ORGANIC CHEMISTRY-HOMOCY CLIC DIVISION.IN the enormous and continually increasing output of investigationsin organic chemistry there is considerable difficulty in recognising andindicating the general lines along which progress is taking place. Alarge number of the investigations carried out in any given year areisolated in character, having as their object the solution of some im-mediate problem, without reference, expressed or implied, to anygeneral theoretical scheme. This fact makes i t impossible to give asurvey of such a field as t h a t of the homocyclic compounds which shallembrace all or even the gre-iter part of the work abstracted during thepreceding year. The most t h a t is possible is to select such researchesas lend themselves to grouping about a few main lines of investigation,especially such as have a wider bearing than on the special problemsunder examination.Although there may be differences of opinion ast o the relative importance of the various tendencies to be observed inrecent work, yet it would seem that certain questions stand out withsufficient prominence to justify their selection.The problems of the relations between colour and chemical constitu-tion, of the nature of isomeric change in aromatic derivatives, of thestructure of the benzene nucleus, and of the conditions of formation ofrings, are not only of great intrinsic importance, but their solution isnecessary before the f undamentlzl principles of organic chemistry canbe ragarded as satisfactorily established. Researches into the constitu-tion and synthesis of certain groups such as the terpenes also possessa more than specialinterest from the remarkable relationships exhibitedby some of their members, and from the ingenuity of the methods towhich their experimental stiidy has given rise.The diazo-compoundsand hydrazones, again, apart from their many practical applications,derive much of their interest from the difficulty of finding formulaecapable of expressing the whole of their chemical behaviour, a difficultywhich has led to prolonged controversy. The same may be said oftriphenylmethyl, the investigation of which has occupied many workersfor a considerable time without entire agreement as to its constitutionhaving yet been reached. It may be said that the insufficiency ofpurely statical formulae is being felt more and more acutely, and suc110 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.conceptions as those of taut omerism and dynamic isomerism are constantlyassuming greater importance, That this theoretical change must ulti-mately lead to a more intimate contact between organic chemistry andphysics can hardly be doubted, but a t present the points of contact ofthe two sciences are comparatively few and scattered.In spite of the increasing use which is made of measurements of suchphysical constants as the refractive index and magnetic rotation asguides in the determination of structure, the application of physico-chemical methods, especially those of chemical statics and dynamics, toorganic problems is still very restricted.Isolated examples during thepast year have been the comparative measurement of the velocity ofesterification of substituted benzoic acids and an attempt to decidebetween the possible intermediate products in the oxidation of naph-thalene t o phthalonic acid by potassium permanganate,2 and of P-naph-thaquinone to phthalic acid by potassium dichromate.3 The methodadopted in each case was to measure the velocity of oxidation, underthe given conditions, of each of the compounds which might conceiv-ably appear as an intermediate stage in the reaction. Determinationsof the velocityof hydrolysis of esters, &c., when the object is not thestudy of structure, belong rather to the department of physicalchemistry.There has been no revolutionary introduction of new reagents t orecord during the past year, and such improvements of detail as needt o be mentioned may be treated under their respective sections, Themultitudinous applications of Grignard’s reaction have been usefullysummarised by A.McKeuzie in a report laid before the British Associa-tion at its Leicester meeting. Among recent tipplications, the action ofGrignard’s reagent on oximes has been shown 4 to be the replacementof hydroxyl by alkyl, followed by the formation of an additive com-pound, which is then hydrolysed :R*CH:N*OH + 2RMgX -+ g,>CH*N<EgX + E,>CHNHR’.The same product is obtained from the O-ethers of the oxime.Xtructure of Benzene.There has been a marked revival of interest in the problem of theconstitution of the benzene nucleus, and the relative advantages of thecentric and Kekuld formulte.On the one hand, the properties of ozoneA. Kailan, Abstr., 1907, i, 849 ; ii, 168, 242, 243, 853,R. A. Daly, ibid., 1907 i, 407,31. C. Boswell, ibid., 407.M. Uusch and R, Hobein, ibid., 535ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 111as a reagent for the detection of ethylene linkings have been utilised.Unfortunately, the results with benzene itself are as yet inconclusive.The existence of Harries’ triozonide of benzene which was cited insupport of the Kekuld formula, is denied by Molinari,G who finds thatbenzenoid compounds do not react with ozone, whilst quinonoid deriva-tives manifest the existence of ethylene linkinga in the nucleus by theformation of ozonides.Phenols react as if quinonoid in structure,which is in accordance with many other facts indicating at least thetransient existence of the tautomeric form. Theoretical arguments infavour of the centric formula for benzene have also been adduced 7 fromthe reactions depending on the polymerisati n of aliphatic compounds.On the other hand, a body of new facts has recently been broughtforward in support of Kekuld’s formula. The old objection to it, thati t indicated the existence of isomeric ortho-di-derivatives, has nowbeen converted into an argument in its favour. It has been known forsome time in technical practice that o-nitrotoluene exists in two distinctcrystalline modifications, having different melting points.It is nowfound * that these are not merely polymorphic crystalline conditions ofthe same substance, but that they also differ in the liquid state, havingdifferent depression constants, and depositing their correspondingcrystals when cooled after long storage in closed vessels. There seems,then, to be some justification for regarding them as true isomerides.Similar modifications have been observed in o-chloro- and bromo-toluene,o-toluidine, o-chlorophenol, o-chloroaniline, and 2 : 4-dinitrophenol.Knoevenagel’s hypothesis of motoisomerism would call for the exist-ence of isomerides in the para and meta series also, and would even in-dicate the possibility of two isomeric mono-derivatives when the sub-stituting group is asymmetric, a suggestion which the author considersto derive support from the fact that freshly distilled nitrobenzene hasa different viscosity from that which has been kept, Although thefacts may be susceptible of more than one explanation, an interestingfield of investigation is opened up by them.There have also been several theoretical attempts 10 to form a con-ception of the benzene nucleus capable of accounting for the orienta-tion of substituting groups, but the views contained in these papers donot lend themselves to summarisation in a report.I n the absenceof any clear physical conception of the nature of valency, there isalways a danger that theories of the valency of Ghe carbon atom,devised ad hoe for the purpose of explaining organic reactions, mayAnn.Report, 1906, 120.7 It. Vidal, ibid., 1020.I. von Ostromisslensky, ibid., 120, 696 ; E. Knoevenagel, ibicl., 202.Abstr., 1903, i, 785.E. Molinari, Abstr., 1907, i, 1039.lo J. Obermiller, ibid., 1907, i, 200 ; B. Fliirscheim, ibicl., 835112 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.prove on examination to be merely rest,atements in other language ofthe problem to be solved.Colour and Constitution.The Annual Reports for the last three years have made referenceto the growing interest in the question of the relation of colour tostructure. The desire to contribute to the solution of this problemhas been the motive of a large number of recent researches, andalthough the relation is still very imperfectly understood, yet withintho limits of certain groups it is possible to establish rules for theappearance or non-appearance of colour. It is now generally recog-nised that a more precise meaning must be given to the idea of colourthan has often been the case.The production of physiological colour,due to the occurrence of absorption in the visible spectrum, is more orless an accidental circumstance. Absorption bands may occur in theultra-violet of equal importance with those in the visible spectrum,I n some cases a change in the frequency of the absorbed rays maycause a band to move from t h e ultra-violet into the visible regionwithout any change of form. A colourless substance may thereforebe converted into a coloured one without any real change in constitu-tion having taken place, the alteration in the molecule being only ofsuch a nature as to cause a certain retardation of those oscillationswithin it which give rise to the absorption.A study of the colour of a substance thus involves the examina-tion of its entire visible and ultra-violet spectrurn,ll and further, asHartley and his successors have shown, of the change of absorptionwith the concentration of the absorbing substance, the results beingbest expressed in the form of curves.The quantitative study of theabsorbing power of a substance, expressed in this manner, makes anexact comparison of different derivatives possible, and relationshipsace made evident which would escape notice i f the examination wereconfined to visual observations of colour.It is therefore not sur-prising to find an increasing use of spectroscopic methods in theliterature of organic chemistry.Several different explanations of the origin of colour are offered atthe present time, bat these explanations are not mutually exclusive,and in the case of certain compounds two or more of the rivalhypotheses meet on common ground. From one point of view, colouris ascribed t o the presence of a quinonoid structure in the molecule(Armstrong, Gomberg, R. Meyer, and others), from another, to theentrance of certain groups, the auxochromes, into a sensitive molecule,l1 The absorption bands in the infra-red appear to have a different origin, andmay be neglected for the purpose of the present discussionORGANIC CHEMISTRY-HOMOCYCLIC DIVISION.113the chromophore (Kauffmann).12 When a coloured compound is pro-duced from a colourless one, as for instance by solution or by salt-formation, this is explained by a particular form of ionisation (Baeyer)or by a molecular arrangement giving rise to a quinonoid or otherlinking characteristic of colour (Hantzsch). The more definitelyphysical theory of Baly regards banded absorption as produced byoscillatory changes of linking within the molecule, due to the con-jugation of groups possessing residual affinity. Whilst, therefore, onthis view the quinonoid structure is only one phase in an essentiallydynamical process, the auxochrome theory is not rejected, but anattempt is made to provide a physical basis for the properties of theauxochromes.Nitroquinol dimethyl ether, the colour of which was disputed,13 hasnow been shown to be yellow when pure.14 Its solution in lightpetroleum is colourless, its other solutions being more and more yellowwith increasing dielectric capacity of the solvent.15 Since both theyellow colour and the molecular weight increase with the concentra-tion, the yellow solutions are regarded as additive compounds.Thecolour is much less intense than that of nitrophenol salts, and is notconsidered by Hantzsch to demand the assumption of molecularrearrangement.A remarkable series of coloured alkali salts of nitro-compounds hasbeen investigated by Hantzsch and his co-workers. Under differentconditions of temperature i t is possible to obtain both red and yellowsalts of the nitropheho1s.l6 The product is often orange, consistingof mixed crystals of the two i+omerides, which may be separated bycrystallisation in absence of water.The red and yellow potassiumsalts of 2 : 4 : 6-tribromo-3 : 5-dinitrophenol both undergo gradualchange in solution until a condition of equilibrium is reached; inother cases one of the salts is only stable at -75’. Structuralisomerism is discussed and rejected, the author preferring to representt.h0 two classes of salts as syn- and anti-stereoisomerides, as :C,H,*O \I NO,MRed.C,H,*O ‘ I \MO,NYellow.Although mononitro-compounds form only colouriess salts, dinitro-compounds, such as phenyldinitromethane, yield both red and yellowl2 For a summary of this question, see H.Kauffmann, “ Die Auxochrome,”Bee also J. Schmidt, (‘ Chinone und Chinoide Ahrens’ Sammlun,g, 1907, 12, 1-3.Verbindungen,” ibid., 1906, 11, 10-11.Ann. Report, 1906, 147.l5 A. Hantzsch, ibid., 513.REP.-VOL. IV. Il4 H. Kauffmann, Abstr., 1907, i, 127.A. Hantzsch, ibid., 207114 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.salts.17 Both modifications retain their colour when dehydrated, butyield identical equilibrium mixtures in solution, which are found to beunimolecular. The presence of the second nitro-group is necessary,since bromonitromethane and phenylcyanonitromethane,CN*CHPh*NO,,form only colourless salts. Evidence is brought forward to show thattrue aci-salts, R*C<g%M, are colourless.When only one of thenitro-groups is in the methane residue, the other being in one of thephenyl groups, as in the nitrophenylnitromethanes, two colourless andfour coloured isomerides are possible. The colourless salts, of whichonly the mercuric salt of m-nitrophenylnitromethane has yet beenisolated, would have the formula :NO,*CGH, *CH:NO,M and N0,M:C6H4: CH*NO,.The compiete series of four chromo-salts, yellow, red, green, andviolet, has been isolated in the case of the potassium salts of p-nitro-phenylnitromethane. In the formation of the coloured derivatives,both nitro-groups must take part, and the following structures areregarded as the most probable, the dinitroparaffins being selected asthe simplest example :0:Bf.o O:N?OR*C==N*OJI R* C*NO,M(111.)I 0 4 II I NO2 1R*C:NO,MColoured.(1.) (11.1Colourless, representedby the ether.Of these, both (11) and (111) may exist in two stereoisomeric modifica-tions. Isomeric chromo-salts have also been obtained from nitro-barbituric acid and similar compounds.l* Since, however, the halogen-phenols, to which such structural formulae cannot be applied, yieldboth coloured and colourless silver salts,lg the isomerism in questionevidently demands further investigation.It will be observed that the formulae which Hantzsch now employs torepresent the coloured derivatives are not strictly quinonold, butcorrespond rather with the 4' peroxide " formula for quinones.Thushe assigns to the coloured salts of hydroxybenzaldehyde and methyldihp droxyterephthalate the respective structures :l7 A.Hantzsch, dbstr., 1907, i, 500.l9 H. A. Torrey and W. H. Hunter, ibid., 1030 ; A. Hantzsch, ibid., 1908, i, 17.1s A. Hantzsch, ibid., 555ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 115The fact that these compounds are fluorescent is opposed to aquinonoid structure and indicates the presence of an unalteredbenzene nucleus.20 In connexion with this point it should bementioned that the tautomeric quinonoid constitution,has been assigned to salicylic acid in order tofrom sodium phenoxide,21 and its formationNHPhoN:C6H4<Co2H.22 Hexplain its productionof a phenylhydrazone,By the action of ethyl iodide on the dry silver salts a t lowtemperatures, coloured quinonoid derivatives of phenolphthalein andtetrabromophenolphthalein have been obtained.28 Hitherto thequinonoid esters have only been prepared by acid alkylation.Theyreadily undergo isomeric change to the colourless Iactone ethers :76"4*s0 o----0\/\\A f I I I:()O E t V \/Phenolphthalein forms a red-+ N\/\OEd,) ()Ethydrochloride at - 30°, which decom-poses as the temperature rises, but stable red stannichlorides arereadily obtained from phenolphthalein (I), the quinonoid ester (11),and the lnctoid dimethyl ether (111) : 24Similar formulae are proposed by Green and King, who representthe coloured alkali salts of the ester thus :25C,H4*ONaOH H>o:c6H4:c<c 6 4 H .CO,Me'It is, however, argued by Baly26 that such oxonium compounds2o H.Kauffmann, Abstr., 1907, ii, 214, 215 ; A. Hantzsch, ibid., 418,21 K. Brunner, ibid., i, 319.22 H. Schrotter and J. Flooh, ibid., 929.23 R. Meyer and K. Marx, Abstr., 1907, i, 421, 932 ; H. Meyer, ibid., 625.24 K. H. Meyer and A. Hantzsch, ibid., 932.25 8. G. Green and P. King, ibid., 933 ; Proc., 1907, 23, 228.would be colourless.According to H. Meyer, however(Monatsh., 1907, 28, 1381), the compound is the phenylhydrazide.PYOC., 1907, 23, 229.1 116 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.As in recent years, the battle of the colour theories has been largelyfought on the field of the triphenylmethane compounds. Whilst Baeyerconsiders that yuinonoid structures are not the specific cause of colour,but merely facilitate its appearance, since they allow weaker auxo-chromes to exert their influence, Gomberg now maintains the quinonoidconstitution of all the coloured derivatives, modifying it to the extentof considering the grouping as the only essential part of thequinone ring, since the linking of the upper carbon atom to a bivalentelement is absent both from these compounds and from the fulvenes.Gomberg finds 27 thak solutions of triphenylmethyl chloride and itsderivatives in benzene react with silver sulphate to form colouredsulphates.When para-halogen is present, one atom of this is alsoremoved, which he considers to be evidence of the quinonoid structure.According to Baeyer,2* however, the slow removal of ring-substitutedhalogen is due to a destruction of the molecule, as is shown by the factthat an odour of quinone is observed, whereas the diphenylquino-methane, O:C:,H,:CPh,, which should be formed on Gomberg’ohypothesis, is without odour.The point was tested by a comparisonof the ferrichlorides of tri-p-chlorotriphenylmethyl bromide and tri-p-bromotriphenylmethyl chloride :II I1II(C,H,Cl!,C:C,H,<~~,FeCl, and (C,H,Br),C:C,H,<GCf,FeCl,.The quinonoid portions of the two molecules are identical, and shouldyield the same halogen with water. The former, however, loses onlybromine, and the latter only chlorine. Similar results were obtainedwith the corresponding stannichlorides by Tschit~chibabin,~~ who alsofound that Gomberg’s coloured solutions of the chlorides in liquidsulphur dioxide, heated at 50°, and then decomposed with alkali, gaveonly chloride without any trace of bromide, whilst the formulawould require the formation of both chloride and bromide.Thequinonoid thoory is, however, supported by other investigttt~rs,~~ andformulz of this type have been applied to the coloured salts of thep-aminocinnamylidene derivatives of acetic and malonic acids,31 as :NH,c~:/=\:cH*cH: CH~CH (co,~),. \=/~7 M. Gomberg, Abstr., 1907, i, 504.A. E. Tschitschibabin, ibid., 1022.30 F. Kehrmann and F. Wentzel, ibid., -601.y1 H. Fecht, ibid., 926.A. von Baeyer, ibia., 691ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 117Whilst Gomberg attributes an exactly similar constitution to the saltsof pararossniline with one equivalent of acid :/=\(NH;C,H,),C:/ \: NH,CI, \=/Baker, from a study of their absorption spectra,32 is led t o representthem as carbonium compounds :An extended study of the hydroxy- and amino-derivatives oftriphenylmethane and their salts has led Baeyer (Zoc.cit.) to the con-clusion that an oscillation between two forms, similar to that describedby Baly as isorropesis, is the origin of the absorption band. If, forinstance, two of the benzene rings, a and b, in triphenylmethane con-tain each a hydroxyl group, then either a or b may become quinonoid,and one form may pass into the other without change of properties.I n the sodium salt of benzaurin, then, there would be a continuousoscillation between the forms :I IPh P hI n Doebner's violet there would be a precisely similar oscillation ofthe chlorine between the two NH, groups, with corresponding changeof linking.The fact that crystal-violet behaves as the salt of a strongrnonoacid base,33 whilst the carbinol is only a weak triacid base, isintelligible if the three amino-groups have the same function and aresaturated alternately.The difficulty of finding any statical arrangement of bonds whichwill invariably produce colour is further illustrated by some investiga-tions of ~ x i r n e s . ~ ~ Phenanthraquinonedioxime and its sodium salt arestrongly coloured, but its dime€hyl ether, diacyl derivatives, andanhydride are all colourless, although the quinonoid configuration isC Hstill intact. Fluorenoneoxime, I '>C:NOH, and its sodium saltC,Hd - -are coloured, the acyl derivatives are paler, but still yellow, but themethyl ether is actually darker in colour.A comparison of thecorresponding derivatives of o-benzoquinonedioxime and P-naphthn-yz F. Baker, Trans., 1907, 91, 1490.$4 J. Schmidt and J. So11, ibid., 1907,, i, 630, 1054.33 A. Hantzsch, Abstr., 1900, i, 365118 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.quinonedioxime35 has shown that the colour is greatly dependent onthe solvent, and the following types of formulae are suggested :Colourless. Faintly coloured. Strongly coloured.The influence of substituents, especially hydroxyl, on the isorropesisof quinones has been studied spectroscopically in the case of naphth-acenequinone.36Benzoic acid has an absorption band in the ultra-violet, which isalso present, though considerably narrower, in the spectra of potassiumand silver benzoates.Phthalic acid bas a wider band, bophthalicacid has a shallow band in the same position, terephthalic acid doesnot show a band.37Triphen y lrnethy l.The controversy, discussed in previous Reports,38 as to the con-stitution of triphenylmethyl still continues. The view that the solidcompound is to be regarded as hexaphenylethane, has been criticisedas being inconsistent with the stability of tetra- and penta-phenyl-ethane. Pentaphenylethane, however, although stable at the ordinarytemperature, becomes unstable when its solution is heated, splitting upinto diphenylmethyl and triphenylmethyl groups, which are thenreadily oxidised.39 The result found by H a n t z ~ c h , ~ ~ who failed toobtain hexanitroethane from iodopicrin, favours the view that theaccumulation of negative groups in the ethane molecule tends towardsinstability, The stable hexaphenylethane, supposed to have beenobtained in an impure state from magnesium triphenylmethyl chlorideand triphenylchloromethane,41 is now considered 42 to be impure tetra-phenylethane.Solid triphenylmethyl is colourless, remaining unchanged forseveral months in dry air.43 Its solutions are yellow, and hence adifferent constitution is required for the solid and the dissolved sub-stance.The evidence is now strongly in favour of the hexaphenyl-ethane formula for the solid compound, but no definite conclusion has35 A.Hantzsch and W. H. Glover, Abstr., 1907, i, 1055,36 E. C, C. Baly and W. B. Tuck, Trans., 1907, 91, 426.37 W. N. Hartley and E. P. Hedley, ibid., 314, 319.38 Anit. Xeport, 1904, 105 ; 1905, 117 ; 1906, 131.39 A. E. Tschitschibabin, Abstr., 1907, i, 204.40 Ibid., 1906, i, 617.41 J. Schmidlin, ibid., 1907, i, 27.42 Tschitschibabin, Zoc. cit.43 A. E. Tschitschibabin, ibid., 691ORGANIC CHEMISTRY-HOMOCY CLIC DIVISION. 119yet been reached as to its condition in solution. The existing rivalviews have been summarised by Tschits~hibabin.~~(1) 66Halochromy,” or colour due t o ionisation, may be assumed.Thus, in accordance with Baeyer’s notation, dissolved triphenylmethylwould be written CPb,-CPh,. This, however, really only restatesthe fact without providing any explanation.(3) Gomberg suggests that the dissolved substance has the formula CPh2:C6€K4<~p,d which partly dissociates into a quinonoid cationand a benzenoid anion.He does not, however, finally decide betweenthis hypothesis and the existence of the free triphenylmethyl radiclein solution, which is in some ways to be ~referred.~5(3) Quinonoid formula, such as those of HeintscheI and Jacobson,46may be written for the coloured modification. The final decisionbetween these views is not yet possible.Against the formulse of Gomberg and Jacobson, it has been urged 47that the compound (I) would be so unstable as at once to change intothe isomeride (II), and in support of this the behaviour of palkyl-Ph2CHH-N *CPh, \=/idenedihydrobenzenes 48 is quoted, the quinonoid configuration in (111)being so unstable as to pass spontaneously into the stable benzenoidform (IV), although in this case the wandering of so heavy a groupas .CHCI, is involved :(III.)The magnesium compound of triphenylmethyl chloride has beendescribed49 as existing in a yellow quinonoid and a colourless benz-enoid modification, but the experiments of Tschitschibabin 50do not confirm the existence of chemical isomerism in this case, a sthe behaviour of the two preparations is found t o be similar, thestatement that the yellow form does not yield triphenylacetic acidwith carbon dioxide being incorrect.44 J.pr. Chenz., 1907, [ii], 74, 340.48 K. Anwers, Abstr., 1907, i, 399.45 M. Gomberg, Abstr., 1907, i, 514.47 K.Auwers, Ber., 1907, 40, 2159.4g J. Schmidlin, ibid., 1907, i, 26, 601.Ann, lieport, 1906, 132.Abstr., 1907, i, 1022120 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.D iazo-compounds.The usual representation of diazonium compounds by Blomstrand’sformula, and of the diazotates, diazo-cyanides, &c., as s tereochemicallyisomeric azo-compounds, has long been recognised as presenting diffi-culties. The Blomstrand formula (I) especially fails to indicate thegreat readiness with which the diazonium salts lose their nitrogen.Since a nitrogen atom directly attached to the benzene nucleus, as inaniline, is generally very firmly retained, it might be expected thatdiazonium salts would only lose one atom of nitrogen instead of bothas is actually the case.The nitrogen atom of aniline is, however,easily removed by oxidation, the product being p-benzoquinone. Sincethis reaction probably proceeds through quinoneimide as an inter-mediate product, there is some justification for considering thataniline reacts in this case in the tautomeric quinonoid form (11) :(1.1 (11.)These facts have suggested the formulation of diazonium salts asquinonoid derivatives,51 diazobenzene chloride having the formulaH\/-\:N*Cl. I\-./ ” --NI n the elimination of nitrogen, as in the action of water, rupturenaturally tends t o occur at the double linking. A considerableamount of evidence is adduced in favour of this formula.Thus the production of a nitrosodiazonium compound by the actionof nitrogen peroxide on thymoquin~nedioxime,~~ is best accounted forby considering the quinonoid configuration as remaining intact :N-OHNO\*.: \-,N* OH NO,*g:N-IFurther, whilst ar-tetr~hydronaphthylamine and 5-aminoquinolineare diazotisable, ac-tetrahydronaph thylamine and 4-aminoquinoline, inwhich the quinonoid formation is no longer possible, do not yield51 J.C. Gain, Tram., 1907, 91, 1049.52 R. Oliveri-Tortorici, Abstr., 1900, i, 653ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 121diazo-salts. The non-existence of aliphatic diazonium salts is alsoaccounted for.Perhaps a greater difficulty is presented by the metallic diazotates(diazo-oxides) and diazo-cyanides. Caio represents the normal orsyn-compounds as quinonoid, but having the hydroxyl, &c., attached tothe tervalent nitrogen atom (I) and the iso- or anti-compounds ashaving the ordinary azo-constitution (11) :This formulation has been but little discussed at present, but it ispointed out that it accords with the different behaviour of the twodiazotates towards phenols, the compound (11) having less tendency tocouple from its having already the azo-constitution, whereas onIIantzsch’s hypothesis the syn-compound might be expected to coupleless readily than the anti- on account of the pxjsibility of sterichindrance :The small number of coloured diazonium compounds hitherto knownhas been increased by the preparation of a remarkably stable seriesof yellow diazonium salts from benzoyl- 1 : 4-naphthylenediamine.53Their aqueous solutions are also yellow, thus differing from the simplerdiazonium iodides and thiocyanates, which yield colourless solutions,and mere regarded by Hantzsch as owing their colour in the solidstate to the presence of the isomeric syn-diazo-compound in solidsolution.The new salts form neutral solutions, couple readily withalkaline P-naphthol, and yield the nitrosoamine with weak alkalis,and hence can only be regarded as true diazonium compounds. Theauthors adopt Cain’s constitution, representing the salts as eitherpara- or ortho-quinonoid :An alternative quinonoid structure might be obtained by assuming63 G. T. Morgan and W. 0. Wootton, Tram., 1907, 91, 1311122 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.the presence of a labile hydrogen atom attached to the nitrogen of thebenzoylamino-group 54 and wandering to the diazonium nitrogen :In the examination of the diazonium compounds from benzene-sulphonylbenzidine and as-benzenes~lphonylmethylbenzidine,~~ it wasfound, however, that coloured diazonium halides were obtained fromboth compounds, although in the second case the labile hydrogen atomis replaced by methyl.The authors therefore adopt the formula :The nature of the transformation of s-trihalogen-substituteddiazonium salts of weak acids into quinonediazides 56 has been morefully elucidated by a study of the mixed chlorobromo-derivatives.57No difference is to be found in the behaviour of chlorine and bromine,so that in the conversion of the symmetrical 3-chloro-5-bromo-p-toluidine, for instance, into quinonediazides, chlorine and bromineare eliminated in equal proportions, the two ortho-positions beingequivalent.When halogen is also present in the para-position, thecomparison of different chlorobromo-derivatives shows that elimina-tion takes place from the two ortho-positions with equal frequency, andfrom the para-position with half that frequency, so that the ratioThe authors do not incline t o accept Gain's formulation of thediazo-compounds. They assume a benzenoid structure for the soliddiazonium salts, but consider that an aqueous solution contains o- andp-quinonoid phases in equilibrium, under such conditions that eachortho-phase occurs twice as often as the para-phase. No explanationof the constancy of this ratio is offered. The qiiinonoid compoundspresent in solution are represented as carbonium derivatives, so thatthe transformation in a particular case may be shown thus, only oneol the two ortho-configurations being given :o : o ' : p = 2 : 2 : 1 .54 J.T. Hewitt, Proc., 1907, 23, 181.55 G. T. Morgan and J. M. Hird, Trans., 1907, 91, 1505.56 Ann. Beport, 1905, 107.57 K. J. P. Orton and W. W. Reed, Trans., 1907, 91, 1554OEtGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 123",*OHBrN2\//\BrfiBrBr OH-+--3Diazonium compounds of this type, although yielding almostexclusively quinonediazides when heated with dilute acids, readilyundergo the normal conversion into phenols under the influence ofsunlight.58 This is shown to be entirely a diazonium reaction, notdependent on the intermediate formation of a syn-diazohydroxide, asit takes place under conditions unfavourable to hydrolysis, as in95 per cent.sulphuric acid solution. Solutions of diazotates are verystable when exposed to light. When alcohols or acetic acid areemployed as the solvent instead of water, the corresponding ethers orphenyl acetates are obtained, the phenetole formation, for instance,being favoured by the exposure of a suspension of the diazonium saltin alcohol, free from acid, to sunlight.Among other investigations in the department of diazo-compoundsmay be mentioned a study 59 of the red oil found by Griess, togetherwith azobenzone, in the product of the action of potassium ferro-cyanide on diazobenzene chloride.Since it yields triphenylhydrazine,NPh,*NHPh, on reduction, and undergoes the semidine transformationto form aminotriphenylamine, NH,*C6H,*NPh,, it is regarded ashaving the constitution C H < NPh I * NPh' I n the case of the diazotoluenesalts, the reaction appears to be more complicated.By the diazotisation of aniline with 0.5 mol. of sodium nitrite indilute acetic acid solution, an orange isomeride of diazoaminobenzenebas been obtained 60 the constitution of which is represented by theformula, NEPh<rPh. Its acetic acid solution probably contains a Ncompound, NH,Ph<h. NPh OAc, which couples with P-naphthol in thecold, yielding a red colouring matter, whilst diazoaminobenzene onlycouples on boiling, and gives a yellow product.Cuprous chloride or58 K, J. P. Orton, J. E. Coates, and F. Bnrdett, Trans., 1907, 91, 35.69 A. Ehrenpreis, Abstr., 1907, i, 453. 6o E. I. Orloff, ibid., 365124 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.molecular copper decomposes the compound at the ordinary temperatureinto phenol, aniline, and nitrogen. The same isomeride is obtained inpresence of other organic acids, but the stronger the acid the lesstendency is there t o form the new compound.The absence of isomerism in the mixed aromatic diazoamino-com-pounds has frequently been investigated. Since in the usual methodof preparation from diazo-salts and amines a considerable part is playedby the water and electrolytes present, an attempt has been made 61 t orealise the isomerism by the synthesis of unsymmetrical diazoamino-compounds from azoimides and magnesium organic derivatives.Theattempt failed, however, the same product being obtained from phenyl-azoimide and magnesium a-naphthyl bromide, and from a-naphthylazo-imide and magnesium phenyl bromide. Several pairs of compoundswere examined in the same manner.Phenylmethyltriazen 62 bas been regarded as tautomeric in character,reacting in the two forms C,H,=N:N*NHMe andC6H,*NH*N: NMe,the second formula being an indirect inference From its behaviour withacids. Its discoverer has now shown63 that it and its homologuesareonly correctly represented by the first formula, as the condensation withdiazobenzene salts gives rise to bisdiazoamino-compounds which arealways symmetrical, having the constitution R*N:N*NR’*N:N*R”.This is proved by the fact that the product from phenylmethyltriazenand p-diazotoluene chloride is identical, and not isomeric, with thatf rom p-tolylmethyltriazen and diazobenzene chloride.An interesting prepnration of certain aminoazo-compounds, whichdiffers from those usually employed in being additive instead of sub-stitutive, consists in the interaction of the p-diazoimides and activeamines.G4 Aniline does not react i n this way, but satisfactory resultsare obtained with the naphthylamines and certain m-diamines. Sincemonoal kylnaphthylamines reac t, only less readily, whilst dialkylnaphthyl-smines do not react, the authors assume that the addition occursthrough the intermediate assumption by the amine of a quinonoidconfiguration :61 0.Dimroth, M. Eble, and W. Gruhl, Abstr., 1907, i, 664.63 0. Dimroth, M. Eble, and W. Gruhl, Zoc. cit.64 G. T. Morgan and F. M. G. Micldethwait, Trans., 1907, 91, 1512.Ann. Report, 1905, 107ORGANIC CHEMISTRY-HOMOCYCIiIC DIVISION. 125N:SO,Ph H.NHA /\/\1 1 + 1 1 1 -\/ \/\/ .. NS02Phi HALN HHS0,Ph*HHydraxones and Hyd~ox?lazo-compounds.The results of a spectroscopic study of the hydroxyazo-compoundsand their derivatives 65 have led to the conclusion that whilst the para-compounds and their hydrochlorides have the azo-constitution, the o-quinonehydrazone formula must be assigned to the ortho-compoundsand their acyl derivatives. The ethers and hydrochlorides of the ortho-series, on the other hand, show absorption spectra indicating that theyhave the azo-structure.The spectrum of the hydrochlorides suggeststhat they are carbonium salts.66 The hydroxyl or ethoxyl group havingsome residual affinity, its presence in the benzene nucleus of an azo-compound modifies the spectrum of azobenzene considerably, but whenthe residual aanity of the substituting group is diminished by replac-ing the hydroxylic hydrogen by acetyl or benzoyl, the spectrumapproaches more nearly to that of azobenzene.Spectroscopic evidence also goes to show 137 that the phenylhydrazonesand osazones exist in neutral solutions in the form of hydrazones, withthe exception of phenylglosalosazone and dextrosazone, which appeart o be partly or entirely in the azo-form.But when sodium ethoxide isadded to the solutions, a part a t least of the substance is convertedinto the azo-compound. The influence of the presence of neighbouringhydroxyl groups, and of the conjugation of :C:N* linkings, was alsostudied.The same problem has also been attacked by a number of workersfrom the chemical side. In favour of the azo-constitution for thepara-compounds is the conversion of p-benzoquinonebenzoyIphenylhyd~-azone (I) into benzoxyazobenzene (11) by contact with potassium65 W. B. Tuck, Trmw., 1907, 91, 449. 66 F. Baker, Gid., 1490.E. C. C. Baly, W. B. Tuck, E. G . Marsden, and M. Gazdar, ibid., 157%126 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.hydroxide in the cold, a transformation which is also undergone by thesubstituted derivatives : 68O:/=\: N*NPh*COPh COPh* O/--\N: NPh.\=/ \-/(1.) (11.1There is less agreement as to the ortho-compounds. Whilst Tuckconsiders these to be hydrazones, the whole of the o-hydroxyazo-com-pounds with their ethers and esters are regarded by Auwers 69 as re-taining the azo-structure. According to this author, the acetyl deriv-ative of benzeneazo-p-cresol is an O-derivative. On reduction, it yieldsthe N-acetyl derivative of the hydrazo-compound, but it is not possibleto obtain the quinonehydrazone on gentle oxidation with ferric chlorideor mercuric chloride, as it at once reassumes the azo-configuration :Me MeThe product of the action of as-benzoylphenylhydrazine on P-naphtha-quinone was found also tVo be an azo-compound, being identical withthat obtained by benzoylating P-benzeneazo-a-naphthol. It undergoesreduction to the hydrazo-compound.Auwers thus considers theseO-benzoyl- and acetyl-derivatives to undergo spontaneously the samechange as that observed by Willstatter and Veraguth in the para-seriesunder the influence of potassium hydroxide. Auwers’ paper is describedas a preliminary note, so that a further experimental investigation ofthis question may be expected.A transformation which is in a sense the reverse of that ofWillstiitter and Veraguth, namely, the isomeric change of azo-com-pounds into hydrazones, occurs on heating tria~ylmethanee,~O thecoloured compounds of the constitution Ac,C*N:NPh passing into thecolourless isomerides Ac,:N*NAc*Ph.The transfer of the labile grouphere is from carbon to nitrogen. One of the acyl groups in the tri-acylmethanes is evidently very loosely combined, When acetyl andbenzoyl are both present, it is the acetyl group that becomes labile.The absorption spectra in the visible region of a number of nitratedp-hydroxyazo-compounds have been examined with the view of studyingtheir constitution.71 The p-nitroazophenols and their salts exhibitabsorptions of radically different type. The authors regard the salts as08 R. Willstatter and H. Veraguth, dbstr., 1907, i, 453.69 K. Auwers, ibid., 554.70 0. Dimroth and M. Hartmann, ibid., 1090.71 J. T. Hemitt and H. V. Mitchell, Trans., 1907, 91, 1251ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION.127being isonitro-derivatives, and they are thus represented as quinonoid.This is in accordance with the views of Baly and Tuck72 on thestructure of the p-nitrophenylhydrazones of aldehydes and ketones. Thequinonoid alkali salts have a purple or blue colour, the absorption beingshifted towards the red in comparison with that of the parent com-pound. The tabulated absorptions indicate that the oscillationfrequency is less the longer the chain of alternate double and singlelinkings in the molecule.As the formation of lakes is known to require the presence oftwo hydroxylic groups (one of which may be carboxyl) in the ortho-position relative to one another, this view of the constitution of thenitro-derivatives was tested by a comparison of the salts of two isomericcarboxylic acids derived from benzeneazo-a-naphthol.Both acids formblue potassium salts, but whereas the onecontaining OH and C0,H inthe ortho-position gives brown precipitates with salts of the heavymetals, thah in which NO2 and CO,H are in the ortho-position givesblue precipitates. The quinohoid structure is therefore intact in thelatter case, and the compounds are related thus :/-\ /-\(Blue. ) (Brown.)\=/(Blue. ) ( Blue. )The nitro-group has a considerable influence on the formation ofphenyl hydrazones. Thus, whils t o -nitrophenyl h y drazine reacts readilywith p-quinones and their monoximes, p-nitrophenylhydrazine onlyreacts readily with the monoximes, and the m-compound is oxidisedwithout forming a h y d r a ~ o n e .~ ~ It is possible that the o- and p-phenyl-hydrazine react in the tautomeric ittonitro-form, which mould be lessreadily oxidisable. Whilst there is little doubt that the derivativesfrom quinones are p-hydroxyazo-compounds, the monoxime derivativeshave different properties. The alternative formulae are :But in their resistance t o acids and in their behaviour on oxidation,72 Trans., 1906, 89, 982. v3 W. Borsche, Abstr., 1908, i, 66128 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.these compounds do not resemble phenylhy droxylamine derivatives,and they are therefore t o be regarded as hydrazones.As an appendix to the hydrazones, some experiments on the oxida-t.ion of aromatic hydrazines by atmospheric oxygen may be menti0ned.7~The main product is the parent hydrocarbon, according to theequation :but this is always accompanied by some hydrocarbon of the diphenylseries :R*NH*NH, + 0 = RH + N, + H,O,2R*NH*NH, + 0, = R*R + H, + ZN, + ZH,O.It is suggested that the first product is a hydroxyhydrazine, whichthen decomposes directly :R H RH k-k --+ NiNH*OH H 6Hor through the intermediate formation of R*N:NH, to which thetransient crimson coloration always observed during the progress ofthe oxidation may be due.The remarkable isomeric changes undergone by the arylhydroxjl-amines in contact with dilute sulphuric acid75 have been furtherinvestigated by Bamberger and his pupils,76 and considerable light hasbeen thrown on the transformation, although certain points stillremain obscure.m-Xylylhydroxylamine yields in this way iminoxylo-quinol, which readily loses ammonia to form xyloqainol. Whenalcohol and sulphuric acid are used, water being carefully excluded,the corresponding ethyl imino-ether is obtained, yielding the quinolether with water :Me OEt Me OEt\/ \//\ .+ II Il Me/\\.( \/NH*OH NH-+ l1111\6e0AMe \/But the change is not arrested at this point. A part of the sub-stance is transformed into ethers of xylorcinol and xylohydroquinone.It is the mechanism of this reaction that has been studied.Experimental evidence is produced t o show that quinols and theirethers exist in solution as hydrates or alcohoiates, and also that wateror alcohol may be added on at a *C:C- linking.The change of position74 F. D. Chattaway, Tmns., 1907, 91, 1323.75 E. Barnberger, Abstr., 1903, i, 83.A ~ s ~ T . , 1907, i, 616, 517, 518, 519, 520, 521ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 129of OH and OEt may be then only apparent, being due to the additionand removal of water or alcohol at different points on the benzenerings, the only truly isomeric change being the wandering of themethyl group. This is explained as being similar to the pinacone-pinacolin change, which consists in an interchange of methyl andhydroxyl :Me Me,C--C<OH Me Me,C--C<OH -+ I 1 Me OH I I OH MeThe production of hydroquinone monoethyl ether and xylorcinoldiethyl ether from 9%-xyloquinol may then be represented by thefollowing scheme :Me OH\/OH.H/\HMe " OH f IIXMe OH OEtMe OHH/\HH O M e \ .. 0H,!,)Ne/\OH OEtOH OHA OH OEtMeOEV'\H -+ H(/MeOEtOHH O M eOEtMe/\HThe Terpene Group.The recent work on the terpene group includes several newsyntheses from the Manchester laboratories. Thus carvestrene, whichis of importance as occupying the same position in the m-cymene groupas dipentene does in the p-cymene series, and to which Baeyerassigned the constitutionCMeAH27 7H\/CH2H,C CH*CMe:CH,has now been synthesised 77 in a manner which fully confirms Baeyer'sconclusions. Starting with cyclohexanone-3-carboxylic acid (I) theethyl ester is caused to react with magnesium methyl iodide, and after'7 W. H. Perkin, jun., and G. Tattersall, Trans., 1907, 91, 480.REP.-VOL.IV. 130 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.hydrolysis and distillation under reduced pressure, the lactone (11) ofcis-1 -methyl-l-cycZohexanol-3-carboxylic acid is obtained :Me0H,/'\H,H,()H*C02Me\-0d ' H 2 I HI ,\/ IH-COBy the successive action of hydrobromic acid and pyricline, this is con-verted into the unsaturated 1-methyl-A1-cycZohexene-3-carboxylic acid(111), the ethyl ester of which reacts with magnesium mgthyl iodide toform a new alcohol, dihydrocarvestrenol or Al-m-menthenol (1V) :Me MeHe/\H H/\HH2 H2H:()H-CO,H H),,)H* CMe,*OH(111.) (IV.)This is the meta-analogue of terpineol, and readily yields carvestreneon digestion with potassium hydrogen sulphate. Other interestingproducts were obtained in the course of the research.The synthesis of terpin hydrate 78 has now been greatly ~irnplified,7~it being found that the treatnient of ethyl cyclohexanone-4-carboxylatewith magnesium methyl iodide, which formed the starting-point of theoriginal process, yields terpin directly under suitable conditions.Three of the products of the oxidation of pinene, namely, terebic,terpenylic, and homoterpenylic acids, have now been synthesised by asimple application of Grignard's reaction to ketonic esters.b0 Thusethyl acetosuccinate and magnesium methyl iodide yield ethylterebate :YOMe Me, ONgI CRIe,---,FH*CO,Et yH*CO,Et tH*CO,Et 0 'y 3 2 --++ YH2 QHz -+C0,Et C0,Et C O p - 1I n an exactly similar way, ethyl P-acetylglutarate is converted intoethyl terpenylate, and ethyl P-acetylndipate into ethyl homoterpenyl-ate.Other investigations in the terpene series have been very numerous,and in some cases highly controversial questions of priority occupy78 Ann.&port, 1904, 117.79 F. W. Kayand W. H. Perkin, jun., Trans., 1907, 91, 372.J. L. Simonseii, ibid., 184ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 131a large space in the polemical discussions, a cause of which is thegreat variety of possible formulae for certain members of the series.A study of the reactions of terpinene nitrosite has led Wallachto the conclusion t h a t it must be represented by the structuralformulaCH, O*NO\/although he does not consider that the compound can be unimolecular,in spite of the results of molecular weight determinations.Thisaccords well with the formula (I) for terpinene suggested by its otherchemical reactions and on optical grounds : 82_ _.-, CHMe, CHMr1.1 (II.) (111. )although Wallach 83 does not exclude the possibility of (II), andassigns the formula (111) to P-terpinene.Synthetical phellandreae, from carvoment hene dibromide, resemblesthe natural hydrocarbon i n all respects except its optical activity andboiling point.64 It is represented by the formulac=2I3 CHMe,Dihydrophellandrene and dihydroterpinene are found 85 to beidentical with the carvomenthene obtained by reduction of limonenemonoh ydrochloride.A comparison of various naturally occurring camphenes with thesynthetical product has yielded evidence for the existence of twoisomeric camphenes.s6Nopinone (I), which has hitherto only been obtained in smallquantities, is readily prepared by the oxidation of nopic acid, which81 0.Wallach, Abslr., 1907, i, 228.83 Abstr., 1907, i, 943.84 I. L. Kondakoff and I. Schindelmeiser, ibid., 329.85 F. W. Semmler, Zoc. cit.82 F. W. Semmler, ibid., 714.b8 0. Wallach, Abstr., 1907, i, 1061.K 132 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ocours in the oxidation products of turpentine.87 When acted on withzinc and ethyl bromoacetate (compare p. 135) a hydroxy-ester is firstformed, which on heating with potassium hydrogen sulphate yields theunsaturated ester (11). On distillation, the acid loses carbon dioxideto form /I-pinene (III), the configuration of which, however, is notidentical with that of the natural hydrocarbon.ss0 CH* C02Et CH,H& I12(>H H2(>HH H H(1.) (11.) (111.)An attempt was made89 to synthesise nopinone by a series ofreactions terminating in the production of 4-/I-bromoisopropylcyclo-hexanone (IV), which is related to nopinone (I) in the same way asdihydrocarvone hydrobromide (V) is to carone (VI) :0 0 0H ) ! , H 2 HJ:p2(IV.1 (V. 1 (VI.but the ring failed to close under the action of alcoholic potash. Anattempt to cause the sodium derivative of ethyl 4-/I-bromoisopropyl-cyclohexanone-2-carboxylate (VII) to undergo internal condensationwith formation of ethyl nopinonecarboxylate (VIII) was0 0/\ HH CBrMe2(VII.) y111.)similrtrly unsuccessful, from which it appears that, in dicyclic systems,the cyclopropane ring is more readily produced than the cyclobutanering.57 0.Wallach and A. Blumann, Abslr., 1907, i, 936.88 0. Wallach, ibid., 1058.8" W. H. Perkin, jun., and J. L. Simonsen, Tyans,, 1907, 91, 1736ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 133Much work has been done on the constitution of sabinene, to whichMeSemmler Q0 assigns the formula FI,~ Hog:, special attention havingC HMe,been given to the products obtained on breaking down the ring so asto obtain cycjopentene and cychpentadiene derivatives. Other inves-tigations have included teresantalic acid 91 and the di- and tri-cyclu-santalols.The method adopted for the synthesis of isolaurolene 93 has beenapplied to campholene.94 ad-Trimethyladipic anhydride yields tri-methylcyclopentanone, from which the tertiary alcohol is obtained withmagnesium methyl iodide, distillation then yielding campholene :YH,-C Me,--CO yH,-CMe, FH,--CMe,CH,* CHMe CO >' C€€,*CHMe>Co CH,*CHMe >CNe*OH(Campholene. )The degree of perfection which has been reached by the hypothesesof structural organic chemistry, and their great adaptability to theexperimental facts, are perhaps nowhere better illustrated than in thesynthetical study of the terpenes, but it has only been possible t omention a few of the very numerous researches in this department ofchemistry, and the closely allied group of the camphors must be leftuniiot iced.I n connexion with the study of natural products, reference shouldbe made to the very careful investigations of chaulmoogric acid 95 andhomoeriodictyol.9G The former is found to be a cyclopentene deriv-ative, and to react in accordance with both of the following formuh :CH/\ A*YH QH*LCH,],,*CO,H ~H-$l*[CH,],2*C0,RCH,*CH, CH,*CH, 9but as the acids obtained from the seeds of three botanically distinctspecies of plants are found to be identical, it is not possible to considerAbstr., 1907, i, 145, 714.91 F.W. Semmler and K. Bnrtelt, ibid., 703, 1062.92 F. W. Semmler and K. Bode, ibid., 431.93 Ann. Report, 1906, 141.94 G. Blaiic, Abstr., 1907, i, 1058.gr; M. Barrowcliff and F. B. Power, Trans., 1907, 91, 557.96 F. B. Power and F. Tiithi, ibicl., 887134 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.chaulmoogric acid as a mixture of stable isomerides, and tautomerismor dynamic isomerism between the two forms must be assumed.Theallied hydnocarpic acid has the same structure, differing only incontaining ten methylene groups in the side chain instead of twelve.Bin3 Syntheses ccnd Po Zycyclic Compounds.The conditions of stability of hydrocyclic compounds containingthree- or four-carbon atom rings have been discussed from variouspoints of view,g7 and it is found that the relative stability is dependentrather on the nature and position of the attached groups than on thenumber of atoms forming the ring. The reaction between tribromo-propane, CH2Br*CHBr*CH2Br, with ethyl sodiomalonate seemed tooffer an equal chance of formation of a cyclopropane or cyclobutanering, but the actual course of the reaction was unexpected, unsaturatedopen-chain compounds being exclusively obtained (see p.92).The series of hydrocyclic hydrocarbons has now been completed bythe preparation of cyclobutane by the reduction of cyclobutene withhydrogen and nickel,gs butane being obtained a t a higher temperature.The general chemical behaviour of cyclobutene indicates 99 that it is thesimplest dicyclic hydrocarbon,CH- CH,1 \ 1 9 CH,* UHand this constitution accords with its physical properties, and with theresults of the reduction of higher dicgclic hydrocsrbons (terpeoes) bySabatier and Senderens' method.The yellow modification of cinnamylidenemalonic acid, when heatedCHCHwith baryta, yields phenylcyclobutene, CHPh< I >CH,, and di-yHPh*QH*QH*QH2CHPh*CH*CH* CH,'phen yl tricy clooctaiie,The white modification of the acid, on the other hand, yields thelatter hydrocarbon together with diphenyldicyclohexane :yHPh*QH* C/H2CHPh.CH *C H,'Another new cyclobutane synthesis is the production, by the actionof sulphuric acid on ethyl s-dimethylacetonedicarboxylate, of the com-Seealso A. Kotz, Abstr., 1907, i, 1018.97 W. H. Perkin, jun., and J. L. Simonscn, Trans., 1907, 91, 816, 840.98 R. Willstiitter and J. Bruce, ibid., 1018.99 N. Zelinsky and J. Gutt, ibid., 1908, i, 14.0. Dobner and G. Schmidt, ibid., 1907, i, 204ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 135pound, CMe<"->CMe*C02Et, which readily hydrolyses andloses carbon dioxide, yielding the hydroxyketone,C(0H)C M e < ~ ~ o ~ C H M e .2The method of synthesis of naphthylenediarnine derivatives describedon p. 135 of last year's Report may also be applied when the benzenederivative used has a side-chain of only two carbon atoms, providedthat a methyl group occupies the neighbouring position to the side-chain.3 Thus o-toluyl chloride and ethyl sodiocyanoacetate give ethylcyano-o-toluylacetate, C,H,Me*CO*CH(CN)*CO,Et, which on heatingwith ammonium acetate gives et h y 1 p-imino-a-cyano-p-o- tolyl-propionate. Acids then convert this into ethyl 1 : 3-naphthylene-diamine-2-carboxylate :1 : 4-Naphthylenediamine derivatives may be synthesised in a similarmanner :WalIach5 has described a new method of enlarging carbocyclicsystems.The cyclic ketones readily condense with esters of bromo-acetic acid to form hydroxy-esters, which after treatment withhydrogen bromide and reduction yield cyclylacetic acids. (Cyclyl issuggested as a general designation for univalent cyclic groups.) Fromthe amides of these acids the cyclylmethylamines are obtained byHofmann's reaction. The nitrites of the cyclylrnethylamines aredecomposed by acids, the first product being probably an unstablediazo-compound :G. Schroetcr and C. Stassen, Abstr., 1907, i, 532.E. F. J. Atkinson, H. Ingham, and J. F. Thorpe, Trans., 1907, 91, 578.J. F. Thorpe, +id., 1004. 5 Abstr., 1907, i, 602, 616136 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.On boiling, nitrogen is removed, and it may be supposed that adicyclic compound results with loss of water :CH,*CH,\ I l)CH>\CH,-CH’?CH,If, in the subsequent hydrolysis, the bond 1 is ruptured, an alcoholof the next higher cnrbocgclic system is obtained :(iH,*CH,*QH,CH2*CH2*CH*OH’and this is the principal reaction. If, however, rupture takes placeat the bonds 2 or 3, which are equivalent, a primary or secondaryalcohol is obtained, which may then lose water to form an unsaturatedhydrocarbon :,>C* CH,.7H2*C H FH,*CHCH,*CH, 2>C:CH2 Or CH,-CHSince the alcohols may be oxidised again to ketones, it is possible inthis way to prepare cyclohexanone from cyclopentanone, cyclooctanonefrom suberone, &c.Of the more complex hydrocarbons with condensed nuclei, aninvestigation of pyrenee has shown that only one of the rings isbenzenoid, the others being quinonoid in structure, the arrangement ofdouble linkings in pyrene and pyrenequinone thus being :cycZoHexanone and methyl alcohol undergo a condensation 7 withsulphuric acid resembling the formation of mesitylene from acetone,the product being dodecahydrotriphenylene (I), which yields tri-phenylene (11) on distillation with zinc dust :(1.) (11.)Attention has been called by Collies to the importance, in the8 J. N. Collie, Trans., 1907, 91, 1806. Compare J. N. Collie and E. R. Chrystall,G. Goldschmiedt, Abstr., 1907, i, 310. C. Mannich, ibid., 205.ibid., 1802ORGANIC CHEMISTRY-HOMOCYCLIC DIVISION. 137building up of rings, of the keten group, *CH,*CO. Compounds inwhich multiples of this group are present, of which ethyl acetoacetateand diacetylacetone are simple types, undergo condensation withremarkable readiness. Owing t o the labile character of the ketengroup, such compounds pass, under gentle treatment, with addition orloss of water or carbon dioxide, into the most diverse homocyclic andheterocyclic compounds. Numerous examples of ring-formation underthese conditions are adduced in illustration, and the importance ofcompounds of this type in the synthesis of substances occurringnaturally in plants is pointed out. Although the photo-syntheses ofthe plant have a t present no equivalent in the laboratory, it isevident that the study of these labile compounds, many of whichundergo condensation even at the ordinary temperature, in presence offeebly acid or alkaline solutions, without the aid of violent con-densing agents, may throw considerable light on obscure questions inplant-physiology.CECIL H. DESCH