218 SUDBOROUGEI : RESEARCHES IN TRE STIT,BENE SERIES. By JOHN J. SUDBOROUGII, Ph.D., D.Sc. BY the action of phosphorus pentachloride on deoxybenzoin, C,H5* CH,. CO*C,H,, Zinin obtained an oily chlorostilbene, which he describes as being unstable and as yielding tolane, C,H,-C i C*C,H,, on distillation (Annnlen, 1869, 149, 375). Some few years ago, I showed (Bey., 1892, 25, 2237) that, when deoxybenzoyn is heated with phos- phorus pentachloride, and the product is distilled, a solid chloro- stilbene, melting a t 54O, is obtained differing materially from the oily compound described by Zinin. It was, however, found possible to obtain an oily compound by the action of the pentachloride in the cold, and decomposing the phosphorus oxychloride by pouring the product on t o ice. This oily compound differs somewhat from that described by Zinin, in that it does not yield tolane on dis- tillation, but is converted, more or less completely, into solid chloro- stilbene.Whether a liquid or solid chlorostilbene is formed by the action of phosphorus pentachloride on deoxybenzok, appears to depend largely on the temperature a t which the action takes place. Thus, when working a t as low zt temperature as possible, and then decomposing the phosphorus oxychloride with ice, an oily product is obtained which does not solidify when cooled to - 15". If, however, the pentachloride and deoxybenzoh are heated together for half-an- hour on the water bath, and the product is then poured into water, extracted with ether, and the ethereal solution well shaken with sodium hydroxide solution, the oil left on evaporating the ether becomes partially solid when cooled in x freezing mixture.The solicl, rapidly separated and recrystallised from alcohol, melts at 54'. Either of the two above-mentioned oils, when distilled under atmospheric, or even under diminished, pressure, is converted into solid chlorostilbene (m. p. 54"). Met(hy1- and ethyl-deoxybenzoin, when heated with phosphorus pentachloride, can also be made to yield oily stilbene derivatives, which, on distillation, are partially converted into solid methyl- and c th yl-chloros t ilbenes. Whether the oily substances are merely slightly impure forms ofXUDBOROUGH : RESEARCHES IN THE STILBENE SERIES. 210 the solid compounds, or whether they are stereo-isomeric forms, Iias, up to the present, not been definitely decided.This question I hope to deal with in a future communication, when the oily compounds have been subjected to a more complete investigation. The present paper contains an account of the solid compounds, chlorostilbene, methylchlorostilbene, and ethylchlorostilbene, and of certain deriva- tives which have been obtained from them. P~*epcwc&oia of Deoxgbenxo!in. For the preparation of these stilbene derivatives, it has been neces- sary to obtain large quantities of deoxybenzob. When the research was first started, some four yearsago, the best method for obtaining deoxybenzob was that recommended by v. Meyer and Oelkers (Zei*., 1888, 21, 1295) ; both this method, and also those described by Wachter (Be?.., 1892,25, 1721) and by Knoevenagel and Weisgerber (Zel.., 1893, 26, 441), have been tried.The simplest process, however, seems to be to treat benzoi'n (1 part) with zinc dust (I part) and glacial acetic acid (6 parts) ; in order to prevent caking, the zinc dust is added, while the mixture of benzoln and acetic acid is kept well stirred ; the whole is then heated on the water bath for about 15-20 hours, or until a portion, when poured into water, gives no flocculent matter, only a yellowish oil which slowly solidifies. The mixture is then poured into cold water and allowed to stand until the oil has completely solidifiecl, the solidified oil being recrystallised several times from alcohol, in which i t is readily soluble. The deoxybenzoyn is thus obtained in the form of colourless plates which melt a t 55'.The yield generally amounts to some 50 per cent. of the theoretical. Extremely interesting is the fact that benzil, C,H,* CO*CO*C,H,, is one of the products of the action of zinc dust on an acet.ic acid solu- tion of benzoin. When the mixture, in the above proportions, is heated for 15-20 hours, practically no benzil is obtained, but it can easily be isolated from the product, if the heating is continued for a shorter period. The mixture, heated at looo for 8 hours, was poured into cold water, and the flocculent precipitate, after being washed and dried, was triturated in a mortar with cold alcohol. The residue, which is almost insoluble in cold alcohol, on recrystallisation from hot alcohol, formed prismatic needles melting at 136O, and, on analysis, proved to be unaltered benzoin.Found, C=79*22 and 78.40 per cent.; H = 5.65 and 5.77 per cent. C,H,* CH(OH)*CO*C,H, requires C = 79.24 and H=5.65 per cent. The cold aIcoholic solution, after standing for several days, yielded a mixture of two different kinds of crystals; as i t was found im- possible to separate these by any chemical means, the solution was allowed to evaporate spontaneously, and the crystals, which were well220 SUDROROUGH : RESEARCHES IN THE STILRENE SERIES. developed, were separated mechanically. One of the compounds ~vhich crystallised in fairly large, flat, colourless, I ustrous plates me1 ting at 55", proved to be deoxybenzoin. The other compound, which was present in smaller quantity, crystallised in long, sulphur-yellow, pris- matic needles, and melted a t 94".It exhibited all the properties of benzil, and its identity was established by analysis. Found C = 79.92 and 80.04 per cent. H=4*73 and 4 5 0 per cent. C,H,*CO*CO*C,H, requires C = 80.00 and H = 4-76 per cent. The formation of benzil, an oxidation product of benzoi'n, by the action of zinc dust and acetic acid on benzoyn, seemed so remarkable, that I hare studied the action of hot acetic acid on benzoi'n, and find that small quantities of benzil are formed under these circumstances if the heating is continued for some 8-9 hours. Pwpccwction of P-Chloi*ostilbene, CGH,* CH: CC1* C,H,. I n order to distinguish the solid chlorostilbene previously mentioned from the oily compound described by Zinin, I propose to call the solid compound P-chlorostilbene, and Zinin's oil a-chlorostilbene ; it may be, however, that further investigation will establish the identity of the two compounds.P-Chlorostilbene is best obtained by treating deoxybenzoh in the cold with a slight excess of phosphorus pentachloride; as the action is somewhat violent, it is advisable to add the pentachloride in small quantities at a time. The mixture is heated for a short time on the water bath, poured into water, and the oil extracted with ether; the ethereal solution is then well shaken with sodium hydroxide, washed with water, dried over calcium chloride, and the ether distilled off. The residue may then be distilled under atmospheric pressure, or, still better, under reduced pressure; in the latter case, the greater part passes over at 189-190" under a pressure of 26 mm., and, as a rule, a considerable amount of hydrogen chloride is evolved during the dis- tillation.When cold, the greater part of the distillate becomes solid ; it is drained on a porous plate, and recrystallised from alcohol. The crystals usually have a yellowish tinge, which can only bc removed by repeated recrystallisation. P-Chlorostilbene cry stallises in large, colour- less, lustrous plates, melts as 53-44', and in general appearance re- sembles stilbene. Under atmospheric pressure, it distils a t 320-324', but a t the same time undergoes slight decomposition. I n the preparation of /3-chlorostilbene, it is not advisable to take the crude product obtained by the action of phosphorus pentachloride and subject it straight away to distillation, either under reduced or atmo- spheric pressure, as in both cases charring is apt to take place, and there is considerable loss.P-Chlorostilbene i s readily soluble in ether,SUDBOROUGH : RESEARCHES IN THE STILBENE SERIES. 221 chloroform, benzene, glacial acetic acid, and hot alcohol ; more sparingly in cold alcohol, and insoluble in water. 0,1640 gave 0.107 AgC1. C1 = 16.14. 0,2085 ,, 0.135 AgCI. C1 = 16.37. 0*1500 ,, 0.069 gram H,O and O*43gramC02. C! = '78.18; H = 5.11. C,,H,,Cl requires c1 = 78.32 ; H = 5.12 ; C1 = 16.55 per cent. Molecular weight determination. I. 0,328 gram dissolved in 1S.61'7 grams of glacial acetic acid lowered 11. 0.2415 gram dissolved in 16,064 grams of glacial acetic acid Theory for C',,H,,Cl, If the solid chlorostilbene is heated for 3-4 hours on the water bath with alcoholic potash, it is converted, by the loss of 1 molecule of hydrogen chloride, into tolane (diphenylacetylene) which melts at 60"; the tolane was further identified by conversion into its dibromide which melts a t 204-205".Sodium ethoxide is also capable of bringing about the same decomposition. When reduced with sodium amalgam and alcohol, P-chlorostilbene is readily converted into stilbene (m. p. 125"). Two grams of chloro- stilbene were dissolved in 30 grams of ethylic alcohol and 40 grams of 2.5 per cent. sodium amalgam were added. The mixture was left overnight, and then neutralised wibh hydrochloric acid, the alcohol evaporated, and the residue extracted with ether.The product left on evaporating the ether crystallised from alcohol in large, lustrous plates, melting a t 124-124.5O ; it gave a dibromide which melted a t 237", thus establishing its identity with stilbene. Cli,Zo+*ostilhae tZiclJok?e (trichlorodibenzyl), C,H,* C'HC11* CCl,. Cf:H5, was obtained by saturating a solution of chlorostilbene in carbon tetrachloride with dry chlorine a t 0" and in the dark. The mixture was allowed to stand for about half an hour, the excess of chlorine re- moved by drawing a rapid current of air through the solution, and the carbon tetrachloride allowed to evaporate spontaneously. The solid residue, washed with a little warm alcohol and crystallised from hot glacial acetic acid, deposited the dichloride in colourless, well-developed, hard prisms melting at 102-103°.It is almost insoluble in alcohol and ether, but readily dissolves in benzene, chloro- form, carbon tetrachloride and hot acetic acid. the freezing point 0.326". lowered the freezing point 0,274". M = 2109. 1% = 214. 31 = 214.5. 0.3668 gave 0-5427 AgC'I. C1 = 36.62 per cent. Theory requires C1 = 37.30 per cent. When the dichloride is heated for half-an-hour with a slight excess of alcoholic potash, it yields two compounds which are best separated222 SUDBOROUCH : RESEARCHES I N THE STILBENE SERIES. by fractional crystallisation from alcohol. The portion which is only sparingly soluble in alcohol crystallises in flat, colourless plates, and melts a t 141'. 0.2051 gave 0.236 RgC1. C1 = 28.46. C,,H,,Cl, requires C1 = 28.51 per cent.It is identical with Liebermann and Homeger's cis-tolane dichloride, which melts at 143' (Bey., lS79, 12, 1971). The other compound is much more readily soluble in alcohol, crystallises in prismatic needles, and melts at 63'. 0.2093 gave 0.2365 AgCl. U,,H,,Cl, requires C1 = 28.51 per cent. It is identical with tmm-tolane dichloride (m. p. 63') (Zoc. cit.). ChZoil*ostiZbene dibyoilnide? C,H,* CHBr C1U1Br*C,H6, or chlorodibromo- dibenzyl, is readily obtained by the action of a chloroform soln- tion of bromine on P-chlorostilbene dissolved in chloroform : 100 grams of a 10 per cent, solution of bromine in chloroform was gradually added to 5Igrams of chlorostilbene dissolved in 40 grams of chloroform. The solution, kept cool by surrounding it with ice- cold water, was allowed to stand for two or three hours, and the chloroform and excess of bromine were then quickly removed by pouring the solution on to a large clock glass, and blowing air over the surface.The solid residue mas washed with a little warm alcohol, and recrystallised from benzene or hot glacial acetic acid. From its solu- tion in acetic acid, it is deposited i n the form of colourless, well-deve- loped prisms ; it melts a t 127", but a t the same time undergoes decom- position. It is readily soluble in chloroform, benzene, and hot ether, but is practically insoluble in alcohol. At 1l0, 100 parts by weight of benzene dissolve 29.18 parts by weight of the dibromide. 0.2056 gram gave 0.286 gram of silver chloride and bromide, which, when reduced, gave 0.1785 gram of metallic silver.Br = 42.69 ; (21 = 9.59 per cent. If boiled for half-an-hour with a slight excess of alcoholic potash, the dibromide yields a mixture of three compounds, which are best separated by fractional crystallisation from alcohol. 1.-The compound least soluble in alcohol crystallised in hard, colourless, flat prisms, and melted at 173-174'. Analysis proved it t o be a chlorobromostilbene obtained by the elimination of a molecule of hydrogen bromide from the chlorostilbene dibromide. 0.183 gram gave 0.205 gram of silver chloride and bromide, which, when reduced, gave 0.1335 gram of metallic silver. Br and C1= 39.07 per cent. 2.-An intermediate product was obtained which melted at 71-78',; it could not, however, be obtained in a pure state.3.-The compound which was most readily soluble in alcohol crystallised in colourless needles or prisms; it melted at 62", and analysis proved it t o be a dibromostilbene. Cl = 27.95. Cl4H,,C1Br2 requires Br = 42.72 ; C1= 9.47 per cent. C,,HloBrC1 requires Br + C1= 39.35 per cent.SUDBOROUGH : RESEARCHES IN TRE STILBENE smm. 2213 0.3 gave 02215 AgBr. The compound must, therefore, be identical with Limpricht and (Be?*., 1871, 4, Br= 47.12. C,,H,,Br, requires Br = 47.33 per cent. Schwanert's dibromostilbene which melts a t 63'. 379). Nit o-de r ivcc t ives. As it was found impossible to obtain any definite nitro-derivative by the aid of nitric acid-concentrated or fuming, alone or in acetic acid solution-the action of the red fumes evolved from nitric acid and arsenious anhydride on chlorostilbene was studied.Solid chlorostil- bene was dissolved in glacial acetic acid, and the solution saturated with the nitrous fumes, while surrounded by ice-cold water. The solution became deep green, and during the course of the action a colourless, crystalline substance was deposited. When saturated, the solution mas kept f o r several hours, and the insoluble product was then collected and recrystallised from hot glacial acetic acid. It forms colourless prisms, melts at 124-125", and is apparently chlorodinitrodibenzyl? C,H,* CH(NO,)*CCl(NO,)*C,H,, formed by the addition of two nitro-groups to the unsaturated compound chloro- s ti1 bene. 0,1221 gave 9.4 C.C. moist nitrogen a t 11" and 764 mm. 0.2542 ,, 0.115 AgC1.C1=11*20 per cent. C1,Hl,CIN,O, requires N = 9.1 3 and GI= 11 *liS per cent. /3-Chlorostilbene thus resembles stilbenc itself, which, according to Gabriel (Bey., 1885,18, 2438), yields an additive compound, C,,H,,N,O,, melting a t about 300". The acetic acid filtrate from this crystalline product was poured into cold water, when an oil was deposited which became semi-solid after a time ; this, after being pressed and recrystallised from hot alcohol, gave fine, sulphur-yellow, prismatic needles, but on spontaneous evapora- tion of its alcohol solution, it is deposited in well-developed monoclinic prisms. It melts at 104-105", and is readily soluble in all organic solvents, but insoluble in water. It was found to be free from chlorine, and analyses gave the following results.N=9*22. 0.3102 gave 0 7124 CO, and 0.0974 H,O. C = 62-63 ; H = 3.48. 0.2414 ,, 0.5501 CO, ,, 0,0864 HiO. C=62.15; H=3.97. 0.236 20.4 C.C. moist nitrogen a t 15" and 760 mm. N = 10.12. 0.298 ,, 25.8 C.C. moist nitrogen a t l4"and 755 mm. N = 10.11. Molecular weight determination. 0.4978 gram dissolved in 16.904 grams of glacial acetic acid lowered the freezing point 0.49". M = 234.4. Theory for C,,HI,N,O,, M = 270. ,, C,,H,,(NO,), requires C = 62 22 ; H = 3.70 ; N = 10.37 per cent.2% SUDBOROUGH : RESEARCHES IN THE STTLBENE SERIES, A11 attempts t'o reduce the nitro-groups, and to obtain amido- compounds proved unsuccessful, the only product formed being a dark brown, tarry mass. That the nitro-groups must be attached to the ethylene carbon atoms, and not to carbon atoms in either of the benzene nuclei, was proved by oxidising the nitro-compound with an acetic acid solution of chromic anhydride.no mono- or di-nitrobenzoic acids could be isolated. If the yellow compound has the composition The only product obtained was benzoic acid CGH,* C(NO2) :C(NO,) COH,, it must be formed from the compound C,H,* CH(NO,)*CCl(NO,)* C,H,, already mentioned, by the elimination of a molecule of hydrogen chloride. Chlorodinitrodibenzyl, when heated with alcoholic potash, at once turns yellow, and Ion acidifying and distilling off the alcohol a yellow oil is obtained, but it could not be induced to solidify. Although the yellow, crystalline compound, according t o the formula just given, contains a double bond, it has been found impossible up to the present to bring about any combination with bromine.A solu- tion of the compound in chloroform does not decolorise a chloroform solution of bromine, and, even after standing for some time, the original compound can be obtained by evaporating off the chloroform. ~et~LyZchZoyostiZ6e.ne, C,H, Cy Cl : C( CH,). C,H,. Methyldeoxybenzoi'n, obtained by v. Meyer and Oel ker's method (Bey., 1888, 21, 1295), was heated with a slight excess of phosphorus pentachloride for about half-an-hour on the water bath. The liquid mass was then poured into water, allowed to stand, neutralised with sodium hydroxide, and extracted with ether; the yellow oil left on evaporating the ether gave the following results on analysis. 0.77 gave 0.539 AgCl. C1= 1'7.32. 0.476 ,, 0.331 AgC1.C1= 17.20. CGH5* CC1:CMe- C,H, requires C1= 15.54 per cent. C,H,- CCI,*.CHMe- C,H, ,, C1= 26.78 ,, ,, When the oil was distilled under diminished pressure, a consider- able quantity of hydrogen chloride was evolved, and the greater part passed over at 178' under a pressure of 18 mm. The distillate, on cooling, deposited a small quantity of colourless, prismatic crystals ; these were drained off and the oil again distilled under diminished pressure, when more of the crystalline compound was obtained. By this process of repeated distillation, a considerable portion of the oil was converted into the crystalline compound, although not 20 per cent, was converted a t any one distillation,SUDBOROUGH : RESEARCHES IhT THE STILBENE SERIES. 225 The crystals, when separated, drained on a porous plate, and re- crystallised from hot glacial acetic acid, formed colourless, lustrous plates melting at 124".0.2 gave 0.5767 GO, and 0.106 H,O. 0.233 gave 0.146 AgC1. C1= 15.51. C,H,*CMe:CCl*C,H, requires C = 78.77 ; H = 5.69 ; C1= 15.54 per cent. Molecular weight determination. 0,2065 gram dissolved in 21-12 grams of glacial acetic acid lowered the freezing point 0.17". M = 224.3. Theory for C,H,*CMe:CCl* C,H,, This crystalline methylchlorostilbene is readily soluble in ether, benzene, chloroform, and carbon tetrachloride, also in hot alcohol and in hot acetic acid. Under atmospheric pressure, it distils at about 311' undergoing but slight decomposition. It differs from the solid chlorostilbene in not being acted on by alcoholic potash, and also by not being readily reduced either by sodium amalgam and water or by zinc and acetic acid.The oily compound which was formed along with this solid methyl- ehlorostilbene on distilling the product of the action of phosphorus pentachloride on methyldeoxybenzoin mas also analysed, and gave numbers which agree with those required for methylchlorostilbene. C = 78.64 ; H = 5.88. M = 228.5. 0.25 gave 0.716 CD, and 0.126 H,O. C: := 78'11 ; H = 5.6. 0.5 ,, 0.307 AgCl. C1= 15.19. 0.58 ,, 0,355 AgCI, Cl= 15-14. Theory, C = 78.77 ; H = 5.69 ; C1= 15-54 per cent. The investigation of this oil is being continued, in order to deter- mine, if possible, whether it is merely an impure form of the solid compound, or whether it is to be regarded as a stereo-isomeride. n~etl~ylciLZol.ostiZbe~~e c l i b i - o t ~ d e , C,H,* Cl\leBr* CClBr* C,H,, or methyl- chlorodibromodibenzyl, obtained by a method exactly similar to that used for the preparation of chlorostilbene dibromide, crystallises from hot glacial acetic acid in well-developed rhombic prisms.It melts and decomposes at 122-1 25", is practically insoluble in alcohol, and only sparingly soluble in ether. 0.2324 gave 0.2966 gram silver chloride and bromide, which, when reduced, -gave 0.185 gram metallic silver. Br = 41 *05, C1= 9.12 per cent. Metl~?/lcl~lo~ostilbene dicldoride, C,H,. CMeCl* CCI,* C,H,, or methyltri- chlorodibenzyl, obtained by passing chlorine into a cold solution of methylchlorostilbene in carbon tetrachloride, crystallises from hot glacial acetic acid in well-developed, colourless prisms.It melts and decomposes at 130°, is readily soluble in benzene, chloroform, carbon Theory, Br = 41-18 and C1= 9.13 per cent.226 SUDBOROUGH : RESEARCHES I N THE STILBENE SERIES. tetrachloride and hot acetic acid, but is almost insoluble in alcohol and cold ether, 0,2722 gave 0.3978 AgCl. C1= 36.15. 0.2433 ,, 0.349 AgCl. Cl=35*49. Theory requires C1= 35.55 per cent. ~ t l ~ ~ Z c ~ Z o l . o s t i Z ~ e ~ e , C,H,* C(C,H,):CCl*C,H,, wits obtained from ethyldeoxybenzoln (Meyer and Oelkers, Zoc. cit.), in exactly the same way as methylchlorostil bene from methyldeoxybenzoi'n. When the crude, oily compound was distilled, the greater part passed over at 1S8-189* under a pressure of 27 mm., and when cold set t o a mass of crystals mixed with a certain amount of oil.The crystals, drained and crystsllised from hot alcohol, in which they are readily soluble, form very large, colourless prisms, which melt a t 60" and distil under atmospheric pressure a t 32s' (corr.). I t is extremely easily soluble in all organic solvents, but insoluble in water. Characteristic is the ease with which i t can be obtained in large, well-developed crystals. 0.1542 gave 0.4448 CO, and 0.0865 H,O. C=: 78.53 ; H=6*24. 0.1435 ,, 0.416 CO, ,, 0.0815 H,O. C-='79*06; H=6*31. 0.298 ,, 0.171 AgCI. C1=14.20. 0.2066 ,, 0.1212 AgCI. CI = 14-52, C,H,* CEt:CYCl*C,H, requires C = 79.1s ; H = 6.18 ; C1= 14.64 per cent. Molecular weight determination. 0.2925 gram dissolved in 20.567 grams of glacial acetic acid lowered the freezing point 0.234". 0.3565 gram dissolved in 19.298 grams of glacial acetic acid lowered the freezing point 0.310".Alcoholic potash has no action on the compound, as was to be expected, It was also found impossible to reduce it either with sodium amalgam and water or zinc and acetic acid. ~tl~~lc~LlorostiZbene dicldo&Ze, C,H,* CEtC11. CCl,* C,H,, or ethyltri- chlorodibenzyl, was obtained by chlorinating LZ solution of ethglchloro- stilbene in carbon tetrachloride, the solution being kept quite cold arid allowed to stand for 2 hours after saturation ; the excess of chlorine was then removed by drawing a rapid current of air through the solution, and the carbon tetrachloride left to evaporate spontaneously. The syrupy, oily residue, which became solid on treatment with cold alcohol, was washed with warm alcohol and then recrystallised from ether..The dichloride crystallises in colourless needles, melts at 90-9 lo, and is readily soluble in chloroform, benzenc, and acetic acid. M = 236.S. M = 232.4. Theory, M = 242.5. 0.214 gave 0.2938 AgCI. C1= 33.96. C!,H,- CEtU1-C'C1l2*U,H, requires C1= 33.97 per cent,SUDBOROUGH : RESEARCHES I N THE STILBENE SERIES. 224 &!h@iZos.ostil6ene clihomide, C,H,* CEtBr* CClBr* C,H,, or ethyl- chlorodibromodibenzyl, was obtained as a thick oil, which readily solidified on treatment with a little alcohol. It melts at 97-99', is insoluble in alcohol, and only sparingly soluble in ether. 0.2 gram gave 0.259 gram of silver chloride and bromide, which, when reduced, gave 0,1615 gram of metallic silver.Br=39.92 and C1= 8.83 per cent. Theory requires Br = 39.75 and C1= 8.81 per cent. The purification and investigation of the oily compounds obtained by treating deoxybenzoin and its alkylic derivatives with phosphorus pentachloride have already been begun, in order t o determine, if possible, whether these oils are only slightly impure forms of the crys- talline chloro-, methylchloro-, and ethylchloro-stilbenes described above, or whether they are to be regarded as isomerides. If the substances are isomeric, beyond a doubt almost they are stereo-isomeric, as the oils are so readily converted into the solid compounds. The best method of attacking the problem, when once the oily com- pounds have been obtained in a pure state, appears to be to study the additive compounds formed by the union of the oils with chlorine and bromine, and then to compare the products with those obtained from the solid compounds.I n the case of rnethylchlorostilbene, for example, the oil and solid, if stereo-isomeric, must bear the same relationship 60 one another as the two tolane dichlorides or as the two crotonic acids. I n other words, one must have the &-configuration A, and the other the tmms-configuration B. C,H,-C;)-CH, C .H,*C*CH, C,H,* c c1 Cl*C*C,H, A, B. ' ' 11 The effect of chlorinating the compound A would be to form an inactive mixture of the two enantiomorphic compounds I and 11. ? C,H,*C*CH, c1 I. C,I€,.b* I c1 c1 I CN; C-C,H, 11. I C1* C*C,H, I c1 or else a racemic compound of the two. compound of I11 and IT.The compound B would also yield an inactive mixture or racemic C1 CH&C,H, Y1 C,H,* C*C€I, I I v. I Cl*C*C,H, C,H,*C*Cl Cl c1228 SUDBOROUGH : RESEARCHES IN THE STILBENE SERIES. It can easily be shown by means of models that compounds I11 and IV are identical with compounds I and 11. This also follows from the fact that the compound C,H,- CCl(CH,)*CCI; C,H, contains only one asymmet'ric carbon atom, and can therefore exist, according t o the van't Hoff theory, in only two optically active modifications, which can combine to form a racemic compound. The result of brominating the compounds A and B would, however, be somewhat different. A would yield us an inactive mixture of the two enantiomorphic compounds V and VI. Sr Br c,H,- ~WH, CH,. 6 C,H, V. I V I. I 7 C,HS*C.C1 Crl*C* C,H, Er Br I I whereas B would yield an inactive mixture of the compounds VII and VIII. Br Br C,H,*bCH:, CH,. b C,H, VII. I VIII. I C1* C C,H, C,H,* C*C1 I I Br Br Compounds VII and VIII, however, are not identical with compounds V and VI. This can readily be established by means of models, and is in accord with the fact that the compound C,H,* CBr(CH,)*CClBr*C,H, contains two asymmetric carbon atoms, and can exist in four optically active and two racemic forms. Based upon these purely theoretical considerations, we should expect both the oily and solid methyl- chlorostilbenes, if they are stereoisomeric in the same way as the crotonic acids, to yield the same dichloride but different dibromides. The same applies to the chlorostilbenes and ethylchlorostilbenes. On the other hand, if the oil is merely an impure form of the solid, then both compounds should Field the same dichloride and also the same dibmmide. A third possibility, that the compounds are structurally isomeric, could be settled a t the same time, for in this case the oil and solid should yield different dichlorides, and also different dibromides. In conclusion, I wish to express to Dr. Irving W. Fay my sincere thanks for his kind assistance in some of the analytical work entailed in this investigation. UNIvEiisITY COLLEGE, NOTTJXGIIAM.