56 CHAPMAN SPTNACENE A NEW HYDROCARBON By A. CHASTON CHAPMAN. IN September 1915 a sample stated to consist of cod-liver oil was submitted 'to me with the request that' I would report on its genuineness. I n the course of its examination the following results were obtained : Specific gravity (15'/15') .......................... 0.8666" Iodine value (Wijs) ................................ 358 Saponification value .................................. 22.5 Unsaponifiable matter.. ............................. Iodine value of unsaponifiable matter ......... 376.2 Free fatty acid (as oleic acid) ..................... 0.42 per cent. I reported that the sample consisted of a mixture of approsirn-ately 89 parts of some unsaturated hydrocarbon oil with approximately 11 parts of some fish oil.Whilst it was clear that I could not have arrived a t any other conclusion as to the general composition of the sample the analytical numbers perplexed me very greatly. I knew of no hydrocarbon or other unsaponifiable substance having the iodine value and the other properties which this appeared to possess and I began to wonder whether the sample might not be an abnormal and not an adulterated specimen of oil. I therefore decided for my own satisfaction to submit it t o a systematic investigation. As bearing on the origin and authenticity of this sample it is necessary to follow the history of the oil a little further before describing the results of its chemical investigation. Some weeks after the work had commenced I received a copy of a report which greatly stimulated my interest in the matter.The vendors of the oil in Lisbon repudiated the suggestion that it was not genuine and submitted a sample to Dr. Hugo Mastbaiun of 89.1 per cent. Bromine precipitate insoluble in ether ......... 76.5 , , FROM CERTAIN FISH LIVER OILS. 57 that city who obtained results similar to mine and expressed the same opinion. As the vendors were still not satisfied they sub-mitted portions of the fish livers to Dr. Mastbaum who again obtained similar results with the expressed oil. Finally Dr. Mast-baum witnessed the actual extraction of the livers of two different species of fish from which the oil in question had been derived. These fish which will be referred t o more fully below appear t o be known in Portugal as " Barroso " and (' Carocho " respectively.The fish according to1 Mastbaum were about 1 metre in length, were quite fresh and the livers were ('smooth and oily and when pricked with a knife considerable quantities of oil immediately separated the bulk of the liver substance being transformed in the course of a few hours into an oily liquid." I n a subsequent communication (Chem. Zeit. 1915 39 139 SSS), the following results are given by Mastbaum for these two samples of oil: '' Barroso." '' Carocho." Specific gravity (15"/15" ) ............ 0.8637" 0-87 11" Butyro-refractometer at 25". ........ 102 93 Polarisation in 200 mm. tube ...... -0.3 - 2.33 still liquid becomes turbid ................... at -7" a t -7" 36.7 -c Solidification point Saponification value ..................15.4 Free acids (calculated as oleic acid) 0.097 per cent. 0.165 per cent. Until quite recently I was unaware that Dr. Mastbaum had published this note and I therefore wrote to him pointing out that I had completed the first part of an investigation of the oil and asking him to give me any information he could in reference to its origin. He replied informing me that the oil in question was obtained from the two species of fish above referred to and added the information that according to a standard Portuguese dictionary the scientific descriptions of the above fish are Ceiztro-phorus granulosus and Scymmus lichia respectively. He also informed me that these fish are caught in deep water off the Moroccan coast and that they have only come into1 the Portuguese market since the employment of steam trawling in those fishing grounds.He suggests that it is owing to this fact that the exist-ence of fish-liver oils containing so large a proportion of unsaponi-fiable matter has hitherto1 escaped observation. Assuming that the fish in question were not in any way abnormal the existence of liver oils containing nearly 90 per cent. of an unsaturated hydrocarbon is a matter of very considerable interest both from the analytical and from the physiological points of view. The fact that samples of oil having approximately the same composition were derived from fish which had been caught during periods extending over several weeks and the fact that the D 58 CHAPMAN SPINACENE A NEW HYDROCARBON consignment represented by the sample submitted t o me amounted t o nearly 5000 litres must I think be regarded as tending t o negative any suggestion of abnormality.It may be recalled that cod-liver oil rarely contains more than 2 per cent. of unsaponifiable matters and that according to Lewkowitsch liver oils obtained from coal-fish tunny skate ling, haddock and hake rarely contain more than about the same amount. On the other hand shark-liver oil appears to contain in some cases as much as 20 per cent. which has been stated to consist largely of cholesterol. I n this .connexion it is of interest to note that the two fish above referred to belong to the Spinacid= or Squalid=) a family of the Selachoidei or sharks. E X P E R I M E N T A L . The sample of oil weighing 330 grams was heated on the water-bath with sufficient sodium hydroxide dissolved in alcohol to effect the complete saponification of the saponifiable portion.After evaporating the alcohol the residue was heated for some time with water and the unsaponified oil extracted by shaking with ether. The ether was distilled off and the remaining oil was heated in a current of steam until the aqueous distillate was free from odour. The oil which did not show any tendency t o distil with the steam, was again extracted with ether the solution dried over calcium chloride and the ether removed by distillation. The residual oil was then submitted to fractional distillation under a pressure of 10 mm. with the following results: ... ... ... 264-268' ...50 grams 268-269' ... ... ... ... 235 ,, 269-2'70' ... ... ... 5 ) ... The residue in the flask welghing about 10 grams was very viscous and dark coloured. From the first fraction there separated on keeping a small quantity of solid which proved t o be cholesterol. The main fraction gave on analysis numbers which indicated that it was a hydrocarbon with a small quantity of some oxygen-containing impurity the percentages of carbon and hydrogen amounting t o 99.6. A preliminary experiment having shown that metallic sodium was without action on the hydrocarbon this frac-tion was distilled over sodium under diminished pressure. There was a little action a t first but this soon ceased and the melted sodium remained quite bright throughout the distillation. The main fraction so obtained consisted of a colourless and fairly mobile oil having a faint and pleasant odour suggestive o FROM CERTAIN FISH LIVER OILS.59 1emOh oil terpenes. terpene-like odour. It burned with a smoky flame producing a C=87.63; H=12*51. It boiled a t 280° (corr.)/l7 mm.: 0.1381 gave 0.4438 CO and 0.1555 H,O. 0.135 , 0.4344 CO , 0.154 H,O. C=87-74; H=12*67. The following results represent the average of five analyses: C = 87.75 ; H:= 12.45. C,H, requires C = 87.80 ; H = 12.20 per cent. The average of four series of estimations by different observers of the molecular weight as determined by the depression of the freezing point in benzene solution was 375 whilst C30H50 requires 410. It will be seen therefore that the determined molecular weight agrees more closely with the formulae CZ7H, or C,,H,, with which the analytical results would also correspond.On the other hand the analyses of the crystalline hexahydrochloride and of the bromine derivative are in better agreement with the higher mole-cular formula. Assuming that the results obtained by the cryo-scopic method as applied to this hydrocarbon are normal there would appear to be a discrepancy which cannot a t present be explained. I would point out that the saturated hydrocarbon gives numbers agreeing better with one of the lower molecular formulae than with the higher. This hydrocarbon does not appear t o be identical with any known hydrocarbon and I propose for it the name spinacene since both Centrophorus granulosus and Scymnus lichia; belong to the natural family of the Spinacidz.It is optically inactive and does n o t solidify when cooled to - 2 O O . Its specific gravity a t 15°/150=0*8641 and a t 2O0/2Oo=0*8616. The following are the results of determinations of its index of refraction a t Z O O : nHa = 1'4932 nD =1.4967 ?LHP = 1.5054 mHy = 1.5130. A t 15O its index of refraction for the B-line is 1.4987. Its specific refraction calculated by the 7z2 - expression is 0.3394, and the molecular refraction 139.1. Taking Conrady's average numbers for the atomic specific refrac-tions (D-line) C30H5 with six ethenoid linkings requires 137.7. Employing the n2 formula the specific dispersive power of spinacene r y - ra is 0.0114 and its molecular dispersion ( r y - m ) M , 4.67.Taking Eisenlohr's numbers for the atomic dispersions for the a- and the y-hydrogen lines the calculated number is 4-33. It (72fT2)d D* 60 CHAPMAN SPINACENE A NEW HYDROCARBON will be seen therefore that the molecular dispersion like the molecular refraction is high. I n this connexion it is worthy of note that both the molecular refraction and the molecular dis-persion of the saturated hydrocarbon obtained from spinacene and described later agree well with the calculated numbers. Viscosity (Time of Efiucc).-Fifty C.C. of spinacene required 78 seconds to flow through the aperture of a Redwood viscometer a t 21° as compared with an average of 370 seconds for rape oil. Absorption of Ozygelz.-1*662 Grams of spinacene were exposed in a flat-bottomed glass dish to an atmosphere of oxygen a t the ordinary temperature.A t the end of two months the hydrocarbon had absorbed 0.397 gram of oxygen and had become so viscous that it would not flow Thin films of the hydrocarbon when exposed to the air formed a hard skin similar to that given by linseed oil. Action of Hydrogen Chloride. Spinacene Hexahydrochloride, C30H50 6 HCl. Dry hydrogen chloride was passed into a well-cooled solution of spinacene in dry ether until the liquid appeared to be saturated. After remaining for some hours a quantity of a well crystalline, colourless solid separated. This was collected washed with ether, and dried in an exhausted desiccator over sulphuric acid and solid sodium hydroxide. More of the solid separated from the filtrate after keeping for twenty-four hours and when this was removed a further quantity formed on remaining for several days.When heated with any of the liquids ordinarily employed for purposes of purification the substance appeared to undergo decomposition but it could be crystallised from a cold mixture of benzene and alcohol. From this solvent it separates in well-defined plates together with some needles the substance being polymorphous : 0.1058 gave 0.2242 CO and 0.0847 H,O. 0.20 , 0.2727 AgCl. C1=33.81. 0.25 , 0.3402 AgCl. C1=33*75. C=57*78; H=8*89. C,H6,,6HCl requires C = 57-23 ; H = 8-90 ; C1= 33.86 per cent. When heated the hexahydrochloride commences to shrink a t 108O and begins to melt a t l l O o . On raising the temperature to about 140° hydrogen chloride is freely evolved.I n a future communication I hope t o deal with the nature of the hydrocarbon left when the elements of hydrogen chloride are removed from the compound either by heating or by the action of reagents FROM CERTAIN FISH LIVER OILS. 61 I propoee also to investigate more closely the properties of spinacene from the point of view of its saturation with other elements or groups. I n particular t,he action of ozone is being studied as i t is expected that this will throw considerable light on the constitution of the hydrocarbon. Spiiuzcene Trihydrochloride CaH,,,3HC1. A current of dry hydrogen chloride was passed through a thoroughly cooled solution of spinacene in ether. As soon as the hydrogen chloride appeared to be in excess the solution was set aside for twenty-four hours and the resulting crystalline hexa-hydrochloride removed by filtration.The ethereal filtrate was shaken with sufficient aqueous solution of sodium carbonate t o remove the excess of hydrogen chloride and was then washed with water dried and evaporated under diminished pressure without the application of heat. A viscous oil remained together with a little of the crystalline hexahydrochloride. The residue was dis-solved in the smallest possible quantity of dry ether and filtered in order to separate the crystals. The ethereal solution after treatment with a little charcoal, was transferred t o a desiccator and the ether allowed to evaporate under diminished pressure. The residue consisted of a pale yellow, viscous oil having a specific gravity 18°/180=1*0137.It is of interest to note that when in one experiment the ethereal solu-tion was treated with sodium carbonate solution immediately after its saturation with hydrogen chloride the resulting oil contained but little chlorine. It is evident therefore that the formation of even the trihydrochloride requires some appreciable time f o r its completion : 0.2607 gave 0.2408 AgCl. C1= 22.90. C3,H,,,3HCl requires C1= 20.5 per cent. The excess of chlorine is due t o the impossibility of separating the hexahydrochloride completely as on keeping a little of the crystalline substance always separates from the oil. When the trihydrochloride is dissolved in ether and the liquid saturated with hydrogen chloride the crystalline hexahydrochloride is formed.Action of Bromine. The action of bromine on spinacene is somewhat complicated, since not only does the saturation of the double bonds appear t o take place in two stages but substitution derivatives are simultaneously formed. Even when a dilute solution of bromine in dry ether or in carbon tetrachloride is added slowly t o a dilut 62 CHAPMAN SPINACENE A NEW HYDROCARBON solution of spinacene in the same solvents a t a temperature of - loo hydrogen bromide is formed in considerable quantity from the very beginning. A solution containing 0.445 gram of bromine in 25 C.C. of carbon tetrachloride was added t o 20 C.C. of a solution containing 0.1996 gram of spinacene in the same solvent'. Both solutions were well cooled in a freezing mixture and immediately after the addition the excess of bromine was determined by titration with sodium thiosulphate after the addition of potassium iodide in the usual way.It was found that 0.39 gram of bromine had entered into the reaction corresponding with 10 atoms of bromine for one molecule of spinacene. The following experiments will illustrate this behaviour. This experiment was repeated with a similar result. I n the next experiment 50 C.C. of the bromine solution (0.89 gram of bromine) were added to the same volume (20 c.c.) of the spinacene solution and t h e excess of bromine determined after keeping f o r three hours a t the ordinary temperature. 0.1996 Gram of spinacene had reacted with 0.566 gram of bromine correspond-ing with rather more than 14 atoms for one molecule.I n these experiments hydrogen bromide was given off in apparently considerable quantities. The following experiments were then made with the object of differentiating between the bromine uniting directly with the hydrocarbon and the bromine existing as hydrogen bromide formed during the operation as the result of substitution. The well-cooled bromine solution was run as before into the well-cooled solution of spinacene in carbon tetrachloride in an apparatus so arranged as t o permit of the estimation of the hydrogen bromide formed. To 0.2007 gram of spinacene 0.557 gram of bromine was added. Immediately after the addition, it was found that 0.427 gram of bromine had reacted, of which 0.0673 gram existed as hydrogen bromide.From this it will be seen that the weight of spinacene taken has united directly with 0.427 - (0.0673 x 2) = 0.2924 gram of bromine which corre-sponds with rather more than 7 atoms for one molecule. I n a further experiment still more carefully arranged so as to prevent the loss of any hydrogen bromide i t was found that the amount of bromine with which one molecule of spinacene united directly corresponded with 6.7 atoms. The above results show that under the conditions obtaining in the last two experiments that is t o say when the temperature was kept low when the solutions of the hydrocarbon and of the bromine were dilute and when the titration of the uncombine FROM CERTAIN FISH LIVER OIL3. 63 bromine was made a t once only three of the six ethenoid linkings are saturated.This fact coupled with the existence of the tri-hydrochIoride and of the other tri-derivatives referred to below, appears t o indicate that three of these linkings are differently situated in the molecule from the remaining three. The fact that the above results are slightly in excess of the six atoms corre sponding with three ethenoid linkings may be due to the com-mencement of the saturation of the remaining linkings or to the union of the spinacene with some of the hydrogen bromide a t the moment of its formation. The most probable explanation of this behaviour with bromine is that three of the ethenoid linkings are in the open chain and the remaining three in some ring system. This view is supported by the way in which spinacene reacts with hydrogen chloride and by some other considerations which will be referred to later.Spinuc erne Dode ca b ro mide C30H50Br12. When a solution of bromine in dry ether is added to a solution of spinacene in the same solvent the colour of the bromine rapidly disappears and after a time a white finely crystalline substance commences to separate. This was filtered washed with ether and purified by the cautious addition of alcohol t o a solution in tetra-chloroethane when it separated in a crystalline condition. From the filtrate which contained an excess of bromine a further quantity of the same compound seEarated on keeping for some days and when this was removed some more crystallised on further keeping : 0.1252 gave 0.123 CO and 0.0416 H20. C=26.79; H=3*69.0.150 , 0.2453 AgBr. Br=69.7. C30H50Br12 requires C = 26-27 ; H = 3-64 ; Br = 70.07 per cent. When heated the substance commences to darken a t about 160° and melts and decomposes at. about 1 8 5 O . It is not appreciably soluble in alcohol or ether is sparingly soluble in acetic acid carbon tetrachloride o r chloroform and is moderately soluble in pyridine or trichloroethylene. It was found that a portion of the bromide which separates from the ether was insoluble in tetrachloroethane the remainder being freely soluble, This contained the same percentage of bromine as the soluble portion so that it would seem that the dodecabromide exists in two modifications. The compound soluble in tetrachloroethane exhibits the pheno-menon of polymorphism the same microscopical preparation often showing needles plates and nodules.I n some cases the trans-formation of one form into the other as for example the forma 64 CHAPMAN SPTNACENE A NEW HYDROCARRON tion of needles from spherical nodules can be observed. When solutions in carbon disulphide are allowed to remain for some time they become converted into a jelly. The filtrate from the above insoluble bromine derivatives was shaken with a slight excess of an aqueous solution of sodium hydroxide. The ethereal liquid was washed with water dried, and evaporated under diminished pressure without the applica-tion of heat. A viscous oil remained which on the addition of a mixture of light petroleum and alcohol and stirring with a glass rod gradually solidified.It was purified by crystallisation from a mixture of ethyl acetate and alcohol or from a mixture of benzene and alcohol. From the latter solvent it forms colourless, spherular masses of radiating needles. This contained 63.7 per cent. of bromine corresponding almost exactly with nine atoms of bromine but owing to the complex character of the reactions involved and to the impossibility of distinguishing by analysis between additive compounds and those in which there has been some substitution i t is inadvisable a t the moment to assign a definite formula to this derivative. The study of this and of other compounds formed during the action of bromine is being continued. Spinacene Trin itrosoch loride C3)H5"( N0C1)3. Two grams of spinacene were mixed with 5 C.C.of amyl nitrite, and to the liquid cooled to -15O a mixture of 6 C.C. of hydro-chloric acid (D 1-19> and 12 C.C. of glacial acetic acid was added. The mixtnre when vigorously shaken became green and in a very short time almost solid. It was allowed to remain in the freezing mixture for one hour and was then treated with water and filtered. An attempt was made to crystallise the air-dried substance from various solvent's but the heating necessary to bring about solu-tion invariably caused decomposition. It was finally obtained in a roughly crystalline form by dissolving it in cold tetrachloro-ethane and gradually adding light petroleum in which it is insoluble. It forms a pale buff-coloured substance which when dry is moderately stable. The above conditions must be closely followed if the solid nitrosochloride is to be obtained and as is the case with the majority of the compounds of spinacene the best results are got by working with small quantities of substance: 0-2533 gave 0.1889 AgC1.Cl=18.48. 0.252 , 16.4 C.C. N2 a t 20° and 752 mm. N=7-36. C3,H,,03N3C13 requires Cl= 17.50 ; N = 6.92 per cent FROM CERTAIN FISH LNER OILS. 65 So far as I am aware this is the first instance of the prepara-tion of a nitrosochloride containing three NOCl groups. The excess of chlorine is due to the impossibility of .obtaining this conipound free from a little of the hexanitrosochloride these com-pounds being unstable and consequently difficult to purify. S 2 r i n ~ r ~ n ~ Dinitrosochloride Mononit?.ol~'p~ridi~p, C3oH,,,(NOC1)2NOC,NH,o.When piperidine was added to the solid spinacene trinitroso-chloride the latter dissolved readily with development of heat. After heating the solution gently in order to avoid decomposition, cold water was added. The precipitated substance was collected, dissolved in dilute sulphuric acid and precipitated by the cautious addition of dilute sodium hydroxide solution. It was then purified by dissolving i t in cold alcohol and precipitating by the gradual addition of water. When dry i t formed a buff-coloured substance which was very readily soluble in dilute acids and in the ordinary organic solvents. When heated it commenced to shrink a t about l l O o and melted and decomposed at about' 146O : 0.2140 gave 17.0 C.C. N a t 15O and 758 mm.This substance forms a crystalline hydrochloride. In the above preparation although the amount of piperidine was in excess of that required to react with the three NOCl groups, it will be seen t h a t only one was attacked. N=9.10. C3,H,;,,0,N,C12 requires N = 8.60 per cent. Sp'nacen~ Trinifroli?ipell.i~~de C,,H,,(NOC,NH,,,),. In the preparation of this compound the nitrosochloride was added to a quantity of piperidine in excess of that required to react with the three NOCl groups. As before there was con-siderable development of heat but' in this case as soon as the reaction had slackened the liquid was heated nearly to the boil-ing point of piperidine. From the resulting dark brown liquid, water precipitated a gummy substance which was purified by dis-solving it in dilute sulphuric acid and precipitating with a dilute solution of sodium hydroxide this operation being repeated three times.The resulting substance which had a pale brown colour, was very readily soluble in alcohol and unlike the mononitrol-piperidide was not precipitated by the addition of water : 0.2041 gave 20.4 C.C. N a t 15O and 758 mm. When dry hydrogen chloride was passed into a solution of this N=11.52. C,,H,,O,N requires N = 11.20 per cent 66 CHAPMAN SPINACENIX A NEW HYDROCARRON compound in ether a crystalline substance separated which how-ever became gummy when attempts were made to dry it. Spinaceite Dinitrosochloride Mononitrolb en2 ylamide, C,,H,,( NOCl),NO*NH*CH,* C,H,. Spinacene nitrosochloride was added little by little to benzyl-amine the latter being in excess of the quantity required to react with the three NOCl groups.There was a vigorous reaction and when this had subsided the mixture was gently warmed and allowed to remain at the ordinary temperature for an hour. On treating the resulting pasty mass with cold water a plastic brown substance remained undissolved. This was purified by dissolving it in cold alcohol and precipitating it by the gradual addition of water the process being repeated several times. Finally the substance after drying was obtained as a buff-coloured powder : 0.2027 gave 15.4 C.C. N a t 20° and 752 mm. N=8*59. 0.1696 , 0.076- AgC1. C l = l l . l l . C,7H,80,N4Cl requires N = 8-27 ; C1= 10.50 per cent. Spinacene Triizitrol b enzylamide C30H,,(NO-NH*CH,*C,H5)3.Spinacene nitrosochloride was treated with an excess of benzyl-amine as in the preparation of the preceding compound but after the first reaction had subsided the mixture was heated to the boil-ing point of benzylamine for a few minutes. On the addition of water to the resulting dark red solution a viscous substance separated. This was purified by repeated solution in alcohol and precipitation with water and finally by dissolving it in glacial acetic acid diluting with water and precipitating with ammonia. The substance thus obtained had when dry a pale yellow colour, dissolved readily in dilute acids and appeared t o form a crystal-line hydrochloride : 0.1653 gave 15.8 C.C. N a t 21O and 740 mm. C,,H,,O,N requires N = 10.27 per cent. It will be seen that the action of benzylamine on the nitroso-chloride is similar to that of piperidine that is to say one mole-cule of NOCl is much more readily attacked than the remaining two.N=10*58. Spina c e n e H e xn n 1 t TOS o c h7oride C,,H (NO C1) 6. I n the preparation of the trinitrosochloride as described above a substance was obtained on one occasion which on analysis wa FROM CERTAIN FISH LIVER o m . 67 found to give the following results corresponding approximately, as will be seen with the composition of a hexanitrosochloride: 0.2381 gave 22.4 C.C. N a t 21° and 754 mm. N=10*62. 0.2305 , 0.255 AgCI. C1=27*42. C,0H5,0,N,C16 requires N = 10.46 ; C1= 26-52 per cent. The experimental conditions attending the preparation of the above compound did not differ intentionally from those which in other cases resulted in the formation of the trinitrosochloride.The percentage of chlorine is high but the compound is not very stable and is difficult to purify. Spins c e n e N i t rosa t e C3,H5 (N 0 *N 0,) 3. A mixture of spinacene with twice its volume of amyl nitrite was cooled in a freezing mixture to -15O and to this solution a well-cooled mixture of nitric acid and glacial acetic acid way added little by little with constant shaking. After remaining at the above low temperature for one hour the mixture which had become viscous was poured slowly into cold water. A yellow substance separated which could not be satisfactorily recrystal-lised from any of the ordinary organic solvents. The substance is practically insoluble in light petroleum sparingly soluble in ether or carbon disulphide somewhat more readily soluble in alcohol benzene chloroform o r carbon tetrachloride and very readily soluble in acetic acid tetrachloroethane or acetone.When heated it decomposes a t 85O with the formation of a considerable volume of gas. When prepared as above described it consisted of a yellow powder but when kept in a specimen phial for several weeks at the ordinary temperature i t underwent gradual decomposition, becoming converted into a dark brown spongy mass: C,HmO,,N requires N = 12.30 per cent. 0.2704 gave 31.0 C.C. N a t 1 9 O and 739 mm. Many attempts were made to prepare a nitroso- or an isonitroso-derivative of spinacene but without success. When to a solution of spinacene in light petroleum cooled to -15O a concentrated aqueous solution of sodium nitrite was added followed by the addition of acetic acid the mixture acquired a bright green colour, indicating the formation of a nitroso-compound but no crystal-line substance could be obtained.N=12.80. W y clr og e n a t i o n of S p i ~ a c e I L e . Fifteen grams of spinacene were introduced into a test-tube together with rather more than 1 gram of freshly prepared an 68 CHAPMAN SPINACENE A NEW HYDROCARBON ETC. very active platinum black. The tube was supported in an air-bath which was so heated that the contents of the tube could be maintained a t a temperature of 180° t o 190° throughout the experiment. A slow current of dry purified hydrogen was passed through the heated spinacene by means of a piece of glass tubing drawn out to a fine orifice and passing to the bottom of the tube, so as to keep the platinum black in a state of suspension.The iodine value which a t the commencement of the experi-ment was 350 (Wijs’s method) fell rapidly a t first and then more slowly until after the hydrogen had been passing for about thirty hours i t had fallen to 18 at which point i t remained almost stationary. The contents of the tube were separated from the platinum black and fractionally distilled over metallic sodium under 18 mm. pressure. Almost the whole of the liquid passed over a t 274-275O (corr.) as a perfectly colourless odourless oil, which burned with a smoky flame and a slightly resinous odour : C=85.23; H=14-67.0.138 gave 0.431 CO and 0.1822 H,O. It will be seen that the above formula is that of a paraffin. That this hydrocarbon is not a normal paraffin is shown by the fact that it remains liquid when cooled to -20° and for several reasons it seems improbable that it is one of the isoparaffins. The specific gravity of the hydrocarbon a t 20°/200= 0.8172O. The following are the results of determinations of its index of refraction a t 20°: C,H, requires C = 85-30 ; H = 14.70 per cent. ttHa= 1.4525 nD =1*454$ 9lHy = 1.4655. n1lp = 1.4607 Employing the n2 formula the specific refraction of this com-pound is 0.3318 and its molecular refraction 140.0. The theoretical number for a saturated hydrocarbon having the formula C,,H,, is 140.1. Its specific dispersion ry-ra is 0.0082 and its molecular dis-persion 3.46.Taking Eisenlohr’s numbers for the specific atomic dispersions of carbon and hydrogen for the a- and y-hydrogen lines, the calculated value is 3.47. The nature of this hydrocarbon like that of spinacene itself, must for the present remain a matter f o r conjecture. That spinacene is a chain compound containing a ring system is exceed-ingly probable but more than this can scarcely be said until the oxidation products have been thoroughly studied. As soon as the opportunity occurs I hope to deal as fully as possible with thi NITRATION OF 2-ACETYLAMINO-Q 4-DIMETHOXYBENZOIC ACID. 69 aspect of the subject. I would merely remark that some pre-liminary experiments in this direction tend to strengthen the view that this hydrocarbon may prove t o be in some way related to the t erpenes. Since this communication was submitted t o the Society I have seen a recent paper ( J . Z d . Eng. Chem. 1916 8 889) by M. Tsujimoto dealing with an unsaturated hydrocarbon obtained from the livers of certain Japanese sharks to which he has given the name “sgualene,” and which if not identical with spinacene, resembles i t closely. My best thanks are due to my assistant Mr. Frederick T. Harry, and to my late assistant Dr. B. Ghosh for valuable help in connexion with this investigation. 8 DUKE STREET ALDGATE E.C. [Received November 25th 1916.