone. The residue from the ether- acetone solution of each fraction is then saponified in two ways either by hot alcoholic potassium hydroxide or by sodium ethoxide a t the ordinary temperature. All six preparations of cholesterol so obtained melted between 148.4' and 149.1". Two specimens of cholesterol obtained from gall stones had m. p. 147.4" and mixtures of the different specimens had 113. p. 147.7-148". Determination of the iodine number according to the methods of Hiibl Havas or Wys gives uniformly too high values for cholesterol so that the usual iodine reagents cannot be employed for the volumetric determination of cholesterol in fat.The author has also examined the various gravimetric methods which have been recommended for the estimation of cholesterol. R i t t e i s process (A. 1902 ii l l l ) depending on the direct weighing of free cholesterol gives unsatisfactory results. Concordant analyses could not be obtained either by use of carbon dioxide or of hydro- chloric acid to neutralise the excess of sodium ethoxide employed for saponification; the yields obtained by use of hydrochloric acid were very bad. Excellent duplicates were obtained by Cappenberg's method (Chem. Zeit. 1909 33 985) but the yield was about 94%.The W. G. Biochemical S t u d i e s on Cholesterol.ORGANIC CHEMISTRY. i. 169 estimation of cholesterol as benzoyl derivative (Dorde and Gardner A. 1908 ii 515) is not quantitative giving as a n average 42*S6% of the cholesterol actually present. The best and most trustworthy method is that recommended by Windaus (A 1910 ii 462) which consists in precipitation and weighing as digitonin-cholesteride. The average yield is 97.3704. isoCholestero1 Coprosterol and the Classification of the Sterols. CHARLES DOR~E (Biochem. J 191 3 7 616-621).-A useful summary of the properties of the ctiief sterols of animal origin (zoosterols) and of plant origin (phytosterols) is given. Animal cholesterol is modified when excreted by the skin glands into isocholesterol or by the intestinal juices into coprosterol.Phyto- sterol is also converted into coprosterol in the intestine and by excretion as in rubber into isocholesterol. The term metasterols is suggested for the derived sterols coprosterol and isocholesterol. Spongosterol (Henze) is probably in the same class. The position of stigmasterol (Windaus and Hauth A. 1907 i 129) and of brassicasterol (Windaus and Welsch A 1909 i 229) is uncertain. Of the chemical relation- ships between cholesterol phytosterol and the metasterols little is known at present. W. D. H. The Oxidation of Coprosterol and Goprostanone. JOHN ADDY- MAN GARDKER and WILLIAM GODDEN (Biockem. J. 1913 7 SSS-595). -When coprosterol (C21rH,80) is oxidised with the theoretical amount of chromic acid the ketone coprostanone (Cz7H,,0) is produced but the yield is only 60% (Dorbe and Gardner T.19OS 93 1628). I n the present work a n excess of chromic acid was used and tha yield was 70%. The sodium hydroxide extracts on acidification also yielded an acid which was crystallisable (m. p. 24c0) and a n oily substance which was not further investigated. Analysis of the acid and of its salts showed i t had the composition C(27H4604. Direct oxidation of coprostanone with chromic acid also gave a small yield of the same acid. After oxidation with ammonium persulphate the products obtained were separated into three fractions all of which were obtained in crystalline form Some preliminary details are given of their properties and further work is in progress.Molecular Rearrangements in the Camphor Series. XII. Derivatives of i8oCamphoric Acid ; Decomposition Products of Aminoisodihydrocampholytic Acid. WILLIAM A. NOYES and LLOYD F. NICKELL (J. Amer. Chem. Soc. 1914 36 118-127).-Noyes and Potter (A. 1912 i 786) have studied the decomposition of aminodihydrocampholytic acid with nitrous acid. Similar work on aminoisodihydrocampholytic acid has been carried out by Noyes and Knight (A. 1911 i 11 l) but the products were not fully investigated. An improved method has now been devised for preparing the latter acid and a further study has been made of its decomposition with nitrous acid acid C02H*CH<c”e2 ?Me*NH2y m. p. 235-236’ [a] - 32.9’ furnishes a hydrochlwide m. p. 296-298’ H. W. IV. D. H. A mi noisodih y drocampholy t ic C H -CH1.170 ABSTRACTS OF CHEMICAL PAPERS. (decomp.) [a? - 45.8' which when heated with acetic anhydride and sodium acetate is converted into the anhydride of d-aminodihydro- campholytic acid. On decomposing aminoisodihydrocampholytic acid with nitrous acid the following products are obtained d-cam- pholytic acid 35.8% ; I-campholytolactone 4.2% ; Z-tvans-hydroxy- dihydrocampholytic acid 19.1% ; and a small quantity of isolaurolene. The Walden inversion occurs only to a slight extent in this decom- position. Z-Campholytolactone has m. p. 114-1 IS' [a] - 8.2" (in alcohol) and on hydrolysis is converted into I-cis-hydroxydihydrocampholytic acid m. p. 117-118' [aID - 53.1' (in alcohol). d-Campholytic acid has Dii 1.006 and [a] +66-35'. The formation of this acid in the decomposition furnishes direct evidence that it is the secondary asymmetric carbon atom of d-camphoric acid which rearranges to form I-isocam phoric acid.E. G. J. J. BLANKSMA (Chena. Weekblad 19 14 1 1 5 9-6 l).-Five halogen-nitro-derivatives of benzoic acid have been prepared. Bromination of 5-nitro-3-amino- benzoic acid yields 2 4 6-tribromo-5-nitro-3-atninobenzoic acid reddish- brown crystals m. p. 117'. The methyl ester of the former acid is converted by the diazo-method into methyl 3-b~omo-5 -nilrobenxoccte colourless crystals m. p. 70° converted by saponification into the corresponding acid. Acetylation of ethyl 5-nitro-3-aminobenzoate produces ethyl 5-nitro-3-acetylaminobenzoate colourless crystal$ m. p. 168'. Saponification of the former ester yields the corresponding 5-nitro-3-aminobenzoic acid orange-red crystals m.p. 208'. Bromina- tion transforms ethyl 5-nitro-3-aminobenzoate into ethyl 2 4 6-tri- 6~om1-1-5 -nitro-3-aminobenxoate light brown crystals m. p. 96' ; and the diazo-method converts it into ethyl 3-chloro-5-nitrobertxoate colourless crystals m p. 54' saponifiable to the corresponding acid Ethyl 3-bromo-5-nztrobenxoate is prepared analogously to the methyl ester and has m. p. 44'. Halogen-nitro-derivatives of Benzoic Acid. Saponification converts it into the corresponding acid. A. J. W. Investigation of Binary Systems Containing Beneoyl Chloride and Various Other Organic Compounds. B. N. MENSCHUTKIN (J. Russ. Phys. Chem. Soc. 1913 45 1701-1709). -The author has investigated the temperature-concentration diagrams of seven systems all of which gave perfectly similar diagrams no molecular compound being formed in any case.The different systems were composed of benzoyl chloride and (1) benzene this sbowed although not very clearly a eutectic arrest corresponding with a eutectic point at - 26.8' the composicion being COPhCI,1*03C6H,; (2) p-xylene the eutectic lies at - 18.5' and corresponds with COPhCl,O*6 5C6H,Me2 ; (3) rnesitylene the eutectic is situated at about - 70° the corre- sponding composition being COPhC1,4*7C6H,Me ; (4) chloro- benzene eutectic. point - 54-2O composition COPhCl,S*lPhCl ; (5) nitrobenzene - 20.5' and COPhCl,l.14Ph*NO2; (6) diphenyl - 8' and COPhCl,c).2C6H5Ph ; (7) diphenylmethane - 15' and COPhCI,0*42CH2Ph2.The detailed results of the measurements are given in tabular form. T. €1. P.ORGANIC CHEMISTRY. i. 171 p-Nitroso - N - phenylglycine and p-Nitroso - N - o - carboxy- phenylglycine. J. HOUBEN (Ber. 19 13 46 39S4-4000).- Previous attempts to cause phenylcurboxymethylnitroso imine NO*NPh*CH,*CO,H to undergo rearrangement into the corresponding p-r;itrosophenylgl ycine have proved unsiiccessfczl (Fischer and Hepp A 1887 1115; Fischer A. 1899 i 349) the isolation of a con- siderable quantity of p-hydroxylaminobenzenediazonium chloride OH*NH*C,H,*N,CI indicating that the greater portion of the starting substance must have lost the nitroso-group without undergoing re- arrangement. The autbor regards the rearrangement which generally occurs as the result of two reactions the first involving the scission of the nitrdso- radicle and the second its introduction into the nucleus and finds that by the action of sodium nitrite and fuming hydrochloric acid the above p-nitrosophenylglycine can be produced in excellent yield.I n a similar manner o-carboxyphenylcarboxymetbylnitrosoamine can be converted into the hydrochloride of p-nitroso-o-carboxyphenylglycine. Pbenylglycine was prepared by boiling 9.5 grams of aniline with 9.3 grams of chloroacetic acid in a solution of 4 grams of sodium hydroxide in 60 C.C. of water for a few minutes; the clear yellow solution obtained when cooled in ice deposited phenylglycine in approx. 85% yield. A solution of nitrosyl chloride was obtained by adding 50 grams of sodium nitrite to one litre of concentrated hydrochloric acid cooled in a freezing mixture keeping the closed flask cold until most of the nitrite was decomposed and a deep reddish- yellow solution was produced. To 650 C.C.of this ice-cold solution 50 grams of finely granular phenylglycine was introduced and the mixture shaken in the closed flask until the substance was entirely con- ver ted into a brownish-yellow powder which was pure p-nitrosophenyl- glycine hydrochloride the yield Frequently attaining 80%. The conversion of phenylglycine into this nitroso-derivative can also be effected by the direct addition of sodium nitrite to a mixture of phenylglycine with concentrated hydrochloric acid the yield being the same. The hydro- chloride readily loses its acid and when washed with water passes into the free p-nitroso~henyZgZyc~ne a sparingly soluble brown substance which decomposes explosively when warmed ; with concentrated ammonia solution i t gives a green colour due to the green amnzonium salt which can be separated by the addition of alcohol. It is interest- ing to note that the isomeric 5-nitroeo-N-methylanthraniiic acid is green and the suggestion is made that the brown colour of the sub- stance just described may be due to a quinonoid configuration N OH:C,H,:N*CH,*CO,H. Preliminary experiments with p-nitrosophenylglycine show that i t may be successfully used for the preparation of dyes condensation with a-naphthol a-naphthylamine m-tolylenediamine gallic acid ethyl gallate and p- naph t hylamine producing respectively reddis h-violet violet-blue blue reddish-violet red and reddish-brown substances.Under the influence of sodium methoxide solution condensation can be effected with p-nitrobenzyi cyanide to an azomethine compound deep red needles from toluene solution. LV-Phenylmethylglycine is conveniently obtained by heating togetheri. 172 ARSTRACTS OF CHEMICAL PAPERS. methylaniline and chloroacetic acid with sodium hydroxide solution for four hours; the resulting base is extracted with ether and precipitated as the hydrochloride; the yield is 74% of the theoretical; hydro- chloride m. p. 215-216' (compare Hinsberg and Rosenzweig A. 1895 i 144). Attempts to prepare a nitroso-derivative were unsuc- cessful this result indicating that the introduction of a carboxyl group into one of the methyl radicles of dimethylaniline checks the formation of a para-nitroso-derivative.This view is supported by the action of sodium nitrite and concentrated hydrochloric acid on methyl dimethylanthranilate which gave rise only to a very small quantity of methyl 5-nitrosomonomethylanthranilate. Treatment of o-carboxyphenylglycine with nitrosyl chloride or sodium nitrite as described above for phenylglycine effects almost quantitative conversion into the yellowish-brown hydpocldoride of p-nitroso-o-cnrboxyphenylglycine which decomposes above 1 00' without- melting The free base NO-C,H,(CO,H)*N H*CH,*CO,H obtained by the action of sodium carbonate or acetate on the hydro- chloride is a green solid which rapidly changes when moist to a brown colour possibly of a quinonoid form.It decompoFes when heated and condenses with benzyl cyanide in the presence of sodium methoxide giving a yellow a zomethine derivative malonitrile cnder similar conditions giving a blood-red azomethine compound ; the azomethine compound CO,H*CH,*NH C,H,( C0,H) N C( CN)*C,H,* NO obtained with p-nitrobenzaldehyde forms red needles m. p. 256-258'. When the above hydrochloride is warmed in acetic acid with a-naphthol P-napht hol gallic acid rn-tolylenediamine and naph t hy lethy lamine substances are produced with blue violet-red violet bluish-violet and greenish-blue colours respectively. [With TR ARENDT.] -Met hyl o-carbomet h ox yphen y I gl ycine pre- pared by the interaction of methyl anthranilate and methyl bromo- acetate when treated with sodium nitrite and concentrated hydrochloric acid gives methyl p-nitroso-o-carbomethozyphenylglycine NO*C,H,(C0,Me)*NH-CH2*C0,Me green needlee m. p.164-165'. Ethyl p-nitroso-o-carbethoxyphenylglycine prepared in an analogous manner has m. p. 131' (decomp.). By heating met'hyl anthranilate with ethyl bromoacetate ethyl o-carbomethoxyphenylglycine m. p. 48O was obtained ; this could be further converted in the usual manner with sodium nitrite into ethyl p-nitroso-o-carbomethoxyphenplglycine green needles m. p. 125' (decomp.). I n a similar manner ethyl p-nitroso-o-carboxyph8nylgl?/cine green needles m. p. 115-116' (decomp.) mas obtained ; this substance when treated in hydrochloric acid with oxidising agents gave with aniline and e t h j laniline bluish-black precipitates and with P-naphthyl- amine a reddish-brown substance ; with benzyl and nitrobenzyl cyanides in alcoholic solution containing sodium methoxide con- densation occurs to yellow azomethine compounds ; the action of sodium methoxide solution itself on the nitroso-ester causes the separation of a yellow substance possibly the sodium salt of ethyl nitrosoindoxylcarboxylate.Methyl o-carboxyphenylglycine is most conveniently prepared byORGANIC CHEMISTRY. i. 173 boiling together in aqueous solution for one hour potassiuni anthranilate and methyl chloroacetate ; the action of sodium nitrite and concentrated hydrochloric acid causes its conversion into methyl p - n ~ l r o s o - o - c a r b o x ~ p ~ 6 n y ~ g ~ y ~ ~ ~ ~ e a green solid m. p. 115 -1 16'. D. F. T. Proof of the Chemical Isomerism of allo- and isoCinnamic Acids.HANS STOBBE and CURT SCEONBURG (Annalen 1913 402 187-259).-Whether GcZEocinnamic acid m. p. 6S0 isocinnamic acid m. p. 58O and isocinnamic acid m. p. 42' are three chemically different isomeric compounds or three modifications of one chemical individual is a problem of very old standing. Since the results of numerous investigations of the physical properties of the three acids lead to the coiiclusion'that they are different in the solid state but alike i n the fluid (dissolved or fused) state Biilmann's theory that the three acids are three modifications of one trimorphous (cis-)cinnamic acid might be accepted were it not for a few observations of Liebermann and of Stobbe which cannot be brought into harmony with the theory.Stobbe in particular advances reasons for his belief that two chemi- cally different isomeric acids exist namely the monomorphous aZlo- cinnamic acid m. p. 6S0 and the dimorphous isocinnamic acid m. p. 58' and 42' (A. 1911 i 859). On the contrary Meyer (A. 1911 i 975) supports Biilmann's theory The objects of the present investigation are to find an explanation of the discordant experimental results obtained by Meyer and by Stobbe (Zsc. cit.) and evidence in favour of one or other of the two theories. h sufficient proof of the chemical isomerism of the three acids mill be obtained if i t can be shown that (i) the acids exhibit differences in the fused or dissolved state; (ii) unlike mixtures of two acids in the fused or dissolved state can be prepared and (iii) regularities obtain during the transformations of the acids.The precautions which must be observed during the manipulation of the acids are emphasised; also i t is essential that the 68O-acid be crystallised from petroleum b. p. 30-33' as well as from water. Evidence of (i) is obtained as follows Saturated aqueous solutions of the 68'-acid prepared at temperatures not exceeding 30° are heated for seven hours a t 3 5 O 55O and 70" in elongated closed glass tubes. The tubes are cooled and after crystals have been deposited each tube is inverted and the m. p. of the damp acid is observed; the tube is then opened in a desiccator and the m. p. of the dry acid determined in the usual manner. Similar experiments are performed with saturated solutions of the 68O-acid in petroleum b.p. 30-33". The results show that the 6So-acid is always obtained from the solutions which have been heated only at 35' but from solutions which have been heated at 56' or 70° generally the 6B0-acid occasionally the 42'-acid never the 58'-acid is obtained. Experiments on the 42O-acid under the same conditions show that the 42O-acid is always obtained from aqueous or petroleum solution. I n a second series of experiments solutions of the 68'-acid in petroleum (b. p. 30-33') of different concentrations are shaken fori. 174 ABSTRACTS OF CHEMICAL PAPERS. half an hour at 35' in a closed vessel (shaped like a three-limbed star- fish) to ensure solution of all crystal seeds. The solution is then collected in one limb and the solvent is distilled at 35' into a second limb.The m. p. of the residue which is crystalline a t once or becomes so after short cooling is determined by immersing the limb in water a t 45O 60° or 70'. The residue is thus found always t o consist of the 68O-acid. Similar experiments on the 42O-acid show that the 42O-acid is always obtained. The results of seventy experiments by the two preceding methods prove that the stable 68O-acid and the metastable 42O-acid in the absence of crystal-seeds can be crystallised unchanged the former however only with certainty when the temperature during the process of solution and during the subsequent heating of the solution does not attain to the rn. p. of the acid or to the m. p. of the acid depressed by the solvent (in other word@ when fusion of the 68O-acid is avoided ; if this occurs the 42O-acid may be produced) Consequentfly each acid retains its individuality in solution under definite conditions and the two therefore are chemical isomerides.Evidence of (ii) cannot be ohtained by an examination of mixtures of the fused 68'- and 42'-acid because the composition of the mixture is uncertain owing t o the ease with which the fused 68O-acid changes t o the 42O-acid. However when a solution of the 42O-acid in petroleum is mixed with increasing quantities of a petroleum solution of the 68'- acid in the t hree-limbed vessel with the necessary precautions against inoculation by crystal-seeds the residue obtained by the distillation of the solvent a t 3 5 O has always rn. p. 42" until the amount of 68O-acid in the mixed solutions has been increased to about 3.5%; then the residue has m.p. 68'. Consequently the relative quantities of the components in the solution determine the character of the solutes and of the residue obtained by evaporation. Once again it is thus shown that the 68O-acid and the 42O-acid are not identical in the fluid state and are therefore chemical isornerides. The fact that the residues obtained by the evaporation ef the mixed solutions always have m. p. 42' or 68' is explained as follows. The residues must be solid solutions of khe two acids because it is known that a heterogeneous mixture of the 42'- and 68O-acids is incapable of prolonged existence. When the solid solution contains about 3.5% or more of the 68'-acid i t is super- saturated with respect to this acid ; solid 68O-acid separates and converts the remaining saturated solid solution entirely into the 68O-acid.When the solid solution contains less than about 3.5% of the 68O-acid t h e mixed crystals are more stable and heterogeneity occurs only after some .possibly a long time when owing t o the spontaneous change 42O-acid -+ 68'-acid so much 68O-acid has been produced that the limit of miscibility is exceeded. Solid solutions of the two acids containing less than about 3.5% of the 68O-acid all have m. p. 42O because the small amount of the 68O-acid present produces no or only an inappreciable depression of the m. p. A crystal having m. p. 42O therefore may be either the pure 42O-acid or a solid solution containing not more than about 3.5% of the 68O-acid ; the more nearly this percentage of 68O-acid is present the more readily will the *' 42O-acid " change to the 68O-acid.I n thisORGANIC CHEMISTRY. i. 175 manner an explanation is found of the conflicting statements of Liebermann Biilmann Stobbe and others regarding the " life " of the '* 42O-acid " a t the ordinary tamperature. The same considerations also serve to account for Stobbe and Reuss's remarkable observation (A. 1911 i 859) that the velocity of transformation of the solid 42O-acid into the 68'-acid increases as the temperature falls ; since the solubility of the solid 68O-acid in the solid 42O-acid decreases as the temperature falls the mixed crystal must the more readily become heterogeneous the lower is the temperature.Evidence in support of (iii) above has also been obtained. A systematic investigation of the conversion of the fused 68'-acid into the 42O-acid has hitherto not been undertaken Biilmann attribntes the transformation to the mere process of fusion (A 1909 i 382) whilst Liebermann ascribes the frequent failure of the transforma- tion to a certain sluggishness of the 68O-acid. Now if the two acids are isomeric the transformation must depend on the time of fusion and on the temperature of superheating. To ascertain the effect of the time of fusion a large number of closed capillary tubes containing a little 68O-acid are heated at 70' for periods varying from 0.5 to sixty minutes. The result of 500 experiments shows that the longer the time of heating the greater is the number of cases in which the 42'-acid is produced. (The contents of a capillary tube after fusion do not form a coherent liquid but a number of short liquid columns separated by air spaces ; after the contents have solidified it is not uncommonly observed that in one and the same tube some of the columns have m.p. 42O and others m. p. 68O.) The effect of super- heating is exhibited in 625 experiments similar to the preceding in which the capillary tubes are heated at 70' loo' l l O o 130' or 150'. The percentage number of cases in which the 42O-acid is produced increases with the temperature and reaches 100% when the tubes are heated a t 130' for not less than two minutes. It is noteworthy that in no case has the 58'-acid been obtained. Since Meyer claims t o have obtained 42'-acid 58'-acid and 68'-acid by the solidfication of fused 42O-acid (A 1911 i 975; 1912 i 32) 175 experiments have been performed with the 42O-acid for different periods of heating and at temperatures from 45' to 130° but in no case has any acid other than the 42O-acid been obtained. The acids m.p. 42' obtained in the previous experiments are not all identical substances because some of them can be kept unchanged for a year,whilst others change to the 68O-acid a t the ordinary temperature after a few minutes or hours. They are in fact solid solutions of the 42'-acid and 68'-acid like those obtained by the evaporation of the mixed solutions in petroleum (see above). According to the results of those experiments the least stable '' 42O-acids " are those which contain the greatest amount (up to about 3.5%) of the 68O-acid.So also in the preceding experiments on the heating of fused 68'-acid i t is proved that the most stable ' L 42'-acids," that is the solid solutions containing the smallest amounts of the 68"-acid are those which have been prepared by heating the fused 68'-acid for the longest periods and at the highest temperatures. The proof is obtained by ascertaining the maximum amount of 68O-acid which can be added to the 42'-i 176 ABSTRACTS OF CHEMICAL PAPERS. acids" without changing the m. p. A second method of proving the same point is based on Stobbe and Reuss's observation that the change of the 42O-acid to the 68O-acid proceeds rapidly a t low tempera- tures (Zoc. cit.). The " 42O-acids " obtained from the fused 68'-acid in the previous experiments are kept a t - 14' for periods varying from 5 to 600 minutes and the m.p.'s are then determined. The results of 110 such experiments show once again that the 68O-acid is obtained most frequently and in the shortest times from those ''42'-acids" which have been obtained by heating the fused 68'-acid for short periods at the lower temperatures (70' and llOo) and which therefore are the solid sdutions containing relatively the greatest amounts of the 68O-acid. Experiments similar to the preceding at - 14O have been performed with "42'-acids" after recry stallisation from water. Such recrgstallised acids behave like the " 42O-acids " obtained from the longest and most highly superheated 68O-acid fusions that is they are solid aolutions contaiuing very little 68O-acid.Biilmann and also Liebermann have observed (Zoc. cit.) that the change liquid 68'-acid -+ 42'-acid depends also on the amount of substance employed. This observation has been confirmed by heating portions of 10 25 or 50 mg. of the 68O-acid a t 70" for varying times and determining the m. p.'s of the solidified products After short heating the 68' is recovered by longer heating " 42O-acids " of varying stability are obtained and the change proceeds more slowly the greater is the quantity of material employed. I n this investigation the formation of the 58O-acid was first observed i n the following series of experiments. Crystalline 42'-acid prepared from the fused 68'-acid is invariably converted into the 68O-acid by keeping at - 75" for two minutes; the same result is obtained with 42O-acid crystallised from aqueous solution.Amorphous '' 42'-acids,'' prepared by heating the 68O-acid for five minutes a t l l O o or for thirty minutes a t 130° solidifying re-fusing at 45' and finally cooling to - 75' yield sometimes the 68O-acid or the 58'-acid or the 42'-acid and occasionally in one and the same capillary tube mixtures of any two or even of all three acids. The results in general are similar to those obtained at - 14O with the very important difference how- ever that the 58O-acid makes its appearance. The presence of 58O-acid in the '' 42O-acids " is due to the spontaneous change of the 42'-acid to the 58O-acid (see below) and its separation from the solid solutions at -75' is due like that of the 68'-acid to the decreased solubility of the 58O-acid in the 42O-acid at low temperatures.The fact that the 58O-acid is produced by the strong supercooling of amorphous but not of crystalline "42'-acids" is in complete agreement with the observations of Stobbe and Renss (Zoc. cit.). The proof that the 58O-acid is a chemical individual different from but isomeric with the 68O-acid and the 42"-acid has been established by the same methods as those employed above in the case of the 68'- and the 42'-acids and t o avoid repetition only the results of the numerous experiments will he recorded. The 58O-acid retains its individuality in solution in water or petroleum and can be crystallised therefrom unchanged provided that the solution has not been heatedORGANIC CHEMISTRY.i. 177 to a temperature at which fusion of the 58O-acid occurs. Determina- tions of the m. p.’s of the residues obtained by the evaporation of mixed solutions of the 58’- and the 68O-acidg show that the residues having m. p. 58’ may be either the pure 58O-acid or solid solutions (of the 68O-acid in the 58O-acid) having m. p. 5S0 (denoted by “58O-acid”). In such solid solutions the maximum percentage of the 68O-acid is about 4 (certainly greater than the maximum solubility of the 68O-acid in the ‘‘ 42O-acids ”) ; when this maximum is exceeded heterogeneity occurs and 68O-acid separates and converts the residual caturated solid solution into the 68O-acid. Since therefore the 58O-acid forms with the 68O-acid fluid mixtures and solid solutions different from those produced by the 42*- and the 6S3-acids the 58O-acid must be isomeric with the 42O-acid and all three acids are isomeric with one another. This conclusion is supported by the facts that the “ 58O-acids ” change into the 68O-acid rapidly at low temperatures and the velocity of the change a t - 75’ is smaller than that of the “ 42O-acids.” It has long been known that the 58O-acid changes into the 42O-acid by fusion.This change has been examined with ‘ k 58O-acids ” crystallised from solution and is shown to depend on time and temperature. The higher the temperature and the longer the time during which the “ 58O-acid ” is kept fused the larger is the percentage number of cases (in 657 experiments) in which the “ 42O-acid ” is produced.The k‘42°-acids” thus obtained are not all alike; some can be kept unchanged for years others change to the 58O-acid at the ordinary temperature in a few minutes. The very important observations are made that a “42O-acid” obtained from liquid 58O-acid changes if it changes at all always to the (solid) 5S0-acid whilst a ‘‘ 42O-acid ” prepared from liquid 68”-acid changes if a t all always t o the (solld) 68O-acid. The ‘‘ 42O-acids” obtained from the two different sources are therefore not identical ; a “ 42O-acid ” prepared from the 68O-acid is a bolid d u t i o n of the 42’- and the 6So-acids only whilst a “ 42O-acid ” obtained from the 58O-acid contains 42’-acid 5S0-acid and a very little 68O-acid. In addition to the proof of the chemical isomerism of the 68O- 58O- and 42’-acids the authors draw the following conclusions from the preceding experiments.Since the 5 8 O - and the 42O-acids are meta- stable i t is probable that the two can never be obtained in a pure state; the relative amounts of the components in each are determined by the source and age of the specimens. The 6S0-acid which has the smallest solubility of the three acids can be obtained practically pure provided it is old enough. I n the solid solution representing any ‘‘ 42’-acid,” the changes 42O-acid -+ 68O-acid and 42O-acid -t 58*-acid are continuously although they may be slowly proceeding. In course of time the solid solution becomes supersaturated with one or the other of the 5 8 O - and 68O-acids ; such acid separates and immediately sets up “ internal inoculation ” of the residual saturated solid solution.The effects of such iiiternal inoculation of different types of “ 42O-acids ” are discussed at length. The phenomenon of “ internal inoculation ” is described. VOL. CVI. i. I Li. 178 ABSTRACTS OF CHEMICAL PAPERS. The paper closes with an explanation in the light of the results of the preceding experiments of the discordant observations of Stobbe and Meyer (Zoc. cit.). c. s. Anhydride Formation with Acylaminocarboxylic Acids. GUSTAV HELLER (Ber. 1913 46 3974-3983).-The author has already observed certain cases of anhydride formation with amino- acids (compare Heller and Tischner A. 1910 i 7 7 0 ; 1913 i 365) and now publishes further results in this direction. m-Nitromandelic acid prepared by the cyanohydrin synthesis from m-nitrobenzaldehyde was reduced by ferrous hydroxide to m-aminoman- delic acid,needles m. p.131-132' (decornp.); when treated with benzoyl chloride in pyridine solution this acid gave a benzoyl derivative granular crystals m. p. 178O unaccompanied by any anhydride. I n the action of benzoyl chloride on rn-aminocinnamic acid in sodium carbonate solution the resulting benxoyl derivative needles m. p. 229O is accompanied by an insoluble substance m. p. 148" which appears to be benxoyl -m-aminocinnamic benxoic anhydride NHBz*CGH,*CH:CH*CO~OBz. This reacts readily with phenyl hydrezine giving benzoyl-m-amino- cinnarnic phenylhydraoide crystalline aggregates m. p. 197-1 9 9 O and wit,h sodium hydroxide solution on warming passes into benzoyl-m- aminocinnamic acid.The formation of the mixed anhydride is attributed to the primary formation of benzoic anhydride which then entere into reaction with the second acid. Similarly to the last cam p-awinobenzoic acid with benzoyl chloride in sodium carbonate solution gave benxoyl-p-amilsobenzoic benxoic anhydride NBBz*C,H4*CO*OBz a granular substance which after melting at 128' resolidifies t o re-melt near 210O. When heated with dilute sodium hydroxide solution the subRtance passes slowly into benzoyl- aminobenzoic acid and when mixed with phenylhydrazine in ethereal solution gives a gradual deposition of p-benxoylaminobenxoic phenyl- hydraside leaflets m. p. 248'. In the light of the above results the substance obtained by Heller and Tischner (loc.cit.) by heating benzoyl-p-aminobenzoic acid with acetic anhydride and described as benzoyl-p-aminobenzoic acid cycloid with acetic anhydride of crystallisation is in all probability benxoyl-p- arninobenzoic acetic anhydride ; it does not react with phenylhpdrazine. The action of boiling acotic anhydride on benxoyl-p-aminocinnamic acid (needlea m. p. 274' with decomp.; obtained by benzoylation of aminocinnamic acid in aqueous alkaline solution) for one hour gives benzoyl-p-aminocinnamic acetic anhydride m. p. 149O analogous t o the last substance. It is slowly converted by hot dilute sodium hydroxide solution into the sodium salt of benzoyl-p-aminocinnamic acid and with phenylhydrazine gives no phenylhydrazide. Treatment with excess of benzoyl chloride in pyridine solution converts p-aminocinnamic acid into (polymeric 0) p-1 -benxoyL1C/-ca~bostyril m.p. 385O. When treated with benzoyl chloride and sodium carbonate solution o-aminocinnamic acid gave a n insoluble product m. p. near 2204 which slowly dissolved in hot dilute sodium hydroxide giving benzoyl-p-aminocinnamic acid and gave no phenylhydrazide withORGANIC GHEMISTRY. i. 179 phenyl hydrazine ; the substance was therefore probably benxoyl-o- aminocinnamic benxoic anhydride but was not pure. Benxoyl-m-amino- benxoic banxoic unhydride obtained in a similar manner from m-amino- benzoic acid and benzoyl chloride likewise could not be obtained pure. By heating benzoyl-o-aminocinnamic acid for half a n hour in boiling acetic anhydride it is converted into benxoyZ-o-iiminocin?zamic anhydride needles m.p. 194-195' which when boiled with ,z mixture of acetic and hydrochloric acids slowly gives car bostyril. Treatment of p-aminohydrocinnamic acid with benzoyl chloride in sodium carbonate solution gave rise to the benxoyl derivative leaflets m. p. 194-195' but the product from benzoylation in pyridine solution is a substance 2CI,H,,03N,C,H,N (compare Heller and Tischner Zoc. cit.) m. p. near 240'. D. P. T. The Stereoisomeric Camphanecarboxylic Acids. PH. BARBIER and V. GRIGNARD (BuZI. Soc. chim. 1914 15 26-37).-Having already investigated the liquid pinene hydrochloride t)y nicans of its organomagnesium derivative (compare A. 1910 i 40C) the authors have turned their attention to the solid hydrochloride but although oxidation of the corresponding organomagnesium compound gives satisfactory results (compare Zoc.cit.; also Hesse A. 1906 i 375) the results obtained by treatment of the same compound with carbon dioxide have proved less decisive (compare Houben A. 1906 i 21). The action of carbon dioxide on the organomagnesium derivative of solid pinene hydrochloride prepared by saturatiog an alcoholic solution of pinene at 75-80' with hydrogen chloride and collecting the crystals which separate from the fraction b. p. 80-82O/13 mm. on distillation gave an acid product b. p. 157-158'/16 mm. The original pinene had [aID - 3 5 * 1 7 O and the hydrochloride m. p. 126' [.ID - 21.66'; the resulting dihydropinenecarboxylic acid CloH17*C02H consists of two isomerides the more abundant forming needlep m.p. 76-78' and yielding a dextrorotatory solution in methyl alcohol whilst the other forms needles m. p. 78-80' and is probably identical with the racemic acid mentioned below. Solid pinene hydrochloride prepared by saturating the same pinene in the cold with hydrogen chloride and therefore comparable with the specimen used by Houben gave the lavorotatory acid m. p 73-74" described by Houben (Zoc. cit.) which is doubtless a complex of two stereoisomerideu. The solid hydrochloride obtained from dl-pinene by the former method when treated successively with magnesium and carbon dioxide gave a racemic acid crystallisiag in needles m. p. 78-80'. The two specimens of active acid obtained above give the same anilide needles m. p. 179O and p-toluidide needles m.p. 185'; also when heated alone with alcoholic potassium hydroxide aromatic amines or especially with mineral acids they undergo isomerisation into one and the same lavorotatory acid m. p. 88-89'; the acid m. p. 73-74' is certainly more resistant to this treatment. These results are interpreted as due to the formation of a more Etable geometrical isomeride which is designated the a- or I-camphane- carboxylic acid whilst the less stable form is the p- or d-isocamphane- n 2i. 180 ABSTRACTS OF CHEMICAL PAPERS. carboxylic acid. The acid m. p. 76-78' is considered to be composed mainly of the /I-acid associated with a small quantity of the a-isomeride whilst in the acid m. p. 73-74' the proportion of the a-isomeride is greater. The a- and /3-acids thus possess the property of forming homogeneous crystalline mixtures.Under the influence of isomerising agents the racemic acid m. p. 78-80" or dl-iso- camphanecarboxylic acid undergoes conversion into a stereoisomeric d-Z-cnmyhanecarboxylic acid m. p. 93-94O. From these results i t would appear that under the conditions of the two methods mentioned for the conversion of pinene into the solid hydrochloride a cettain amount of isomerisation similar to the above is effected but it is also demonstrated that such isomerisation can also occur in the organomagnesium compound as already suspected by Houben and Doescher (A 1911 i 61) the properties of the acid product beiug found to vary with differing conditions for the interaction of magnesinm with the hydrochloride.Preparation of Chlorides of Hydroxy-acids. EDUAKD KOPETSCHNI and LADISLAUS KARCZAG (Bey. 1914 47 235-237 ; D.R.-P. 262883 266351).-When the salts of hydroxycarboxylic acids are treated wit,h thionyl chloride (Be?.. and 1st patent) or carbonyl chloride (2nd patent) the acid chlorides are obtained. SaZicyloyl chloride forms white needles m. p. 17-5-18'. m- and p-Hyd~oxyben2oyl chlorides are oils having penetrating odours which do pot solidify a t - 15' to - 20'. GZycolEoyl chloride is a n oil of similar properties. These chlorides are very reactive giving for example with methyl alcohol the corresponding methyl esters of the acids. Preparation of Derivatives of Aromatic Hydroxycarboxylic Acids. RICHARD WOLFFENSTEIN (D.K.-P. 267980. Compare this vol. i 45).-/3/3/3-l'richloro-tert.-butyZ 4-i~ydroxy-m-toEuate prepared by the interaction of ,3/3/3-trichloro-tert.-butyl alcohol and 4-hydroxy-m- toluic acid has m.p. 97'. /3/3P-~richloro-tert.-butyE vanillate has m. p. 1 30° and PPP-tric?~loro-tert.-butyl 1-naphtho1-2-cccrboxyZ~te has m. p. 133'. These compounds have properties similar to those described in the chief patent (loc cit.). R. LESSER and R. WEISS (Ber. 2913 46 3937-3946).-When heated with formic acid tetrachloroanthranilic acid (compare Villiger and Blangley A. 1909 i 922) gives aformyl derivative CO,H* C,CI NH* COH colourless needles m. p. 304-306' soluble in dilute sodium carbonate solution; the product obtained on heating with acetic anhydride is a very pale yellow needles m. p. lactonic compound C6C1,< 198-199* which when heated with aqueous acetic acid becomes hydrated to the corresponding acid tetrachloro-o-acetyZccm~nobenxo~c acid CO,H*C,CI,*NH*COMe needles m.p. 240-241' which easily reverts in the atmosphere t o the original lactone (compare Mohr and Kohler A 1910 i 11 6). Tetrachloroanthranilic acid is only D. F. T. J. C. C. J. C. C. Halogen S u b s t i t u t e d P h t h a l i c and Anthranilic Acids. CO*? N=CMe'ORGANIC CHEMISTRY. i. 181 diiiicultly and iocornpletely diazotisablo (compare Orndorff and Nicholls A 1913 i 99); the solution of the diazo-compound was converted through the xanthate by treatment with chloroacetic acid into tetrachloro-o-carboxyphenylthiolacetic acid CU2H*C6CI,*S*OH2*co2H indistinct crystals m. p. 225' (decomp.); this substance of which the yield was poor gives ' I octacl~lorothioindigo " when warmed with fuming sulphuric acid.The action of bromine in acetic acid solution on isatoic acid gives rise to 3 5-dibromoanthranilic acid colourless needles m. p. 234-235' (compare Dorscb A. 1886 359) which is converted by acetic anhydride into an internal anhydride of the acetyl derivative (compare above) colourless needleg m. p. 174-175' and can be made to pass through the stages diazo-compound and xanthate to 3 5-dibromo- 2-carboxyphenylthiokucet~c acid C02H*C6H2Br2*S*CH,oC~2H colour- less prisms m. p. 186-187' (decomp.). I n the action of excess of bromine in hot acetic acid solution on isatoic acid there is obtained occrtsionally a tribrol72oanthranilic acid colourless needles m. p. 245-246' (decomp.) ; this acid when warmed with ammonia solution gives a red insoluble substance and also gives a crystalline barium salt.When treated with boiling acetic anhydride tribromo- anthranilic acid yields tribromoanthranil needles m. p. 145 -146'. The authors were unable to produce tetrabromoanthranilic acid by the action of bromine on isatoic acid (compare Dorsch Zoc. cit.) but were able to obtain i t by the action of potassium hypochlorite in alkaline solution on tetrabromophthalimide. Tetrabromoanthvanilic acid forms colourless needles m. p. 204-205'; barium Ralt crystal- line ; silver salt amorphous. When heated with acetic anhydride it gives the internal anhydride of the acetyl derivative (see above) needles m. p. 257-2589 It is already known that phthalic anhydride can be converted by the theoretical amoiint of bromine into 4 5-dibromophthalic acid m.p. 208-209'. This substance has now been converted through the imide m. p. 245-246' into 4 5-dibromoanthranilic acid colaur- less needles m. p. 228-229' (decomp.) ; barium salt needles ; calcium salt needles ; silver salt amorphous and insoluble; the internal anhydride of the acetyl derivative forms microscopic needles m. p. 184-185'. The diazotised solution of 4 5-dibromoanthraoilic acid slowly deposits colourless needles of the diazo-compound ; when boiled with alcohol and dilute sulphuric acid respectively the diazo-solution gives rise to 3 4-dibromobenzoic acid needles m. p. 234-235' and to 4 5-dibromosaZicylic acid m. p. 217-218' qrobably identical with the isomeric .acid of unknown constitution (Hubner A.1578 148). The diazo-compound was also converted through the xanthate by means of chloroacetic acid into 4 5-dibromo-2-carboxyphenylthiolacetic acid C02H*C,H2Br2*S-CH2-C02H needles m. p. 240-241' (decomp.). I n one experiment on the bromination of phthalic anhydride a n exceptional result was obtained in the formation of another dibromo- phthalic acid needles m. p. 194-195'; anhydyide needles m. p. 152-153' ; imide small rods m. p. 280-281'. 3 6-Dichloro-4 5-dibromophthitlic acid is already known ; itsi. 182 ABSTRACTS OF CHEMICAL PAPERS. anhydride m. p. 269-270° can be converted into the irnide yellow needles m. p. 362-364' which on treatment with alkaline sodium hypochlorite gives 3 6-dichloro-4 5-dibromoc~nt~ranilic acid colourless needles m.p. 193-194' (decomp.); barium salt small rods ; the anhydride of the acstyl derivative C9H302NC12Br2 forms needles m. p. 213-214". Isomeric Phenylphthalimides and Some Allied Compounds. 111. MITSURU KUHARA SHTGERU KOMATSU and ROEUHACHI NISHIYORI (Mem. Colt. Sci. Eng. ILy6t6 1913 5 343-363).-1n an earlier paper (A 1911 i 205) i t was shown that a mixture of a phthalamide with a colourless normal and a yellow unsymmetrical phthalimide results when phthalyl chloride reacts with an aromatic primary bass a t a low temperature. To account for this it was assumed that the chloride exists as a mixture of the two forms at the temperature of experiment. Pitwlewski (A. 1895 i 134 414) and Dobrev (A 1895 i 360) pre- pared the nitrophenylphthalimides but they obtained different substances.It is now shown that Pawlemski had isolated the white symmetrical compounds whereas Dobrev had obtained the coloured as-isomerides. Both compounds are formed in the same operation according to the above viems. The as-nitrophenyiphthalimides were also obtained by the action of acetyl chloride on the nitrophenyl- phthalamic acids They are more soluble in alcohol than their isomerides into which they are transformed by mineral acids. The three nitrophenylphthalamic acids give coloured soliitions in alkalis those obtained from the 0- and p-nitro-compounds being deep orange-red whilst the meta-derivative gives a yellow solution. The ortho- and para-salts are therefore represented by the quinonoid formula CO,K.C,H,*CO*N UGH4 NO-OK whilst the meta-salt is derived from the enolic form of the acid and is represented by C0,K*C,H,*C~OK):N*C,H4*N02. The behaviour of the phthalamic acids a8 pseudo-acids was investigated by the preparation of a number of salts. A t low temperatures disilver sa1t.s were obtained from which dimethyl esters were prepared. ALI ethereal solution of o-ni troaniline reacted with phthalyl chloride on warming and gave as-o-?titswophenylphthulinzide The acid was not diazotisable. D. F. T. v6H4-y :N* C,H,*NO co-0 canary-yellow needles m. p. 152-153' the s-isomeride white m. p. 1 98-200° and o-nitropheny?phthalamide pale yellow lustrous cryst.als m. p. 187-189' (probably identical with Pawlewski's o-nitrophthalanilide). as-p-Nitrophenylphthalimide m.p. 191 -1924 and the as-?n-derivative m. p. 234-235' were identified wit,h Dobrev's compounds and the symmetrical isomerides with Pawlewski's. Disilvey phthakc6mate CO,Ag*C,H,*C(OAg):N H was obtained by adding silver nitrate (2 mols.) to ammonium phthalamate at - 5' as a white precipitate. It reacted with methyl iodide and the oily ester gave ammonia on hydrolysis. Ammonium phenyl phthalamate was ? C,H,( CO*NH*C,H NO,) 2 .ORGANIC CHEMISTRY. i. 183 converted into the silver disilver barium and lead salts the last three being derived from the acid in its enolic form. Pinzethyl phenylphthalamate CO,Me*G,H,*~(OR/le) NPh was obtained in plates m. p. 90' which were hydrolysed to aniline and phthalic acid. Silver and disilver salts of m- and p-tolylphthalamic acids; silver and disilver salts of m-xylylphthalamic acid ; a silver and orange lustrous dipotassium salt of o-nitrophenylphthalamic acid ; a silver and yellow lustrous scaly dippotassium salt of m-nitrophenylphthalamic acid ; and a silver and lustrous orange-red dipotassium salt of p-nitro- phenylphthalamic acid are described.Most of these phthalamic acids were discovered by Tingle and Rolker (A. 1909 i 28). Isomeric Phenylphthalimides and Some Allied Compounds. IV. h!fITsuRu KUHARA and SHIGERU KOMATSU (Mern. toll. 8 c i . Eng. KyGt6 19 13 5 363-366).-Phthalyl chloride has been condensed with 4-nitro-m-toluidine and with 6-nitro-o-toluidine and three J. C. W. products isolated in each case (compare preceding abstract). as-4-Nitro-m-tolyZphthaZimide %H4'F):N* c6H'Me'No2 forms canary- co-0 yellow needles m.p. 156-158' ; the symmetrical isomeride crystal- lises in white needles m. p. 180-18 I ' and 4-nitro-m-tolylphthalamic acid gives alkaline solutions which are orange-yellow when cold and orange-red when hot. The eilver salt was annlysed. as-6-Nitro-o- tolylphthalimide forms yellow needles m. p. 183-184' 231-232' after solidification ; the symmetrical compound has m. p. 231-232' and 6-nitro-o-tolylphthcclamic acid forms yellow crystals and a silver salt. Condensation Products of Phthalic Anhydride with m- and p-Aminobenzaldehyde and their Derivatives. P. GELMO (J. pr. Chem. 191 3 [ii] 88 810-828).-m-Aldehydophenylphthalimide C,H,:C,O,:N*C,H,*CHO prepared by heating phthalic anhydride with anhydro-m-aminobenzaldehyde at 150° crystallises in lustrous white felted needles m.p. 177' (corr.j and forms a phenylhydraxone crystallising in small yellowish-white needles which become brown and decompose at about 224'; the oeime forms lustrous white leaflets m. p. 212O (decomp. corr.). When heated with aqueous alkali hydroxides it is hydrolysed to m-aminobenzaldehyde and phthalic acid ; wit'h aqueous sodium carbonate containing a little sodium hydroxide i t yields m-aldehydopheny!phthaZamic acid CO,H*C,H,*C@ * N H* C,H,*CHO. This crgstallises in flat yellowish-white needles m. p. 157' (decomp. cow.) and yields a phenylhydraxone as a light yellow crystalline powder m. p. 2 16-217' with previous darkening and decomposition at 207' ; the oxime cryst'allises in small yellowish-white leaflets which begin to decompose at 130° m.p. 177-178'; the salts with the alkali-metals together with the barium calcium zinc and silver salts are readily soluble in water. J. C. W.i. 184 ABSTRACTS OF CHEMICAL PAPERS. p-Aldehydophenylphthalimide prepared from phthalic anhydride and p-aminobenzaldehpde at 1 60° forms yellowish-white clusters of broad needles m. p. 202' (corr.) and resembles the meta-isomeride in being readily hydrolysed by strong alkalis to phthalic acid and the original arninoaldehyde thephernylhydraxone crystallises in long slender citron- yellow needles m. p. 235-236" with previous darkening and decomposition a t 225' ; the oxime forms yellowish-white leaflets m. p. 227' (decomp. corr.}. It is hydrolysed by aqueous sodium carbonate t o p-aldehydophenylphthcclamic acid This forms a citron- yellow crystalline powder m.p. 139-140O (decomp.) and yields a deep orange phenylhydrazone which begins t o decompose at 210° m. p. 231-233' ; the oxime crystallises in yellowish-white needles m. p. 176" with previous decomposition at 170". F. B. Preparation of Dia,nthraquinonylaminecarboxylic Acids. BADJSCHE ANILIN- & SODA - FABRIK (D.R.-P. 2682 19).-Dianthraquin- onylaminecarboxylic acids are prepared by heating halogen- or amino- anthraquinonecarboxylic acids with amino- or halogen-anthraquinones respectively in a solvent of high boiling point in the presence of a catalyst such as copper and an oxide or salt of the alkaline-earth metals Examples are given illustrating the preparation of a dianthraquinon ylaminecarboxylic acid from 1 -chloroanthraquinone-2- carboxylic acid and 1-aminoanthraquinone or alternately from 1-aminoanthraquinone-2-carboxylic acid and 1-chloroanthraquinone of 2 2'-dianthraquinonylamine-3-carboxylic acid from 2-chloroanthra- quinone-3-csrboxylic acid and 2-aminoanthraqninone and of 2 1'- dianthraquinonylamine-3-carboxylic acid from 2-aminoanthraquinone- 3-carboxylic acid and 1 -chloroanthraquinone.J. C. C. Constitution of Naringenin. Phloroglucinol Esters of Phenol- carboxylic Acids. ADOLF SONN (Ber. 191 3 46 4050-4059).- Will (A. 1885 i 906; 1887 i 715) supposed naringenin to be a n ester compounded of phloroglucinol and p-coumaric acid whereas Tutin (T. 1910 97 2054) from analogy t o eriodictyol and hesperitin assumed that i t has a ketonic structure. The ester from the above phenol and acid synthesised by Fischer's method is not identical with naringenin which indirectly supports the alternative constitution. p-Coumaric acid was prepared by Perkin's reaction at 175' in an open vessel pressure being unnecessary.It was dissolved in N-alkali treated with methyl chloroformate in a freezing mixture and then the solution was acidified when p-rnelhylcarbonatocinnamic acid C0,Me*O*C6H4*CH:CH*C0,H separated. It crystallised in thin prisms from acetone m. p. 198-199'. The chloride CllH904Cl formed white silky prisms or needles m. p. 337-138" from which the anilide was obt.ained in pearly leaflets m. p. 165-166' which yielded p-cournarundide thin lerrflets m. p. 202' when hydrolysed by dilute sodium hydroxide.Phloroglucinolcarboxylic acid was digested with N-sodium hydroxide (2 mols.) and then treated with the above p-methylcarbonatocinnamoyl chloride in acetone. The solution was acidified poured into water and the crude product treated with cold N-alkali (3-4 rnols.) when the carbonato-group was hydrolysed. OnORGANIC CHEMISTRY. i. 185 acidifying the deep yellow solution p-coumaroyZphZorogZucinolcarboxyZic [4(or 2)-p-coumaroyloxy-2 6(or 4 6)-dihydroxybenxoic] acid OH*C,H,*CH CH*CO*O*C,H,(OH),*CO,H was isolated. It crystallised with 1$H20 1n long thin needles m. p. 194' and gave fa deep violet coloration with ferric chloride. It was converted into p-coumaroyZphZorogZuc~noZ [ 3 5-dihyd~oxypAenyl p-coumarate] OH C,H,-CH C H* CO* 0 C,ET,( OH) by heating in vacuum at 200-220'.A yellow indefinite powder m. p. about 200° mas thus obtained almost insoluble in benzene whereas naringenin forms well-defined crystals m. p. 248' and is soluble i n benzene. Similar phloroglucinol esters of the p-hydroxy benzoic acid derivatives were also prepared. p-Methylcarbonatobenzoylphloroglucinolcarboxylic acid was obtained from p-methylcarbonatobenzoyl chloride (E. Fischer A. 1908 i 892) and hydrolysed by cold sodium hydroxide to p- h ydroxy henxoylp hlorog Zucino Zcar box y lic [ 4( or 2 ) - p- hy droxy benzo y Zoxy - 2 6(or 4 6)-dihydroxybenzoic] acid OH*C,H,*CO*O~C,H2(OH),*C0,H. The latter crystallised in thin tufted needles m. p. ZlS' and gave a deep violet-red coloration with ferric chloride. When maintained for a short time in vacuum above its melting point it yielded p-hydroxybenxoylph ZorogZucinoZ[3 ; 5 -dih ydrox?l;lhenyZ p- h y d ~ o x y benzoate] OH*C,H,*CO-O*C,H,(OH) which was obtained in white moss-like masses of needles m.p. 218'. p - Coumaric acid was also reduced t o @-p-h y d roxy ph enyl propionic acid and the crude ethereal extract of the latter was condensed as above with methyl chlorocnrbonate. /3 - p - illethylcarbonntophengz- propionic acid CO,Me* 0. C,H,*CH,* CH,*CO,H forms thin glistening tablets m. p. 86-87O. The crude chloride was condensed with phene- tidine in the cold. The phenetidide C19H2105K crystallised in prisms m. p. 142-1 43' and was hydrolysed to @-p-hydroxyphenyl~opiorto- phenetidide OH*C,H,*CH,*CH,*CO*NH*C,H,*OEt which formed thick prisms m. p. 175-176'.The above chloride was also condensed with glycine in the presence of the theoretical amount of N-alkali hydroxide. The colourless oil obtained on acidifying was extracted with ether and hydrolysed by cold alkali and finally @-p-hydroxy- pheny Zpropion yZgZ ycine OH*C,H,* CH,*CH,*CO*N fl CH CO H was obtained iu glistening prisms from hot water m. p. 179'. It may be a constituent of normal urine since P-p-hydroxyphenylpropionic acid is. J. C. W. Oxides and Isomerides of Trithiobenealdehyde. 0. HINSBERG (J. p r . Chem. 1913 [ii] 88 800-810).-Baumann and Fromm (A. 1890 25) have shown that trithiobenzaldehyde exists in two isomeric forms termed p- and y-trithiobenzaldehydes which are considered to be stereoisomerides of the cis-trans-ty pe. By the successive oxidation of the @-form with hydrogen peroxide and reduction with hydriodic acid the author has obtained a third isomeride which he terms 6-trithiobenzaldehyde.The new isomeride is converted into the @form by crystallisation from glacial acetic acid benzene or chloroform. The author suggests t h a t the p- and S-forms are sulphur isomerides of the same type as the a-.and P-trimethylene sulphides (compare A,i. 186 ABSTRACTS OF CHEMICAL PAPERS. 1913 i SlS) whilst the /3- and y-forms are to be considered as cis-trans-isomerides. The different behaviour of the P- and y-forms on oxidation is however difficult to reconcile with this view ; on oxidation with hydrogen peroxide P-trithiobenzaldehyde is converted into a tetroxide belonging t o the &series whilst the y-isomeride gives rise t o a pentoxide.Further both the 8-tetroxide and y-pentoxide on oxidation with potassium permanganate yield the same trisulphone. 6- Trithiobenxaldeh yde tetroxide ( F- tribenzylidenedisuZphoxid~sulphone) CHPh<SO~CHPh>~~2 prepared by oxidising P-trithiobenzaldehyde with hydrogen peroxide for several days at the ordinary temperature in glacial acetic acid solution cry stallises in slender colourless needles m. p. 256-257' (decomp.) but slowly decomposes on pro- longed heating at 130-140'. When kept in contact with hydriodic acid (D 1.96) for several days it is reduced to &trithiobenzaldehyde which forms a pale brown crystalline powder m. p. 180'. y- Trithiobenzaldehyde pent oxide ( y-tribenz ylidenes21lphoxidedisulphone) C H P h < ~ ~ ~ ~ ~ ~ ~ > S O obtained by oxidising y-trithiobenzaldehyde under the same conditions as given above for the oxidation of the P-isomeride crystallises from glacial acetic acid in small colourless needles (decomp 254") and is reduced by hydriodic acid to /3-trithiobenzaldeh yde. ~ibenxylidenetrisulp~one CHPh<so~.Caph>S02 SO *CHPh prepared by oxidising the S-tetroxide or y-pentoxide with potassium permanganate and sulphuric acid crystallises from ethyl acetate in small needles which are not fused at 300'. It differs from the 6-tetroxide and y-pentoxide in not being reduced by hydriodic acid and in forming a sodium salt which undergoes no change when boiled with aqueous sodium peroxide.Methylation of Ketones. J. D. RIEDEL AKT.-GES. (D.R.-P.266405).-Rlethyl derivatives of ketones aldehydes and esters are readily prepared by reducing the corresponding hydroxyrnethylene compounds with hydrogen by the aid of colloidal palladium as catalyst. Examples are given of the preparatim in this way of 1-methylcyclo- hexan-2-one methylcamphor and ethyl methylacetoacrtate [compare following abstract]. J. C. C. A New Method of M e t h y l a t i o n Methyl Compounds from Hydroxymethylene Compounds. A. KOTZ and E. SCHAEFFER (J. pr. Chem. 1913 [ii] 88 604-640).-The method is only applicable for the methylation of aliphatic and hydroaromatic ketones and esters of ketonic acids. It consists in converting the ketone into the hydroxymebhylene derivative (I) and reducing the latter by means of hydrogen in the presence of eolloidal palladium to the corresponding methyl derivative (IV) SO*CHPh F. B.-q:0 y:0 y o -+ q:0 + -+ -C:CH*OH CH *CH,*OH C:CH CH-CH; (1.) w.1 (In.) (IT. 1ORGANIC CHEMTSTRY. i 187 I t is probable that ketone-alcohols of the type (11) are f'ormed as intermediate products in the reduction and that these readily lose water yielding unsaturated compounds (111) which then undergo further reduction. Evidence in support of this view is furnished by the isolation of a ketone-alcohol of this type in the reduction of hydroxymethylenedihydrocarrone although attempts to reduce P-h;rdroxybutaldehyde which contains the grouping sh?wn in (11) were unsuccossful. On Rccoiiut of their marked acid character the hydroxymethylene compounds often precipitate the palladium from its colloidal solutions and in these cases the reduction cannot be carried out by the usual methods of Pita1 and Skita.The following method however was found to be very serviceable where precipitation of the palladium occurred a methyl-alcoholic solution of the hydroxymethylene com- pound is placed in the absorption vessel and an aqueous solution of palladous chloride (without any protective colloid) caref iilly intro- duced below this solution. On vigorously shaking a rapid absorption of hydrogen takes place and although the palladium is very soon pre- cipitated the reduction is complete provided that only small quantities of the hydroxymethylene compound are employed ; for large quantities successive treatment with palladous chloride in this manner is necessary.The following compounds have been prepared by reduction of hydroxymethylene compounds in this manner or according to Skita'B method. 1 3-I>imethylcycZohexan-2 -one from 3-hydroxymethylene-1 -methyl- cyclohexan-2-one ; methylcyclohexnn-2-one from hydroxymethylene- cyclohexnn-2-one ; 1 4-dimethylcyclohexan-%one from 4-hydroxy- met h j lene- 1 -methylcycZohexan-3-on e ; met hylmenthone 1 -methylcycZo- pentan-2-one and rnethy Icam pbor from the corresponding hydroxy- met hyleue compounds ; rnethy5sothujone ( 1 3 5 - t ~ . i m e t h y l - 2 - i s o ~ ~ o p ~ ~ - CO-y Me A1-cycloiue?ztcn-4-one) CMe< anoil b. p. 103-105°/11 mm. CMe CPrp ' and 229-330° under ordinary memire. The reduction of hydroxyieihy lenedi hydrocnrvone yields hydroxy- CO*CH(CH,*OH)>CH.CMe CH CH methyldihy cl r o c a r vo n e CH Me<CH - a - a which was isolated in the form of i t s benzoyl derirative'(1ustrous crystals m.p. 11 1 O) and methyldihgdroccwvone CHMe<Co'CHMe>CH*CMe:CH CH,-C H a liquid b. p. 226-227O h a v i n i an odour resembling t h a t of menthone (sernicarbnzone m. p. 168O). Hydroxymet hylene cornpounds of diketones and ketonic esters can be reduced by this method but not those derived from esters o€ dicarboxglic acids. OEt*CH:C( COMe)2! and ethyl hydroxymethyleneacetoacetate yield aa-diacetylethane and ethyl methylacetoacetate respectively whereas the hydroxymethylene derivatives of ethyl malonste and etbyl slwcinate do not undergo reduction. Thus ethoxymethyleneacety lacetone,i. 188 ABSTRACTS OF CHEMICAL PAPERS.O€ the ethers and esters of hydroxymethylene compounds only those which are readily hydrolysed by water can be reduced by this method ; the ethyl ethers of ethyl hydroxymethyleneacetoacetate and hydroxy- methyleneacetylacetone being readily hydrolysed by water are quanti- tatively reduced to the corresponding methyl derivatives whilst methoxymethylenecamphor and ethyl ethoxymethylenemalonate which are not readily hydrolysed cannot be reduced. Similar results were obtained in the case of the acetyl derivatives. The reduction of the iron and copper salts aud also the amide imide and anhydride of hydroxymethylenecamphor was investigated but with negative results. On the other hand the chloromethylene derivatives of camphor and ethyl succinate are readily reduced to methylcamphor and ethyl methylsuccinate respectively.The behaviour of acetplacetone acetylcamphor and 4-acetyl-1-methyl- cycZohexan-3-one which in the enolic form have a structure CO-C:CMe*OH similar to that of the hydroxymethylene compounds has also been investigated but no reduction was found to take place; a similar negative result was obtained with the enolic form of dibenzoylacetyl- methane F. B. Influence of the Substituent on the Interaction of Benzene and its Derivatives with Benzoyl Chloride . i n Presence of Antimony Trichloride. B. N. MENSCHUTKIN (J. Buss. Phys. Chem. Soc. 1913 45 1710-1738).-1n preceding papers (1911 and 1912) the author has shown for a number of subdituents the immense influence these exert on the capacity of aromatic compounds of combining with antimony trichloride and tribromide and on the properties of the products of such combination.In order to obtain a greater insight into this influence the in- vestigation has now been extended to a more complex reaction that of benzoyl chloride on benzene and its near derivatives in presence of antimony tri-haloids. This reaction proceeds with less ease than in presence of aluminium chloride and all the experiments were carried out a t 150-155' at which temperature evolution of hydrogen chloride proceeds with moderate rapidity . The tem perat ure -concen tration diagrams of binary systems composed of benzoyl chloride and a number oE aromatic hydrocarbons indicxte in no case the formation of a molecular compound (this vol. i 170). Preliminary experiments were made at 150-155' in absence of antimony tricbloride but only with mesitylene did the reaction hydrocarbon + benzoyl chloride = ketone + hydrogen chloride take place to an appreciable extent and even in this case the velocity was very small.I n the presence of antimony trichloride these reactions proceed readily and give good yields of ketone. The first phase in the reaction consists in the formation of a compound of antimony trichloride with the hydrocarbon 2SbC1,,C6H,R ; this is then acted on by the benzoyl chloride the result being a compound OF antimony chloride with the ketone SbCI,,C,H,*CO*C,H,R which decomposes into i t s constituents at the temperature of the experiment; theORGANIC CHEMISTRY. i. 189 antimony trichloride thus liberated may then react with fresh quantities of hydrocarbon and benzoyl chloride.The total reaction expressed by the equation ArH+ C,H,*COCl+ SbCI = Ar*CO*C6H + HCl + SbCI proceeds to an end and gives results in accord with the equation representing a bimolecular equation the hydrocarbon and the benzoyl chloride being taken always in molecular proportions. The velocity of the reaction varies directly as the square of the concentration of the antimony trichloride. With benzene and its monoalkyl derivatives the velocity con- stants of the reaction in presence of BSbCI and 1SbC1 are as follows 2SbC1,. lSbCI,. - - C6H6 .................. 0'000224 1 0*0000546 1 C,H,Me .............. 0.00272 12 0.00067 12 C6H5Et ............... 0'00466 21 0~00111 20 C6H,Pr ............... 0.0051 23 0,0012 22 - - C&,'C,H,,............ 0 *0052 23 These figures indicRte the great influence of the benzene sub- stituent on the velocity of the reaction The identity of the results for propyl- and isoamyl-benzenes seems to indicate the attainment of a limiting velocity of reaction but the explanation perhaps lies in the marked resinification accompanying the reaction in these two cases. With dialkylbenzenes the results are as follows 2SbC1,. 1 SbCl,. - - C6H6 .................. 0'000224 1 0*0000546 1 p-C:,H,Me ............. 0.00476 21 0 -001 12 21 o-C,H,Me ............ 0'00725 32 0.00176 32 m-C6H4Me,. ........... 0'0178 80 0.00446 81 p-C H,MePr ......... 0 '0089 40 0'0028 40 The influence of the position OF the second substituent is very marked (compare Menschutkin A.1898 i 186 ; 1900 j 335 341 ; de Bruyn and Steger A 1899 i 744) whilst the increasing magnitude of the influence as the size of the substituent increases is peen on comparing the results for p-dimethyl- and p-metbylisopropyl- benzenes. The effect of a third substituent is shown by the following figures 2SbC1,. 1 SbC1,. - - CeH6 .................. 0'000224 1 0.0000546 1 C&f,Me .............. 0.00272 12 0.00067 12 m-C6H4Me ............ 0'0178 80 0.00446 81 1 3 5-C6H,Me ... 0.107 478 0.0266 493 1 2 4-C6H&h?3 ... 0.0316 141 0.00760 139 The constants for 1 3 5- and 1 2 4-trimethylbenzene are in the ratio 3.4 with BSbCl 3.5 for lSbCl and 3.6 for 0.5SbC1,. The constant for mesitylene is lowered from 0.107 at 155' to 0.0049 at 84" when 2SbC1 is taken and from 0.0269 a t 155'to 0.0169 a t 125' and to 0.0059 a t 105' when 1SbC1 is used.i.190 ABSTRACTS OF CHEMICAL PAPERS. I n presence of 2SbC1 the following constants were obtained diphenylmethane 0.0 175 ; diphenyl 0*0035 ; triphenylmethane 0.0030 but in these cases marked resinification and blackening occur. For benzene cblorobenzene bromobenzene and nitrobenzene the constants are in the proportions 1 0.1 0.023 0102. The results of the series of measurements made with each compound are given in full. T. H. P. Synthesis of 2 3-Dihydro xyacetophenone. HEINRICH VON KRANNICHFELDT (Ber. 191 3 46 40 16-4025).-The synthesis of 2 3-di hydroxyacetophenone has been accomplished by the methods of Klages (A 1904 i 45) and Posner (A. 1912 i 453). Incidentally typical derivatives of 2 3-dimethoxybenzaldehyde have been prepared [with M.HAARSMA]. 2 3-Dimethoxy-1-a-hydroxyethylbenzene (Klages Zoc. cit.) mas added to ;t solution of potassium dichromate and sulphuric acid at 30' when the resulting yellow oil was immediately removed by steam. 2 3-Dimethoxyucetophcnone C10H1203 was thus obtained as a n almost colourless refractive mobile faintly smelling liquid b. p. 143-144'/14 mm. The semicarbazone formed white prisms m. p. 166-167' and the oxime crystallised in white glistening scales m. p. 96-97'. The methoxy-gronps were replaced by hydroxyl by heating with hydriodic acid or with powdered aluminium chloride in chlorobenzene. 2 3-Dihydroxyucetophenone (v-ucetocuteclhol) CsHsOq formed dark yellow prisms m.p. 97-98' which gave yellow solutions in alkalis and concentrated sulphuric acid. I t is best purified by the hydrolysis of the acetyl derivative C12H,,0 which crystallises in white leaflets m. p. 109'. The above oxime was also obtained by boiling 2 S-dimethoxycin- numic acid Cl,H120 with hydroxylamine (Posner's method). This acid was obtained from o-veratraldehyde by Perkin's method in white silky crystals m. p. 180-181° which sublime in beautiful needles. The dibrornide forms colourless prisms m. p. 152O and the ethyl ester crystallises in colourless glistening scales m. p. 44-45'. p-2 3- Dimet hoxyphenylpropionic acid C1?H,,O was obtained by reduction in colourless crystals m. p. 63'. l'he chloride is a pale yellow mobile oil b. p. 165-166O/15 mm. and the unilide forms white needles m.p. 106-107'. 2 3-Dimethoxybenzaldehyde was also converted into the bisulphite compound and this was filtered washed with alcohol and treated with a saturated solution of potassium cyanide. On pouring the product into water 2 3-dimethoxymundelonitrile solidified. It crystallised in white microscopic tablets and prisms m. p. 76' and was hydrolysed to 2 3-dimethoxymandelic acid C,H,(OMe)2*CH(OH)*C02H by solu- tion in concentrated hydrochloric acid. This acid crystallises in microscopic white prisms m. p. 93' and forms an acetyl derivative m. p. 114'. The nitrile was also heated with hydriodic acid but the 2 3-dihydroxyphenylacetic acid could not be purified and was there- fore methylated by methyl sulphate and the ester hydrolysed. 2 3-Dimethoxyphenylacetic acid CI0Hl2O4 was then obtained in white needles m.p. 82-43'.ORGANIC CHEMISTRY. i. 191 [With E. ROSSELET.1-Attempts were also made to obtain 2 3-dihydroxyacetophenone by the action of aluminium chloride on guaiacol chloroacetate (compare Fries and Finck A. 1909 i 42)) but rearrangement into the para-position resulted the product being o-chloro-3 4-dihydroxyacetophenone (Mannich and Hahn A. 1911 i 649). The above compound was prepared by heating guaiacol with chloroacetyl chloride and obtair?ed in white needles m. p. 60' (compare Einhorn A. 1903 i 90 and Dzierzgovsky A. 1894 i 74). Creosol chloroacetate (Einhorn loc. cit.) was pre- pared by the same method. When heated to 100" with aluminium chloride i t yields 3 4-dihydroxy-o-tolyl chloromethyl ketone C,H,03Cl which crystallises in almost colourless microscopic prisms m.p. 130° provokes to sneezing and gives dark green colorations with ferric chloride. 3 4-Dihydroxy-o-tolyl methyl ketone 0 H1003 was prepared from the compound by the action of zinc dust in acetic acid. It formed white needles m. p. 168-169' and gave a dark green coloration with ferric chloride. The chloroacetyl group was eliminated from the above chloroacetates and also from p-creosol chloroacetate (Fries and Finck Zoc. cit.) by the action of aniline in cold ethereal solution when w-chloroacetanilide was obtained. J . C. W. Dinaphthgl Ketones and Dinaphthyl Ketimines. A. E. TSCHITSCHIBABIN and S. I. KORJAGIN (J. Buss. Yhys. Chem. Xoc. 191 3 45 1823-1829).-The authors have prepared the three dinaphthyl ketimines by the method given by Moureu aud Mignonac (A.1913 i 873)) and have converted them by boiling with dilute hydrochloric acid into the corresponding ketones which are obtained in good yields. Di-/I-naphthyl ketimine C(C,0H7)2:NH crystallises in white plates m. p. 121*5-122*5O and is a moderately strong base; its hydyo- chloyide and picrate m. p. 230.5-232" (decomp.) mere prepared. a-Nc6phthyl P-nuphthyl ketimhte C(C,,H,),:NH crystallises in hemispherical masses of prismatic needles m. p. 103-104" ; the hydrochloride and picrate m. p. 223.5-225-5' (decomp.) were prepared. Di-a-naphthyl ketimine C(CloH7),:NH forms white prismatic needles m. p. 87-88'; its hydrochloride and picrate m. p. 191-192.5' (decomp.) were prepared.Di-a-naphthyl ketone forms a picmte crystallising in large prisms m. p. 121*5-122'. T. H. P. Benzyl Derivatives of Anthraquinone and 05 Indigotin. M. TSCHILIKIN [with E. PAULSEN] (J ~ u s s . Phys. Chem. soc.,''1913 45 1834-1845).-The author has investigated the interaction of antfira- quinone and phenyl benzyldimekhylammonium chloride in presence of sodium hydroxide and hyposulphite. According to the results of Hole (Fdde Zeitung 1910 244) and of Porai-Koschic (J. Buss. Phys. Chem. Soc. 1910 42 1079) this reaction would be expected to lead to ethers of anthraquinol (oxanthranol) the latter being formed by the reduction of the anthraquinme. The product obtained is however,i. 192 ABSTRACTS OF CHEMICAL PAPERS. analogous to that obtained by Liebermann (A 1882 855) and must he regarded not as an ether of oxanthranol but as an alkyl derivative namely benzyloxanthrone.I n its formation the anthraquinone is first converted into the disodium derivative of anthraquinol which is a red unstable compound and readily undergoes re-oxidation t o ant hraquinone. thus -+ This comDound From this disodium compound <he reaction proceeds ONa CH,Ph OH CH,Ph \/i .\/ /\A/\ /\/\/\ I I/ I I -+ I I/ I1 I \/\/\/ \/\//\/ /'. ..* I1 ONa C1 0 C C C C has the properties ascribed to it by Liebermann (Zoc. cit.),ievi (A. 1885 1140) and Bach (A 1890 1144) and on treatment with concentrated sulphuric acid the >C(CH,Ph)*OH group is transformed into >C:CHPh the compound thus formed having m. p. 117' and not 127' as stated by Levi (Zoc. cit.).I n addition t o beuzyloxanthrone the above reaction gives yellow needles o€ 9-hydroxy-9-benxyl-9 10-dihydroanthracene C(OH)( CH Ph) m. p. 135-136' which Bach (A. 1890,1425) described as 10-hydroxg- 9-benzpl-9 10-dihydroanthracene m. p. 130-140'. The above results throw a new light on the structure of the benzyl- derivatives of leucoindigotin and ethers of indigotin in general as well as on the whole reaction of benzylation which is of great technical interest. That two isomeric leucoindigotins exist may be regarded as indicated by the existence of two diacetyl derivatives to which the author attributes the structures C6H4<-CH 2 > C f 3 H 4 ? and only one tetra-acetyl derivative is known corresponding with the /?-form. Leucoindigotin is assumed to consist of an equilibrated mixture of the ketonic and enolic isomerides corresponding with the above diacetyl compounds the eiiolic modification always being formed in the larger quantity.Etherification by the action of benzyl chloride either alone or in presence of dimethylaniline or by the action of phenylbenzyldimethylammonium chloride (leucotrope) yields only one benzyl derivative for which the constitution is proposed. Evidence such as the identity of this compound with that obtained from the organo-magnesium compound is quoted in support of this structure. T. H. P.ORGANIC CHEMISTRY. i. 193 Preparation of P-Nitro-derivatives of Acyl-p-diaminosnthra- quinones. FARBENFABRIKEN VORM FRIEDR. BAYER & Co. (0.R.-P. 267445).-When 1 4-dibenzoyldiziminoauthr q u i n ~ r l e IS treated wich nitric acid in nitrobenzene suspension a t 90° it yields 2-nitro-1 4- dibenxoyldiaminoanthrapuinone orange needles which on hydrolysis gives 2 - mti-o - 1 4 - diaminoanthraquinone greenish-blue needles.Similarly 1 4-diamiuoanthraquinoneurethane gives 2-nztro-1 4-di- arniikoanthraquinoneurethune orange-red prisms which on h y drol J sis yields the above 2-nitro-1 4-diaminoanthraquinone. The urethane of 1 4 5 8-tetra-aminoanthraquinone yields a 2-nitro-derivative which on hydrolysis furnishes %nit&- 1 4 5- 8-tetra-aminoanthraquinoae. J. C. C. The Constituents of Oil of Calamus. H. THOMS and R. BECKSTROEN (Ber. 1913 46 3946-3948).-The divergence between the results of the authors (A 1901 i 396; 1902 i 809 810) and thobe recently published by Seumler and Spornitz (this vol.i 69) is attributed to the different origin of the samples investigated that of the latter investigators being a Russian oil whilst that of the authors was Japanese. D. F. T. The Constituents of Ethereal Oils. (The Sesquiterpene Fraction of Java Citronella Oil.) P. W. SEMNLER aud K. E. SPORNITZ (Bdr. 191 3 46 4025-4029).-A specimen of the sesquiter- pene fraction of Java citronella oil was freed from methyleugenol by washing with dilute alcohol. It then had the constauts b. p. 153-166'/ 15 mm. D 0,8659 n 1,50386 and the analysis suggested the . presence of some sesquiterpene alcohol. After boiling with sodium followed by several fractionations sespuicitronellene C H24 was obtained ; b. p. 138-140'/9 mm. D 0.8489 n 1.53252 a + 0'36'.The exaltation of the molecular refraction 7453 instead of 69.6 the reduction by sodium and alcohol to inactive dihydrosesquicitronellene C15H26 b. p. 131-133O/12 mm. D 0,8316 ~n 1.4800 and the reduction by hydrogen in presence of platinum to inactive octahydro- sesquicitronellene CI5H3 b. . p. 115-1 17O/Y mm. D 0.7789 n 1.43518 indicate that the oil is an aliphatic sesquiterpene with two conjugated double linkings homologous with the aliphatic terpene ocimene (Eoklaar A. 1906 i 377). cycloSespuicitronelZene C15H249 was obtained by heating the aliphatic oil with concentrated formic acid. It had b. p. 129-138"/15 mm. D2,,- 0.8892 n 1,5069 a,+56'. I t could no longer be reduced by sodium and alcohol but reduction with hydrogen and platinum did not conclusively prove whether the compound is moao- or di-cyclic.The mashed starting material was also ozonised and amonq the products of subsequent hydrolysis was found a parafin C20H38 b. p. 165-167°/15 mm. D 0,8387 n 1.46370. It 1s uncertain whether the substance is a normal constituent or due to sophistication of the citronella oil with petroleum. The above-mentioned sesquitsrpene alcohol C,,H,,O was isolated. It is tertiary contains conj ugitted double linkings and readily submits to ring formation. J. C. W. VOL CVI. i. 0i 194 ABSTRACTS OF CHEMICAL PAPERS Review of the Pioneer Work on the S y n t h e s i s of Caout- chouc. FRANCIS J. POND (J. Arner. Chem Xoc. 1914 36 166-199). -The literature relating to the early work on the synthesis of caoutchouc is reviewed in detail The author bas repeated Bouchardst's experiment but has failed to obtain any product which could be regarded as caoutchouc.E. G. T h e Oleo-resins of Jeffrey and Singleleaf Pines. A. W. of Pinus monophylla contains from 80 to 85% of d-a-pinene; 4 to 5% of I- or i-limonene; 4 to 6% of d-cadinene ; losses by polymerisation etc. 4.5%. The colcphony contains 7.22% of resen and resin acids isomeric with abietic acid. The volatile oil of Pinusjefreyi consists of about 95% of n-heptane and 5% of an aldehyde apparently citronellal. The colophony contains 12.5% of resen and resin acids isomeric with abietic acid. w. P. 8. SCHORGER (J. Ind. Eng. Chew. 1913 5 971-973).-The volatile Oil The Main Constituent of Japanese Lac. IV. Some Derivatives of Hydrouruehiol.RIKO MA JIMA and IKUYA NAKAMURA (Ber. 1913 46 4080-4088. Compare A 1912 i 883). -Further derivatives of hydrourushiol confirm the assumption that i t contains two ortho-hydroxyl groups and a straight hide-chain para t o one of the hydroxyl groups. The nature of this alkyl substituent is shown by oxidation with peruanganate in cold aqueous acetone t o palmitic acid The side-chain should therefore be C15H31 and not C14H20 as was originally 6uggeskd. I n another paper (this vol. i 167) the preparation of 1-tetradecyl- and 1-pentadecpl-3 4-di- methoxybenzenes is described. Unfortunately neither of these products is identical with dimethoxyhydrourushiol although they seem to be nearly related to it. The following derivatives of hydrourushiol are described lead salt ; dibenzoyl derivative colourless needles m.p. 59-60.5' ; di-m-dinitro- benzoyl derivative white spherical tufts m. . p. 93' ; 6romodiacetyZ- hydrourushiol colourless m. p. 35-43' dzbro~~ohydrourucshiol a brown powder m. p. 60-66'. Dirnethoxyhydrourushiol was nitrated and f u i ther derivatives were obtained from the nitro-compounds. Fuming nitric acid in acetic acid solution yielded nitrodinzethoxyhydl.c,~ydroui.ushioZ in long needles m. p. 72-73' which were dissolved in fuming nitric acid aud thus con- verted into a &nitro-derivative which formed almost colourless long flat crystals m. p. 53'. The mononitro-derivative was readily reduced by zinc dust and glacial acetic acid to aminodiinethozyhydrourushiol which yielded small purple granules m.p. 65-66.5'. The white hydyochloride had m. p. 152-153' ; the platinichloride formed slender yellow prisms m. p. 17S0 and the acetyl derivative had m. p. When warmed with a mixture of fuming nitric acid and sulphuric acid it yielded dilz~trornonocccetyEhyd~ourush~o2 which formed colourless needles m. p. 69-70*5' and gave a yellow potassium salt. It was hydrolysed to 66-67' Diacetylhpdrourushiol was very difficult to nitrate.ORGANIC CHEMISTRY. i. 195 dinikohydrourushiol pale yellow needles m. p. 122-1 22.5') and this was converted into the diacetyt derivative pale brown granules m. p. 72-5-73.5'. The analyses of all these derivatives give values which agree with either C,,H or C,,H for the side-chain. On heating hydrourushiol over a free flame or a t 350-400' in a sealed tube only catechol could be isolated from the decomposition products.J. C. W. Synthesis of Picein the Glucoside of the Pine (Pinus picea); New Artificial Glucoside. F. MAUTHNER (J. pr. Chm. 191 3 [ii] 88 764-770).-The gluco-p-hydroxyacetophenone obtained by the condensation of /?-acetobromoglucose with p-hydroxyaceto- phenone and hydrolysis of the resulting tetra-acetogluco-p-hydroxy- acetopheuone (A. 1912 i 574) is now found to be identical with picein the glucoside isolated by Tanret (A 1894 i 616) from the needles of the pine. The preparation of tetra-acetogluco-p-hydroxyacetophenone is best accomplished by t h e gradual addition of aqueous sodium hydroxide to a solution of p-hydroxyacetophenone and P-acetobromoglucose in acetone the temperature not being allowed to rise above 16'.Tetra - acetogluco - p - hydroxybenxophenone C27H28011 . prepared by shaking an ethereal solution of /3-acetobromoglucose with p-hydroxy- benzophenone dissolved in aqueous sodium hydroxide crystallises in colourless needles m. p. 167-168O and is hydrolysed by barium hydroxide at the ordinary temperature to gluco-p-hydroxybenzophenme which forms colourless needles m. p. 178-1779' [a] - 55.58' in alcoholic solution. Methyl tetra-acetogluco-m-hydroxybenxoate C22H26012 prepared from /?-acetobromoglucose and methyl rn-hydroxybenzoate iu acetone solution in the presence of sodium hydroxide has m. p. 114-115' and is hpdrolysed by aqueous barium hydroxide t o gluco-m-hydroxybenxoic mid C13H1608 crys tallising in colourless needles m.p. 143-1 44O [ax - 68.41' in aqueous solution. biCrystallised Rhein" of Commerce. 0. A. OESTERLE and E. R. HAUGSETH (Arch. Phnrm. 1913,25 1,550-552).-" Crystallised rhein " of commerce is insoluble in cold aqueous sodium carbonate and does not contain a methoxy-group and therefore does not contain rhein (1 8-dihydroxyanthraqiiinone-3-carboxylic acid) or emodin methyl ether. Its m. p. is 192' and after recrystallisation from benzene 196O ; its acetyl derivative has m. p. 208'. These constants are those of chrysophanic acid and its acetyl derivative respectively. '' Crystallised rhein," therefore is almost pure chrysophanic acid ; the analytical data correspond with the formula C,,H,&IeO,(OH),. F. B. c. s. Comparative Examination of the Constitution of Tannin from Turkish and from Chinese Gall-nuts.K. FEIST and H. HAUN (Arch. Phcwm. 19 13 25 1 468-526).-A comprehensive survey is given of the chemistry of tannin since its discovery by Scheele in 1787. 0 2i. 196 ABSTRACTS OF CHEMICAL PAPERS. Finely powdered Chinese gall-nuts are dried at 100' and then extracted successively with chloroform benzene and ether. (The last solvent removes a crystalline substance which proves to be gallic acid.) The tannin is then extracted from the galls by means of acetone. The tannin is obtained from the acetone extract as an amorphous powder and yields by hydrolysis with boiling dilute sulphuric acid 93.5% of gallic acid and about 5.5% of dextrose (identified by its reducing properties rotatory power fermentibility and in the form of its OFazone m.p. 205'). Hence tannin from Chinese gall-nuts like that from Turkish contains chemically bound dextrose ; the latter contains more dextrose than the former. Turkish tannin freed as completely as possible from glucogallic acid yields by hydrolysis gallic acid and about 9% of dextrose. Since tannin therefore is apparently a compound (or several com- pounds) of gallic acid and dextrose the most rational method of ascertaining its constitution is an examination of the '' simplest tannio," that is the crystalline substance glucogallic acid which is extracted by ether from Turkish galls after they have been extracted with chloroform and with benzene. Glucogallic acid has a molecular weight about 320 (by titration and by the ebullioscopic method in acetone) and yields about equal molecular quantities of gallic acid and dextrose by hydrolysis with boiling dilute sulphuric acid.By methyl- ation by an excess of ethereal diazomethane glucogallic acid yields a substance m. p. 7' sofltening at 60° yellow crystals which does not develop a coloration with ferric chloride does not reduce Fehling's Eolution even after boiling for a short time contains 49"L of methoxyl and is hydrolysed not by boiling dilute sulphuric acid but by methyl alcoholic potassium hydroxide yielding gallic acid trimethyl ether. Hence in glucogallic acid the dextrose cannot have condensed with the gallic acid at a phenolic hydroxyl group. Glucogallic acid there- fore is an ester of dextrose and gallic acid. The following considera- tions determine a t what point of the dextrose molecule ester formation has occurred.Glucogallic acid reduces Fehling's solution (this does not prove the presence of an aldehyde group because gallic acid con- trary to statements in the literature also reduces Fehling's solution) but does not form an osazone ; moreover methylated glucogallic acid does not reduce Fehling's solution. Hence glucogallic acid does not con- tain a free aldehyde group and therefore has the constitution 1--0- I C,H,(OH),*CO*O*CH*[CH*OH]2-CH*CH(OH)*CH2*OH. It is quite different in its behaviour from Fischer and Strauss's /3-glucosidogallic acid 1 C0,H*C,h',(OH),*O*CH~[CH*OH]2*CH*CH(OH)*CH2*OH. Attempts have been made to synthesise glucogallic acid by means of bromotetra-acetylglucose. I n the first place i t is shown t h a t bromo- tetra-acetplglucose in ether reacts with an excess of silver phosphate or arsenate in the sense of the equation C,H70Br(OAc)4+ AgA = AgBr + C,H70(0Ac),A (where A denotes an acid radicle).Silver triacetylgallate cannot be prepared but by shaking an ethereal solutionORGANIC CHEMISTRY. i. 197 of bromotetra-acetylglucose and triacetylgallic acid with silver car- bonate silver bromide and silver acetate are formed and an amorphous substance is produced which is probably incompletely acetylated gluco- gallic acid ; the evidence for this deduction however is by no means satisfactory. Attempts to acetylate glucogallic acid have been unsuccess- ful as also have been attempts to hydrolyse the preceding amorphous substance both decomposing during the respective operations.By treatment with an excess of ethereal diazomethane Chinese and Turkish tannins hare been converted into methylated derivatives which have been separated into fractions containing different per- centages of methoxyi. This result is regarded as yet another proof of the heterogeneity of tannin. The same conclusion is drawn from the results of an investigation of the rotatory powers in water and in 96% alcohol of various samples of commercial tannin. Although glucogallic acid cannot be extracted from Chinese gall-nuts and is not produced by the hydrolysis of Turkish tannio the authors regard it as highly probable that this acid is the basis of the Turkish and the Chinese tannin molecules. Both are esters of gallic acid and dextrose; in Chinese tannin the percentage of gallic acid is milch greater than that of dextrose.I n agreement with this the molecular weight of Turkish tannin in .acetone is very much smaller than that of Chinese tannin and all the fractions of methylated Turkish tannin contain a greater percentage of methoxyl than the corresponding fractions of methylated Chinese tannin. The fraction of methylated Chinese tannin which is insoluble in cold absolute alcohol is identical with Herzig's methylotannin; it is not an individual substance however its percentage of methoxyl correspond- ing with that required by a mixture of methylated undeca- and dodeca-galloyldextrose. The fraction of methylated Turkish tannin which is Eoluble in cold absolute alcohol is excluding glucogallic acid the only crystalline derivative cf tannin hitherto obtained ; i t again however is not an individual substance but is a mixture of methylated mono- di- and tri-galloyldextrose.c. 5. Combination of Caffeine with Phenols. A. BAUMANN (GIbem. Zentr. 1913 2 2036-2037; from Arb. Pharrn. Imt. U&v. Berlin 10 127-147).-Chlorocaffeine has been condensed with a number of phenols reaction being carried out in the presence of alkali hydroxide in aqueous solution under reflux sometimes at higher temperatures under pressure and sometimes in the presence of xplene. The phenol compound of caffeine CI4Hl4O3N forms needles m. p. 143' which are soluble in concentrated hydrochloric acid without decomposition and are stable towards 15% aqueous sodium hydroxide. The guaiacol compou.nd prepared by heating t'he requisite materials in aqueous solution a t 140° consists of colourless needles of faintly bitter taste m.p. 151-152'. The thymol compound C,8H2203N4 needles m. p. 130-132' is prepared in xylene solution at 155'. With the three cresols in aqueous solution chlorocaff eine yields the 0-cresol compound C,,H,,O,N needles m. p. 150-151" the m-cresol compound needles m. p. 144-145' and the p-cresol compound needles m. p. 138-139Oi. 198 ABSTRACTS OF CHEMICAL PAPERS. The eugenol cornpound C,,H,,O,N consists of needles m. p. 11 9-120° whilst the carvuwol compound needles has m. p. 145-146'. The cntechol-monocaffeine compound [o-hydroxyphenoxycafleine] C1,Hl,O,N crystallises in needles m. p. 215' which are soluble in dilute sodium hydroxide and yield an acztyl derivative CI6Hl6O5N4 needles m.p. 186-1 87". The corresponding di-caffeine compound has m. p. 280-281' and is insoluble in dilute sodium bydroxide. I n a similar manner resorcinol yields a monocaff eine compound needles m. p. 197-198' (acetate m. p. 123-124') and a dicaffeine compound m. p. 288-289'. Analogously quinol gives a monocaffeine com- pound m. p. 206-207' (acetate m. p. 210-211") and a dicaffeine compound m. p. 285". p-Hydroxybenzoic acid likewise condenses with chlorocnff eine to yield the substance Cl,fI,,O,N m. p. 213-244' needles of slightly bitter taste soluble in dilute sodium carbonate solution. o-Hydroxybenzoic acid and methyl salicylate could not be condensed with chlorozaffeine in the presence of alkali whilst hydroxy- caffeine could not be condensed with phenols. The phenol and thymol compounds of caffeine are stable towards boiling coricentrated hydrochloric acid and are not attacked a t 160' by a solution of hydrogen chloride in ether or xylene.They are however decomposed by concentrated aqueous hydrochloric acid a t 160" into pbenol or thymol and hydroxycaffeine. A similar decompo- sition is caused by heating with aqueous more rapidly with alcoholic sodium hydroxide. H. W. Fermentative Decomposition of Creatinine. 11. D. A CKER- MANN (Zeitsch. Biol. 1913 63 78-82. Compare A. 1913 i 1049).- N-Methylhydantoin could not be detected as a product of the decompo- sition of creatinine by means of macerated cat's liver in the presence of toluene. When creatinine is decomposed by putrefactive bacteria for a period of four weeks sarcosine is formed.It is considered that the sarcmine is derived from 117-methylhydantoin but other possibilities of its formation are suggested also X. I?. A. Preparation of Salts of Hgdrastinine and its Homologues. HERMAN DECKER (D.R.-P. 267699. Compare A 1911 i 906).-Salts of hydrastinine and its homologues of the annexed general fornula (R = hydrogen alkyl aryl or alkylaryl ; CH2 X = acidic group) are obtained by con- densing N-acylalkyl derivatives of homo- (). /\/'\CH CH2<O*t INX*Alkyl piperonylamine \/\/ CR CH, 0, c,B,*cH,*cH,*N<~~~ and converting the dibydroisoquinoline derivatives so obtained into their salts. N-Form ylmethylhomopiperonylamine when condensed by means of phosphoric oxide yields hydrastinine ; N-formy lethylhomopiperonyl- amine gives N-ethy lnorhydras tinine (platinichlorids brown prisms m p.1 95-1 97' decomp.) and beneoyl-N-methylhomopiperonyl- amine m. p. 80-81' furnishes 1 -p~~enylAydrastinine which formsORGANIC CHEMISTRY. i. 199 white needles m. p. 152-153' (decomp.). base forms silky needles m. p. 98-99' The picmte of the latter J. C. C. Strychnos Alkaloids. XIX. Some New Oxidation Pro- ducts of Bruoine. HERMANN LEUCHS and HUBERT RAUCH (Be?. 1913 46 3917-3922).-1n addition to brucinonic acid and dihydro- brucinonic acid (Leuchs and Brewster A. 1913 i 210) the oxidation of brucine by potassium permanganate in acetone solution gives rise to a mixture of nehtral substances which amounts to 10% of the brucine used. By a complex process of extraction and reaytallisation with various solvents the amorphous mixture could be made to yield three new crystalline substances.The chief of these appears to have the composition C23H2407N2 and the name trihydroxydehydrobrucine is suggested ; it crystallises in cubes or prisms which are very sparingly soluble in chloroform and benzene and has [u] +87.4' (in hydro- chloric acid D 1.19) and m. p. 336' (decomp.). I t is of interest that the solution in acetic acid exhibits no optical activity The second product which has the composition C,,H,,P,N of a trihydroxybrucine and so is possibly a dihydro-derivative of the previous compound crystallises from benzene solution in needles or prisms m. p. 130-135O (decomp.) which contain benzene of crystallisa- tion; the product recrystallised from water has m.p. 240' (decomp.). I n acetic acid solution the substance has [u] + 5 - 6 6 O . The third substance C,,H,,O,N crystallises from acetic acid in pale green quadratic pyramids m. p. 290' (decomp.) [ag + 72.8' (in acetic acid). All three substances give the well-known colour reaction x i t h nitric acid. D. F. T. Synthesis of Hygrine. 11. Synthesis of Racemic Hygrine. A New Oxidation Method. Further Remarks on the Catalytic Reduction of Pyrrole Derivatives. KURT HESS (Ber. 1913 46 4104-4115. Compare A. 1913 i 1378).-In the former paper an account was given of the preparation of two isomeric 1-methyl- pyrrolidylpropanols one of which should yield hygrine on oxidation to the ketone. Difficulty was then experienced in the methylation of the pyrrolidy 1 derivative.For this purpose formaldehyde has now been employed with the unexpected result that oxidation of the hydroxyl group also took place during the one operation so that almost quantitative yields of the desired ketones were obtained. A comparison of these compounds with hygrine establishes the alkaloid although the resolution of the synthetic substance a i d the rac>matioa of the natural product are yet to be accomplished. Willstatter's assumption (A 1900 i 405) is therefore justified. The oxidation of hydroxy-aminee to amino-ketones (according to the scheme :C*NHR.**CH(OH).** + CH,O = :C*NMeR***CO..* + H,O) seems to be of general application and is the subject of future worki. 200 ABSTRACTS OF CHEMICAL PAPERS.by the author. The preparation of diacetonmet hylamine from diacetonalkamine is now described. a-2-Pyrrolidylpropan-a-01 was acidified with hydrochloric acid and heated with 40% formaldehyde at 115-120' in a sealed tube The CH,*r hle product a-l-met?~ylpyrrolidyI-2-iuroiunne-a-one was dried over barium oxide and obtained as a colourless mobile hygroscopic oil b. p. 70-75"/12 mm. bath 90-95' 69-74'/14 mm. bath 85-90' 80-82'/21 mm. bath 95-100'. I t has a powerful basic and narcotic odoiir is very soluble and gives the pyrrole reaction. It reduces alkaline but not acid permanganate and also silver nitrate and neutral gold chloride but from an acid solution of the latter the auricldoyide was obtained in yellow microscopic needles m. p. 106' (corr.). The oxime C,H,,ON was obtained as a clear syrup b.p. 140'116 mm. which did not form a picrate. The picrate C1,H,,O8NJ4 of the ketone formed long yellow needles which sintered at 95' (corr.) and had m. p. 103' (corr.). Similarly a-2-pyrrolidylpropane-P-ol was converted by formaldehyde in to a- 1 -met?~yZpyr~oZidyl-2-propane-/3-one (formula above) which forms a colourless oil b. p. 79-83'/14 mm. bath 90-looo 89-92'/ 21-22 mm. bath 100-110'. It is not so mobile as the isomeride is not discoloured by light and its poppy-like odour is not so unpleasant. The picrate forms slender yellow needles which sinter at 162' (corr.) and have m. p. 176' (corr.). The picrate from a specimen of natural hygrine with b. p. 79-81'/11-12 mm. bath 90-95' [a];l*2' had m. p. 158' (corr.). The oxime had m.p. 125' (corr,) and the picrate of the oxime agreed exactly with that of the oxime of natural hygrine (Liebermann and Kiihling A. 1893 i 446). Attention is again called to the influence of traces of oxygen on the spongy platinum used in the catalytic reductions. Much better results are obtained when the preparation is not dried in vacuum but is simply wabhed with the solvent which is subsequently employed Admission of air t o the catalyst causes a rise of temperature and cohesion of the finely divided metal. Using spongy palladium as catalyst complete reduction of the pyrryl aloohols t o a-2-pyrrolidyl- propane [2-propylpyrroZidine] C7H15N was effected. This compound is very volatile in steam and could thus be fractionated from the less volatile partially-reduced products.It is a clear mobile narcotic oil b. p. 145-150'/765 mm. Diacetonamine prepared by Heintz's method (A. 1874 1080) was successfully distilled in a high vacuum; b. p. 25'/0*14 mm. I t was reduced by sodium and alcohol to diacetonalkamine (Kahan A. 1897 i 494) b. p. 70- 75'116 mm. and this was heated with 40% formalde- hyde at 115-1 20'. ~iacetonmelhylomine C7H,,0N was then obtained as a colourless mobile very soluble oil b. p. 50-53'/15 mm. It has a peculiar menthol-li ke overpowering odour. A New Method for the Preparation of 1-Substituted Isatine. R. STOLL~ (Ber. 1913 46 3915-3916).-A preliminary account of the formation of 1-substituted isatins by the intra- CH'<CH,*CH*COEt J. C. W.ORGANIC CHEMISTRY. i. 201 molecular condensation of oxalic amide chlorides [chloroglyoxylamides] of the type COCl*CO*NRPh which are obtainable by the interaction of oxalyl chloride and mono-substituted anilines under suitable conditions.C/~l~roglyolc:ylod~p~en~lamide COCl*CO*NPh needles m. p. 70" when treated with water forms diphen$oxamzc acid needles m. p. near 146" (decomp.) and with alcohol produces the corresponding ethyl ester platelets m. p. 87'. If a bimolecular proportion of diphenylamine i p taken with oxalyl chloride the product is OXCdO- dipi~en?/Zaniide plates m. p. 1 6 9 O 'which often accompanies the amide chloride as by-product. The intramolecular condensation of chloroglyoxylodiphenylamide gives 1-phenylisntin which when treated with alkalis is converted into the sodium salt of the corresponding isatic acid ; this regenerates 1 -phenylisatin on the addition of acids.When heated with phosphorus pentachloride phenylisat in gives a dichloro-derivative leaflets m. p 1 00'. Ethylaniline reacts in carbon disulphide solution with oxalyl chloride in the presence of aluminium chloride forming 1-ethylisatin m. p. 95". D. F. T. o-Aminopropiophenone its Acyl Derivatives and Their Condensation to 2- and 4-Hydroxyquinolines. E. WOHNLICH (Arch. Yharm. 1913 25 1 526-550).-o-Nitrop~opiop?~enone b. p. 175"/25 mm. a viscous yellow liquid is obtained i n 42-46% yield by treating a n ethereal solution of o-nitrobenzoyl chloride with a n ethereal suspension of ethyl sodiomethylacetoacetate and hydrolys- ing the resulting ethyl o-nitrobenzoylmethylacetoacetate by boiling dilute sulphuric acid.By reduction with tin and concentrated hydrochloric acid on the water-bath i t is converted into o-amino- pi*opiophenone m. p. 45-46' yellowish-white leaflets from which the following acyl derivatives bave been prepared by the action of the respective acid anhydrides acetyl derivative NHAc*C,H,*COEt m. p. 71" colourless plates; propionyl derivative m. p. 51° stout rhombohedra ; butyryl derivative m. p. 39-40' large rhom bohedral crystals and benzoyl derivative m. p. 130° colourless ueedles. In acordance with Camps' rule (A. 1900 i 115 310; 1902 i 178 336) the preceding o-acylaminopropiophenones except the last yield 2- and 4-hydroxyquinoline derivatives by condensation with the calculated quantity of sodium hydroxide in boiling aqueous alcohol. The 4-hydroxyquinoline derivatives exhibit distinctly basic and acidic properties and develop red colorations with ferric chloride whilst 2-hydroxyquinoline derivatives do not give a ferric chloride reaction and are feebly acidic and basic.Thus o-acetylarninopropiophenone yields IConrad and Limpach's 4-hydroxy-2 3-dimethylquinoline (A. 1892 78) and 2-hydi.oxy-4-ethytpuinoline XI. p. 197" m a l l prisms (hydrochloride long needles ; platinicJ&ride 2C,,H 10N,H,PtCI,,H,O decomp. 199-ZOO" pale yellow needles ; yicrate m p. 149-150° yellow need ies). o-Pro pion y laminopropio phenone yields 4-h ydroxy- 3-me1l~~l-2-ethylpuino2ine m. p. 297" (hydrochloride m. p. 180-1 85" ; NO,*C,H:,*COEt,i. 202 ABSTRACTS OF CHEMICAL PAPERS. picrate m p. 184' yellow needles ; ptatinichboride 2C,,H,,0N,H2PtCI,,2H20 decomp.21 8' [rapidly heated] reddish-yellow needles) and 2-hydroxy- 3-methyZ-4-ethyZquinoZine m. p. 1 88' colourless needles (picrate m. p. 183" ; hydrochloride m. p. 120-1 30'). o-Butyrylaminopropiophenone yields 2-hyd~oxy-3 4-dicsthplquinoline m. p. 174-175' (picrate m. p. 156-1 57' small yellow needles ; platinichloride 2C SH ,ON H,Pt Cl,,H20 m. p. 194-1 95' red needles) and 4-hydroxy-3-methyZ-2-propyZ- puinoline m. p. 275 (picrate m. p. 179'; pZatinic?2oride 2 C,,H,,ON H,PtCI 6 2H,O m. p. 201-202°). Only one quinoline derivative can be formed by the condensation of o-benzoylaminopropiophenone. 4-Hydroxy-2-phenyE-3-math~t~~inoZine m. p. 276O develops a coloration with ferric chloride but is only feebly basic. By treatment with phosphorus pentachloride the preceding hydroxy- quinolines are converted into the corresponding chloro-derivatives ; by reducing the latter by hydrogen iodide in glacial acetic acid or by dis- tillation with zinc dust the corresponding alkylated quinolines are obtained.It is found that chlorine in position 4 is more difficultly re- moved than chlorine in position 2 ; the difficulty increases with the number of carbon atoms in the alkyl groups so that the halogen can be eliminated from the higher 4-chloroalkylquinolines only by distillation with zinc dust. 4-Chtoro-2 :3-dimethylpuinoZine has m p. 78' (decomp.). 2-ChZoro-4-ethylquinoline m. p. 76-77' crystallises in colourless prisms. 4- ChZoro-3-meth yZ-2-ethyZquinoZine and the corresponding io do- compound have m. p.22-23' and 64-65' respectively. By distilla tion with zinc dust under diminished pressure 4- hydroxy-3-methyl- 2-propylquinoline yields two substances one of which is probably 3-methyl-2-propyZqeinoline m. p. 59' (platinichloride m. p. 228O) whilst the other m. p. 65-70' is possibly an indole derivative since i t responds to the pine-shaving test. 2-C'hloro-3-methyZ-4-ethyZquinoZine m. p. 7 2-7 3O is converted into 2-iodo-3-rnethyZ-4-ethyZquinoZine9 m. p. 103O by hydriodic acid potassium iodide and amorphous phosphorus but is reduced by hydrogen iodide in glacial acetic acid a t 275' to 3-methyZ-4-ethyZpuinoZ~ne ?. p. 172-1 73'127-28 mm. (picrate m. p. 196-197O ; platinichlorzde m. p. 230' reddish-yellow needles). In a similar manner 2-chioro-3 4-diethylquinoline b.p. 203-205°/24-25 mm. is reduced to 3 4-diethylquinoZine b. p. 177-178'/25 mm. (picrate m. p. 179" yellow needles ; pEatinichZoride m. p. 230' [decomp.] yellowish-red needles). Equal molecular quantities of o-aminopropiophenone and ethyl acetoacetate react a t 160-165O to form only Z-hydroxy-3-acetyk 4-ethyZpuirnoZine m. p. 198-199' colourless needles which is quite devoid of basic character whilst under similar conditions o-amino- propiophenone and ethyl benzoylacetate yield 2-hydroxy-3-benxogl- 4-ethylquinoline m. p. 213O about 8-10% of the intermediate sub- stance CH,Bz*CO*NH*C,H,*COEt m. p. 94-95O also being obtained ; the latter is converted into the former by aqueous alcoholic sodium hydroxide. c. s.ORGANIC CHEMISTRY. i 203 Preparation of Anthraquinoneacridinecarboxylic Acids.FARBENFABRIKEN VORM. FRIEDR. BAYER & Co. (J).R.-P. 262469 j.- 1-Anilinoanthraquinone-2-carboxylic acid when heated with sulphuric acid (60°B6) t o 90-looo yields a mixture of anthraquinone-l :2- acridone and l-acridineanth~*upuinone-2-carboxylic acid Similarly 1 -p-chloroanilinoant~1raquinone-2-carboxylic acid gives 1-p-chloro- ncridineanthraquinone-2-carboxylic acid and a corresponding acid is produced by the condensation of 1 -p-toluidinoanthraquinone-2-carboxylic acid. J. C. C. The Decomposition of Phenylbenzylideneisooxazolone by Phenylhydrazine. ANDRB MEYER (Bull. SOC. chim. 19 13 [iv] 13 1106-1107. Compare D i n s and Griffin A. 1913 i 1086).- l-Phenyl-4-benzylideneisooxazol-5-one is decomposed by boiling in alcoholic solution with phenylhydrazine giving phenylbenzylidene- hydrazone.Hydroxylamine and bases capable of combining with aldehydes and ketones can take the place of the phenylhydrazine. The reaction is general to such isooxazolones as the above. W. G. Preparation of Anthraquinone-N- 1 1 '-oxazines. FARBWERKE Antbraquinone-N 1 1'-oxazines are obtained by heating a mixture of 1 -halogen-anthraquinones having the 2-position free and 2-halogen-1 - aminoanthraquinones in nitrobenzene solution with sodium acetate (to fix mineral acid) and a catalyet. VORM. MEISTER LUCIUS & BRUNIXG (D.R.-P. 266945 266946).- Anthrapuinone-N-1 1'-oxaxine C,,H,02<-O->C,,H,02 NH is pre- pared from 1-chloroanthraquinone and 2-bromo-l-aminoanthraquinone using copper acetate as catalyst.It separates from quinoline as a brownish-violet crystalline powder. The compound is also obtained (second patent) by heating a mixture of 1-nitro-2-hydroxyanthra- quinone and 1-chloroanthraquinone in nitrobenzene solution with metallic copper. J. C. C. Preparation of Anthraquinonethiazoles. BADISCHE ANILTN- & SODA-FABRIK (D.R.-P. 267523. Compare this vol. i 87).-When 2-aminoanthraquinones containing a free ortho-position are heated with sulphur or sulphur-yielding substances and benzylidene chloride (instead of benzotrichloride as previously described Zoc. cit.) anthra- quinonethiazoles are produced. These are yellow crystalline powders which dye cotton in yellow shades from the hyposulphite vat. J. C. C. P r e p a r a t i o n of Phenazselenonium Colouring Matters.FARB- interaction of potassium selenocyanate and o-nitrodiazobenzene yields o-ni troselenocyanobenzene which with alkalis gives o-nitroselenophenol and this on reduction furnishes o-aminoselenophenol. The latter condenses with di- or tri-nitrohalogenbenzenes to nitroselenazines which on reduction and subse- quent oxidation give phenazselenonium colouring 1 3-ljinitro~~~nosel~ncczi.ne (annexed formula) prepared from picryl chloride and o-aminoseleno- WERKE VOHM. MEISTER LUCIUS dZ BRUNIM (D.R.-P. 261969).-The NH I I 1 'NO matters \/\/\/ Sei. 204 ABSTRACTS OF CHEMICAL PAPERS. phenol is a deep red powder m. p. 195'. It yields by the above treatment 1 3-diaminophenazseZenoniun chloyide dark olive needles. On nitration dinitrophenoselenazine gives 1 3 6-lrinitro33heno;eZenaxine a red powder m.p. 180° and the 1 3 G-triaminophencczseZeno~i~m chloride prepared from it forms needles with a metallic lustre. J. C. C. The Mechanism of the Rearrangement of Amiaobenzyl- a n i l i n e s i n t o Dipbenylmethane Bases. A Contribution to the Partition of Chemical Valency. J. VON ERAUN and 0. KRUBER (Bey. 1913 46 3952-3965).-p-Aminobenzylaniline and it.s deriv- atives when warmed with aromatic amines in acid solution become con- verted into pp'-diaminodiphenylmethane derivat.ives in the manner represented by the equation N~le,*C,H;CH,*NH*C,~,~e + C6H5NMt.2 = NRle2* C,H4*CH2*C,H,*NMe2 + C,H,Me*NH but I ittle i s known of the mechanism of the change A suggestion bas been made (von Eraun A. 1908 i 684) that the action occurs in two stages the first involving the fission of the molecule of tbep-amino- benzylaniline derivative by the action of water the second stage being the condensation of the p-aminobenzyl alcohol derivative formed with the aromatic amine producing the diaminodiphenylmethane com- pound; taking the above example the rquations would be NMe,*C',H,*CH,*NH*C,H,Me + H20 = and NMe,*C,B,*CH,-OH + C6H,*NMe - NMe,*CGH,*CH,~C,H,*NMe2 + H,O. This view of the change has proved n useful working hypothesis (com- pare von Braun and collaborators A.1912 i 968; 1913 i 1327 1330) but as aminobenzyl alcohol compounds with a tertiary nitrogen atom cannot be made to give the condensation assumed above the explanation must be sought in an entirely different direction.New facts have now been discovered concerning the rearrangement. The solutions used must not be too acidic otherwise the yield will be far from satisfactory; the amount of acid should he that required for dissolving the reagents. It is found that such a solution for example of the dihydrochloride of a p-aminobenzylaniline has characteristic properties and on warming or adding a little alkali a reddish-yellow colour appears which disappears when a large excess of acid is intro- duced. m- Aminobenzylaniline derivatives do not produce this colour change neither do they give the rearrangement with aromatic amines. The conclusion is therefore drawn that it is those molecules of the p-arninobenzylaniline derivative which are less rich in acid which enter into the action and that tbe lack of acid must exert a weakening affect on the linking C*N in the centre of the molecule.'I'he colour reaction mentioned above also indicates that the same influence cause8 a tendency t o alteration to some quinonoid configuration. The only explanation which appears to be compatible with the evidence is that represented by the scheme NMe,* C6H4* CH,*OH + C,H41\IXe*NH NMe,(HCI) C,H,*CH,*NH* C,H 4Me -+ or NMe,*C6H,-CH2*NH(HCI)*C6H4Me -+ NBle,C1~-C,H,~-CH,*~~NH~C,H4Rle 1 .... . .............. . H __...__.._.______.__..... ! NMe,-=C,H,==-CH2***NH2*C6H4Me. i-... ..... _.-__- ..-. a1 -.-_....._...__.....- IORGANIC CHEMISTRY. i. 205 The weakening thus produced a t the CON linking in the centre of the molecule renders possible the interaction with a n aromatic amine to produce a diaminodiphenylmethane derivative.It is considered that the balance of the evidence is decidedly in favour of the latter of the two possibilities for the first stage. It is possible that the conversion of diazoamino-compounds into aminoazo-compounds in the presence of the hydrochloride of an aromatic amine depends on a similar preliminary disturbance of the N*NH NPh:N*NHPh,HUl -3 NPhEN***NH,Ph The mole- linking thus I ,,_.__......._. CI--- cule then becomes able to react with the aniline hydrochloride NPh-N***NH,Ph + C6H5NH,,HCl + ! _______.._.. C1 ........ i NPh:N*C,H,*NH,,HCi + NH,Ph,HCl (compare Goldschmidt and Reinders A. 1896 ii 525 556). The following substances are described nz-Aminobenzylaniline hydro- chloride has m.p. 63" (compare Purgotti and Monti A. 1901 i 22). p-Dimethy laminobenzy 1-p-toluid ine di hydrochloride forms colour- less needles m. p. 186-187O and can be converted into a platini- chloride m. p. 189'; when heated in the crystalline conditi0.n or in solution the dihydrochloride assumes a reddish-yellow colour ; this is not due to a fission of the molecule into p-toluidine and a qriinonoid substance of the type NMe,Cl:C,H,:CH for no appreciable amount of p-toluidine is formed. By the action of P-naphthalenesulphonyl chloride on a n aqueous acetone solution of dirnethylaminobenzyltoluidine i n the presence of alkali at the ordinary temperature the ncbphthalenesulphonyl derivative NMe2=C,H,*CH,*N(S0,*C,,H7).C,H4~~e needles m. p. 137" is obtained ; its hydrochloride and sulphate are sparingly soluble in water.If finely divided dimethylaminobenzyltoluidine is shaken with a n equal quantity of powdered cymogen bromide dinretl~ykcniinobenxyl- cymo-p-lohidine NMe,*C,H,-CH,* N(CN)*C K,Me a colourlees crys- talline mass m. p. 1 3 5 O is obtained (hydrochlwids leaflets m. p. 165' ; platinichloride m. p. 1 16 -1 1 So) which is accompanied by some methgl- cyanoantino6enzylcyccno-p-to?uidine CN *NMe *CGH4*CH,*N( CN)*C,H,Me m. p. 140'; this is also obtainable by the action of cyanogen bromide on the monocyano-derivative. The monocyano-compound is easily hydrolysed by warm dilutja hydrochloric acid giving as-p-tolyl- d~meth~~urn~nobenxg~carbam~de NMe2*C,H,*CH,*N(CO*NH2)C~H4Me m. p. 178-1'79'; like the cyano-compound this is a mono-acid base ; platinichloride m.p. 169". p-Dimethylttminobenzyl-p-toluidine gives a p-nitrobenzoyl deriv- ative deep red needles m. p. 144O; hydrochloride m. p. 65-66" (decomp.) ; platinichloride m. p. 153-154O. The above acyl derivatives of p-dimethylaminobeuzyl-p-toluidine combine fairly readily with methyl iodide ; the monocyano-derivative gives a methiodide NMe,I*C,H,*CH,*N(CN)*C,H,Me m. p. 95-looo (decornp.) ; the corresponding methiodides of the nitrobenzoyl and of the benzoyl derivatives have m. p. 120-121" and 150-151O respect- ively. The methiodide colourless leaflets m. p. 16P of the acetyl derivative was converted by silver chloride into the correspondingi. 206 ABSTRACTS OF CHEMICAL PAPERS. methochloride m. p. 188'; platinichloride m. p. 210' (decomp.).On repeatedly evaporating its solution in concentrated hydrochloric acid to drynees this methochloride compound loses its acetyl group with production of p-dimethy~minobe~nxyl-p-toluidine methochloride hydro- chloride NNe3Cl*C,H,*CH2*NH( HCl)*C H Me a colourless powder m. p. 177O ; pZatinichZoride m. p. 1996 !Chis methochloride hydro- chloride as also the acyl derivatives of p-dimethylaminobenzyl-p- toluidine do not give a colour reaction when their aqueous solutions are warmed or treated with alkali and they do not undergo the condensation with aromatic amines thus confirming the authors' view as to the mechanism of this condensation. D. F. T. Aminohydrazines. 11. Benzylidene-p-aminophenylhydrazine ( Benzaldehyde-p-aminophenylhydrazone). HARTWIG FRANzEN'and B.VON FURST (Bey. 1913 46 3965-3974. Compare FrmzeD A 1 S07 i 321).-It has already been shown that benzaldehyde-o-nitro- phenylhydrazone can be reduced by sodium hy posulphi te in ammoniucal alcoholic solution to benzaldehyde-o-aminophenylhydrazone but this product could not be hydrolysed to the free hydrazine on account of its conversion under the influence of acid into 2-phenylbenziminazole. As the para-isomeride could not be capable of this condensation it has now been submitted to investigstion. When an aqueous solution of sodium hyposulphite is gradually introduced into a boiling alcoholic solution of benzaldehyde-p-nitro- phenylhydrazone to which one-fifth its bulk of a cold saturated alcoholic solution of ammonia has been added reduction is effected tbo benzalde~~yde-p-aminop~enylhydraxone yellow needles or leaflets m.p. 133-134' (decomp.) which rapidly darken. This substance when treated with benzaldehgde in boilkg alcoholic solution is converted into its benxylidene derivative CHPh:N*C,H,*NH*N:CHPh golden- yellow leaflets m. p. 163-164' which is also obtained occasionally as a product of the above reduction. I n a similar manner p-anisaldehyde and m-nitrobenzaldehyde condense with benzaldehyde-p-aminophenyl- hydrazone producing the p - anisylidene and m-nitrobenxylidene derivatives a yellowish-brown crystalline powder m. p. 14S0 and an intense yellow powder m. p. 15b0 respectively. With phenylthio- carbimide the benzaldehyde-p-aminophenylhydrazone yields a thio- carbanilide derivative NHPh*CS.NH*C,H,=NH*N :CHPh an almost colourless crystalline mass m.p. 157O. Unlike the ortho-isomeride (loc. cit.) benzaldehyde-p-aminophenpl- hydrazone when added in hot alcoholic solution to dilute hydrochloric acid at - 10' gives a violet-red .powdery hydrochloride m. p. 179-18lo which is unstable in a moist condition; on the addition of concentrated ammmia solution the hydrochloride regenerates the original benzaldehy de-p-aminophenyl hydrazone. The sdphate can be obtained in a similar manner and shows similar properties. The suggestion is made that them strongly coloured salts are possibly of an azo-structure for example NH,(HCI)*C,H;N:N*CH2Ph or a - - - quinonoid structure 3>4H4:N( HCI)*X CHPh. When benzaldehyde-p-aminophenylhydrazone is treated in boilingORGANIC CHEMlSTRY.i. 201 alcoholic solution with 2N-snlphuric acid a deep-coloured solution is obtained mbich after the addition of water and subsequent cooling gives a deposit of almost colourless leaflets of a sulphate ; the lack of colour is indicative of a normal structure and confirms the occurrence of a rearrangement in the formation of the coloured salts just described. Under the action of a hot alcoholic solution of hydrcigen chloride the behaviour of benzaldehyde-p-amioophenylhydrazone is quite different; instead of a hydrochloride or of the likely hydrolytic products benzaldehyde and p-aminophenylhydrazine there is obtained p-phenylenediamine together with benzaldehyde and ammonia. This result indicates a fission of the molecule a t the N-N linkiug instead of at the C:N linking as might be expected and throws light on the intramolccular condensation of the ortho-isomeride under the influence of acid I n the latter case the primary products are probably o-phenylenediamine benzaldehyde and chloroamine of which the two former condense t o a dihydrobenziminazole; this is then oxidised by the chloroamine t o benziminazole itself D.F. T. Crystallography of Some Aromatic Nitrogen Compounds. JUL~EN DRUGMAN (Zeitsch. Kryst. A4in. 191 3 53 266-270).-The following compounds prepared by F. D. Chattaway were measured. Bismethylphenylazimethylene (monoclinic a b c = 1.3335 1 1.2667 ; p = 103’44’). N-Dichlorobenzylidenediacetamide (T 1912 101 1207). N-Chlorophthalimide (rhombic-bisphenoidal a b c = 0*3000 1 0 9 7 5 ) . I!-Dichlorobenzenesulphonimide (monoclinic a b c = 0 72 I3 1 1 -2009 ; p = 100029’).1,. J. S. C o n s t i t u t i o n of Acetylacetonecarbamide 14 6-Dimethyl- dihydropyrimid-2-oneI. W~LLIAM J. HALE ( J . Amer. Chena. Xoc. 19 14 36 104-1 15).-Acetylacetonecarbamide exists in two modifica- tions one colourless the other yellow. It has been suggested by Haan (A. 1908 i 577) that the former is 4 6-dimethyl-2 3-dihydro- pyrimid-2-one CHGCMeeNH>CO and the latter 4 6-dimethyl- CMe=N 2 5-dihydropyrimid-2-one CH2<U;\le CMe:N>Co. 0 t h compounds crys- tallise with 2H,O and have m. p. 1 9 7 O . A potussiunz. salt can be prepared from a n aqueous solution of either of these compounds and is completely decomposed by the carbon dioxide of the atmosphere. On adding silver nitrate t o a solution of either modification a white silvei.salt is precipitated which reacts with methyl iodide with formation of a red crystalline methyl iodide additive compound of a trimethylpyrimidone. Mercuric chloride also yields a n insoluble salt with the dimethyldihydropyrimidone. When R solution of the yellow acetylacetonecarbamide in methyl alcohol is boiled with methyl iodide the additive cornpound C,H,ON,,MeI is produced in the form of colourless crystals which gradually become red when left in the air. If diazomethane vapour is passed into a cold solution of acetyl-i. 208 ABSTRACTS OF CHEMICAL PAPERS. acetonecarbamide in chloroform 3 4 6-trimsthyl-2 3-dihydropyrimid- a-one m. p. 63' is obtained which forms colourless crystals. The same compound can be prepared by the condensation of methylcarbamide with acetone ; its hydrochloride crystallises in colourless prisms.The conclusion is drawn that acetylacetonecarbamide does not contain a hydroxyl group and that the tautomeric forms must have the constitutions ascribed to them by Haan. E. G. Preparation of 5-Mono- and -di-allylbarbituric Acids. GESELLSCIEAFT FUR CHEMISCHE INDUSTRIE (D.R.-P. 268158). -The interaction of ally1 bromide and barbituric acid according to the quantities used leads to the production of 5-nllylbarbituric acid m. p. 1 6 2 O and 5-dic~llyZbarbituric acid. The latter forms colourless leaflets m. p. 169-170' and has a stronger hypnotic action than the corresponding diethyl derivative (veronal). J. C. C. Constitution of Benzoylhydrttzicarbonyl. R.STOLLE and K. 0. LEVERKUS (Ber. 1913 46 4076-4080. Compare A. 1913 i 898).-The authors have carried out the proposed condensations of nitrobenzoylhydrazicarbonyl with benzoyl chloride and benzoylhydrazi- carbonyl with nitrobenzoyl chloride and have obtained different products. This argues against the hydrazicdrbonyl structure for these derivatives. The formation of 2 5-diphenyl-1 3 4-oxadiazole by the action of heat on dibunzoylhydrazicarbonyl and the preparation of ethyl dibenzhydraziodoformate by the action of sodium ethoxide (A. 19 13 i 97) agree with the alternate oxadiazole formula. The tautomeric formulEe most nearly correspond with the behaviour of '' benzoylhydrazi- carbonyl " and its substitution products. On this basis the compound is re-named 2-keto-5-phenyl-2 3- dihydro-1 3 4-oxadiazole ; " dibenzoylhjdrazicarbonyl " is called 2-keto-3- benzoyl-5-phenyl-2 3-dihydro-1 3 4-oxadiazole ; and Sches- takov's methyl derivative of '' benzoylhydrazicarbonyl " (A.1913 i 97) which yields berizoylhydrazine on heating with water and therefore exists in the tautomeric form corresponding with (11) is re-named 2-methoxy-5-phenyl-1 3 4-oxadiazole. 2-Keto-3-p-nitrobenxoyl-5-phen$-2 3-dihpdro- 1 3 4-oxndiaxole CPh:? O<CO-N* coo C6H,.N0,' o<C*--NBz was obtained by heating '' benzoylhydrazicarbonyl " with p-nitro- benzoyl chloride in pyridine in leaflets m. p. 193'. 2-Keto-3-benxoyl- 5-p-nitrophenyl-2 ; 3-dihydi-o-l 3 4-oxadiccxoZe C( C6H;NO2):T formed glistening needles and leaflets m.p. 196O. A mixture of the isomerides me1 ted about 30" lower. Both compounds yielded benzogl- p-nit~obenxoylhydrazine C1,H,,O,N on warming with dilute sodiumORGANIC CHEMISTRY. i. 209 hydroxide and precipitating with acids. The compound was also prepared from p-nitrobenzoylhydrazine and benzoyl chloride. It is a white powder m. p. 236'. On heating alone they also gave the same phenyl-p-nitrophenyl-1 3 4-oxadiaxoZe C,,H,O,N the evolution of carbon dioxide being quantitative. This compound was also obtained by heating benzoyl-p-nitrobenzoylhydrazine with thionpl chloride. It forms glistening leaflets m. p. 209'. J. C. W. The Formation of Methenylbis[phenylmethylpyrazolone] from Phenylmethylpyrazoloneglyoxylic Acid. WILHELM WISLI- CENUS and OTTO BILFINGER (Ber.1913 46 3948-3949).-A correc- tion. The substance recently obtained by heating a n alcoholic solution of l-phenyl-3-methyl-5-pyrazolone-4-glyoxyl~c acid ( Wislicenus Elvert and Kurtz A. 1913 i 1387) and described as l-phenyl-3-methyl-4- methylene-5-pyrazolone is in reality methenylbis[phenylmethyl- pyrazolone] ~ p h * C o > ~ ~ ~ ~ ~ ~ < c N N=CMe (compare Betti and CO-rPh Mundici A. 1907 i 543). D. F. T. Preparation of 1 7-Dimethylguanine. FARBENFABRIKEN VORM. FRIEDR. BAYER & Co. (D.R.-P. 262470).-2 4 5-Triamino-6-pyrimi- done is converted into the 5-formyl derivative by treatment with dilute formic acid at a moderate temperature the latter methylated and the 1-methyl derivative boiled with concentrated formic acid whereby 1 -methylguanine XFii:&:gZ$A3H crystallising in needles is produced.m. p. 337-339' is obtained. On mkhylhion 1 7-dimethylguunine needles J. C. C. Carbamides of the Rosaniline Series. GUIDO MAYER (J. pr. Chena. 1913 [ii] 88 699-730).-An account of the preparation and properties of carbamide derivatives of the following three types from pararosaniline rosaniline and tri-p-aminophenyldi-m-tolylcar binol the dye-base of new-magenta I. NH2*CO*NH*C,H,*C(C6H,~~H2)2~OH. 11. NHPh*CO*NH*C,H,*C(C,H,*NH,),*OH. C,H,-NH*CO*NH*C H III. OH~C/C,H~*NH*CO.NH*C~H:~C*OH. \C,H,*NH*CO*NH~C,H,/ The compounds of the first type are obtained in the form of their hydrochlorides (1) by the addition of potassium cyanate to aqueous solutions of the hydrochlorides of the dyes and (3) by fusing the dye- bases with carbamide and heating the product with hydrochloric acid The following compounds were prepared in this manner 4' ; 4"-Diumino-4-carbamidotripheny~carbinol hydrochloride ~H,C~:C6~~,:C(C6H4'NH2)*C6~,~NH'CO~~~ from pararosaniline ; 4' 4"-diamino-4-carbc~midodiphenyl-m-tolylcarbinol hydrochloride from rosaniline and 4' 4"-diamino-4-carbamidophenyL di-m-tolylcurbinol hychochloride from the dye base of new-magenta.VOL. CVI. i. Pi. 210 ABSTRACTS OF CHEMICAL PAPERS. The corresponding bases which probably have a quinonoid structure are obtained by heating the hydrochlorides withdilute aqueous potassium hydroxide for several days at 90-95'. The compounds described above decompose when heated with the evolution of ammonia and aniline give red colorations on treatment with strong mineral acids and yield diazonium compounds which couple with resorcinol and R-salt to form yellow azo-dyes. They are reduced to the corresponding leuco-compounds with zinc and hydrochloric acid and give characteristic colorations with potassium chlorate and hydrochloric acid and other oxidising agents ; the picrates and nzcrcurichlwides are mentioned.When the sulphate of the carbamide derived from pararosaniline is treated with acetic acid and solid sodium nitrite and the resulting solution diluted with water an unstable flesh-coloured diazonium oompound fi H,*CO*N( ~ O ) * C ~ 4 * C ( O H ) ( C H ~ " C I is precipitated. The diazonium compound rapidly loses nitric oxide yielding a product from which 4"-amino-4'-hydroxy-4-carbamidotri- phenylcarbinol hydrochloride NH,Cl C,H, C( C,H,* OH) *C,H,*NH* CO*NH is obtained by boiling with water and subsequent treatment with hydrochloric acid.When boiled the filtrate from the above unstable diazonium compound yields aurin and 4' 4"-dihydroxy-4-carbamido- t riphen ylcarbinol 0 C6H4 C( C,H4*OH)*C6H,*NH* CO-NH a black substance which gives intensely violet solutions in dilute aqueous potassium hydroxide. The phenylcarbamides of type I1 are obtained in the form of their hydrochlorides by fusing the dye bases with phenylcarbamide arid dissolving the product in hydrochloric acid. The hydrochlorides of 4' ; 4"-d~am~~zo-4-phe~ylcarba~nidot~~p~nylcarbino~ NH2C1:C,H,:C( C,H4jNH,)~U6H,~NH.C0.Nf-3;Ph from pararosaniline of 4' 4 -diamino-4'-phenylcarbamidod+henyl-m- tolylcarbinol from rosaniline and 4' 4"-diamino-4-phe.izylcaybamido- phenyldi-m-tohjlcarbinol from the dye base of new-magenta were pre- pared in this manner; the corresponding imino-bases are formed by heating the hydrochlorides with aqueons potassium hydroxides.The above phenylcarbamides form picrates and mercurichlorides can be diazotised and give characteristic colorations on treatment with strong mineral acids and oxidis- C,H,:NHCl*CO*NH*C,H ing agents. When heated they C//C,H4*NH-co*NH*C,~~~c decompose without melting. The tricarbamides of type 111 are obtained in the form of their hydrochlorides (annexed formula) by passing carbonyl chloride into a hydrochloric acid solution of the dyes ; the hydrochlorides are sparingly soluble in water and decompose when heated with evolution of aniline and ammonia.The free bases are also described. F. B. \C6H,*~Ho~O*NHCI:C6H4~ Some Thiocarbamides of the Rosaniline Series. SIEGFRIED HILLER (J. pr. Chem. 1913 [ii] 88 731-743. Compare preceding abstract).-When heated with carbon disnlphide in alcoholic solutionORGANIC CHEMISTRY. i. 211 in the presence of a little sulphur pararosaniline in converted into a trathiocarbccmide of the annexed con- /C6H4*NH*CS*NH*CGH4\ stitution ; similar compounds bGH4:N--GS-N :C6H C-C,H,*N H*CS*NH*C,H4zC C43H32N6S3 and C45H34N6S3 have also been pre- pared from rosaniline and 4 4' 4"- triaminophenyldi-rn-tolylcarbinol. The three compounds are brownish-red to blackish-red substance$ having a metallic lustre yield in acetic acid solution metallic precipi- tates with picric acid copper sulphate and mercuric chloride and are decomposed by hot acids with liberation of hydrogen sulphide.On treatment with sodium hydroxide their solutions in acetic acid yield the corresponding red carbinol bases. When heated with benzoic acid and aniline the trithiocarbamides from pararosaniline and rosaniline yield blue phenyl derivatives which yield reddish-brown solutions in strong acids. Attempts to prepare the trithiocarbamide derived from para- rosaniline by the addition of ammonium thiocyanate to a hot alcoholic solution of pararosaniline hydrochloride gave tri-p-aminotriphenyl- carbinol thiocyanate C2,H,8N4S as ft lustrous green metallic precipitate. 4 ; 4'-Diarnino-~''-phanylthiocnrbarnidotri~?~enylca~binol NH C6H4 C( C6H ;NH2) C H4*NH CS* NHPh 4 4'-diarni~ao-~''-phenyZbhiocn~bamidodiphenyl-m-tolylcarbinol NH:C,H,:C( C6H40NH2)*C H,Me*NH*CS*NHPh and 4 4'-diamino-4"-p?~enylthiocar bamidophen yldi-m -tol y Zcarbinol are obtained by heating phenylthiocarbimide with alcoholic solutions of pararosaniline rosaniline and tri-p-aminophenyldi-m-tolylcarbinol respectively.They are amorphous violet substances which give orange- red solutions in sulphuric acid and when heated with dilute acids or water are partly resolved into their components and partly converted into the corresponding cltrbamides with evolution of hydrogen sulphide. When heated with benzoic acid and aniline they yield bluephenyl derivatives. The hydrochlorides (BHCI) prepared by the direct union of phenylthiocarbimide and the hydrochlorides of the original dye bases form red aqueous or alcoholic solutions from which the corresponding carbinol bases are precipitated on the addition of sodium hydroxide.NH:C,H4:C(C,H,Me*NH2)*C6H~Me*NH*CS*NHPh F. B. The Kinetics of the Sandmeyer Reaction. P. WAENTIG and JOHANNA THOMAS (Ber. 1913 46 3923-3937. Compare Heller A. 1910 i 240; Heller and Tisehner A. 1911 i 243).-The cuprous hdoid employed in the Sandmeyer reaction is f airlg generally believed t o form an intermediate additive compound with the diazo-compound although various views are held as to the details of the mechanism of the change. Cryoscopic experiments with a solution of cuprous chloride in dilute hydrochloric acid give probability to the view that it is the double molecule Cu2C12 and not CuCl which is involved in the change.P 2i. 212 ABSTRACTS OF CHEMICAL PAPERS. By adding a concentrated solution of benzenediazonium chloride at - 17" gradually to a saturated solution of cuprous chloride in 25% hydrochloric acid at -60° the authors have obtained a n unstable bulky red precipitate of composition C,H,N2CI,Cu2C1 analogous to the cuprous bromide compound isolated by 'Hantzsch (A 1895 i 516). Measurements of the velocity of the decomposition of benzenedi- azonium and p-toluenediazonium chloride solutions at 0' in the presence of an approximately equimolecular quantity of cuprous chloride dissolved in hydrochloric acid by means of the volume of liberated nitrogen show that the p-tolueneaiazonium salt decomposes much more slowly than the benzencdiazonium salt.The constants calculated for a unimoleculsr change in any one experiment were satisfactory but the higher the initial concentration of diazonium Ralt and cuprous chloride the greater the constants obtained. This effect appears to be due to the cuprous chloride and the deduction may be drawn that of the two successive reactions that expressed by N,PhCl + Cu,Cl = N,P~CI,CU,CI is of measurable velocity and seems to be unimolecular on account of the constancy of the concentration of the cuprous chloride during an experiment whilst the change N,PhCl,Cu,Cl = PhCl + N + Cu,CI2 is very rapid. The simplicity of this view is somewhat discountenanced by the impossibility of obtaining satisfactory constants when the diazonium Ralt and cuprous chloride are present i n other than equimolecular proportions and by the authors' discovery that with equimolecular proportions of diazonium and cuprous salts the velocity of the reaction is very considerably depressed by increased concentration of the hydrochloric acid present.A rise OF loo in the temperature of the reaction increases the velocity to twice its original value This result indicates that in the preparation of a phenol from a diazonium salt (for which the tempera- ture-coefficient is above 3*5) a high temperature will be advisable whilst a lower temperature should prove more satisfactory for the formation of an aryl haloid. It is suggested that the decomposition of the diazonium haloids may to some extent follow the course Ar C1 Ar C1 Ar C1 NiN+ I -+ N-N -+ N I I I I c1 Ii HC1 I 61 H analogous to the well-known explanqtion of phenol formation for the investigation of which it will be necessary to investigate the velocity of decomposition with various concentrations of acid in the absence of the disturbing effect of cuprous chloride D.F. T. The Nature of the Free Amino-groups in Proteins. DONALD. VAN SLTKE and FREDERICK J. BIRCITARD (J. Biol. Chsm. 1914 16 639-547).-1n various native proteins (haemoglobin cweinogen gelatin edestin gliadin etc.) the amount of free amino- nitrogen is equal to one-half the lysine nitrogen ; in glisdin there is a difference of 0.7%. The period required for complete reaction of theORGANIC CHEMISTRY. i. 213 proteins with nitrous acids (thirty minutes) is longer than that required by a-amino-acid s (three to four minutes) but corresponds with that found for lysine with an a-amino-group free. The conclusion is drawn that the same group is free in the protein molecule; and this group practically represents the entire amount of free NH determin- able in native proteins by this method All the others are condensed into peptide linkings.With the primary proteoses the relations are different the free NH exceeding half the lysine nitrogen by 3 to 5%. W. D. H. The Colloidal State of the Mixed Protein and Gold Sols. FRIEDRICH JACOBS (Biochem. Zeitsch. 1913 58 343-351).-The effect of the addition of gold sols to protein solutions was investigated by measurement of the changes in the viscosity and osmotic pressure of the solutions. Under the conditions of the experiments this addition caused no appreciable change in the viscosity.It produced however a lowering of the osmotic pressure which mas 7.4% in the case of albumin and 3.9% in that of ha?moglobin. I n the case of globulin no change could be detected owing to the small absolute pressure. S. B. S. The “Gold Numbers” of the Proteins of the Blood. W. HEUBNER and FR. JACOBS (Biochein. Zeitsch. 1913 58 352-361). -The “ gold numbers ” of various fractions of albumin globulin and bzemoglobin were determined. The variations in different prepara- tions of the same substance were however somewhat large. The “gold numbers ” cannot be used therefore for the identification of auy given blood-protein. Some Protein Derivatives. KARL LANDSTEINER (Biochem.Zeitsch. 19 13 58 362-364).-The products obtained from proteins of horse-serum precipitated therefrom by alcohols by treatment with acetic anhydride with alcohol in the presence of hydrochloric and sulphuric acids and with diazomethane are described. The first and third of these products when tested by the complement-deviation method were found to have lost their species specificity. S. B. S. S. B. S. The Clotting of Caseinogen Solutions. SAMUEL BARNETT SCHRYVER (Biochem. J. 1913 7 568-575).- Casein prepared by the use of pepsin (or rennin) differs from metacaseinogen a product obtained by the action of water a t 3’7” on caseinogen in that it cannot be converted by solution in alkali hydroxide and reprecipitation into a more soluble product which dissolves in calcium hydroxide t o yield clottable solutions.Psrncreatin clots caseinogen only in the presence of soluble calcium salts; the casein produced has only half the aolubility in half-saturated lime water of the casein prepared by pepsin. Caseins are regarded as combinations of the protein and enzyme. The action of the enzymes is not an ordinary proteoclnetic one; the same view has also been expressed by van Slyke and Bosworth. “ Natural caseinogen ” is probably not a calcium salt. W. D. H,i. 214 ABSTRACTS OF CHEMICAL PAPERS. The Hexone Bases of Caseinogen. DONALD. VAN SLYKE (J. Biol. Cham. 1914 16 531-538).-The following figures were obtained by Kossel’s and by the author’s methods Kossel’s Nitrogen distribu- method. tion method. Histidine ..................... 4 -1-4 *5 6-2 Arginine ........................75-79 7‘4 Lysine. ......................... 8.7-9 -3 10.3 The loss of histidine in Kossel’s method is ascribed to adsorption by barium phosphotungstate. The results with arginine are practically the same in both methods. The Kossel-Patton method as modified by Osborne Leavenworth and Brautlecht gives more consistent results for histidine. W D. H. Ovo-ruucoid and Metallic Hydroxides. J. NEUMANN (Zeitsch. physiol. Chem. 191 4 89 149-150).-0vo-mucoid can be completely precipitated by zinc hydroxide in the presence of potassium hydroxide or sodium carbonate. This is attributed to the formation of a hydrate; other metallic hydroxides act in the same way and there is in the quantity used an inverse proportion between it and the moleculnr weight of tho metallic hydroxide.Primary proteins and proteoses act in the same way but not peptone. Enzyme Action Facts and Theory. HENDRIK PIETER BARENDRECHT (Riochem. J. 191 3 7 549-561).-A criticism of some of the recent researches on the kinetics of enzyme action. The radiation theory of enzyme action (Barendrecht A. 1904 ii 551) is extended. Enzyme action spreads like radiation from a centre. The radiation may be absorbed by the substrate by the products of action or by any other foreign substance. The enzyme particle extends its catalytic action in a sphere as regards both hydrolysis and synthesis A new isomeride of dextrose enzyme made is postulated the transformation of which to stable dextrose is a balanced action. W. D. H. E. F. A. The Auto-catalysis of Trypsinogen.H. M. VERNON (J. Physiol. 1913 47 325-338).-When once trypsinogen is activated that is converted into trypsin by enterokinase the view is taken that further activation is mainly due to trypsin itself (auto-catalysis) or rather to a variety of trypsin called ‘‘ unstable try psin.” Behaviour of Diastase in the Presence of a Specjflc Precipitate. AGNES ELLEN PORTER (Biochem. J. 1914 7 599-603). -The action of diastase is accelerated by the presence of serum. In the presence oE antigen (dilute egg-white solutions) the enzyme’ is more active with normal rabbit serum than when rabbit serum immunised against egg-white is used hence the immune serum exercises the greater absorption. A similar difference is observed when horse- serum is used as the antigen. Taka-diastase displayed no tendency to become absorbed by a specific precipitate. Serum from which the globulins had been precipitated by saturation with carbon dioxide and removed after centrifuging retained most of the original enzyme activity.E. F. A. W. D. H.ORGANIC CHEMISTRY. i. 215 Hydrolysis of Glycogen by Diastatic Enzymes. 11. Influence of Salts on the Rate of Hydrolysis. ROLAND VICTOR NORRIS (Biochern. J. 1913 7 622-629. Compare A. 1913 i 308)-A dialysed extract of pig's pancreas has practically no hydrolysing action when added to a dialysed glycogen solution. The addition of small quantities of neutral salts restores the activity of the enzyme. The most effective salts are those of the halogen acids the activity diminishing in the order chloride bromide iodide.PotasRium sodium and also calcium barium and magnesium chlorides produce quantitatively the same acceleration. Nitrates have a slight acceler- ating action but sulphates are almost without action in restoring activity to the dialysed enzyme. They have no depressing action neither do they hinder the acceleration produced by sodium chloride. The concentration of salt required to produce a maximum degree of hydrolysis rises with increasing enzyme concentration but appears to be independent of the glycogen concentration. The anion is much more concerned in the reaction than the cation and the action of the salts is chiefly confined to the enzyme. The Preparation of Protein-free Emulsin. KOHSHI OHTA (Biochem. Zeitsch. 1913 58 329-338).-1f the commercial prepara- tion of emulsin is digested with one-tenth part of its weight of pancreatin in 100 times its weight of water rendered slightly alkaline with ammonia a considerable amount of protein passes into solution.If the digest is dialysed concentrated and precipitated with alcohol a product is produced which is somewhat more active than the original emulsin and which is quite free from proteins. The substance thus obtained contains calcium magnesium and phosphorus and yields a reducing substance on hydrolysis with acids. Whilst the protein-free emulsin hydrolyses amygdalin and salicin it is free from other ferments which are present in the crude product. S. B. S. The Optimum Temperature of Salicin Hydrolysis by Enzyme Action is Independent of the Concentration of Substrate and Enzyme.ARTHUR COMPTON (Proc. Roy. Soc. 1913 B 87 245-254).-The temperature of the greatest activity of sweet- almond emulsin acting on salicin is determined under different conditions according as the concentration of the substrate and that of the enzyme are varied separately or together. The optimum is shown t o be constant at about 34'. The Reduction Ferments. V. The Go-ferment of Per- hydridase. Formation of Aldehydes from Amino-acids. A. BACH (Biochem. Zeitsch. 1913 5 8 205-212).-1t has been shown t h a t erepton (a commercial digestion product of protein) acts as a co-ferment to perhydridase. It is now found that this preparation yields aldehyde when its solution is submitted t o distillation. The aldehyde appears entirely in the first third of the distillate.If the regdue is kept for twenty-four hours and again distilled a quantity cjf aldehyde equal to that given in the first distillation is obtained. If the distillation is now continued in a current of air aldehyde is again produced each fraction of the distillate containing approximately E. F. A. E. F. A.i. 216 ABSTRACTS OF CHEMICAL PAPERS. the same amount of aldehyde. Strecker’s reaction will explain the formation of aldehyde from erepton which can be imitated very nearly when aniline is distilled with p-benzoquinone. The oxidation of the amino-acid is effected by the water and the hydrogen thus set free reduces the “acceptor.” I n a curreut of air the latter is again oxidised. The action of erepton as a co-ferment to perhydridase can therefore be explained by the fact that under certain conditions it can give rise to a simple aldehyde which is the true co-ferment. Perhydri- dase is a true aldehydase. S. B. S. The Conditions of Action of Rennin. L. MICHAELIS and A. MENDELSSOHN (Biochem. Zeitsch. 19 13 58 315-328).-The optimal precipitation point of caseinogen by acid from either pure solution or milk lies at [H’]=2*5 x 10-5. I n presence of lime it is shifted in the acid direction and is then less distinct a t about [H’] = 3 x 10-4. The optimal coagulation point by rennin in the presence of lime lies between [H’] = 4 x 10-7 and 1 x but cannot be determined with greater accuracy. and [H’) = 1 x 10-7 are certainly outside the region of optimal activity. Between the optimal point of acid precipitation and of rennin action there exists a zone which lies outside the optimal conditions for both these actions This indicates that the rennin action and the acid precipitation are distinct actions conclusion which is confirmed by the fact that the caIcium ion action in rennin clotting cannot be replaced by increased hydrogen-ion concentration. Preparation of Organic Arsenic Compounds. HEINRICH BART (D. R. -P. 2 6 7082).-3 - ~arbethoxyam~no-4-hyclroxyp~eny~ar8in6 (annexed formula) prepared by the electro- lytic reduction of the corresponding arsinic acid is a colourless crystalline powder m. p. \-/ 155-160O ; on treatment with sulphurous ASH acid and subsequent hydrolysis it furnishes a base which is useful for the preparation of therapeutically active compounds HEINRICH BART (D.R.-P. 268 1 72).-3-Carbethoxyamino-4-hydroxyphenylarsinic acid is prepared by adding a solution of diazotised 5-amino-2-hydroxyphenyl- urethane (m. p. 1 3 0 O ) to a solution of sodium arsenite and sodium hydroxide to which copper paste has been added. Sodium p-hydroxy- phenylarsinate is similarly obtained from p-aminoyhenol. Preparation of Aromatic Stibino-compounds Aromatic Stibine Oxides and their Derivatives. CHEMISCHE FABRIK VON HEYDEN AKT.-GES. (D.R.-P. 268451).-Aromatic stibino-compounds and stibine oxides are prepared by reducing the corresponding mono- arylstibinic acids. Stibinobenzene C,H,*Sb:Sb*C,H is obtained by reducing pheaylstibinic acid Kith sodium hyposulphite in sodium hydroxide solution ; m-aminophenylstibine oxide NH2*C6H,*Sb0 by reducing m-nitrophenylstibinic acid with stannous chloride (the intermediate compound m-aminophenylstibine chloride hydvochlorids The points [H’] = 3 x S. B. S. OH C02Et.NH/-\ J. C. C. Preparation of Organic Arsenio Compounds. J. C. C.PHYSIOLOGICAL CHEMISTRY. i. 217 HCI,NH,*C,H,*SbCI has m. p. 215") ; di-m-aminostibinobenzene from the intermediate compound just mentioned and sodium hyposulphite or hypophosphite and mm'-diamino - pp' - dihydroxystibinobenxene from m-nitro-p-hydroxyphenylstibinic acid and sodium hyposulphite. J. C. C. Preparation of Secondary Aromatic Stibine Oxides and their Derivatives. CHEM ISCHE FABRIK VON HEYDEN AKT.-GES. (D.R.-P. 269206).-Secondary aromatic stibine oxides are prepared by reducing the chlorides of the corresponding stibinic acids with suIphur dioxide in methyl-alcoholic solution. m-Aminodiphenylstibine oxide NH2*C,H,*SbPh*O*SbPh*C6H4*NH2 is a faintly-coloured powder which sinters a t 70". J. C. C. Preparation of Aromatic Nitrohydroxystibinic Acids. CHEMISCHE FABRIK VON F. HEYDEN AKT.-~ES. (D.R.-P. 262236. Compare A. 1913 i 1122).-Nitrohalogenarylstibinic acids are treated with a1 kali hydroxides and furnish nitrohydroxystibinic acids. p Chlorophenylstibinic acid prepared from diazotised p-chloroaniline and antimony oxide yields on nitration 4-chloro-3-nitrophenyZstibinic acid the aqueous solution of which with hydrochloric acid gives the chloride N02*C,H,Cl*SbOC1,. OR warming the acid with potassium hydroxide solution potassium 3-nilro-4-hydroayphenyZ-l-atibinate is produced as a scarlet crystalline powder from which the yellow 3-nitro- 4-hydroxyphenyl-l-stibinic acid is obtained on acidification (compare loc. cit.). J. C . C. CHEMISCHE FABRIK VON HEYDEN AKT.-GES. (D.R.-P. 269205. Compare A. 1913 i 41 6).-Secondary and tertiary stibinic acids are prepared by treating an aromatic diazonium salt with a mono- or di-aryl substituted antimony oxide in the presence of alkali hydroxide. The interaction of benzenediazonium chloride and rrt-aminophenylstibine oxide fur - nishes m-aminodiphenylstibiizic mid a faintly-coloured powder. Chloro-m-pheny2enestibinic acid a brown powder is obtained by de- composing diazotised p-chloro-m-aminophenylstibine chloride with cold sodium hydroxide and by a similar method phenyl-m-phenylenestibinic acid C,H,:SbPh(OH)2 a &own powder is prepared from nz-amino- diphenylstibine oxide. J. C. C. Preparation of Aromatic Stibinic Acids.