140 JONES APU’D WHEELER : THE COJIPOSITION OF COAL.x I x . - Th e Coinposit ion Coa,Z.By DAVID TREVOR JONES and RICHAKD VERNON \VHF,ELF,P,.THE methods most promising in results t h a t have been appliedtowards attempting to elucidate the question of the chemical com-position of coal have been either to treat the coal with differentsolvents, in the hope of being able to extract some simple substancetherefrom, or to distil the coal destructively under various con-ditions in order to examine the products of decomposition.The method employed t o obtain the results described in thepresent paper was that of destructive distillation of untreated coalin a vacuum a t a low temperature, in the manner already described.Recently Pictet and Bouvier (Compt. relid., 1913, 157, 779) havepublished a preliminary note regarding experiments that they havecarried out, which are practically identical in character with ourown.The results obtained by these investigators differ, however,from ours in sevel-al importmt respects which will be discussedlater.It may be remarked a t the outset that the destructive distillationa t low temperatures of coal in its entirety cannot be expected toreveal the nature of all the various classes of compounds thatcompose the coal substance. I n previous papers it has been shownthat only the (( resirous subst mces ” in bituminous coals decomposeto any great extent a t temperatures below 500O. It is the “ resinoussubstances” only, therefore, in the coal conglomerate that must beregarded as having undergone examination by the method of attackdescribed in the present paper.The work on which we are atlpresent engaged, namely, that of destructive distillation in a vacuumof the separate portions into which coal can be divided by thesolvent action of pyridine and chloroform, will, it is hoped, enableconclusions to be drawn as t o the character and composition of the“ humus substances ” in the coal conglomerate as well as of the“ resinous substances.”Bituminous coals, wllen distilled in a vacuum (5 to 40 mm.) a ttemperatures up t o 430°, yielded, besides gaseous products andwater, about 6.5 per cent. of their weight of tar. On distillinJONES AND WHEELER: THE COMPOSITION OF COAL. 141this tar, about half remained as pitch, boiling above 300O.Theoils boiling below 300° contained the following substances :1. Unsaturated (ethylenic) hydrocarbons of indeterminate com-position, for the most part richer in carbon than the mono-olefines,(CnH2n).2. Naphthenes (C,H,,) and liquid paraffins, the former greatlypredominating, forming together about 40 per cent. of the oils.3. Phenols, chiefly cresols and xylenols (between 12 and 15 percent:).4. Aromatic compounds (about 7 per cent.), apparently homo-logues of naphthalene, although naphthalene itself did not appearto be present.5. A solid paxaffin in small quantity, of which the melting pointlay between 52'5O and 54O, and of which the molecular weight was373-7 (that is, intermediate between the values required for C,,H,,and c27H56)*These formed between 40 and 45 per cent.of the oils.6. Pyridine bases in traces only.Benzene, anthracene, carbon disulphide, and solid aromatic hydro-carbons were absent, nor was there any evidence of the presenceof toluene or other homologves of benzene in more than minutequantity.The water distilled from the coal contained hydrochloric acidand traces of ammonium chloride.Picht and Bouvier (Zoc. cit.), using a French gas coal, which theydistilled under 15-17 mm. pressure in an iron retort placedvertically and with its mouth upwards, obtained about 4 per cent.of a light tar a t temperatures not exceeding 450O. This tar theyre-distilled, a t atmospheric pressure, between 120° and 300O.Contrary to our results, no phenols were present in their distillates,iior were ammonia or solid paraffins detected.Their preliminaryexperiments also indicated that aromatic compounds, if not absent,existed in traces only. On the other hand, Pictet and Bouvierobtained appreciable quantities of menthol-like compounds and ofbases, chiefly secondary bases.The marked difference between the oils that we have obtainedand those described by Pictet and Bouvier may, perhaps, beattributed to differences in the apparatus employed or to the muehgreater rapidity with which their distillations were effected. (Adistillation was completed in five hours, whereas ours took as manyweeks.)It seems more probable, however, that the differences observedin the oils are due t o differences in the nature of the coals employed,for i t is inconceivable that, assuming the coals t o be similar incharacter, Pictet and Bouvier should find no phenols in their oils142 JONES AND WHEELER: THE COMPOSI'I'ION O F COAL.whilst ours contain as much as 12 per cent.* Further, differencesin character between the Frenca and the British coals are indicatedby the absence from the distillates from the former of solidparaffins which, although small in amount, seem to be essentialproducts of the distillation of all the bituminous coals that we haveexamined.This brings 11s t o the question as to whether free hydrocarbonsnecessarily exist in all coals.The wide distribution of hydro-carbons, and particularly of Dolymerisable hydrocarbons, in plantlife aff orda strong argument for the existence of such polymeridesin coal, despite their known tendency towards rapid oxidation.The fact that paraffins have been found oozing from coal seams,usually in the neighbourhood of a fault (compare Cohen and Finn,J .SOC. Chern. Ind., 1912, 31, 12), is no criterion of what ordinarilyobtains; for such oils may very well have been formed locally fromthe original coal by increased temperature, due to earth-movement,for example, a t some period in the history of the seam.More direct evidence of the existence of hydrocarbons as usualconstituents of the coal substance has been obtained by Pictet andRamseyer (Ber., 1911, 44, 2486), who have isolated hexahydro-fluorene, C13H16, from a portion of a gas coal soluble in benzene.Pictet and Ramseyer regard the existence of hexahydrofluorene intheir coal as lending support to the views of Donath, who considerscoal to consist in part of compounds, originally liquid, that havegradually solidified owing to progressive polymerisation. The factthat hexahydrofluorene was Ihe only hydrocarbon that they couldextract from coal withL benzene, Pictet and Ramseyer attribute tothe possible preferential solvent action of benzene for that sub-stance ; or, alternatively, to the possibility of other hydroaromaticcompounds that may have been present originally having poly-inerised to a greater extent, and thereby escaped solution inbenzene.It should be noticed, however, that Pictet and Ramseyerextracted traces of phenol and bases by benzene simultaneously withthe hexahydrofluorene.We have found that phenols and bases areinvariably products of the low temperature distillation of coals,so that their extraction by a solvent from the particular coal usedby Pictet and Ramseyer suggests the possibility of that coal, atsome time in the history of the seam, having been raised t o atemperature of incipient decomposition, when hexahydrofluorene,phenols, and bases were set free. It is possible, therefore, thathexahydrofluorene, instead of being a survivor from polymerisedcompounds that compose part of the coal substance, may have had* We obtain phenols also in the oils distilled from coal under a pressure of 5 mm.at temperatures as low as 325JONES AND WHEELER: THB COMPOSITION OF COAL.143its origin after the coal substance was formed, and may be anincidental constituent of some rather than an essential componentof all coals.We have obtained evidence of the existence of free, solid paraffinsin several British coals by treating the extract obtained by thesolvent action of pyridine and chloroform (T., 1913, 103, 1704)with pentane. The pentane solution yields crystals of paraffin wax,nleltiiig between 5 5 O and 59O, similar in composition to thoseobtained by the destructive distillation of the coal. This wax formsabout 0.10 per cent. of the total weight of the coal. The argumentsalready advanced against the supposition that hexahydrofluoreneis present in coals generally are, of course, applicable in the presentcase also-free solid paraffins may not be present in all coals.It is necessary now t o discuss the manner in which the varioussubstances that are found in the oils obtained by distillation ofcoal in a vacuum at low temperatures have been formed.It is impossible to explain the formation of liquid (or gaseous)paraffins by assuming a thermal decomposition of free solid paraffins,for these are obviously present in the coal in insufficient quantity.At the same time it is difficult to conceive of their formation, or,for that matter, of the formation of other types of hydrocarbonsfound in the coal distillates, by rapid and complete pyrogenicsynthesis either immediately before or a t the moment of distillation.It seems more reasonable to suppose that the paraffins must bepresent in the coal substance in such a manner that whilst,in a sense, structurally complete, some change in their state, suchas can be produced by moderate heating, must take place beforethey can be set free.The most likely condition of existence of the parailins inaccordance with this supposition, that seems t o us to explain theirappearance in coal distillates, is as alkyl or p a r a h o i d groupsattached chemically to another, non-alkyl, group R.H.Thus wehave the paraffin in what may be conveniently called a “bound”condition, as a component part of a molecule represented by thegeneral formula RH-C,H,,+,, where N may have any value upto 32 or even higher (compare Cohen and Finn, who isolated thehydrocarbon, C32H667 from a mixture of paraffins occurring in acoal seam).The rapid distillation of I‘ free ” paraffins from these ‘ I bound ”molecules when coal is decomposed thermally can now be explainedaccording to the following scheme : ** Compare Engler on the formation of petroleum (Zeitsck.nngew. Chem., 1908,30, 1585)144 JONES AND WHEELER: THE COMPOSITION OF COAL.orRH*C,q?,+, --+ R + CnH,,,, + G,H*?I1 * * (2)It will be seen that with some modification this hypothesis canbe applied to explain the formation of other types of cornpoundsfound in the distillates. Thus, in the case of the naphthenes,whilst we cannot rule o u t the possibility of ring-formation takingplace during the distillation, it is more probable that they existin the coal as “ bound” molecules.F o r the results of recentinvestigations, notably those of Ipatiev (Ber., 1911, 44, 2928), onthe isomerisation or condensation of ethylenic compounds t o formnaphthenes under the combined effects, of heat and pressure, renderit more than probable that under the conditions of the formationof coal saturated ringed carbon compounds have resulted. Theformation of “free” naphthenes from the “bound ” moleculesduring the distillation of the coal can readily be explained accordingto a scheme similar to that given for the paraffins :RH-C,H,,-, + R + CNH2,v . . . . . . . . . . . . . (3)RH*CH,.(:H,.. . CH2*C,H2,-,--+ R+CH3*CH,. . . CH,*C,H:N-l (4)The presence of etlzylenic compounds in the oils may beexplained in part by the breaking down of highly polymerisedethylenic compounds that can reaso,nably be assumed to be presentin the coal substance, the process being comparable t o that whichtakes place when caoutchouc breaks down and yields isoprene; orwe may apply our hypothesis and assume a “bound” ethylenegroup :RH*C,H2,_* -+ R + C,H,, .. . . . ( 5 )Equation 2 also affords a possible explanation of the formationof a portion of the ethylenic compounds.Naphthalene homologues were present in the oils distilled fromcoal in a vacuum up t o 430°, but benzene was absent. There isno reason, according to our hypothesis, why ‘ I bound ” naphthalenegroups and “bound” benzene groups should not both be presentin coal, capable of yielding free naphthalene and free benzene onheating a t low temperatures.The presence of the former type ofcompounds and the absence of the latter in the oils thereforerequires some explanation. I n this connexion the behaviour onheating of their hydrogenated derivatives, concerning which weare able t o record observations made during the course of anotherinvestigation, is suggestive : dihydronaphthalene suffers decom-position a t 400°, yielding chiefly naphthalene and hydrogen ;hexahydrobenzene, on the other hand, remains undecomposed a t450°, the maximum temperature reached in our distillations ofcoal. I f therefore we infer that hydrogenated naphthalene groupJONES AND WHEELER: THE COMPOSITION OF COAL. 145as well as hydrogenated benzene groups are present in coal in the“bound” condition, it can be understood that only the formerwould eliminate hydrogen under the conditions of our experimentsbefore becoming (‘ free,” and that benzene would not appear.Wethus have as a general scheme similar to those already given:In the case of the phenols the assumption of the presence incoal of “bound ” groups, su‘ch as -C6H8<gE3, is rendered doubtfulby the fact that very little of the coal substance is soluble inpotassium hydroxide ; whilst, although (‘ bound ” hydrogenatedgroups might be assumed, as in the case of the naphthalene homo-logues, our knowledge regarding the thermal decomposition ofcompounds of that type is insufficient to enable us to judge whetherhydrogen would be eliminated from them in preference to water.*Russig (Ghem.Zeit., 1902, 26, 190, 344) suggested that phenolsmay arise from the decomposition of polymerised coumarones con-tained in coal, Eraemer and Spilker (Ber., 1900, 33, 2257) havingobserved that pcoumarone, (C8H60)4, begins to decompose a t 300°,’when it yields 17 per cent. of phenol and 51 per cent. of coumarone.As already stated, we find phenols in the oil distilled from coal a t325O, but, if Russig’s theory be accepted, it is difficult t o understandwhy no coumarones, corresponding with the experiments of Kraemerand Spilker, can be detected.EXPERIMENTAL.Distallation in a Vacuum-The general arrangement of thedistillation apparatus was similar to that described in the pre-ceding paper ((( The distillation of coal in a vacuum ’7 ; that is tosay, it consisted of an inverted glass ((retort” connected throughcondensing tubes to a rapidly-acting Sprengel mercury pump.Theretort contained abodt 1’25 kilos. of coal broken into fragmentsand sieved through a 10- and on a 60-mesh sieve, all fine dust beingcarefully winnopred out. The coal was dried for one hour at 105Obefore being put into the retort.For the experiments described in this paper two bituminous* The observation by Ipatiev (Be?.. , 1910, 43, 3383) that 1-methylcyclohexan-2-01,on heating t o 350” eliminates water in preference to hydrogen, has reference onlyto what occurs in the presence of alumina.VOL. cv. 146 JONES AND WHEELER: THE COMPOSITION OF COAL.coals (one from Scotland and one from Durham) were employed,the proximate and ultimate analyses of which were :Proximate Analyses.' I Volatilematter." Fixed carton. Ash.Scotch coal ......26'36 70-36 3.28 per ceut onDurham ,, . . . . . . 30 '81 65.09 4-10) dry coalU 1 t imn t e A na I y ses.Carbon. Hydrogen. Oxygen. Nitrogen. Sulphur.Scotch coal ...... 86.92 4.98 5-56 1.75 OT9 per wnt. on ash-Durham ,, ...... 86-88 5'41 4'71 1-75 1-25) free dry coalI n early experiments fractional distillations of the coal a tdifferent temperatures were conducted in the manner described inthe preceding paper. It was found, however, that there was noappreciable difference in the character of the liquids distilled a tthe different temperatures, ,so that in the later experiments thecoal wits raised slowly direct to 430° and maintained a t thattemperature.The bulk of the distillate collected in the first receiver, whichwas not surrounded by any cooling-mixture.It consisted of athin, brown, pleasant-smelling tar, which was covered with a layerof paraffins, The specific gravity of the tar was 0*999:.;, which ismuch iower than that of ordinary coal tars.The second and third condensers, which were cooled by a solutionof solid carbon dioxide in ether, contained a reddish-brown oil witha lower aqueous layer containing hydrochloric acid and ammoniumchloride."A certain quantity of oil passed through the Sprengel pumpwith the gases evolved during the distillation and condensed underatmospheric pressure ilz the connexions leading t o the apparatusfor collecting the gases.This oil was of a pale straw colour, boilingbetween 116O and 154O/760 mm., and had a specific gravity of0.781g. On combustion analysis, 0.1430 gave 0.4547 CO, and0.1712 H20. C=86.72; H=13.30 per cent. This oil was not furtherexamined separately, but was added to the main bulk.E x a m i n a t i o n of t h e Volatile Oils.-Oils of very low boiling pointwere removed from the gases collected by passing the latter slowlythrough a condenser cooled by solid carbon dioxide and ether. The* A fraction obtained a t 350" fumed strongly in the air and lied n specific gravityof 1.135. The supernatant oil contained traces of a chloro-derivative, presumablyformed hy the addition of the elements of hydrogen chloride to some unsaturatedcompoundJONES AND WHEELER: THE COMPOSITION OF COAL. 147liquid so obtained was warmed to 35O to free it from dissolvedgases,* and was found to boil between 35O and 125O.It had aspecific gravity 0*699:,5, and 0.1732 on combustion gave 0.5394 CO,and 0-2272 H,O.Itwas clear and limpid, and remained so on long keeping. It rapidlydecolorised potassium permanganate solution, and a solution ofbromine in chloroform.In order to remove and estimate the quantity of olefines present,a portion of the crude oil was first washed with concentratedsulphuric acid in a small, graduated separating-burette, rise intemperature being guarded against. A considerable quantity of“ acid tar ” was formed, and the volume of the oil was reduced from5.1 C.C.to 3.0 c.c., whilst its specific gravity rose slightly (from0.69 to 0.71).The ‘‘ acid t a r ” was heated in steam a t 180°, and the distillateexamined for benzene. A negative result was obtained. The oilitself, with the ‘( acid tar ” removed, was then examined for benzenein the following manner. The oil was washed with fuming nitricacid (D 1-50>, and the acid washings were diluted with water. Thereddish-brown deposit thus formed was washed and dissolved inalcohol and treated with (1) zinc and sulphuric acid, and (2)potassium hydroxide solution. The mixture was then distilled insteam, and the first runnings tested for aniline with a fresh solutionof bleaching powder. The negative result obtained demonstratedthe absence of benzene in the oil.After having been washed with nitric acid, as stated above, theoil was treated with weak fuming sulphuric acid, whereby the lasttraces of olefines were removed.+ It was then freed from acid bywashing with water, with a solution of sodium carbonate, and,finally, with water again, and distilled. The distillate was collectedin two fractions, the first boiling between 48O and looo, and thesecond between looo and 135O.On combustionanalysis the following figures were obtained : 0.1280 gave 0.3964CO, and 0.1797 H,O.Comparison of these figures with the known values of carbon andhydrogen f o r naphthenes and paraffins boiling over about the rangeof temperature of the oil shows conclusively that the oil was amixture of these two classes of compounds:* These gases contained : olefiiies, 23 5 per cent., and paraffins, 75.4 per cent.byvolume, the ratio n/V for tho paraffins being 2.5.T The acid used was of the strength recomniended by Zeliuski for similar work(Ber., 1912, 45, 3678), and consisted of a mixture of two parts of sulphuric acid(D 1’84) and one part of a 7 per cent. fumiag sulphuric acid.C = 84-94 ; H = 14.57 per cent. (C + H = 99.51).This liquid had an extremely unpleasant, garlic-like odour.The first fraction had a specific gravity 0*691:,5.C = 84.46 ; H = 15-60 (C + H = 100-06).L 148 JONES AND WHEELER: THE COMPOSITION OF COAL.C,Hlo n-pentane (b. p. 36') requires C=83.3 ; H= 16.7 per cent.C,H,, n-hexane (b. p. 6 9 O ) ,, C=S3*7; IS= 16.3 ,,C,H,, n heptane (b.p, 98') ,, C=S4.0 ; H = 16.0 ,,C,H,, n-ocbine (b. p. l Z S o ) ,, c1= S4.2 ; €1 = 15.8 ,,C,H2, naphthenes require C = 85.7 ; H = 14.3 per cent.The second fraction, boiling between looo and 135O, was analysed.Here again it is clear that a mixture of naphtAhenes and paraffins0.1353 gave 0.4234 CO, and 0.1846 H,O.was present:C,H,, n-nonaneC,H,, n-octane (b. p. 1 2 6 O ) ,, C = S 4 - 3 ; H=15.8 ,,C,H,, 9%-heptane (b. p. 98O) ,, C = 81 0 ; 11 = 16.0 ,,C = 85-34 ; H = 15-16.(h. p. 150') requires C = S 4 . 4 ; I { = 15% per cent.C,H,, naphthenes require C = 85.7 ; H = 14.3 per cent.We have not as yet attempted any direct examination of theolefines contained in the crude oil, nor of those found in thehigher boiling oils described later.I n the present instance thehydrogen content of the olefines, as calculated indirectly from theanalyses of the oil before and after washing with sulphuric acid,was about 13.3 per cent.; that is, cansiderably lower than that ofthe mono-olefines, C,H2, (I3 = 14.3 per cent.). Whilst some open-chain (mono-olefines) were very probably present, it would seemthat there were also unsaturated compounds cyclic in structure.Oils of Higher Boiling Point.-The oils now examined wereobtained from the two cooled condensers and from the pump con-nexions. To these was added that portion of the tar collected inthe first receiver that distilled below 200°/30 mm.These oils were washed with dilute sodium hydroxide solutionto remove phenols, and with dilute sulphuric acid to remove bases.The basic compounds recovered were too small in quantity to beexamined.After removal of phenolic and basic compounds, the oils werewashed with a solution of sodium carbonate and with water, and,after having been dried over anhydrous magnesium sulphate, weredistilled.The composition and specific gravity of successivefractions of the neutral oils are given in the table below :Fraction. I. 11. 111. 1v. -The phenols are described briefly later.Boiling point ................ 160--2UO" 200-2.50" 250-275" 275 --300"Cnrlwl ......................... 86-28 87.88 88.13 88 '24Hytlrogcn .................. 13.1 1 11 %5 11.09 10-36C - F H ........................ 99.39 99 53 99.22 58 60Sl~eciBc gravity, 15/15 ..,......0.804 0.556 0.890 0'928These neutral oils thus consisted almost entirely of hydrocarbons.They were colonrless, and had a disagreeable, penetrating odour.They rapidly decolorised potassium permanganate solution and JONES AND WHEELER: THE COMPOSITION OF COAL. 149solution of bromine in clilorof orin, and they also readily addedon the elements of hydrogen chloride. On boiling with anhydrouszinc chloride they polymerised but slowly, nor did they absorboxygen with any rapidity.The oils were now washed with concentrated sulphuric acid(U 1-84), whereby unsaturated etliyleiiic compounds were removed ;and afterwards, for the removal of aromatic compounds, eitherwith weak fuming sulphuric acid or with nitration mixture.It was found that washing with concentrated sulphuric acidgenerally reduced the volume of t h t oils by about one-half, andthat there was a concomitant lowering of their specific gravity.The oils obtained after further exhaustive washing with weakfuming sulphuric acid and with nitration mixture were freed fromacid, dried over sodium, and distilled from sodium in a vacuum,care being taken to avoid their decomposition.The refined oilswere clear liquids with a pleasant odour resembling that of refinedpetroleum. On analysis it was found that the carbon percentageswere too high for the paraffins of corresponding ranges of boiling.They approximated from a slightly lower figure up to 85-7, whichis the value for the percentage of carbon in the naphthenes, C,H2,.The same results were obtained with 211 the oils after exhaustivewashing with weak fuming sulphuric acid and nitration mixture.It is noticeablethat the fractions of lower boiling point appear t o consist ofmixtures of paraffins and naphthefies, whilst the fractions of higherboiling point must either consist of naphthenes only, or, what isless likely, of mixtures of paraffins with hydrocarbons richer incarbon than the polymethylenes :Fraction. I.11. 111. I v. v.Boiling point .........Specihc gravityl5/15 - 0 7841 0.7844 - -Cnrbon .................. 8 5 - 4 i 85-42 85.26 85.56 85-82Hydrogen ............ 14.46 14.73 14'56 14'65 14-36C + H ............... 99-93 100.15 99-82 100-21 100.18The analyses are set forth in the table below.150-190" 190-260" 200-250" 225-260" 160-260°/30mm.CloH22 (b.p. 173O) has D 0.747, and requires C=84.5; H=15*5.C1,H,, (b. p. 330O) ,, D 0.777 ,, C = 85.09 ; €I = 14.91.Naphthenes, require C = 85.7 ; H = 14.3 per cent.It will thus be seen that whilst the carbon-content of any mixtureof paraffins that may conceivably be present would fall below 85.0per cent., that of the refined oils is invariably higher and approxi-mates t o 85.7, the value for the naphthenes.Phenols.-The washings from the oils with sodium hydroxide werefirst treaked with ether t o remove dissolved oil, and afterwardsacidified with dilute sulphuric acid and extracted with ether. Theethereal solution was dried with anhydrous magnesium sulphate150 JONES AND WHEELER: THE COMPOSITION OF COAL.and filtered, and the ether evaporated.The phenols remainingwere distilled in a vacuum. A fraction collected between looo and145O/30 mm. had a specific gravity of 1.0372, and on analysis gavethe following results, which show it t o consist essentially of amixture of cresols and xylenols:0.2464 gave 0.7078 CO, and 0.1758 H,O. C=78*34; H=7*93.Cresols (C,H,O) require C = 77-77 ; H =I 7-40 per cent.We were unable to determine whether phefiol itself was present.Aromatic Compo liiids.-If the neutral oils boiling between 200°and .300° are heated and their vapours aspirated through ailaqueous solution of picric acid, crystals of picrates are deposited,those from the fractions of lower boiling point being yellow, andthose from the fractions of higher boiling point dark red.A quantity of apparently pure picrate, but insufficient foridentification, was obtained from a fraction of the oils boilingbetween 250° and 300O.This picrate was dark red, and meltedsharply a t 1 1 8 O .On treating that part of the neutral oils boiling between 170°and 300° with picric acid and warming, the liquid became darkred, but no precipitate was obtained on cooling. On the additionof pentane, picric acid mixed with a picrate was precipitated. Thisprecipitate war decomposed in a current of steam, the distillatecollected being a viscid oil of specific gravity 1%019::, and having anodour resembling that of crude naphthalene. 0.1905 gave 0.6420CO, and 0.1326 H,O.Dimethylnaphthalene (CI2Hl2) requires C = 92.31 and H = 7.69per cent.The values obtained from the combustion analyses, taken inconjunction with the physical characteristics of the oil and itspicrates, which are stable in the presence of water, point to theoils being homologues of naphthalene.Naphthalene itself couldnot be detected.Solid Parafii?s.-Of the tax contained in the first condenser, thatportion boiling abave 200°/30 mm. was distilled in a vacuum. Thedistillate obtained between 200° and 260°/ 30 mm. was analysed :0.2032 gave 0.6370 CO, and 0.1741 H,O. C=85*49; H=9.52(C + H = 95*01), showing that compounds other than hydrocarbonswere present.The portion of the oil that boiled a t 200-240° in a vacuumwas a pale yellow liquid, whilst that boiling a t 260° cooled to ared-coloured paste. The fraction boiling between 220° and 240°in a vacuum partly solidified on cooling, needle-shaped crystalsbeing deposited. Those crystals were recovered by removing anyXylenols (C8Hl,0) ,, C=78*67; H=8.20 ,,C = 91-91 ; H = 7.73 (C +- H = 99.64)INTERACTION OF GLYCEROL AND OXALIC ACID. 151resinous matter by charring with concent rated sulphuric acid andextracting with light petroleum. On t ctrystallising twice fromacetone, pure crystals were obtained, which melted between 52-5Oand 54O. On analysis: 0.1554 gave 0-486.5 CO, and 0.2097 H,O.C = 85-38 ; H = 14.99. A molecular-we~gl~t determination gave avalue 373.7, that is to say, intermediak between the values requiredfor C,,H, and C2J€,,.Searchwas made in the fractions of higher boiliug point for anthraceneby oxidation with chromic acid, but no trace of anthraquinone wasobtained. There is no reason t o believe that aiithracene or othersolid aromatic compounds were present.Pitch-The residue remaining from the coal distillate when thelatter was distilled in a vacuum up to 260° consisted of a soft,stringy, shining pitch of specific gravity 1.128;;. This pitch wasentirely soluble in chloroform, aad was free from carbon. 0.2101gave 0.6782 CO, and 0.1487 H,O. C=88-04; H=T.87 (C+H=95.91). This pitch is being further examined.Committee of the Home Office for permission to publish thispreliminary account of their wcrk,No other solid compound was found in the distillate.The thanks of the authors are due t o the Explosions in 1 11. 1nesESKMEALS.CUMBERLARD