ORGANIC CHEMISTRY.THE present report embraces the subject matter of four separatereports i n previous annual volumes, dealing respectively withaliphatic, homocyclic, and heterocyclic compounds, and with stereo-chemistry. A glance at3 the Abstracts for the year 1908 will showthat the space allotted to summaries of organic research is equal to thatdevoted t o the rhumbs of all other sections of chemical activity. Itis therefore inevitable that references to many valuable reseaches havebeen omitted, for only by making a selection of the year’s output inorganic chemistry has it been found possible to keep this reportwithin reasonable bounds.The theoretical views implied in the conventional structuralformuh have held their own with remarkable success throughout thedevelopment of organic chemistry, but their insufficiency to expressthe whole of the facts is becoming more obvious every year, and greatuncertainty still prevails as to the nature of the changes to be madein order to bring certain of these formulae into closer accordance withthe experimental evidence.There can be little doubt that ideasderived from the consideration of plane formulae require to be con-stantly verified by reference to the steric relationships of thecomponent parts of the molecule of organic compounds. I n certaincases the uncertainty or ambiguity attaching t o the use of staticconjugations has led t o the adoption of dynamic formule. Thedirect application of such a physical hypothesis as that of the electronto organic chemistry would seem to be premature, probably becausethe hypothesis is still imperfectly developed, even in its simplerapplications t o the question of chemical affinity, but it has alreadyfurnished several suggestive qualitative ideas.The recent developments in the bheory of chemical valency impliedin such conceptions as those of principal and subsidiary valencies andco-ordinated compounds have been adopted only in very few instances.Many of this year’s researches have an important bearing on the valencyof oxygen, and afford ample confirmatory evidence as t o the quadri-valent nature of this element, particularly as manifested in the cyclicoxonium salts.On the other hand, no definite conclusion has bee74 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.reached in regard to the assumed tervalency of carbon in triphenyl-methyl.A striking feature-of the year's work is the number of investiga-tions dealing directly or indirectly with the influence of unsaturation,especially of the nature and arrangement of double linkings.Theideas put forward by Thiele, in 1899, recur frequently in recent work,notably those of partial valency and of the conjugation of doublelinkings. The first will be discussed below in connexion with thequestion of colour and fluorescence; the second finds an importantplace in many discussions as to the origin of colour and the reactivityof tautomeric compounds.The Mechccnism of Reactions.The great advances made in recent years in our knowledge of thestructure of organic compounds have directed attention all the morevividly t o the gaps which remain. This is especially the case withregard to reactions.The chemical equation only represents, as a rule,the initial and final stages, the mechanism by which the resultis obtained remaining obscure, and only to be filled by the assumptionof hypothetical intermediate products, the nature of which it is oftenvery difficult to define. Much recent work has been directed to theelucidation of the mechanism of familiar reactions, such as thecatalytic hydrogenation and dehydrogenation of organic compounds bymetals, the reactions depending on the employment of magnesiumorganic compounds, the condensations in presence of aluminiumchloride and of sodium ethoxide, and the formation of esters, etc.Catalytic Action of Metals and Inorganic Substances.The remarkable results achieved by Sabatier and Senderens on thehydrogenation of organic compounds in the presence of finely-dividednickel at moderately high temperatures have led to numerousresearches on the influence of catalysts in the synthesis and decom-position of organic substances.Many aromatic compounds have been converted by the Sabatier andSenderens' method into the corresponding hydroaromatic derivatives.With hydrogen a t 200' in the presence of nickel, p-benzoquinone andits homologues are reduced quantitatively to quinols. A t 220-250' thequinols are decomposed into the corresponding phenols, and water isliberated, but a t lower temperatures the qninols undergo furtherhydrogenation, becoming converted into cyclohexane-1 : 4-diols.Quinol itself a t 160-170' gives a mixture of phenol, cyclohexanolORGANIC CHEMISTRY.75and cis- and trccns-cyclohexane-1 : 4-diols ; at 130' this cis-modificationis formed exclusively.Catechol under similar conditions gives cis-cyclohexane-1 : 2-diolexclusively, resorcinol does not yield definite products, but pyrogallolis reduced to cyclohexane-1 : 2 : 3-triol.1Other unsaturated rings may be hydrogenated by this method.yH,*CH,CH,*CH, Furan at 170' gives rise to tetrahydrofuran, >0, togetherwith a small amount of butyl alcohol and some saturated hydro-carbons.2I n some cases the reactions have showed signs of reversibility.The reverse change has been observed at 200-300' with indene andacenaph t hene when reduced to hy drindene and tetrahydroacenaph thene,but decomposition products also arise.The hydrogenation of naph-thalene to tetrahydronaphthalene which occurs at 200' is reversed a t300'.I n certain instances, hydr~genation,~ under ordinary pressures inthe presence of nickel, leads to the formation of secondary products,and even to the total suppression of the primary reaction. Quinoline,for example, is transformed into methylketol. When nickel oxide isemployed as the catalyst a t high temperatures (200-300') and underhigh pressures (100-200 atmospheres), the hydrogenation runs itsnormal course, and the method has proved remarkably successful inyielding highly hydrogenated products, which are obtained only withconsiderable difficulty by other pr;ocesses.Aniline yields principallyhexahydroaniline, together with cyclohexylaniline and dicyclohexyl-arnine. The last of these is the chief product of the hydrogenation ofdiphen ylamine.Quinoline is readily reduced, first t o tetrahydroquinoline, and thent o decahydroquinoline, an almost quantitative yield of the latter beingobtained. Anthracene can be reduced in three successive stages t odihydroanthracene, tetraliydroanthracene, and, finally, perhydro-ant hracene.Phenanthrene, which requires a higher temperature (320-370°/150-170 atmos.), is similarly reduced in stages to its dihydro-,tetrahydro-, and octahydro-derivatives, and, finally, perhydrophen-anthrene ; the yield of the last is so good that the method is recom-mended for the preparation of the compound.When potassium phthalate is mixed with nickel oxide and heated in1 P.Sabntier and A. RIailhe, Compt. rcizd., 1908, 146, 437, 1193; A . , i, 278,2 A. Eourgnignon, B i d . Soc. chiliz. Bch~., 1908, 22, 87 ; A., i, 280.3 M. Padoa and U. Ir'abris, Atti R. Accad. Lincei, 1908, [v], 17, i, 111 ; A., i,529.25576 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.hydrogen at 300°/100 atrnos., an excellent yield of potassium trans-hexah ydrop hthala t e (ti*alzs-cycZo hexane- 1 : 2 - dicar boxyla t e) is produced.Sodium benzoate is readily converted into sodium hexa hydrobenzoate(cyclohexanecarboxylate), the potassium salt being less affected.Experiments conducted under ordinary pressures show that thenickel oxide acts as a catalyst only under high pressures.On theother hand, the presence of the catalyst is essential in the foregoingreductions, for heating with hydrogen alone under high pressure doesnot lead to the hydrogenation of the organic compounds.4The use of metals as catalysts at high temperatures has led to therepetition and extension of certain old experiments on the use ofplatinum-black in reductions carried out on substances in solution.Compounds containing the ethylene linking are readily reduced withhydrogen in the presence of this catalyst even at the ordinary tem-perature. Ethyl oleate in ethereal solution gave ethyl stearate inquantitative yield, and oleyl and erucyl alcohols were similarlyreduced to the fully saturated alcohols. Geraniol was convertedinto a mixture of P(-dimethyloctane and yy-dimethyloctanol.Cholesterol in ethereal solution furnished a dihydrocholesterolidentical with /3-cholestanol.Aromatic substances can be hydro-genated to hydroaroma,tic derivatives by this method, but somewhatless readily than by the high temperature processes; benzoic acid, f o rexample, gives about 12 per cent. of cyclohexanecarboxylic acid.5A powerful catalytic reducing agent has been prepared by addinghydrazine to a solution of palladium chloride and sodium protoalbinate,when the hydrosol of palladium is formed. The colloidal solutiontakes up a large amount of hydrogen, and when unsaturated sub-stances are dissolved therein, and the liquid left in contact withhydrogen, the gas is quickly absorbed, and an energetic reduction iseffected.Fumaric (maleic) and cinnamic acid in the form of theiralkali salts are converted into succinic and P-phenylpropionic acidsrespectively. The method is applicable to the unsaturated glyceridesoccurring in many fats and oils ; these substances are quantitativelyconverted into the saturated glycerides, this being the only processyet discovered which will effect this reduction at the ordinarytemperature.6Hydrogenation is not the only chemical change promoted by thecatalytic influence of inorganic materials. Dehydration of alcoholsW. N. Ipatieff, Ber., 1908, 41, 991 ; W. N. Ipatieff, W. Jakowleff, andL. Rakitin, *id., 1908, 41, 996 ; W.N. Ipatieff and 0. Philipoff, ibid., 1908, 41,1001 ; A., i, 330, 332, and 342.R. Willstiitter and E. W. Mayer, ibid., 1908, 41, 1475, 2199 ; A., i, 383, 636.C. Paal and J. Gerum, ibid., 1908, 41, 805, 2273 ; A., ii, 392, i, 599 ; C. Paaland-K. Roth, ibid., 2283 ; A., i, 599ORGANIC CHEMISTRY. 77and similar substances is induced by certain oxides. Precipitatedalumina and silica, when dried and gently calcined, promote thedecomposition of ethyl alcohol into ethylene and water at temperaturesmuch below those at which heat alone would be effective in decom-posing this substance. At 280' the alcohol is quantitatively decom-posed into water and ethylene. Prolonged calcination diminishesvery considerably this catalytic pbwer, for after being heated a t a whiteheat, the oxides have no effect on the alcohol below 400°, and thenonly a small amount of ethylene is produced.At 300' in the presence of gently-calcined alumina, ether isdecomposed into ethylene and water, the reaction being recommended asa process for preparing this hydrocarbon. Acetic and propionic acidsat 350' give respectively acetone and methyl ethyl ketone, togetherwith carbon dioxide and water.Alumina also induces the eliminationof hydrogen halides : a t 280' chloropropane gives propylene, andethylene dichloride yields vinyl chloride.Carefully-calcined gypsum promotes the decomposition of alcohol at420°, 90 per cent. of the gas evolved being ethylene and the resthydrogen. Anhydrite has no effect below 460".Aluminium silicateand kaolin have been shown to possess in some degree this catalyticpower, which is somewhat diminished by calcination.7When heated alone theprimary aliphatic alcohols decompose at a red heat into (i) water andan olefine and (ii) hydrogen and an aldehyde, but they show noappreciable decomposition below 400'. I n contact with finely-dividedcopper, nickel, cobalt, iron, and platinum, the alcohols undergodehydrogenation at about 350°, the products being hydrogen andan aldehyde. At the same temperature, certain oxides, and particularlyalumina, induce dehydration, the products now being water andan olefine.The Grignard Reaction.Although every year sees a great increase in the applications of theGrignard reaction in the synthesis of organic compounds, the exactmechanism of the process is even yet imperfectly understood, especiallyin respect of the true nature of the intermediate additive products.The complexes which the reagent forms with ether are geperallyuncrystallisable, but it has now been found * that by the employmentof amyl ether in place of ethyl ether, crystalline compounds, such as(C,H,,),O,CH,MgI, may be prepared.Whilst magnesium alkylhalides can also add on a second molecule of ether, the additive com-pounds with tertiary amines never contain more than one mol. of theJ. B. Senderens, Compt. rend., 1908, 146, 125, 1211 ; Bull. Soc. chint., 1908,[iv], 3, 633 ; A., i, 494, 496, i, 168.T. Zerewitinoff, Ber., 1908, 41, 2244; A., i, 616.A few general principles are apparent78 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.latter.They are, however, capable of combining also with a moleculeof ether, yielding mixed ether-amino complexes. Replacement ofether by amine, or conversely, may take place, the change proceedingin the one or the other direction according to the relativestability ofthe resulting additive compounds, as shown by thermochemicalmeasurements. Ethers are not capable of expelling aliphatic amines,the final product being one containing one mol. of ether and one mol. ofz~mine.~ It is concluded from these results that addition takes place attwo dissimilar positions in the molecule of the magnesium organichalide.The Grignard reaction may also take place, although less readily andless smoothly, in the absence of ether or even of a solvent.Aromatichalogen compounds combine with magnesium almost quantitatively onheating, and the products react with water, yielding the parent hydro-carbon. Aryl chlorides, and the halides of alkyls lower in the series thanamyl, must be heated in a sealed tube a t 270'. Bromo- and iodo-benzene yield some diphenyl as well as benzene, and the lower alkyliodides give rise to hydrogen and unsaturated hydrocarbons in additionto the principal product. These results indicate that the first stage inthe reaction is the direct addition of magnesium to the alkyl or arylhalide, and that the formation of an oxonium compound in the presenceof ether is a subsequent stage.10Other metals behave in a simi1a.r manner to magnesium when heatedwith organic halides.The products obtained with aluminium, indium,and thallium yield the hydrocarbons on treatment with water.l1The varying degree in which metals of different groups are capableof replacing magnesium in the Grignard synthesis has been studiedand tabulated.l2From the action of sodium and alkyl halides on certain compounds,the intermediate formation of sodium alkyl seems to be established.Thus acetophenone, isoamyl iodide, and sodium yield phenylmethyliso-amylcarbinol, OH*CPhMe-C,H,,. A mixture of sodium with amercury dialkyl, which must contain a mercury alkyl compound, mayalso be used. Thus mercury diethyl, sodium, and benzaldehyde reactin ether to form phenylethylcarbinol, identical with that obtained bymeans of Grignard's reagent.Carbon dioxide forms the correspondingcarboxylic acid with a mixture of sodium and mercury alkyl.Aromatic acids may be prepared by the action of carbon dioxide onthe same mercury alkyl-sodium mixture in presence of an aromaticW. Tschelinzeff, Ber., 1908, 41, 646; A., i, 254.lo J. F. Spencer and E. M. Stokes, !Z'mns., 1908, 93, 6 8 ; J. F, Spencer andl1 J. F. Spencer and M. L. Wallace, ibid., 1908, 93, 1827.la J. Zeltner, J . pr. Chem., 1908, [ii], 77, 393 ; A!., i, 401.M. S. Crewdson, ibid., 1821ORGANIC CHEMISTRY. 79hydrocarbon. When a side-chain is present, the carboxyl group entersthis, so that, for instance, m-xylene gives na-tolylacetic acid, and ethyl-benzene gives a-phenylpropionic acid.13Magnesium aryl chlorides, which cannot be prepared by theordinary method, are readily obtained when the reaction betweenthe metal and the aryl chloride is induced by the introduction ofa more reactive halide, such as ethyl iodide.14The chlorine in chlorodimethyl ether is not replaceable bymagnesium, even when iodine is added, but this ether reacts vigor-ously with magnesium phenyl bromide, giving benzyl methyl ether,and this reaction is found to be general.15Among new Grignard syntheses to be noted are the production of/3-ketonic esters from a-halogen fatty esters :JfgCHRBr*CO,Et -+ CH,R*CO*CHR*CO,Et ;the production of propiolic acid by the action of carbon dioxide onmagnesium acetylene bromide l7 :CHiC*MgBr+ CO, -+ CHiC*CO,H,and the preparation of thienyl derivatives from iodothiophen.ls Thus2-iodothiophen and acetone condense with magnesium in etherealsolution, yielding thienyldimethylcarbinol and P-thienylpropylene :C,SH,*C ICTe,-OH -+ C,SH,*CMe:CH,.Anhydrides of dibasic acids generally react with Grignard's reagentin such a way that both the carbonyl oxygen atoms are replaced by alkylor amyl groups, succinic anhydride, for instance, giving compounds ofthe type CH2*CR2*oH ICH,-CR, OH'Camphoric anhydride, on the other hand,reacts aboormally, only one oxygen atom being replaced, giving amixture of the isomeric dialkyl-campholides l9 :C H,-CH-CR, CH,-CH-CO I bMe, >O and 1 bMe,>.CH,*CIlEe* CO CH,.CMe*CR,I Il 3 P. Schorigin, Bcr., 1908, 41, 2711, 2717, 2723 ; A., i, 866, 881, 886.l4 A.Hesse, D.R.-P. 189476 ; A., i, 592.l5 A. Reychler, Bull. SOC. chim., 1907, [iv], 1, 1198 ; A., i, 159 ; J. L. Hamonet,l6 J. Zeltnbr, Ber., 1908, 41, 589 ; J. pr. Chem., 1908, [ii], 78, 97 ; A., i, 243,l7 B. Oddo, Gnxzctta, 1908, 38, i, 625 ; A., i, 748.ibid., 1908, [iv], 3, 254 ; A . , i, 242.759.V. Thomas, Compt. rend., 1908, 146, 642 ; A., i, 360.J. Houben and A. Hahn, Ber., 1908, 41, 1580 ; A., i, 53980 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.o-Phthalaldehyde forms dihydroxyalkylbenzenes, which on distilla-tion pass into 1 : 3-dislkylphthalans 20 :Since C-alkylated primary bases may be prepared by the action ofmagnesium organic compounds on Schiff’s bases, the hydramides,CHRT*CHR*N:CHR, which contain the characteristic grouping ofSchiffs bases twice, have been examined from the same point of view.The reaction is, however, abnormal, addition at the unsaturatedgroupings being accompanied by the resolution of a carbon-nitrogensingle linking, according to two different schemes :(I). CHPh(N:CHPh), with MgMeI, followed by water, -+CHMePh-NH, + NH(CHMePh),.(11). CHPh(N:CHPh), -+ CHO*Ph + NH,*CHMePh,magnesium aryl halides reacting only according to the second scheme.21A reducing action of magnesium organic compounds has also beenobserved in the aliphatic series.Thus ethyl hydroxypivalate,OH*CH,*CMe,*CO,Et, reacts with magnesium ethyl bromide, theketone formed being partly reduced, especially at low temperatures,with evolution of ethylene, forming pp-dimethylpentane-ay-diol,C H,Me CH(0H) * CMe,* CH,* OH.The Friedel and Crafts Reaction.In the condensation of phthalic anhydride with benzene, it is foundthat the presence of exactly one (double) molecule o€ aluminiumchloride is necessary to bring about the reaction, and the formationof an intermediate product is thus rendered highly probable :Naphthalene will not readily condense with phthalic anhydride inan indifferent solvent under the influence of aluminium chloride, butwhen benzene or one of its homologues is added, the whole of thenaphthalene reacts to form naphthoylbenzoic acid, even although thebenzene may be in great excess.Naphthalene (or anthracene) isthus capable of displacing benzene from the additive product.22When nitro-compcunds are condensed with hydrocarbons, the nitro-group is eliminated, and some oxidation takes place.Thus trichloro-nitromethane and benzene react in presence of aluminium chloride,giving triphenylmethane and some triphenylcarbinol, the latter2o F. NelkenandH. Simonis, Ber., 1908, $1, 986; A., i, 348.21 M. Busch and L. Leefhelm, J. pr. Chem., 1908, [ii], 77, 1, 20; A., i, 151,22 G . Heller and K. Schiilke, Ber., 1908, 41, 3627 ; A., i, 994.153ORGANIC CHEMISTltY. 81produced by oxidation. Ethyl nitrate under similar conditions actsas a nitrating agent, giving nitrobenzene with benzene. Benzeneamyl nitrite, and aluminium chloride give small quantities ofnitrosobenzene.23When benzene is condensed with carbon tetrachloride, the chlorineatoms are replaced in pairs, the first product capable of isolationbeing dichlorodiphenylmethane, no trichloro-compound being detected :A study of the behaviour of halogen derivatives of benzene in thisreaction, the condensation product being subsequently boiled withsulphuric and acetic acids, leads t o the conclusion that the substitu-tion is not simultaneous, but that the velocity of replacement of thesecond chlorine atom is very great.24CCl, + 2C6H6 -+ CCl,Ph2.The Claisen Condensation.Although several different explanations of the Claisen condensationhave been suggested by different workers, that put forward by Claisenin 1887 is still the most generally accepted, namely, that the ester firstforms an additive compound with sodium ethoxide. A study of thevelocity of formation of ethyl acetonyloxalate from acetone, ethyloxalate, and sodium ethoxide in alcoholic solution, as measured by thecolour produced by ferric chloride after neutralisation, is interpretedby Clark as favouring this explanation when the dependence ofthe velocity on the concentration of the reacting substances is takeninto account.25 A quantitative comparison of different reacting esters,and of the influence of catalysts shows that, as required by thishypothesis, the addition of alcoholic sodium ethoxide greatlyaccelerates the reaction between ketones and esters in presence ofsodium.26On the other hand, experiments with menthone and pulegone,reacting with amyl nitrite (instead of a carboxylic ester) in presenceof sodium ethoxide, shows that the group 'CH2*CO*CH: is attacked,not at the *CH,*CO* point, as might be expected, but at the *CO*CH:gro~p.~7 A discussion of the mechanism of this reaction leads to theconclusion that, as suggested by several authors, the metallic derivativeof the ketone must react to some extent in the C-form.In other words,there must be in the solution a real or virtual equilibrium (that is,either an equilibrium of the ions or one within the molecule itself,Q E. Bodtker, BuIl. Soc. chin?., 1908, [iv], 3, 726 ; A., i, 621.24 J. Boeseken, Rec. trav. ehim., 1908, 27, 5 ; A , , i, 189.25 R. H. Clark, J. Physical Cheqn., 1908, 12, 1 ; A., i, 124.26 J. R. Tingle and E. E Gorsline, Amr. Chenz.J., 1908, 40, 46 ; A., i, 732.27 R. W. L. Clarke, A. Lapworth, and E. Wechsler, Trans., 1908, 93, 30.REP.-VOL. V. 82 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.without actual dissociation) between the C-sodium and theO-sodium derivatives ::C:&*ONa Z :CNa*d:O.These C-metallic derivatives must react in a similar manner toorgano-metallic compounds, and the aceto-acetic ester synthesis thusbecomes a particular case of the reaction between esters and organo-metallic compounds. The authors give a theoretical discussion of thewhole question on this basis.The reversibility of the acetoacetic condensation, that is, thedecomposition of the diketone by sodium ethoxide, is found to dependon the acidity of the compound, the more acid its character thogreater being its stability.Thus methyl y-acetyldimethylaceto-acetate, CH,*CO*CH,*CO*CMe,*CO,Me, is hardly attacked by sodiumethoxide even on boiling, but when the acidity of the centralmethylene group is reduced by methylation, as in ethyl trimethyl-acetonedicarboxylate, CO,Et*CHMe*CO*CMe,*CO,Et, this power ofresistance disappears.28Esterijcation and Catalysis.Perhaps the most interesting problems in connexion with theformation of esters from acids and alcohols are the influenceof catalysts on the velocity of the reaction and the effect ofsteric hindrance, both of which have received attention during thepast year. The influence of catalysts in tha formation of ethylbenzoate has been the subject of a very extensive series of measure-rnent~,~g the results of which prove that a large number of factorsenter into the process, but do not allow any general conclusion to bedrawn.The efficiency of the catalyst could not be shown to dependon its degree of ionisation. The effect of neutral salts is also acomplex one, and dehydration plays a considerable part, as urged byH. E. Armstrong in numerous papers.An important theoretical paper has quite recently appeared dealingwith the general catalytic function of hydrogen ions.30 The treatmentof the subject is mainly physical and mathematical; so far asesterification and ester hydrolysis are concerned, the most strikingsuggestion is that the addition of water to an acid in a less basicsolvent diminishes the availability of the acid, or diminishes the28 W.Dieckmann and A. Kron, Ber,, 1908, 41, 1260 ; A . , i, 388.cg I. K. Phelps, M. A. Phelps, E:. A. Eddy, H, E. Palmer, R. JV. Osborne, andR. Smillie, Amer. J. Xci., 1908, [iv], 25, 39 ; 26, 281, 290, 296 ; A . , i, 166, 789,790.30 A. Lapworth, Tmas., 1908, 93, 2203 ; E. Fitzgerald and A. Lapworth, ibid.,2163ORGANIC CHEMISTRY. 83concentration of those u ions ” which bring about catalysis. Theseions are not identical with the “hydrogen ions” to which theelectrical conductivity is due, which are probably complex. Theideas propounded are likely to form the basis of an interestingdiscussion of the nature of catalysis.Many experiments on the catalytic function of acids in the hydro-lysis of imino-esters,31 the formation of 0ximes,3~ and the acetylation ofamino-groups 33 have also been made, and the importance of consideringthe basic character of the substances undergoing change is in allcases pointed out.An examination of the absorption spectra of certain unsaturatedketonic compounds, of which cinnamylideneacetic acid and cinnamyl-ideneacetone are types, leads to conclusions of some interest inconnexion with the different rate of esterification of different acids.%It has been conjectured that the process of esterification depenas onthe residual affinity of the carbonyl group in the acid, to which thealcohol adds itself.The spectroscopic evidence shows that theresidual affinity of a carbonyl group is greatest when the atomattached t o the hydroxylic oxygen is not ionised, the effect of addingan acid, and thus diminishing the ionisation of the organic acid,therefore being to increase the residual affinity of the carbonyl.I nconnexion with this, the effect of small quantities of hydrogen chlorideon the spectra of aromatic amino-aldehydes and -ketones has beenexamined. The development of a new band indicates that theresidual affinity of the arnino-group is increased by such addition, andthis would account for the catalytic action of acids i n the acetylationof such compounds.35The effect of steric hindrance is illustrated by experiments witharylated acetic acids.36 Triphenylacetic acid is much more difficult t oesterify than trialkylated acetic acids, the introduction of the thirdphenyl group having also a much greater effect than that of thefirst or second.The same thing is shown by a comparison ofbenzilic acid, CPh,*C(OH)*CO,H, with glycollic and mandelic acids,the phenyl group having a far greater influence than hydroxyl inretarding esterification.31 J. Stieglitz, Anter. Chem J., 1908, 39, 29, 166 ; A . , i, 167, 168.32 S. F. Acree, ibid., 300; A., i, 169.Zd A. E. Smith and K. J. P. Orton, Trans., 1908, 93, 1212.3 . ~ E. C. C. Baly and K. Schaefer, ibid., 1808.35 E. C. C. Baly and E. G. Marsden, ihid., 2108.36 J. Gyr, Ber., 1908, 41, 4308; A., 1909, ii, 33.a 84 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Simple Hydrocurbons and Aliphatic Derivatives.The complicated and much discussed problem of the behaviourof the simpler saturated and unsaturated hydrocarbons a t high tem-peratures has been submitted to further experiment, with the resultthat methane is found to be always the main product of the thermaldecomposition of ethane, ethylene, and acetylene. Methane itselfdecomposes directly into carbon and hydrogen, mainly at the contactwith the walls of the vessel, whereas the other hydrocarbons undergodecomposition throughout their mass.Acetylene is only producedin notable quantity from ethylene.37 The authors suggest that suchresidues as :CH2 and iCH have a t least a momentary existence, andgive rise to methane by combination with hydrogen. All theseprocesses take place slowly in comparison with those of combustion.The direct formationrof methane from carbon and hydrogen, whichmas questioned by Mayer and Altrna~er,~* has now been placed beyonddoubt by the production of 73 per cent.of the theoretical quantity ofmethane by the action of hydrogen on purified carbon.39Aliphatic derivatives of great complexity are of frequent occurrencein recent chemical literature, and the need of further classification hasbeen felt. Meldola 4O has proposed the grouping of open-chain systemsinto those composed respectively of similar and dissimilar atoms, andhas suggested for these, from analogy to homocyclic and heterocycliccompounds, the designations “ homocatenic ” and ‘‘ heterocatenic,”the latter including, for instance, the long chain-systems of thepolypeptides.The interesting and highly reactive group of the ketens has beenfurther studied.The parent substance, keten, CH,:CO, the isolationof which was announced last year 41 as a product of the thermaldecomposition of acetic anhydride, has been proved to have theconstitution then assigned to it. It has also been found possiblet o prepare it by the general method for the production of ketens,namely, by the action of zinc on bromoacetyl bromide :CH,Br*CO*Br + Zn = CH,:CO + ZnBr2.42The simplest polymerisation product of keten is acetylketen,CH,*CO*CH: GO,a colourless liquid, which reacts with water to form acetoacetic acid,37 W. A. Bone and H. F. Coward, Trans., 1908, 93, 1197.38 Ann. Report, 1907, 74.39 W. A. Bone and H. F. Coward, Trans., 1908, 93, 1975.4* Trans., 1908, 93, 1665, footnote.41 Ann.Xeport, 1907, 85.42 H. Staudinger and H. W. Klever, Ber., 1908, 41, 594 ; A., i, 246ORGANIC CHEMISTRY. 85and with aniline to form acetoacetanilide. With phenylhydrazine itforms a phenylhydrnzone-phenylhydra~ide.~~A new method for the preparation of dimethyl- and diethyl-ketenhas been found in the decomposition of di-substituted malonicanhydrides by heat :The anhydrides required for the above reaction are prepared byheating the semi-chlorides of the acids ; they are highly polymerised.A second method of preparation is that of treating the acid dichlorideswith aqueous pyridine.45When, instead of dialkylmalonic acids, malonic acid is treated by amethod for producing the anhydride, as by the action of acid chlorideson eilver malonate, or of silver oxide on malonyl dichloride, keten isnot produced, but in its place carbon suboxide is obtained, thehypothetical intermediate anhydride thus decomposing according tothe equation :c H , < ~ ~ > O = C<Eg 3- H,O.Dibromomalonyl dichloride, CBr,(COC1)2, readily reacts with zinc,giving a good yield of carbon sub~xide.*~These methods of preparation indicate that carbon suboxide isitself to be regarded as a keten.It combines with formic and aceticacids to form compounds in which the characteristic structure appearsto be preserved, the formic acid compound, fok instance, reacting inaccordance with the formula c02H>C:C:C<g2H. OH The acetic acidcompound breaks up in a remarkable manner on heating, yieldingacetic anhydride and a syrup which is completely converted by waterinto malonic acid.Whilst this substance may be only a polymerideof carbon suboxide, it is a t least possible that it may represent thehitherto unknown malonic anhydride, CH,<c0>o.47A comparison of the properties of the ketens now known hassuggested their classification in two groups, of which the first includesketen and its monoalkyl derivatives and carbon suboxide, all of whichare colourless, incapable of autoxidation, and are polymerised bypyridine. They are termed aldoketens. The members of the secondgroup, consisting of the di-substituted ketens, are coloured, undergoco43 F. Chick and N. T. M. Wilsmore, Trans., 1908, 93, 946.44 H. Staudingar and E. Ott, Bw., 1908, 41, 2208, 3829 ; A., i, 602, 939.45 A.Einhorn, Annalen, 1908, 359, 145 ; A., i, 312.46 H. Staudinger and S. Bereza, Ber., 1908, 41, 4461 ; A., 1909, i, 83.47 0. Diels and L, Lalin, i6id., 1908, 41, 3426 ; A., i, 9386 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.autoxidation, yield keten bases with pyridine, and, unlike the aldo-ketens, yield additive products with benzylideneaniline and quinonecontaining the groups C:N and C:O respectively, They are termedketo-ketens. Thus diphenylketen combines with quinones to formp-lactones, such as 0: C , H , ~ ~ ~ > C O , which decompose whenheated, giving derivatives of diphen ylquinonemethane. The additivecompound with dibenzylideneacetone loses carbon dioxide sponta-neously, the product being aediphenyl-y-diphenylmethylene-Aaa-pentadiene,CH:CHPhCPh,:C< C H : C HP h'a compound belonging to a class yhich may be described as acyclicfulvenes, from their resemblance to the fulvenes :CH,:C< CH:$?HCH:CH'Their colour is much less intense than that of the fulvenes, as mightbe expected from the absence of the ring structure.4sPrevious attempts to prepare the aldehyde of lactic acid, thesimplest methyl sugar, have failed, and it has been conjectured that itmust isomerise to acetol :CH,*CH(OH)*CHQ -+ CH,*CO*CH,*OH.It has now been found that its acetal may be prepared by thereduction of methylglyoxalacetal :CH,*CO*CH~OEt), -+ CH,*CH(OH)*CH(OEt),,and the latter compound, which may be regarded as the methyl ketoneof diethoxyacetic acid, may be prepared by the action of magnesiummethyl iodide on the amide of that acid.It was found better toemploy the piperidide, from which the acetal was readily prepared. Thelactaldehydeacetal obtained on reduction was readily decomposed byacids, yielding the bimolecular crystalline form of lactaldohyde, whichslowly dissociates in solution, as shown by determinations of themolecular weight .49The condensation of malonic acid with acetone in the presence ofacetic anhydride leads to the formation of the p-lactone of P-hydroxy-isopropylmalonic acid (I). It is remarkable that the only otheraliphatic /3-lactone known is an isomeride of this (11) derived fromas-dimethylmalic acid : 5O7Me,*7H*C02H ?Me,* 7"- C0,H(I) o--CQ (11) (70-048 H.Staudinger, Be?., 1908, 41, 906, 1355, 1493; A., i, 318, 410, 411.49 A. Wohl, ibid., 1908, 41, 3599 ; A., i, 941.50 A. N. Meldrum, Tram., 1908, 93, 598ORGANIC CHEMISTRY. 87The oxidation of butyric acid to acetone is probably typical of anumber of oxidations which take place in the living plant, and anexamination of this reaction proves that it is readily effected in thecase of the higher acids. Thus lauric acid iwoxidised’by hydrogenperoxide: an unstable ketonic acid is the first product, readilypassing into methyl n-nonyl ketone.50aCarbohy dyates.Further evidence for the y-oxidic formula for glucose and otherreducing sugars has now been brought forward. Thus glucoseanilideexists in two stereoisomeric modifications, and on methylation yieldstetramethyl glucoseanilide, identical with that ,prepared from tetra-methyl glucose.Similarly, glucoseoxime is methylated to tetramethylglucoseoxime methyl ether. That the tetramethyl derivatives possessthe y-oxidic structure is shown by the impossibility of furthermethylittion, since an aldehydic derivative would allow of theintroduction of two more methyl groups. The oxime therefore hasthe constitution :OMe*CH,*CH(OMe)*CH*CH(OMe)*CH(OMe)*CH*NH*OHand exists i n two stereoisomeric modifications. A similar result wasobtained with the anilides. The results of acetylation and reductionof glucoseoxime are ditficult t o reconcile with the formula proposed,but it is pointed out that the reagents used in these cases are moreliable to cause isomeric change than the methyl iodide and silver oxideused in m e t h y l a t i ~ u .~ ~The mutarotation of lactose has been shown to be due t o thegradual establishment of equilibrium between the hydrated a-modifi-cation and the anhydrous P-modification. I n the scheme below, thefirst equilibrium is attained instantly, the second only slowly : 52I I ? 0a-form + H,O z=? hydrated form Z H,O + p-form.The change of dextrosephenylhydrazone into the isomeric modifica-tion is accelerated by acids and retarded by alkalis. It is suggestedthat, if not syn- and anti-modifications, one of the isomerides may bethe true hydrazone and the other one of the stereoisomeric forms ofthe y-oxidic derivative, as suggested by Irvine for the oxime andanilide : 53OH*CH,*CH(OH)*CH C H (OH)*CH(OH)*CH*NH*NHPh.I A I50a H.D. Dakin, J. Bid. Chem., 1908, 4, 221 ; A., i, 134.51 J. C. Irvine and A. M. Moodie, l’&xns., 1908, 93, 95 ; J. C. Irvine and59 C. 8. Hudson, J. Amer. Chem. Soc., 1908, 30, 1767 ; A., i, 952.53 R. Rehrend and F. Lohr, Annulen, 1908, 362, 78 ; A., i, 765.R. Gilmour, ibid., 142988 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.When pure glycerose is polymerised by contact with colloidalbarium carbonate in methyl alcohol, a pentose is formed. It thusappears that polymerisation does not take place directly, but that thesimple sugar is first depolymerised to f~rmaldehyde.~~ The view hasalso been taken that alcoholic fermentation consists in a breakingdown of the sugar molecule to formaldehyde, followed by the conversionof this into alcohol and carbon dioxide.Formaldehyde is convertedby zinc carbonate into formic acid, methyl alcohol, acetol, and methyl-ketol. Zinc dust reacts in the same way, the product then includingseveral sugars, among which p-acrose could be detected, and poly-hydroxy-acids. Dextrose and zinc dust yield formic acid,diacetyl, methylglyoxal, and polyhydroxy-acids, but no methylThe action of dilute alkalis on the hexoses has been further studied.Since alkalis have been found to convert aldoses partly into thecorresponding hexoses, it should be found that Z-gulose and 2-idoseshould be converted into 2-sorbose, and this was confirmed. As mightbe expected from the fact that dextrose and lrevulose form an equi-librium mixture in alkaline solutions, these two sugars are found tobehave similarly on prolonged treatment with dilute alkali.Theprincipal product of decomposition is i-lactic acid, but poiyhydroxy-acids and small quantities of formic acid, carbon dioxide, and alcoholare also formed.56 The lactic acid formed in these experiments isalmost certainly produced from methylglyoxal, dihydroxyacetone,or glyceraldehyde. An examination of the polyhydroxy - acids,however, leads to conclusions which are not in accordance withthose of Nef,b7 saccharins with less than six carbon atoms beingundoubtedly present. Nef's views, also, would indicate that iso-saccharinic acid should be readily obtained from lzevulose, but this isnot the case, only a very small yield being obtained.On the otherhand, lactose gives a considerable yield of isosaccharinic acid.58Of other carbohydrate problems, that which has led to the mostdiscussion is the question of the chemical behaviour of cellulose,especially on esterification. Analyses of several anhydrous poly-saccharides indicate that the generally-accepted formula for starch,cellulose, etc. (C6H1,,05)n, is incorrect, and should be replaced by(C6Hlo05)n,H20. I n raffinose, melezitose, and mannasaccharide, = 3 ;in inulin, f i = 6.5954 C. Neuberg, Bwchem. Zeitsch., 1508, 12, 337 ; A., i, 765.55 U'. Lob, ibid., 78, 466 ; A., i, 715, 765.56 J. Meisenheimer, Ber., 190%, 41, 1009 ; A.: i, 319.57 Ann. Report, 1907, 87 ; also Annnlen, 1907, 357, 214 ; A., i, 5.58 H.Kiliani, Ber., 1908, 41, 469 ; A., i, 246.69 H. Kiliani, Chent. Zeit., 1908, 32, 366 ; A., i, 320ORGANIC CHEMISTRY, 89Cellulose formate is prepared by the action of formic and sulphuricacids on cellulose, but the number of formyl groups introduced is asyet undetermined.60 A new acetylating agent, by means of whichthree acetyl groups are introduced, consists of 100 grams of glacialacetic acid, 50 grams of zinc chloride, and 100 grams of aceticanhydride. It does not react with starch.6l When cellulose isheated with nitric acid and acetic anhydride, the latter being inexcess, only nitrates are obtained, unless sulphuric acid is alsopresent, when some of the nitrate groups are removed and aceto-nitrates are produced.62 When mixtures of anhydrous sulphuricand nitric acids, the latter in large excess, are used, the cellulosenitrate always contains less nitrate groups than when more dilute acidis employed.63Agreement has not yet been reached between different observers ast o the true nature of ‘‘ soda-cellulose.” Whilst one observer finds adistinct break in the curve representing the partition of sodiumhydroxide between cellulose and water, at the composition correspond-ing with the formula C,,Hl,0,0Na,64 another, making use of theformer data as well BS of new measurements, concludes that no defi-nite composition can be assigned to the solid substance, but that theprocess is one of adsorption, closely resembling the behaviour ofpalladium towards hydrogen, and probably susceptible of the samephysical e ~ p l a n a t i o n .~ ~Isomeric Change and Tautomerism.The influence of catalytic agents in bringing about isomeric change,as shown by the mutarotation of nitrocamphor, has been investigatedin various solvents.66 That the presence of a catalyst is essential, andthat an ionising soivent is not itself capable of bringing about thechange, is shown by the complete arrest of mutarotation when carbonylchloride is added. This reagent combines with the traces of ammoniaor amines present in most of the solutions, and thus destroys theircatalytic function. The addition of acids does not have the sameeffect, as neutral salts also exert an accelerating action.A systematic review of the different forms of structural isomerismhas been given by the original author of the theory of tautomerism.6760 J.P. Bemberg, D.R.-P. 189836, 189837; A . , i, 322.81 D. ,T. Law, Chem. Zeit., 1908, 32, 365 ; A., i, 321.62 E. Berl and W. Smith, jun., Ber., 1908, 41, 1837 ; A., i, 505.63 B. Rassow and W. v. Bong&, Zeitsch. angew. Chem., 1908, 21, 732 ; A., i, 394.64 W. Vieweg, Ber., 1908, 41, 3269 ; A., i, 857.135 0. Miller, ibid., 4297 ; A., 1909, i, 13.66 T. M. Lowry and E. H. Magson, Trans., 1908, 93, 107, 119.67 C. Laar, J. pr. Chm., 1908, [ii], 78, 165; A,, i, 74990 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The thirty-nine types are classified by him according to the number OFchanges of linking involved, and, further, into those which respectivelydo and do not involve change of valency.The scheme affords a usefulreview of the labile types of isomerism.The equilibrium between ketonic and enolic modifications has beenstudied from many points of view. Michael, in a theoretical review ofthe whole subject,0* insists on the necessity of distinguishing clearlybetween desmotropy, the reversible transformation of two isomeridesinto each other through the wandering of a labile hydrogen atom, andmesotropy, the process of irreversible isomerisation during theformation of derivatives. I n subsequent papers, he and others haveinvestigated the suitability of various reagents for distinguishingbetween enolic and ketonic substances. The best modification ofHantzsch’s ammonia reaction consists in employing triethylamine ortripropylamine, and in noting the development of heat rather thanthe formation of a precipitate.69 This is quite satisfactory when thecompound is mesotropic, only enolic substances then reacting, butdesmotropic compounds are, as might be expected, generally convertedinto the more stable modification by the amine.The results obtainedwith phenylcarbimide 70 and with acetyl chloride or acetic anhydride71are also indecisive in such cases, for similar reasons.Ethyl oxalosuccinonitrile, CN*CH,*CH(CN)*CO*CO,Et, prepared bycondensing ethyl oxalate with succinonitrile, has been obtained 72 in anenolicnnd a ketonic modification, both of which are crystalline. Thelatter form dissolves alcohol in to a violet, fluorescent solution. I fconfirmed, this would be the first case of fluorescence observed in thealiphatic series, and Kauff mann has therefore suggested the presenceof a ring ; the chemical reactions, however, indicate that the compoundhas the simple structure assigned to it.The enolic (I) and ketonic (11) modifications of ethyl methylcyclo-hexenonedicarboxylate :CH( C02E t) CMe>CE CH(CO,Et)*CMeSCHCHf%( C0,Et) : C( OH) CH2<C€€(C02Et)--C0(1.1 (11.1are stated 73 to give different vapours when distilled, although bothmodifications have the same boiling point.Desmotropy has notpreviously been observed to persist in the state of vapour.6* A. Michael, Annaleqt, 1908, 363, 20 ; A., i, 943.6o A. Michael and H. D. Smith, ibid., 36 ; A . , i, 943.7O A.Michael and P. H. Cobb, ibid., 64 ; A . , i, 947.71 A. Michael and A. Murphy, jun., ibid., 94 ; A., i, 949.72 W. Wislicenus and P. Berg, Ber., 1908, 4.1, 3757 ; A., i, 965.73 P. Rabe, Annaben, 1908, 360, 289; A., i, 530ORGANIC CHEMISTRY. 929-Nitrofluorene has been isolated in two desmotropic forms, ofwhich the mi-form, (?6H4>C:NO*OH, is comparatively stable. It is ct3 *4prepared from fluorene; ethyl nitrite, and potassium e t h ~ x i d e . ~ ~The well-known tautomerism of cyclic ketones, such as phloro-glucinol and dihydroresorcinol, is also exhibited by monoketones con-taining a simple ring i f a sufficiently powerful reagent, such as anacid anhydride, be employed to detect the enolic. hydroxyl g r o ~ p . ~ 5 I nthis way, acetyl derivatives of the enolic forms of cyclohexanpe andits three methyl derivatives, and of menthone, cyclopentanone, andsuberone, have been prepared.Camphor gave no indication of anyacetylation.An addition t o the many physical properties which have beenutilised to give indications as t o the ketonic or enolic condition oftautomeric substances is made in a recent communication, which dealswith the viscosity of liquids of this class, these compounds being mixedwith various solvents.76 The measurements show t h a t ethylacetoacetate is partly enolised, both alone and in solution.The addition of piperidine has a marked effect in increasing theviscosity.The peculiar isomerism of di-o-derivatives of benzene, referred to inlast year’s Report (p. ill), and there considered in relation t oKekulB’s formula for benzene, has received surprisingly little attention,and it is therefore uncertain how far the formation of such isomeridesis a general one.A certain number of nitrated derivatives of benzene,it is true, have been shown to exist in two modifications, but this isnow attributed by the author77 t o isomerism of the nitro-group.Thus l-chloro-2 : 4-dinitrobenzene exists in a stable and in a highlylabile form, the two forms being chemically identical. The fact thatboth modifications have the same colour excludes the possibility thatone of them has the mi-constitution.The two modifications of 2 : 4-dinitrophenol, however, differ incolour, one of them being pale green in solution or when fused.Thetwo substances give a eutectic mixture, they form similar salts, andare formulated as :OH OH74 W. Wislicenus and M. Waldmiiller, Ber., 1908, LEI, 3334 ; A., i, 973.75 C. Mannich and V. H. HAncu, ibid., 1908, 41, 564; A . , i, 275.A. E. Dunstan and J. A. Stubbs, Truns., 1908, 93, 1919.I. Ostromisslensky, J. pr. Chem., 1908, [ii], 78, 263 ; A., i, 86892 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The very remarkable changes undergone by the aromatic fulgides, ofwhich triphenylfulgide,is a type, when exposed to light have been fully ~tudied.7~ Thechange from orange to brown, brought about by light, is strictlyreversible, but after frequent repetition of the change, an irreversibletransformation occurs, resulting in the formation of ‘‘ photo-anhydrides,” the investigation of which hits not yet been published.A thewetical discussion is at present hardly possible, but it may benoted that the triarylated fulgides (with one exception) and certainof those containing two aryl groups are phototropic, whereas tetra-phenylfulgide is not.Certain compounds outside this group behavein a similar manner.An isomeric change of a more definite character, also taking placeunder the influence of light, is that of carv0ne.7~ Whilst solutions ofdihydrocarvone were hydrolysed by exposure to sunlight, thusbehaving similarly t o other cyclic ketones, carvone was converted intoan isomeride, resisting the action of permanganate. It is suggestedthat a process of ‘‘ internal polymerisation,” such as that representedbelow, takes place :I CH,:C*CH3CW.==-C--COCH3The migration of acyl groups in certain instances will be discussedlater in connexion with the structure of hydroxyazo-compounds. Themigration of aryl groups in iodohydrins during the elimination ofhydrogen iodide has been exhaustively studied.80 Thus a-l-naphthyl-propylene, C,,H7=CH:CH*CH3, on treatment with mercuric oxide andiodine, is transformed into a-1 -naphthylpropaldehyde, the iodohpdrinisomerising and losing hydrogen iodide at the same time :CloH,*CH(OH)*CHI*CH3 --+ CHO*CH(CloH7)*CR,.Similarly, P-l-naphthylpropylene passes into a-naphthylacetone :CloH7*CMe:CH, -+ CloH7*CMe(OH)*CH21 -+ CH3*CO*CH2*CloH7.A case of wandering of bromine has been observed in the trans-formation of nitroamines.81 2 : 6-Dibromo-l-nitroaminobenzene under-CH,-CH-CH,III I 1co CH-C-- i P H 2 .Y *‘=3 --3 II ICH,cH2-cH--PH,78 H. Stobbe, Annalen, 1908, 359, 1 ; A., ii, 339.79 G. Ciamician and P. Silber, Ber., 1908, 41, 1928 ; A., i, 555.80 M. Tiffeneau, Bull. Soe. chim., 1907, [iv], 1, 1205; Compt. rend., 1908, 146,29 ; 147, 678 ; A., i, 165, 166, 972.K. J. P. Orton and C. Pearson, Trans., 1908, 93, 725ORGANIC CHEMISTRY. 93goes rearrangement in the usual way t o 2 : 6-dibromo-4-nitroaniline,but at the same time a part of it forms the isomeric 2 : 4-dibromo-6-nitroanilineY the migrating nitro-group actually expelling a bromineatom, which re-enters the nucleus in a different position :BrBrII \ RrWith s-tribromo-1-nitroaminobenzene, the same change takes place,but the para-position now being occupied, the bromine is unable to re-enter the nucleus, and is found in the solution.I n concluding this section, reference should be made to theinteresting method of studying the process of isomeric change byobserving the change in rotatory power of an active solvent, not itselfundergoing change, as described in last year’s Report (p.184). Themethod is particularly applicable to the oximes, and one case has beeninvestigated, that of p -iodobenz-sym-aldoxime,:in which no other methodfor measuring the velocity of change is available. The change inrotation of n-propyl tartrate brought about by the presence of W-iso-nitromethane undergoing change to the stable nitro-compound :is less than that due to the oximes, but is still considerable.Thechange of ammonium cyanate and thiocyanate into carbamide andthiocarbamide respectively may be followed by dissolving a portion ofthe substance which is being heated from time to time in an aqueoussolution of ethyl tartrate and measuring the rotation, or in theformer case, by allowing the change to take place in the ethyl tartratesolution.82Oxonides.C,H,*CH:NO* OH --+ C6H5*CH2*N02,The value of the ozone method for recognising the presence ofethylenic linkings continues to be disputed. Whilst Molinari statesthat such linkings only are attacked by ozone, Harries finds that sub-stances containing triple linkings also react readily with ozone to formunstable products having the properties of ozonides.It is suggestedthat the differences observed may be due to the fact that Molinariemploys ozonised air, and Harries, the presumably more energeticozonised oxygen. The simpler olefines yield very stable ozonides,which may be distilled in a vacuum, when treated with ozone in an82 T. S. Patterson and A. McMillan, Trans., 1908, 93, 104194 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.indifferent s0lvent.~3taining a chain of three oxygen atoms :The formulation of such compounds as con->y-y<0 0\/0is not in accordance with their optical constants, and it appearsrather that only two of the oxygen atoms have an ether linking, thethird being “carbonyl oxygen,” but further study is required before adefinite formula can be assigned to them.The decomposition productsof ozonides of oleic acid and triolein by water and alkalis are verycomplex, although the saponification number I’ of such compounds isa fairly definiteAZiphatic Diaxo- and T~iuxo-compounds.I n 1907 an investigation of the action of sodium azide on variousdiazonium salts led to the production of a number of aromatic azides,of which the most noteworthy mere the hydroxyphenylazoimides,obtained from the aminophenols, and the t riazo-derivatives of thenaphthalene series, prepared from the isomeric nitronaphthalene-diazoniurn ~ a l t s . ~ 5An extension of this research to the aliphatic series has resulted inthe isolation of triazo-derivatives of the simplest organic compounds.By double decomposition, ethyl chloroacetate and sodium azide havefurnished ethyl :triazoacetate, N,*CH,*CO,Et, from which triazoaceticacid, N,*CH,*CO,H, and triazoacetamide, N,*CII,-CO*NH,, have beenprepared by the usual methods. The simplest triazo-ketone, acetonyl-azoimide, or triazoacetone, N,*CH,* CO* CH,, has been obtained, andcompared with its cyclic analogue, camphorylazoimide. Triazoacet-oxime has been isolated, and the isomeric ethyl a- and p-triazo-propionates have been prepared, but only the former ester could behydrolysed into a-triazopropionic acid, CH,*CH(N,)*CO,H, the p-esterbeing decomposed by caustic alkalis, and even by ammonia, with theelimination of the triazo-group.Triazoethyl alcohol, N,*CH2*CH,0H,and triazoacetaldehyde, N,*CH,*COH, mere obtained from ethylenechlorohydrin and chloroacetaldehyde ; the former mas a fairly stablesubstance, giving rise to esters, such as triazoethyl acetate,CH,*CO,*CH,*CH,N,,isomeric with ethyl triazoacetate, and the latter was an extremely explo-sive and unstable liquid, which was decomposed by hydroxylamine,phenylhydrazine, and other reagents for aldehydes.C. D.Harries and K. Haeffner, Ber., 1908, 41, 3098 ; A , , i, 846.84 E. Molinari, ibid., 585, 2782, 2789, 2794 ; A., i, 244, 849.s5 M. 0. Forster and H. E. Fierz, Tram., 1907, 91, 855, 1350, 1942ORGANIC CHEMISTRY. 95The refraction and dispersion of certain of these triazo-compoundswere determined, so also were the dissociation constants of triazo-acetic and a-triazopropionic acids, the latter determinations showingthat the effect on the strength of acetic acid produced by theintroduction of a triazo-group is less than that due to a bromine atombut greater than that due to an iodine atom.s6Bistriazo-compounds were also prepared : 1 : 2-bistriazoethane,N,*CH,*CH,*N,, and ethyl bistriazoacetate, CH(N,),*CO,*C,H,, thelatter being distilled and analysed in spite of its explosive properties.1 : 1-Bistriazoethane, CH3*CH(N,),, which was produced by thegeneral method from ethylidene dichloride and sodium azide, could notbe distilled even under greatly reduced pressure, owing to the violencewith which it explodes a t temperatures below 50".a-Triazopropionic acid, which is a racemic compound, mas resolvedby means of brucine, and the hvorotatory component reduced tod-alanine.57Ethyl diazoacetate, first discovered by Curtius, gives rise to a com-plicated series of transformation products under the influence of alkalis.A tabulated summary of some of these derivatives was given in lastyear's Report (p.158). Among the points studied since this tablewas published is the action of hydrlzziue on ethyl diazoacetate andbisdiazoacetate and their imides.Triazoacetylhydrazide (I) is produced either by the action ofhydrazine hydrate on diazoacetamide or by treating ethyl diazoacetatewith anhydrous hydrazine ; its constitution is demonstrated by thefollowing synthesis :CH,I*CO,Et Asf'3 N,*CH,*CO,Et 22 N,*CH,*CO*NH*NH,.(1.1The six-membered dihydrotetrazine ring in ethyl bisdiazoacetate (I),N,* co c<;;;,N>c-co*N3.(V.18(i J.C. Philip, Trans., 1908, 93, 918, 925.yi M. 0. Forster and H. E:. Fierz, ibid., 72, 669, 1070, 1174, 1859, 186596 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.is not affected by hydrazine, the action of which is effective only onthe carbethoxy-groups, thus giving rise to ethyl bisdiazoacetatehydrazido (11) and bisdiazoacetic acid dihydrazide (111).These hydrazides are converted respectively by nitrous acid intoethyl tetrazinedicarboxylate azide (IV) and tetrazinedicarboxylbisazide (V).88The action of cold concentrated aqueous potassium hydroxide onethyl diazoacetate leads to the production of tripotassium +-diazo-acetate (I), a compound which is transformed by this alkali at 100'into potassium bisdiazoacetate :This tripotassium salt corresponds with the amide (11) produced bythe action of liquid ammonia on ethyl diazoacetate.On treatmentwith alkali nitrite and acetic acid, this potassium salt yields potassiumnitrosodihydrotetrazinecarboxylate (III), from which reduction withhydrogen sulphide leads to dihydrotetrazinecarboxylic acid (IV), a sub-stance which, even at 60°, loses carbon dioxide and passes intoN-l-amino-1 : 3 : 4-triazole (V).*gCO,K* c < ~ ~ ~ ~ > o N o ----, CO,KG<~L.G~>CH -+8-fiT (111.) (IV. 1CH C H\/NI*=2(V. 1The compound obtained by the action of potassium ethoxide on ethyldiazoacetate is not, as was formerly supposed by Hantzsch andLehmann:o a potassium ethyl isodiazoacetate, C02Et*C<xg, but a$-diazoacetate derivative (potassium ethyl $-diazoacetate) containinga molecule of potassium ethoxide :CO,E t C<g',Ng>CK C02Et,C2H, OK.The corresponding sodium derivative is known, and both substanceshave all the properties of t,b-diazoacetates (CN-dihydro-1 : 2 : 4 : 5-tetr-azine-3 : 6-dicarboxylates).glT. Curtius and E.Rimele, Ber., 1908, 41, 3108 ; A., i, 921.89 Ernst Muller, ibid., 3116 ; A., i, 922.yo Ber., 1901, 341, 2506 ; A., 1901, i, 678.y1 T. Curtius, A. Darapsky, and Ernst Muller, ibid., 1908 41, 3140 ; A., i, 923ORGANIC CHEMISTRY. 97Further details of the transformations of ethyl diazoacetate will befound in a recently-published &sum6 of the ~ o r k . 9 ~The Terpene Group.Great activity has been shown in the investigation of the terpenes andallied substances during t b past year, and important progress hasbeen made in the establishment of the constitution of several membersof the group.The great services rendered to this branch of chemistry,as to so many others, by the discovery of Grignard's reaction areobvious on an examination of the papers dealing with terpenesyntheses. I t is in connexion with the study of the terpenes, also,that the optical method, the determination of the refractive index anddispersion, has proved of the greatest value. Several cases of apparentexception to the regularities in the relation of refractive index toconstitution have been removed recently, as the result of a morecomplete purification of the substances supposed to be anomalous.There still remain some marked apparent exceptions, some of whichmay be due to.unnoticed isomerisation during the preparation. Suchisomeric changes, usually involving the shift of a double linking, arefrequent. I n a recent instance, the process of heating with quinolineand quinoline hydriodide, a procedure sometimes adopted, is found tocause the rearrangement of methylenecyclohexane to methyl-Al-cyclo-hexene, the double linking changing its position : 93The synthesis of carvestrene94 has now been supplemented by thesynthesis of an isomeride, also of the na-menthadiene series, for whichthe name isocarvestrene is proposed, the two terpenes differing only inthe position of the double bond i n the ring.95 Starting with ethylcyclohexanone-2 : 4-dicarboxylate (I), methylation gives ethyl l-methyl-cyclohexan-6-one-1 : 3-dicarboxylate (II), which loses a carbethoxylgroup on hydrolysis, yielding l-methylcycZohexan-6-one-3-carboxylicacid (111).By reduction to the hydroxy -acid, treatment with hydrogen bromide,and subsequent removal of hydrogen bromide by diethylaniline, an* Ber., 1908, 41, 3161 ; A., i, 924.93 A.E. Faworsky and I. Borgmann, Ber., 1907, 40, 4863; A., i, 15.94 Ann. Report, 1907, 129.95 K. Fisher and W. H. Perkin, jun., Trans., 1908, 93, 1876.REP.-VOL. V. 98 ANNUAL REPOKTS ON THE PROGRESS OF CHEMISTRY.H C0,Et Me C0,Et H Me\/ \/ \/H Hd\/ C0,Et(1.)H2(111.)*2(11.1acid is obtained which is shown to be 1 -methyl-A6-cycZohexene-3-carboxylic acid (IV).Magnesium methyl iodide converts the un-saturated acid into A6-m-menthenol (V), the terpineol of the series,from which isocarvestrene (VI) is obtained by removing water withmagnesium methyl iodide.H2Me Me MeThe new terpene is found to possess somewhat remarkableproperties, the high refraction and dispersion, and the formation of adibromide instead of a tetrabromide, causing it to resemble a terpenewith conjugated double linkings, a constitution which appears to beexcluded by the conditions of the synthesis.The important synthesis of terpineol\ by W. H. Perkin, jun., in1904,96 has been completed by the production of the two activeterpineols, the original product having been inactive. This has beeneffected by the resolution of &I-1 -methyl-A3-cycZohexene-4-carboxylicacid,by crystallisation of its strychnine and brucine salts.97Another interesting synthesis from the same laboratory has beenthat of a terpineol, terpin, and terpene containing a five-memberedring.Qs Ethyl cyclopentanone-3-carboxylate (I) reacts with magnesiummethyl iodide, forming 1 -methyl- 3-isopropenol-A5-cycZopen tene (11)the terpineol of the five-carbon series, which yields the correspondingterpin (111) with acids :co CMe OH-CMe/\ /\VH VH2 p z/\p 2 p 2CH,-CH-CO,Et CH2-CH*CMe2*OH CH,-C K*CMe,*OH(1.) (TI.) (111.)96 Ann.Report, 1904, 116.97 K. Fisher and W. H, Perkin, jun., Trans., 1908, 93, 1871.98 W. N. Haworth and W. H. Perkin, jun., Trans., 1908, 93, 573ORGANIC CHEMISTRY.99A second method of synthesis leads equally to members of thisseries. Ethyl 2 -met hylcyclopentan- 2 -one- 3-dicarboxylate (IV) maybe broken down by hydrolysis to pentane-pyc- tricarboxylic acid (V),the ethyl ester' of which condenses under the influence of sodium,forming ethyl 2-met8hylcycZopentanone-3 : 5-dicarboxylate (VI). Byreduction and addition and subsequent removal of hydrogen bromide, aCHMe*CO,Hco I co CH*CO,HI A(1V.j (V. 1 (VI. )CC),Et YH QH Me /\ ?€I2 ?Me*CO,Et ?HaCH,-CH*CO,Et CYH,*CO,H CH,-CH*CO,Etmixture of unsaturated esters is obtained, from which ethyl l-methyl-A4-cycZopentene-2-carboxylate (VII) has been isolated by an ingeniousprocess. Magnesium methyl iodide converts it into the terpineol (VIII)The corresponding terpene, l-methyl-2-isopropenyl-A4-cycZopenteneCHMe CHMe CHMe/\ /\ /AfiH yH*C'O,Et g€€ QH*CMe,*OH EH FH-CMe:CH,CH--CH, CH-CH, CH--CK,(VII.) (VIII. ) (1X.j(IX) is obtained from this by dehydration [with succinic anhydride.This hydrocarbon is possibly identical with that obtained by Semrnlerfrom sabina ketone.It is only possible to make a selection from the very numeroilspapers dealing with terpene chemistry. Terpinene, although rarelypresent in natural products, is of interest from the fact that it isfrequently produced by the action of acids on other terpenes, aridtherefore appears to be one of the most stable members of the group.It has been previously suggested that terpinene is identical withcarvenene, and this is confirmed by the conversion of carvenone (I) byway of the oxime into 2-amino-A3-menthene (11), distillation of thephosphate of the latter yields a pure A' '3-menthadiene (carvenene)(111), which, from its conversion into terpinene nitrosite, appears to beidentical with ter~inene.9~ Whilst fornier preparations of terpineneMe Me Me(1.1 (11.1 (111.)have failed to show the optical exaltation corresponding with the con-99 C. D. Harries and R. Majima, Ber., 1908, 41, 2516 ; A., i, 733.H 100 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.jugated double linkings, i t is exhibited by the carvenene prepared bythis method.1 A study of the oxidation products of a- and P-terpinenealso leads to the conclusion that the hydrocarbon yielding the nitrositeis A" "menthadiene (a-terpinene).2The new terpene obtained from Cyprus origanum oil, and describedas origanene, has also the properties of A1'3-p-menthadiene, and showsoptical exaltation.Its properties agree fairly well with those ofcarvenene or a-terpinene.32-Methylcarvenene has been prepared by the action of magnesiummethyl iodide on carvenone.*The first synthesis of a-phellandrene from a substance containing asmaller number of carbon atoms has been accomplished by treating~sopropyl-A2-cycZohexen-4-one, prepared by the isomerisation of sabinaketone, with magnesium methyl iodide.5Turning now to bicyclic terpenes, a comparative study of the actionof hydration in breaking down bridged linkings has been made byWallach.6 The addition is very commonly accompanied by molecularrearrangement, sabinene and pinene hydrates, for instance, probablyfirst losing water, which is then added on in a different position,yielding terpinene-4-01 and a-terpineol respectively.The relativebehaviour of three- and four-atom rings on hydration confirmsPerkin's conclusion, derived from experiments of a different kind,that the relative stability of cyclopropane and cyclobutane ringsdepends much more on the nature and position of the attached groupsthan on the number of carbon atoms in the ring.Of the two possible formula for umbellulone proposed by Tutin (I)and by Semmler (11) respectively, the first is preferred, since theCH,-CH-GOCHRle,I CH,*C-- coI/ I CH--CMe=CHsecond would require that tetrahydroumbellulone should be menthone,whereas the mixture of stereoisomeric tetrahydro-derivatives actuallyobtainediis unlike menthone.1 J.W. Briihl, Ber., 1908, 41, 3712; A., ii, 1002.2 0. Wallach, Annalen, 1908, 362, 261, 285; A., i, 811, 813 ; Semmler,however (Rer., 1908, 41, 4474 ; A., 1909, i, 110), dissents from this conclusion, andiuaintains the A1:4 constitution for terpinene.8 S. S. Pickles, Trans., 1908, 93, 862.4 H. Rupe and F. Ernmerich Ber., 1908, 41, 1750 ; A., i, 556.6 Artnalen, 1908, 360, 82; A., i, 429.7 F, Tutin, Trans., 1908, 93, 252 j F. W. Semmler, Ber., 1908, 41, 3988 ; A.,0. Wallach, Annalen, 1908, 359, 266 ; A., i, 424.1909, i, 38ORGANIC CHEMISTRY. 101Santene has received the constitution (I) on the ground that gentleOxidation with permanganate yields a glycol (11), further oxidationgiving a diketone (111), the constitution of which is proved by itsoxidation to trans-yclopentanedicarboxylic acid.8CMe*CH-CH, OH- CMe- CH-CH, I\ie*CO*CH-CH,I I IMe*COdH-CH,(1.1 (11.1 (111.)A further attempt has been made to elucidate the nature of theisomerism of the two modifications of isonitrosocamphor by the studyof the action of diaz~rnethane.~ This reagent converts the unstableinto the stable modification, and the latter into the AT-ether, which ispossibly C,H,,< C:NMe:o. I The presence of a nitroso-group is im-probable, from the absence oE a blue or green colour and of theLiebermann reaction, and the production of the N-ether is in betteragreement with the formulation of isonitrosocamphor ascoSeveral investigations have been directed to the establishment ofthe constitution of fenchone.Three formula have been proposed, duerespectively t o Wallach (I), Semmler (11), and Glover (111). Of these,CH,*CH-CHMe CH,*CH--C!Me, CH,*CH,-CMeCH,.CH-CO CH,*hMe* CO CH,*C'H--CO(1.1 (11.) (111.)the first formula is very similar to that of a-methylcamphor, and thesecond to that of aa-dimethylcamphor. Both are open to the objectionthat fenchone behaves very differently from camphor in many of itsreactions. The oxidation of fenchene and fenchone lo leads to inde-cisive resuIts, it being difficult to reconcile the reactions observed witheither of the proposed formuls.The production of dihydrofencholen-I 4 I I CMe, I CMe2 I I GI12 I I Iamide, NH2*Co*?Me* CH2>CH*CHMe,, by the action of sodamideCH,-CH,on fenchone, and the formation of apofenchene,fiMeoCH2>CH#CHMe, CH--CH,(the constitution of which is established by its oxidation through a8 F. W. Semmler and I(. Bartelt, Ber., 1908, 41, 385, 866 ; A , , i, 195, 355.M. 0. Forster and IT. Holmes, Trans., 1908, 93, 242.lo 0. Wallach, Annalen, 1908, 362, 174 ; A., i, 809102 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ketonic acid t o P-isopropylglutaric acid) from the latter, is consideredto favour Semrnler’s formula.aa-Dimethylcamphor (111) should be capable of preparation fromdimethylcampholide (I) by addition of potassium cyanide, hydrolysis todimethylhomocamphoric acid (II), and distillation of the calcium salt.A comparison of this compound with fenchone should then be ofconsiderable interest.It mas found, however,12 that although theCH,*CH-CMe, CH, *C Ha CMe,*CO,H CH, CH-CMe,CH,*CMe --CO CH,*CMe*CO,H CH,*CMe-GO(1. ) (11.1 (111.)required dimethylcampholide was readily obtained by the action ofmagnesium methyl iodide on camphoric anhydride, it mas impossiblet o cause combination with potassium cyanide, isomerisation to anunsaturated acid always taking place.The preparation of a number of derivatives of a-methylcamphor hasshown 13 that this compound is entirely similar to camphor, and con-sequently unlike fenchone. or-Methylcamphor forms bromo- andsulpho-derivatives closely resembling those of camphor.The synthesis of an isomeride of /3-pinene from nopinone, describedin last year’s Report (p.132), has been modified, and decomposition ofthe unsaturated acid is now found to yield fenchene or P-pinene,according to the conditions of dehydration of the nopinolacetic acid,The fenchene is probably formed by intramolecular change of P-pinene.Since P-pinene yields bornyl chloride with hydrogen chloride, bothcamphor and camphene may be prepared from it, this being the firstsynthesis of camphor from a compound (nopinone) containing a smallernumber of carbon atoms.14A cyclooctadiene was obtained by Willstatter and Veraguth in 1905from the alkaloid $-pelletierine, and its ready polymerisationsuggested the presence of conjugated double linkings in the mole-cu3e.15 This compound derives its principal interest from the fact thatcaoutchouc is supposed to be a polymerised dimethyl derivative of thesame hydrocarbon.Since, however, the A1:5-members of this seriesalso polymerise readily, this constitution also becomes a possible one.An examination of the diozonide, which is hydrolysed by waterto succindialdehyde, proves that the double linkings are in the1 : 5-position : 16I I I I UMe, 1 >. I y e 2 I y e 2 I11 L. Bouveault and Levallois, Compt. rend., 1908, 146, 180 ; A., i, 193.l 2 G. Komppa, Ber., 1908, 41, 1039 ; A.: i, 352.l 3 W. H. Glover, Trans., 1908, 93, 1285.l4 0. Wallach, Aiznalcn, 1908, 363, 1 ; A., i, 997.19 Ann. Report, 1905, 121. C. D. Harries, Ber., 1908, 41, 671 ; A ., i, 254ORGANIC CHEMISTRY. 103QH,*CH:CH*$!H, -+ CH,*CH:CH' CH,The investigation ofFH,-CH-~H.~H, CH,*C'HO CHO-~H,--f I CH,*CH*CH*CH, CH,*CHO CHO*CH,'\/0,the polymerisation products showed that therewas no direct connexion between the constitution of the hydrocarbonand that of caoutchouc. The simplest polymeride, dicyclooctadiene,appears from its behaviour with ozone to have the constitution :~HDCH2*CH,*~-~H-CH,*CH,*QH,CH*CH,*CH,*CH CH,*CH,*CH=CH.Connected with the ethereal oils, although not a member of theterpene group, is elemicin, isolated by Semmler from elemi resin.17This is proved to be 3 : 4 : 5-trimethoxy-1-allylbenzene (I). Whendistilled over sodium, i t is converted into the isomeric propenylCH,-CH:CH, CH: CH-CH,/\ /\\/ OMd lOMe OMe(/OMeOMe(1.) (11.)OMederivative, isoelemicin (11).Permanganate oxidises it to 3 : 4 : 5-tri-methoxybenzoic acid. Sodium and alcohol reduce both elemicin andisoelemicin to 3 : 5-dimethoxy-1 -n-propylbenzene, the p-methoxyl groupbeing eliminated,CrystaZZine Liquids.The property of existing in a liquid phase, which exhibits certain ofthe optical properties of crystalline solids, for example, double refrac-tion, is possessed by many organic substances, and the relationshipwhich in all probability exists between their chemical constitutionand the development of this crystalline liquid condition is at presentunder investigation. .In p-methoxycinnamic acid, which exhibits this peculiarity, theproperty is probably due to the presence of the groupCH,*O*C,R,*CH:CR,for dianisyltetrylene, which consists of two of these radicles, stillretains the power of forming an anisotropic liquid.p-Methylamino-benzaldehydephenylhydrazone, CH,*NH*CBH,*CH:N*NH*C~H~, s-di-F. W. Qemmler, Eer., 1908, 41, 1768, 1918, 2183, 2556 ; A , i, 557, 558, 664,734104 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ethylbenzidine, C,H,*NH*C,H,*C,H,*NH*C,H,, and p-methoxy-cinnamaldazine,CH,*O* C6H,*CH :CH* CH:N*N: CH*C H: CH* C,H,*O*CH,,which likewise display this phenomenon, are either para-substitutedcompounds or possess long, straight chains.ls Cholesterol derivatives,which frequently exhibit the crystalline liquid condition, probablycontain an asymmetric group and a long, straight chain.The turbidity of many of these crystalline liquids is not due tosuspended impurities, emulsions, or inclusions of foreign substances,but is a natural consequence of the presence of differently situatedanisotropic, crystalline fragments. Moreover, perfectly clear crystalline liquids have now been prepared, some of which maintain theirtransparency in any position, whilst others are clear or turbidaccording to the incidence of the light.The chemical constitution of substances forming crystalline liquidphases warrants the view that a linear structure favours the develop-ment of the property, whilst a cruciform or many-branched configurationinhibits this condition.Para-substituted benzylidene-p-amino-a-alkylcinnamates having thefollowing general formula have been prepared and examined fromthis standpoint :para-x.C6H, * CHON C,H,*CH: CO,R.YThe tendency to develop the crystalline liquid condition reaches itsmaximum when an ethyl or a n-propyl group is introduced at R (theester radicle); it is also increased by the replacement of methoxyl atX (the para-substituent) by ethoxyl or phenyl. The lengthening ofthe side-chain Y in the order methyl, ethyl, and phenyl inhibits to anincreasing extent the property of exhibiting the liquid crystallinecondition. The property of circular polarisation is developed to aremarkable extent by the introduction of an active amyl group intoposition R.The ethyl p-azoalkylcinnamates and p-azoxyalkylcinnamates re-semble the foregoing azomethine derivatives in respect of thisproperty of assuming the crystalline liquid condition.19The simpler azoxy-derivatives (azoxybenzene, p-azoxyphenetole,the three isomeric azoxytoluenes, and azoxyanisoles) can also existin the anisotropic liquid condition, and it is stated that these modifi-cations differ from the ordinary varieties in certain chemical properties.Among other anomalous reactions, they give the Liebermann nitroso-coloration, and are not transformed into hydroxyazo-compounds.20l9 D.Vorlander, ibid., 2033 ; A., i, 641.2o T. Rotarski, zbid., 865 ; A., i, 374.T. Rotarski, Ber., 1908, 41, 1994 ; A . , i, 640ORGANIC CHEMISTRY. 105These differences are sufficiently remarkable to warrant a moreextended investigation.Optical Activity.In recent measurements of the optical activity of organic compounds,the theoretical connexion with the degree of asymmotry of the mole-cule, as expressed by Guye’s “ asymmetry product,” 21 has beenfrequently referred to and discussed, without, however, any verydefinite conclusions having been yet reached.A recent physicalinvestigation 22 leads to the conclusion that the variation in the valueof the expression with the temperature and the wave-length of thelight used must be taken into account. The author states that if asubstance containing a single asymmetric carbon atom could be found,the temperature of reversal of sign of which could be determined, ameans would be provided of testing the validity of the modified Guye’sequation, but this test has not yet been applied.The difficulty of finding any simple formula for the relation hasbeen further illustrated by the examination of a series of salts, all ofwhich contained p-bromophenyl, methyl, and allyl, the remaining groupbeing ethyl, n-propyl, isopropyl, or isoamyl.As in similar seriesexamined previously, no simple relation was dis~overed.~a Theinfluence of the constitution of the substituting groups may alsobe so great as to outweigh that of mass.24The influence of the introduction of unsaturated groups into themolecule has also been further studieda25 The alkaloid salts of anumber of acids were examined, and the rule that the change fromthe saturated to the ethylenic linking produces an increase in therotatory power was confirmed.The triple linking, however, maygive rise to a higher or a lower value in different cases, the directiondepending mainly on the asymmetric part of the molecule. Neithercould any rule be found for the comparative influence of cis- and trans-configurations. The presence of several unsaturated groups stillfurther increases the optical activity, the relative nearness of theunsaturated linkings having an important effect. The investigationwas extended to include sulphur derivatives in which the valency ofthe sulphur, and therefore the amount of its residual affinity, wasvariable. The progressively increasing unsatnration’ of the sulphurin the sulphone, R,SO,, the sulphoxide, R,SO, and the sulphide, R,S,is accompanied by a small increase in the rotation, but the conjugationAnn.Report, 1907, 178.23 E. Bose, Physikal. Zeitsch., 1908, 9, 860 ; A., 1909, ii, 2.23 H. 0. Jones and J. R, Hill, Trans., 1908, 93, 295.24 R. W. Everatt, ibid., 1225, 1789.T. P. Hilditch, ibid., 700, 1388, 1618106 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.of two bivalent sulphur atoms in the disulphide, RSoSR, produces arelatively enormous effect. The results obtained with alkaloid saltsof aromatic sulphonic and sulphinic acids follow the same rule if thelatter are assumed to contain sexavalent, and not quadrivalent,sulphur, an assumption for which there is chemical evidence. If bothseries oE compounds contain sexavalent sulphur, the sulphonic acidswill be the less saturated, owing to hydroxyl, possessing some residualaffinity, replacing hydrogen : R*SO,*H -+ R*SO,*OH.No certain occurrence of optical activity has yet been observed ina compound in which the asymmetric carbon atom is not attached toother carbon atoms.Two compounds of very simple structure, chloro-sulphoncetic acid, SO,H*C€tCl*C!O,H, and chlorobromomethane-sulphonic acid, CHClBr*SO,H, one containing only two and theother only a single carbon atom, have been prepared and combinedwith various active bases, but without any resolution into opticalisomerides being obtained. If isomerism in such a case is possible,the separation of the isomerides is evidently a very difficult one.26The suggestion of van't Hoff, that enantiomorphous forms mayexist of compounds containing no asymmetric atom, but derivingtheir asymmetry from the general structure of the molecule, as inallene derivatives,a,>c:c:c<;,and that such isomerides should exhibit optical activity, is one ofgreat theoretical interest, which has not yet been put to an experi-mental test.An attempt to prepare allene derivatives of the above type,capable of combining with active acids or bases, having proved un-successful, compounds have been selected in which symmetrical closedrings replace the double linkings, of which 1 -methylcycZohexylidene-4-acetic acid,Me ~ > c < ~ ~ : ~ ~ > c : c < , " , H92is a typical example. The synthesis of this acid27 has proved to be adifficult problem, and the product obtained is impure; the test ofresolution has therefore not yet been applied.The problem is ofsuch importance for stereochemical theory that it seems advisableto mention the investigation in this place, in spite of its incompletestate.The problem of . a complete asymmetric synthesis, that is, the arti-ficial production of an optically active substance without the interven-tion of an optically active reagent, has been frequently attacked from26 W. J. Pope and J. Read, Trans., 1908, 93, 794.27. W. H. Perkin, jun,, and W. J. Pope, ihid., 1075ORGANIC CHEMISTRY. 107the physical as well as from the chemical side. The use of circularlypolarised light has often been suggested and employed for this pur-pose, so far with entirely negative results.I n the latest attempt inthis direction 28 the authors point out some of the conditions whichmust be fulfilled in order that such an attempt may have a prospectof success. The reaction by which the substance is produced must beone which is brought about by light. A reaction which proceedsindependently of illumination is not likely to be affected by polarisa-tion of the light falling on the reacting substance. A suitable reactionwas found in the removal of carbon dioxide from substituted succinic,malonic, and cyanoacetic acids by light in the presence of uraniumsalts, as, for instance :Me>c<g;2H Et -+ ",> c<"$"'CMeCl* CO,H E>c<H Me CMeCl*CO,H -+C1>C<Co,RNo optical activity of the resulting product was, however, observedwhen the light was polarised before entering the solution, and theattempt is therefore so far unsuccessful.A remarkable series of observations is recorded with reference tothe separation of active components from a dZ-mixture.29 A super-saturated solution containing, for instance, dZ-sodium ammoniumtartrate may be caused t o crystallise by the addition of a crystal ofthe active modification of a tartrate isomorphous or isodimorphouswith the dissolved salt, and the crystals separating will have the samesign as the crystal used for inoculation, !This is not surprising whenthe similarity in crystalline structure is taken into account. It isfurther stated, however, that it is not necessary that the crystal usedfor inoculation should be optically active.The experiments wereprincipally made with glycine, and when added to a supersaturatedsolution of dl-asparagine, deposition of active asparagine was broughtabout. It was impossible to predict whether the d- or the Z-formwould separate, but the same crystal of glycine always brought aboutcrystallisation of the same isomeride. It is suggested that glycinecrystals are hemihedral. If these experiments should be confirmed,and are not found to be due t o the presence of optically activesubstances in the glycine employed, a great advance will have beenmade in the artificial production of active compounds by an asymmetricsynthesis.Another line of attack which has been adopted by several workers,38 F. I-Ienlc and H. Haakh, Ber., 1908, 41, 4261 ; A., 1909, i, 6.29 I.Ostromisslensky, ibid., 3035 ; A . , ii, 913108 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.is that of preparing an asymmetric compound in an active solvent,the molecules of which might reasonably be expected to exert adirecting influence on the course of the reaction. An attempt of thiskind has now been made in the nitrogen series, but without success.Benzylmethylaniline was combined with ally1 iodide in a number ofoptically active solvents, such as d-limonene, Z-menthol, and Z-menthylchloromethyl ether, but in no case was an active ammonium iodideobtained.30I n the continued series of investigations of the influence of solventson the optical activity of ethyl tartrate, Patterson and his co-workershave examined a large number of halogen and nitro-compounds.31Amongst the halogenated solvents, such as alkyl iodides, chloroform,acetylene tetrabromide, etc., many exceptions were again found to therule that specific rotation varies inversely with solution volume.Witharomatic halogen derivatives, well-marked minima of rotation wereobserved at certain concentrations. The changes of rotation withtemper'ature were also examined, and solutions in a-bromonaphthalenewere found to have a maximum rotation at 94", the case resemblingthat of dilute solutions of alkyl tartrates in water, the rotation-temperature curve of which also passes through a maximum. A widerrange of activity was observed when aromatic nitro-compounds wereused as solvents, the specific rotation at infinite dilution in a-nifro-naphthalene being about + 65", and in s-trinitrobenzene about - 30'.The relation between the maximum rotation and the temperature atwhich it occurs is fairly independent of the nature of the solvent andof the concentration.As in the problem of the relation of activityt o asymmetry, it is evident that in spite of the vast quantity ofexperimental material that has been accumulated, the completetheoretical explanation is far from having been attained.Two forms of asymmetry have been observed in nitrogen compounds,the first, in substances containing tervalent nitrogen doubly linkedwith carbon or nitrogen, as in oximes and diazo-compounds; thesecond, in substances containing quinquevalent nitrogen, So far,activity has only been observed in the latter class when the five sub-stituting groups are all different, although four different groups shouldsuffice to produce asymmetry.A new class of active substances hasnow been found, in which two valencies of the nitrogen atom are unitedto the same, or similar, atoms. Methylethylaniline oxide, for example,has been resolved into two active components by conversion into the30 E. Wedekind and 0. Wedekind, Ber., 1908, 41, 456 ; A., i, 255.31 T. S. Patterson and D. Thoinson, Trans., 1908, 93, 355 ; T. S. Patterson andD. P. McDonald, ibid., 936; T. S, Patterson, ihid., 1836ORGANIC CHEMISTRY. 109d-camphorsulphonate.very weak base, must containThe solution of the active oxide, which is ain either case, the number of different groups united with the nitrogenatom is only four. If the compound present in the solution of the basehas the second formula, the positions of the two hydroxyls cannot beidentical, since the addition of hydrochloric acid produces a salt havingthe same activity as the base. It is concluded that four of thevalencies of a quinquevalent nitrogen atom are directed, like those ofa carbon atom, to the angles of a tetrahedron, the fifth (ionisable)valency being mobile.Isomerism would then occur whenever thegroups attached by the four fixed valencies were different.32Active piperidine derivative’s having a large substituent in position2, and an alkyl attached to the nitrogen atom, have been found to yieldtwo stereoisomeric quaternary salts with alkyl haloids, only a singleproduct being obtained when the substituting group in position 3 isabsent or small.This has been further confirmed by the examinationof U- and P-pipecoline derivatives, benzyl haloids being added to theZ-bases.Turning now from the production of active substances to theirresolution and racemisation, the preparation of the two active forms ofbenzoin has now been accomplished. Z-Mandelic acid is converted intothe amide, and this is combined with magnesium phenyl bromide,yielding Z-benzoin. The d-isomeride is prepared in a similar way.With the exception of laevulose, no keto-alcohol had previously beenresolved into itts optically active c0mponents.3~The autoracemisation of active ammonium salts in solution has beenthe subject of but it now appears certain from cryoscopicmeasurements and from comparative determinations of the velocity ofdissociation and of change in rotation, that the observed alteration inrotation is due to the breaking up of the ammonium salt in solutioainto tertiary amine and alkyl halide.A case of partial racemisation has been observed in the hydrogentartrate of hydroquinaldine.36The Walden inversion has been the subject of several investigations.33 11.Schaltz, Cibid., 2005 ; A., i, 678.a4 A. McKenzie and H. Wren, Trans., 1908, 93, 309.35 E. Wedekind, 0. Wedeliind, and F. Yaschke, Ber., 1908, 41, 1029, 2659 ; A.,36 A. Ladenburg and W. Hemnann, ibid., 966 ; A., i, 364.I n neither case were optical isomerides obtained.33J. Meisenheimer, Ber., 1908, 41, 3966 ; A., 1909, i, 20.i, 334, 722 ; 8.von Halban, ibid., 2417 ; A., i, 627110 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The replacement of halogen by hydroxyl or methoxyl in phenylchloro-acetic acid has been examined from this point of view, the resultsbeing expressed in the two following schemes :1.11.d-OH GHPh*CO,H4I NaoHd-CHClPh*CQ,HPCl5$? E-CHClPh*CO,HI AgaOPClS +NILOH ?$ I-OH*CHPh*CO,HAg20AgaO and Me1d-OH*CHPh*CO,H -- 3 d-OMe*CHPh*CO,HPI5MeONaZ-CHC1Ph*C02H -3 I-OMe*CHPh*CO,HIt appears from this that in the interconversion of the mandelic acids,sodium hydroxide behaves abnormally and silver oxide n0rmally.3~E. Fischer has also continued his investigations on the samesubject,38 and finds that valine (a-aminoisovaleric acid) remainsoptically unchanged after conversion into the bromo-fatty acid andback again into the amino-compound. It is suspected, however, thatthis result is due rather to a double Walden inversion than to itsabsence.This exceptional behaviour is attributed to the influence ofthe isopropyl group. Active aminophenylacetic acid is racemised sorapidly by nitrosyl bromide or nitrous acid that pure active productscould not be obtained.It is oniy possible t o refer to a few more papers dealing with stereo-isomerism. A method of determining the configuration of a-dioximesis furnished by the fact that of the possible isomerides, only the syn-modification has the property of forming complex metallic d i o ~ i r n i n e s .~ ~The method has been tested in a number of cases, arid has beenapplied to several a-dioximes of previously unknown configuration.No marked difference is to be found between the dielectric constantsof d-, L, and i-modifications of asymmetric compounds. Ethylracemate, however, has a much greater absorptive power for electricwaves than the tartrate, and, since hydroxylic groups are theprincipal cause of such absorption, this suggests that the hydroxyls aremainly concerned in the formation of the racemic compound.40The viscosity of a solution of a racemic salt is always less than thaty7 A. McKenzie and G. W. Clough, Tmns., 1908, 93, 811.3a E. Fischer and H. Scheibler, Bey., 1908, 41, 889, 2891 ; A., i, 324, 857 ;3g L.Tschugaeff, ibid., 1678 : A., i, 554. * A. W. Stewart, Trans., 1908, 93, 1059.E. Fischer and 0. Weichhold, ibid., 1286 ; A., i, 419ORGAXIC CHEMISTRY. 111of the active components, but the difference is very small. Thetransition point of the racemate also appears as a break in thetemperature-viscosity curve.41Some Reactions o j the Cyclic Hydrocarbons and theiv Derivatives.1. Aromatic from Hydroaromatic Compounds. --The conversion ofaromatic substances into hydroaromatic derivatives by hydrogen anda catalyst has already been mentioned. An interesting case of theconverse change has recently been worked out, in which phenolhas been produced from cyclohexanol in such a way that the inter-mediate products could be isolated, showing the gradual transitionfrom the saturated to the aromatic ring.cycZoHexanone, obtained by the oxidising action of bromine orchlorine from the saturated alcohol, was brominated to 1 -bromocycZo-hexanone. The elimination of hydrogen bromide led to h2-cyclo-hexenone ; addition of bromine t o this substance, followed by removalof hydrogen bromide from the resulting dibromide, finally gave phenol :Ethyl cyclohexane-2-carboxylate is converted into salicylic acid by aprecisely similar series of changes.422. Oxidation of Aromatic Substances.-Although the aromatichydrocarbons (benzene, naphthalene, etc.) are not themselves oxidisedto definite products by chromyl chloride yet, the correspondingaldehydes are readily obtained with this oxidising agent from the m-and p-nitrotoluenes and the three isomeric chlorotoluenes.I n asimilar manner, diphenylmethape and triphenylmethane are convertedquantitatively into benzophenone and triphenylcarbinol respectively.43Caro’s acid has been successfullyapplied to the preparation of tertiaryamine oxides ; tetramethyldiaminodiphenylmethane and hexamethyl-triaminotriphenylmethane having been thus converted into tetra-methyldiaminodiphenylmethane dioxide (I) and hexamethyltriamino-triphenylmethane trioxide (11) respectively.(1.1 (11;)41 A. E. Dunstan and F. B. Thole, Tmns., 1908, 93, 1815.la A. Kotz and C. Gotz, Annulen, 1907, 358, 183 ; A., i, 173.43 H. D. Law and F. M. Perkin, Trans., 1908, 93, 1633112 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The reactions of these amine oxides are of interest ; sulphur dioxideand nitrous acid convert them respectively into the sulphonic acids andnitro-derivatives of the original tertiary bases.In these reactions theoxygen atoms of the di- and tri-oxides become involved, and the sub-stituents take up ortho-positions with respect to the nitrogen atoms.44The oxidation of acetyl-p-phenylenediamine by Caro’s acid results inthe formation of p-nitrosoacetanilide (green plates, m. p. 1 7 5 O ) , whichis transformed by water into its colourless, bimolecular polymeride(m. p. 181°).45The technically important conversion of p-nitrotoluene and itsderivatives into stilbene compounds has been further investigated, theresults showing that the following scheme of condensation in two stages,accompanied a t each stage by aerial oxidation, furnishes an explanationwhich is generally true of these condensations.condensation1st stage2CH3*CGH4*N02 -- 3$lH,*C,H4*N02 condensation EH*C,H,*NOCH,*C6H4*N0 211dstage CH*C H *NO __+ a 4$oxidation j,oxidatiuny3,*C6H4*N0, EH*C,H,* NO2CH2*C,H4*N0, CH*C6H,-N024-Ni tro-o-xylene and 4-nitro -2 -met hoxy + duene with methyl -alcoholicpotash and aerial oxidation gave chieflj 4 : 4’-dinitro-2 : 2’-dimethyl-stilbene and 4 : 4‘-dinitro-2 : 2’-metboxystilbene.Similar changesoccurred with 4-nitro-o-toluic acid and 2 : 4-dinitrotoluene, but in theformer case sodium hypochlorite, and in the latter, iodine wereemployed as oxidising agents.4G3. Reduction of Aromatic iVitro-conLiuozcnds.-A new phase in thereduction of nitro-groups has been observed by G.Heller andA. S o ~ r l i s , ~ ~ who also correct an earlier statement of Bamberger’s,4*that his so-called ‘‘ agnotobenzaldehyde ” has the formulaCOH* C,H,*N(OH)* O=N(OH)*C,H,*COH.This substance is really a molecular compound of o-nitrobemaldehydeand o-hydroxylaminobenzaldehyde,COH~C,H,*N0,,COH*C6H,*NH OH.They find, however, that the reduction of o-nitromnndelonitrile withE. Bamberger and L. Rudolf, Ber., 1908, 41, 3290 ; A . , i, 1011.J5 J. C:Cain, Trans., 1908, 93, 686.46 4. G. Green and J. Baddiley, ibid., 1721.47 Ber., 1908, 41, 373 ; A., i, 208.48 Ibid., 1906, 39, 4252; A., 1907, i, 163ORGANIC CHEMISTRY. 113zinc dust leads to the formation of a molecular compound of hydroxyl-aminomandelonitrile and dihydroxylaminomandelonitrile,CN*CH(OH)*C,H,*NH*OH,CN *CH(OH)*C,H,*N(OH),.This product on treatment with phenylhydraxine yields hydroxy-isatinphenylhydrazone and the a- and /3-phenylhydrazones of isatin.The reduction of nitro-compounds by alcoholic ammonium sulphide,which was first practised by Zinin in 1842, has been further developedin recent years.I n 1902 and subsequently, J. B. Cohen and othersshowed that hydroxylamino-derivatives were produced by the partialreduction of trinitrobenzene and trinitrotoluene with hydrogensulphide in the presence of a small amount of ammonia. They alsofound that chloronitro-compounds and alkyl dinitrobenzoates likewisegave hydroxylamines under similar condition^.^^ The fact that thischange, R*NO, -+ R*NH*OH, occurs generally, has been furtherdemonstrated by reducing the simpler nitro-compounds with alcoholicammonium sulphide in the cold, when excellent yields of the correspond-ing arylhydroxylamines were obtained.For example, a-nitronaphtha-lene gives a-naphthylhydroxylamine, from which a-nitrosonaphthaleneis easily produced by oxidation with silver oxide or lead peroxidein anhydrous solvents.50It has generally been assumed that the azoxy-compounds formedduring reduction by a condensation of the nitroso- and hydroxylamino-derivatives can only arise in neutral or alkaline solutions. But incertain cases this condensation may occur even in the presence ofmineral acids. The reduction of certain substituted nitro- anddinitro-compounds has been systematically examined, the resultsshowing that (1) condensation and reduction .of the inter-mediate nitroso- and hydroxylamino-derivatives both proceed at ameasurable rate whether the solution be acid, neutral, or alkaline ; (2)condensation is induced by the presence of tervalent nitrogen in thefree arylhydroxylamine, the quinquevalent nitrogen of the hydroxyl-amine salt being incapable of condensation.514. Pormation of Aromatic Hydroxylic Compounds.--In practice theintroduction of hydroxyl into the nucleus of an aromatic hydrocarbonrequires several operations, although small amounts of nitratedphenols are produced during many nitrations.According to a recentpatent, a mixture of benzene, strong nitric acid, and mercuric nitrategives a fairly good yield of picric acid, together with smaller amountsof nitrobenzene and o-nitrophenol.52The nitration process may be divided into two phases, in the second49 Trans., 1902, 81, 26 ; 1905, 87, 1257.5o R.Willstatter and H. Kubli, Ber., 1908, 41, 1936 ; A . i, 522.59 B. k’liirscheim and T. Simon, Tram., 1908, 93, 1463.51 R. Wolffenstein and 0. Boters, D.R.-P. 194883 ; A , , i, 629.REP.-VOL. V. 114 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.of which water is usually eliminated from the additive compoundformed in the initial phase : -’” ‘E” nitro-compound.\ €3 C*NO,/ NO,‘<OH--3 I \CHC<HThe alternative elimination of nitrous acid leads to the productionof phenolic compounds.The direct introduction of hydroxyl groups into the anthracenemolecule has been effected, and alizarin, free from by-products, maybe produced by heating anthraquinone with potassium chlorate andaqueous alkali hydroxide at 200°.53Similarly, anthrarufin (1 : 5-dihydro~yant~hraquinone) and chrysazin(1 : S-dihydroxyanthraquinone) yield respectively the 1 : 2 : 5- and1 : 2 : 8-trihydroxyanthraquinones when they are heated at 180-185Owith sodium nitrate and aqueous alkalis, a mixture of potassium andsodium hydroxides giving the best result.54Ring Pownation.The general tendency which undoubtedly exists for the formationof five- or six-membered rings in preference to those containing feweror more components, is usually explained in terms of von Baeyer’sstrain hypothesis, to which reference was made in last year’s Report(p.138). The new cases of ring formation observed during the pastyear furnish further confirmation of the validity of this general-isation.A striking illustration of the way in which the above tendencyaffects the properties of straight-chain compounds is afforded by astudy of the aminoketones having the general formulaC,H,*CO*[CH,],*NH,.A general method has been devised for preparing these substances,which consists in converting the phthalimino-aliphatic acids into theirchlorides, and then condensing these with benzene in the presence ofaluminium chloride : 55CO,H* [CH,],*N:C,H,O, -+ COC1*[CH27,*N:C,T~,0:, -+C,~~*CO*[CH2~2*,~:C,H40, -+ C,€15*CO*[CH,]z*NH, (I).Acid hydrolysis of the ketone leads to fission a t the double linking,63 D.R.-P.116526 ; A., i, 191.64 D.R.-P. 196980 and 195028 ; A., i, 807.65 S. Gabriel, Ber., 1907, 40, 2649 ; 1908, 41, 1127 ; A., 1907, i, 625 ; 1908, i,464ORGANIC CHEMISTRY. 115when phthalic acid and the salt of the aminoketone (I) are set free.It was at once seen that the stability of these compounds variedconsiderably with the value of x. The a-aminoketones correspondingwith x = 1 are only known in the form of their salts. When liberatedtherefrom, they undergo simultaneously condensation and oxidation, sothat a six-membered pyrazine ring is produced.N J 3 2 N\/NH, NThe P-aminoketones, such as diacetoneamine,CH,*CO*CH,*C(CH,),*NH,,are stable, and can be isolated without showing any tendency toundergo cyclic condensation into four- or eight-membered rings.They-aminoketones are so unstable that even in the form of theirhydrochlorides, condensation occurs with the production of a five-membered ring ; thus phenyl y-aminopropyl ketone gives rise t o2-phenylpgrroline :The 6-aminoketones are as unstable as the y-compounds. I npreparing one of these, the following series of operations mas carriedout :C,H,O,:N* [CH,],*CH( CO,*C,H,), -% C8H402: N*[CH2],* C0,H z?(C6H6 A1C13) C,H,O,:N*[ CH, J,*COCI --+- C8H402 :N* [CH,],*CO*C,H,.Hydrolysis of the final product, &phthaliminovalerophenone, gaverise, not to &aminovalerophenone, but to %phenyltetrahydropyridine,the six-membered cyclic condensation product :C,H,* F==N---CH,CH,-CH,-CH, --3 CH,*CH,*CH,'C,H5*7:0 H,N*FH, IThe eaminoketones, for example, eaminocaprophenone, which mightgive rise to seven-membered rings, are quite stable under conditionsin which the y- and 6-aminoketones condense to pyrrole and pyridinederivatives respectively.56The stabiIity of P-bubstitnted ketones disappears in the presence ofa reagent capable of condensing so as to form a five-membered ring.Methyl P-chloroethyl ketone condenses in this way with hydroxyl-amine, phenylhydrazine, or any of the reagents used in detectingcarbonyl oxygen.6G S.Gabriel arid J. Colman, Ber., 1908, 41, 2010, 2014; A., i, 648, 649.1 116 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.An isooxazoline, N< CH*(?H2 is produced with hydroxylamine, 0 --C*CH,’whilst hydrazine and phenyl hydrazine yield pyrazolines,67N :CH* FH2XN-- C* CH,.A methyl group situated in an aromatic nucleus generally preservesits inert paraffinoid character, but the tendency for ring formationmay become sufficiently great to overcome this inertia, and themethane carbon atom then becomes involved in the formation of itnew ring,When the nitrosoacyl-o-toluidines are gently warmed in an anhydroussolvent, a condensation of this kind occurs and an indazole isDroduced.58\I--CH,The intervention of a methyl substituent also occurs when a mixtureof 2-methyl-1 : 2’-dianthraquinonylamine and lead oxide are added toaqueous potassium hydroxide at 170’.co CO\A/\/\I I I I v\/\/ cous-Dianthraquinon yl-p-acridone thus produced is a red substance,which on reduction gives an oxidiaable, violet leuco-derivative.59Several ortho-acridones of the anthracene series have been obtainedby the condensation of acetylmethyl-1-aminoanthraquinone and itsderivatives.CH,*?O HE--YOco N*CH, C N-CH,/\/\A -3- /\/\/’\I I I I I I I \/\/v \/\A/ co co57 RI.Maire, Bull. SOC. chim., 1908, [iv], 3, 272 ; A., i, 290.5t3 P. Jacobson and L. Huber, BET., 1908, 41, 660 ; A., i, 298.59 D.R.-P. 192436; A,, i, 456ORGANIC CHEMISTRY. 117I n these substances the methyl radicle is in an aliphatic group, but,nevertheless, the same tendency to the formation of six-memberedrings underlies the condensation.60o-Phenylenediacetonitrile (I), when warmed in alcoholic solutioncontaining a trace of sodium ethoxide, undergoes moleciilar rearraage-ment into P-imino-a-cyanohydrindene (11) :(11.1This product gives all the reactions of an imino-derivative, but whenthe cyanogen group is replaced by CO,H, C02*C2H5, or CO-NH,, thenthe structure changes to that of an aminoindene.The acid (111) doesnot lose its nitrogen until after carbon dioxide has been eliminated,when P-hydrindone (IV) is produced :C , H 4 < ~ ~ o ~ > C * N H 2 -+ C6H4<CH2>>C0. c=2(111.) UV.1The above p-imino-a-cyanohydrindene is hydrolysed by dilute acidsinto a-cyano-P-hydrindone (V) ; this compound yields a phenyl-hydrazone and a C-methyl derivative (VI), but also reacts in its enolicform (VII) to give rise to acyl and o-alkyl derivatives :(VII.)These results afford an interesting example of the formation of Lfive-membered ring, the structure of which is greatly influenced by thenature of its substituents.61The interaction of the primary aromatic amines and 2 : 3 : 5-trinitro-4-acetylaminophenol, a substance containing a singularly mobilenitro-group in position 3, leads to the production of iminazoles :OH OH OH/\NO, /\NO, /\NO,I-+ NO,(/NHPh --+ NO,j/-NPhNH*COMe lN:CMe NH-COMeDinitrohydroxy-1 -phenyl-methylbenziminazole.I n some cases the intermediate product can be isolated, but6o D.R..-P. 192201 ; A , , i, 456.61 C.W. Moore and J, F, Thorpe, Trans., 1908, 93, 165118 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.generally the ring formation occurs concurrently with the firstcon densation.62The disruption of an unstable four-membered ring, and the subse-quent formation of a more stable six-membered system, is illustratedby the condensation of primary amines and 4-nitroacetoanthranil :The process is general, for condensation occurred with extremereadiness with ammonia, methylamine, benzylamine, aniline andP-naphthylamine, and several other bases.63The well-known Slraup reaction has been utilised in the productionof technically important quinolines of the anthracene series.1-Amino-anthraquinone, condensed with glycerol in the presence of sulphuricand nitrobenzenesulphonic acids, yields anthraquinonyl-1-quinoline,Anthraquinonyl-1 : 5-diquinoline is prepared in a similar mannerfrom 1 : 5-diaminoanthraquinone : G4The Skraup synthesis, applied for the first time in the phenanthreneseries, has led to the production of 9 : 10-phenanthraquinoline from9-aminophenanthrene,G5A t the present time a special interest attaches to carbazole and itshydrogenated derivatives, owing to their possible relationship with thealkaloids of the strychnine group.Starting from the cyclohexanones,which are now readily procurable, a fairly general method has been82 R. Meldola and J. G. Hay, Trans., 1908, 93, 1659.63 M. T. Bogert and W. Klaber, J. Amer. Chem. SOC., 1908, 30, 807 ; A . , i,G1 D.R.-P. 159234 ; A . , i, 365.65 F. Herschniann, Ber., 1908, 41, 1998 ; A., i, 683.466ORGANIC CHEMISTRY. 119worked out for the synthesis of tetrahydrocnrbazole and its homo-logues :H H NHTetraliydrocarbazole.The aromatic hydrazones of these cyclic ketones undergo condensationwhen gently warmed with dilute acids.The reaction is general forall aromatic hydrazines containing one free ortho-position, but onlytakes place with simple saturated cyclic ketones, and not with thosecontaining either unsaturated or bridged rings.66Tetrahydrogenated acridines have been obtained from the cyclicketones by the following methods.1. Condensation with aromatic o-amino-aldehydes and ketones :3.\/\\/OCRNH,/l I -+H CR2. Condensation with isatin and alkali hydroxides :3. Condensation of the o-acylketohexamethylenes with aniline andThis process leads t o a mixture of n tetrahydro- its homologues.acridine with a tetrahydrophenanthridine :9133H CHHI H&JO*C 1.0 H3 NH2-!{/ /\ I --3 H A / \ / \HI I /I I\/\/\/ \ / * HH NNH2\/\1 if\/-+G6 W.Borsche, Annnlen, 1908, 359, 49; d., i, 365120 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The foregoing hydroaromatic carbazoles and acridines can betransformed into the corresponding aromatic compounds by heatingwith lead oxide.67Diphenanthracridine (I) and phenophenanthracridine (11) havebeen prepared, the first by condensing 9-aminophenanthrenewith methylene halides, and the second, by treating the same basewith o-nitrobenzyl chloride and stannous chloride.68The foregoing cases of ring formation are all examples of thegeneral tendency to the production of five- and six-membered cyclicsystems. This rule is not, however, without important exceptions,and seven- and eight-membered rings do occasionally make theirappearance, The cyclooctadienes, mentioned in connexion with thechemistry of the terpenes and indiarubber, contain a ring of eightcarbon atoms.The following remarkable series of reactions, published towards theclose of 1907, illustrate the exceptional case of the conversion of abenzene nucleus into a seven-membered ring.Reference has already been made, under the heading of aliphaticdiazo-compounds, to the great variety of products obtainable fromethyl diazoscetate. One of the most remarkable properties of thisester is its reaction with benzene and its homologues.When gentlywarmed with one of these hydrocarbons, nitrogen is evolved, and thebivalent residue, CO,Et*CH<, attaches itself to two contiguouscarbon atoms of the aromatic nucleus with the formation of a bicyclicsystem, consisting, as a wholo, of a seven-membered ring, but dividedinternally by a bridged linking into two cyclic components, one contain-ing three, and the other six, members.The case of m-xylene was des-cribed in the last communication on this subject, and as it resemblesthose of toluene and benzene, it may be taken as an illustration of thiscondensation.Inasmuch as the bicyclic structure is like that existing in carone,the initial product receives a name derived by transposition from thatof the analogously constituted ketone. This methyl 1 : 3-dimethyl-norcardienecarboxylate (I), so far, retains an aliphatic character, that6' Borsche, Ber., 1908, 41, 2203 ; A ., i, 682.68 P. C. Anstin, Tm?u., 1908, 93: 1760ORGANIC CHEMISTRY. 121it yields the amide (11) on treatment with ammonia, Alkalinehydrolysis, however, converts it into a monocyclic acid (111),3 : 5-dimethylcyclo-A2'3 ' 5-heptatriene-l-carboxylic acid. The amidegives rise to the isomeric 3 : 5-dimethylcycEo-A3 : : 7-heptatriene-l-carb-oxylic acid (IV) when hydrolysed with alkalis, but under the influenceof acids the heptatriene ring is transformed again into a benzene ring,and 3 : 5-dimethylphenyl-4-acetic acid (V), isomeric with the twoheptatrienecarboxylic acids, is produced.6g(?H:CMe'$H fl>CH*CO,Me --+ C H : C M e * ? H > ~ ~ * ~ ~ , ~ e I --+ CMe:CH*CH N CMe:CH*CHCMe:CH-CH(111.)Mordant Colours as Heterocyclic Complexes.The theory that the coloursproduced on mordanted fabrics owetheir existence to the formation of heterocyclic complexes in whichthe metallic oxide or hydroxide becomes a component of the ring, hasbeen discussed by A.Werner and C. Liebermann. Both agree onthe general principle that the peculiar properties of these lakes aredue t o a ring structure involving the metallic base. The greatstability and sparing solubility of lakes, the remarkable differencesbetween their colours and those of the ordinary salts containing thesame metal, and the singular fact that in many cases the metal doesnot exhibit certain of its characteristic analytical reactions, all justifythe theory of a heterocyclic constitution for these dyes.Werner adduces many instances in which ring structure is possibleonly on the assumption that both the principal and secondaryvalencies of the metal are involved in ring formation.Benzoylacetone and similar compounds are shown to behave asweak dyes on mordanted cotton. The lakes thus produced must besimilar in structure to the closely allied metallic acetylacetonates, towhich on account of their great stability a co-ordinated constitutioni s ascribed.Accordingly these lakes are also formulated as co-69 E. Buchner and E. Delbruck, Annalcn, 1907, 358, 1 ; A., i, 87122 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ordinated compounds (I), where R is an alkyl or acyl radicle and M isthe metal.O*CRM< >CH‘0:CR,C:CH*O(1. ) (11.)This view of their constitution is justified by the fact that hydrouy-methylenecamphor, a compound in which the keto-enolic structure isthe stable form, also behaves as a mordant dye, the lakes of whichmust have the above formula (TI).Liebermann, on the other hand, assumes that in those cases wherea carbonyl group is contiguous to the hydroxyl or isonitroso-radicle,it is only necessary to make the very reasonable assumption that thisgroup reacts in its hydrated form (I11 and IV), and then the oldertheory becomes sufficiently comprehensive to include all cases ofmordant dyeing without recourse to the hypothesis of supplementaryvalency.\ H.0’ OH(111.) (IV- 1Werner, in reply, points to the fact that the ordinary theory ofvalency does not account for the great difference in colour andstability between the lakes and the colourless salts of certaindibasic acids, many of which also possess a cyclic structure :According to Werner the production of a metallic heterocycliccomplex leads to the development of colour and other propertiespeculiar to lakes only when both the supplementary and principalvalencies of the metal are involved in the structure of the ring.70Quuinmes.The interest attaching to the aromatic diketones and theirderivatives is due to the fact that the existence of these compoundsaffords justification €or the assumption so frequently made that the firstphase in the interactions of aromatic substances and various reagentsis the formation of an additive product in which the valencies of thearomatic nucleus have undergone a certain rearrangement.Whenthe new arrangement persists in the final product, this substance isregarded as a, quinone derivative or quinoid.1436 ; A., i, 441.7O A. Werner, Ber., 1908, 41, 1062, 2383 ; A., i, 440 ; C. Liebermann, ibid.ORGANIC CHEMISTRY. 123I n the simplest case of benzene, we may have the change fromC,H,II to p- or o-C6Htv.fC\/C\/C\/C v c'\/CI /\ /\Corresponding with these two types of rearrangement there existthe long known p-benzoquinone and the more recently discoveredo-benzoquinone.A t first only quinones of the ortho- and para-series were known,but lately the possibilities of quinonoid rearrangement have been con-siderably extended by Willstatter and his collaborators.In 1907 itmas shown that 2 : 6-dihydroxynaphthalene could be oxidised so as toyield the corresponding 2 : 6-naphthaquinone,71c1whilst 1 : 5-dichlors-2 : 6-dihydroxynaphthalene furnishes 1 : 5-dichloro-2 : 6-naphthaquinone, a substance possessed of considerable stability,I n continuing the study of o-benzoquinone it has been found thatwhen catechol is rapidly oxidised (15 seconds) with pure silver oxide indry ether a colourless modification of the quinone can be isolated,This colourless variety, which is very unstable and changes quickly intothe red modification, is also obtained when an ethereal solution of thelatter is cooled, The two forms are therefore in equilibrium insolution.These isomerides are represented by the following formulae,the colourless variety being regarded, not as a quinone, but as abenzene peroxide.72H HC C/\vHQ $XOHC C:OcHRed.\/ CHColourless.71 R. Willstatter and J. Parnas, Ber., 1907, 40, 3971 ; A., 1907, i, 1056.72 R. Willstatter and F. Miiller, ibid., 1908, 41, 2580 ; A., i, 731124 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The colour of m-nitroaniline and its derivatives was long agoascribed by H. E. Armstrong to the existence of a dynamic formhaving a meta-quinonoid structure, and more recently Baly has re-ferred to the possibility of a meta-qninonoid rearrangement. Thesespeculations have now been justified by the revision of the constitutionof tribromoresoquinone, a yellowish-red substance originally discoveredin 1872 by Liebermann and Dittler, who prepared i t by removing twobromine atoms from tribromoresorcinol dibromide, and gave it theformulaMolecular-weight determinations in boiling benzene have now shownthat the substance has the simpler formula C,HO,Br,, and as itliberates iodine from potassium iodide, it is regarded as a meta-quinone, namely, 2 : 4 : 6-tribromo-m-benzoq~inone,~30 ..This substance gives neither oxime nor hydrazone, for hydroxyl-amine and phenylhydrazine behave towards it as reducing agents andconvert it into tetrabromodiresorcinol, C,HBr2(OH),*C,HBr2(0H),.One of the most characteristic reactions of quinones is the formationof intensely coloured additive compounds with their reduction products ;quinhydrone produced by the combination of p-benzoquinone and quinolin molecular proportions being perhaps the best known example ofthis class of substances.That this property is not restricted topara-quinones is shown by the fact that tetrachloro-o-benzoquinonecombines with tetrachlorocatechol to form octachloro-o-quinhydrone,C,C1,O2,C,H,C1,O2,~H20, a substance separating in lustrous, blackneedles.T4 Willstatter proposes to call these additive compounds,which consist only partly of a quinonoid complex, meriquinoids,whilst the quinones themselves and their derivatives (imines, etc.)are termed holoquinoids. The formation of the meriquinoids isattributed to the residual affinity of the quinonoid oxygen atoms, orin the case of the quinoneimines to that of the imino-group. Theintense colour of the additive product is considered to be due to anoscillation (isorropesis) of the quinone linking between the two7y R.Meyer and K. Desamari, Be?., 1908, 41, 2437 ; A . , i, 658.74 C. L. Jackson and P. W. Carleton, Amer. Chern. J., 1908, 39, 493 ; A., i,427ORGANIC CHEMISTRY. 125components of the molecular compound. I n this way the colour of themeriquinoids is referred to the same cause as that which is assumedby von Baeyer t o be operative in the di- and tri-phenylmethanecolouring matters, namely, the oscillation of the quinonoid conditionbetween two or three aromatic nuclei.These views on the colour of quinhydrones and meriquinoids arenot accepted unreservedly by F.Kehrmrznn,75 who objects to theparallel drawn between quinoneimines and their meriquinoid salts,on the one hand, and the triphenylmethane-imine bases and theirsalts on the other. I n the first case, the intensification of colour isdue to the introduction of auxochromic groups (NH2 or OH) withoutany modification,of the chromophore ; in the second, the bases andtheir salts are quite different in constitution, and the variation ofcolour is due to this complete change in the configuration of thechromophore. The difference in stability between the ordinaryquinhydrones and the meriquinoids (Wurster's salts, etc.) is merelyone of degree, and there is no reason for assuming that these twotypes of partial quinoids differ essentially in constitution.The coloured salts formerly obtained by 0.Wurster 76 on oxidisingalkylated p-diamines are regarded as meriquinoids by Willstiitter.The red salt produced by the action of bromine on ns-dimethyl-p-phenylenediamine is meviquinonedimethyldi-imonium bromide,I II II I/\/ \/NMe, .---------- NMe2Br( NO,)The nitrate has also been obtained, and the blue salt fromtetramethyl-p-phenylenediamine contains a sulphate which is onlyone-third quinonoid,C6H4[NMe,*HS0,],,2C,H4(NMe),,€I,S0,.Benzidine gives rise to two meriquinonoid chromates, whilst bothholo- and meri-quinoids have been obtained from tetramethyl-b e n ~ i d i n e . ~ ~Further investigations on the oxidation products of benzidine,diphenyline, and tolidine have resulted in the production of dipheno-quinonedichlorodi-imide, NC1:C,H4:C6H4:NC1, ditoluquinonedichloro-BcT., 1908, 41, 2340; A., i, 698.R.Willstatter and J. Piccard, ibid., 1908, 41, 1458 and 3245 ; A., i, 475,'13 B i d . , 1879, 12, 1803 ; 1887, 20, 2071.915126 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.di-imide, h010- and meri-dichloroditoluquinonedi-imonium chlorides,the last of these having the composition 78r 01 -lNH,<-\-/-\NH, --/ \-/Me Mec1 c12HC1, xH,O.The action of alkali hydroxides on 1-phenylisoquinolinium meth-iodide and papaverinium methyl halides leads to the formation of2-phenyl-a-naphthol and 6 : 7-dimethoxy-2-mp-dimethoxyphenyl- a-naphthol, which on oxidation yield binuclear quinones having theappearance of indigotin.790 00 0The residual affinity of quinones is also manifested by the formationof additive compounds with certain inorganic chlorides.The followingare representative examples of this series : p-benzoquinone stannicchloride, C,H,0,,SnC14 (red), a-naphthaquinone antimonic chloride,C,oH,02,2SbCI, (red), P-naphthaquinone stannic chloride (green),phenanthraquinone mercuric chloride, 2C,,H,02,HgC1, (red).*OAromatic ketones behave similarly, and the following yellow com-pounds have been obtained,. benzophenone antimonic chloride,CPh20,2SbC1,,and benzil stannic chloride, (COPh),,SnCl,.F. Kehrmann refers the production of these compounds to the salt-forming capacity of quinonoid oxygen, and points out that he hadpreviously shown that phenanthrnquinone and chrysoquinone aredibasic substances forming two series of salts.81An ingenious use of the properties of pheriyliminoquinones has beenmade in the study of the tri-, tetra-, penta-, hepts-, andocta-hromo-7a W.Schlenk, Annalen, 1908, 363, 313 ; A., 1909, i, 36.79 H. Decker, ibid., 362, 305 ; A . , i, 806.so K. H. Meyer, Ber., 1908, 41, 2018 ; A., i, 731.*l Ber., 1908, 41, 3396 ; A., i, 993ORGANIC CHEMISTRY. 127derivatives of p-hydroxydiphen ylamine. I n each case, oxidation withchromium trioxide gives rise t o the corresponding quinoneanil (phenyl-iminoquinone), the colour of which becomes more intense as the pro-portion of bromine increases, t h e shades varying from scarlet to apurple so dark as to appear black.These substances readily undergo hydrolysis, so that the number ofbromine atoms present in each ring, and, in some cases, also theirorientation, are readily ascertained : 82Br Br Br Br Br BrBr Br BrB/'-\NH, + o:/=\:o.\-/ \=/Br Br BrAn extremely interesting synthesis of quinones from straight-chaincompounds has been effected by a modification of the process for pro-ducing ethyl alkyloxalacetates. A mixture of ethyl oxalate and anester of a monobasic fatty acid is treated with sodium instead ofsodium ethoxide, whenbenzene is produced :R*CH,* C0,EtCO,Et*CO,Et$.!O,EtL'O*CORCH2 \ CH,R \co*yoC0,Eta hydroxyquinone derived from a p-dialkyl-vO,E tAtmospheric oxygen intervenes in the last step of this condensa-tion, which takes place so readily that it may be employed as a lectureexperiment to demonstrate the formatio of a coloured quinoiie fromcolourless, comparatively simple, aliphatic esters.I n the foregoingscheme, R may be methyl, ethyl, isopropyl, n-butyl, phenyl, or benzyl.These hydroxyquinones are yellow, whereas their alkali salts arebl uish-viole t.Equally noteworthy is the unique hydrolysis of these dihydroxy-dialkylbenzoquinones under the influence of boiling aqueous alkali82 A. E. Smith and K. J. P. Orton, Trans., 1908, 93, 314128 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.hydroxides, when the ring structure disappears with the production ofsymmetrically disubstituted succinic acids,CO,H* CHR*CH( CH,R) *CO,H,which are always obtained in their two stereoisomeric modifications,so that in some cases the reaction may withadvantage be employed inthe preparation of these isomerides.I n this hydrolysis, the hydroxy-quinones react in their tautomeric form :co*co RYH* C0,HCH,R CH C0,H C(0H)-CH,R --+ RCH<C-b -+R CH<CO' C02H/\C0,H CH,k C02HThe triketocarboxylic acid containing an a-diketonic group under-goes the benzilic acid change to a dicarboxylic acid which condenses toa lactonecarboxylic acid ; this compound then takes up water and losescarbon dioxide to form the disubstituted succinic acid.The penultimate product, the lactonecarboxylic acid, was in oneinstance prepared by an independent method and shown to undergo thelast step in the hydrolysis under the influence of alkali hydr0xides.~3I n connexion with the production of quinones, reference may bemade to the more recent of Zincke's long series of researches on thechlorination and bromination of phenols.84Di-p-hydroxydiphenylmethylethylmethane yields a tetrabromo-derivative, HO*C6H,Br,~CMeEt*C6H,Br,~OH, which on furtherbromination is resolved into s-tribromophenol and the +-hexabromideof p-sec.-butylphenol ; a t 100' a $-heptabromide is produced. These$-bromides, which unlike the less brominated phenols are insoluble inaqueous alkali hydroxides without decomposition, readily losehydrogen bromide when treated with sodium acetate or sodiumcarbonate, and give rise t o quinones :C,H, CHBr,\/H CBr\/Br A B rBr" "Br \/ ..0+-p-Tribromo-sec. -butyltetra-bromophenol.(Colourless hemiquinone.)-+C H CHBr, </ C ..Br/\BrBrll I ' B ~ \/ 0p-Dibromo-sec. -butylidenetetra-bromoquinone.(Yellow quinone.)83 F. Fichter, Annalen, 1908, 361, 363; A., i, 658.84 T. Zincke and J. Goldernaim, ibid., 362, 201 ; A., i, 780; T. Zincke andE. Birschel, ibid., 221 ; A., i, 781ORGANIC CHEMISTRY. 129The tetrwhloro- and tetrabromo-dihydroxybenzhydrols (I), whentreated with hydrogen chloride and hydrogen bromide respectively,y iel d the $-pen tac hloro- and +-pen tabromo -di hy drox ydiphenylmet h y 1chloride and bromide (11). These halides, when shaken with aqueousacetone, lose hydrogen halide and furnishbromo-hydroxybenzylidenequinones (111).OH OHABr()BrIIAB r U B rOHBr()Br I0(1.1 (11.)derivative.hemiquinone.Colourless benzenoid Colourlessthe tetrachloro- andOH0(111.)qcunone.Yellowtetra-These examples suffice to show the close relationship between theappearance of colour and the development of the complete quinonoidconfiguration. Of interest in this connexion is the preparation ofcoloured hydrocarbons of the quinodimethane series, of which thefollowing synthesis is an example :Benzoyltriphenylmethane, COPh°C6H4*CHPb2, when subjected tothe Grignard reaction with magnesium a-naphthyl bromide, yields asubstituted benzhydrol, HO*CPh(C,oH7)*C,H4*CHPh2, from whichp-benzhydryldiphenyl-a-naphthylmethyl chloride (I),C1*CPh(CloH,)*C,H4*CHPh2,is readily obtained by the action of hydrogen chloride. This chloride,when heated with quinoline, loses hydrogen chloride, giving rise tothe orange-red hydrocarbon, triphenyl-a-naphthylquinodimethane (11).This hydrocarbon absorbs halogen halide, regenerating the chloride (I).These changes constitute a reversible reaction, which may berepresented as follows : 85(I.) Colourless benzenoid derivative.(I I. ) Coloured quinoid.Recently several methods of oxidising aromatic hydrocarbons toquinones have been patented, based on the indirect employment ofelectrolytic methods. For example, manganic alum, produced electro-lytically from an aqueous solution of manganous and ammonium85 A. E. Tschitschibabin, Ber., 1908, 41, 2770 ; A., i, 872.REP.-VOL. V. 130 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.sulphates, is recommended for oxidising naphthalene, anthracene, andphenanthrene to their respective quinones.s6Colour and Constitution.Contributions to the study of the relation of colour to structurehave again been very numerous, and in several of these a comparisonhas been made between colour and fluorescence. Starting from themuch-discussed case of nitroquinol dimethyl ether, an examination ofa large number of coloured and fluorescent substances 87 has shownthat the colour of the light emitted in fluorescence is quite commonlydependent on the solvent employed.The fluorescent band of the solidsubstance lies furthest towards the nltra-violet, then follow the solu-tions in indifferent solvents, then those in dissociating solvents, thefluorescence of the vapour exposed to Tesla-radiation lying nearest tothe red.This is true even of compounds in which it is impossibleto assume a change of constitution to be brought about by dis-sociating solvents. Changes of colour in general run parallel withfluorescent changes. The theory of partial valencies is invoked inexplanation of these facts. Considering the ordinary statical formulaeas representing an ideal constrained state of the molecule, the solidsubstance and its solutions in indifferent solvents are supposed toapproach most nearly to this, whilst a greater freedom of the partialvalencies is possible in dissociating solvents and in the state of vapour.Nitroquinol dimethyl ether is then written in the form (I), theMe0I()-NO2\/IMe0formula (11) representing the ideal limiting condition.The paler thecolour of the solution the more nearly the molecule approaches to thelimiting condition, which is, however, never completely attained.m-Nitrodimethylaniline is a good example of a compound showingvariable fluorescence. The results are unfavourable to the assumptionof a quinonoid structure. The influence of solvents in modifying thedirection of the lines of force which constitute the partial vnlenciesis illustrated and discussed.*6 D.R.-P. 189178 ; A . , i, 350.87 H. Kauffmann, Ber., 1908, 41, 4396 ; A., 1909, i, 96ORGANIC CHEMISTRY. 131Somewhat different results have been reached by a very exhaustivestudy of the triphenylmethane (or 6‘, tritane ”) series.88 It is shownthat the parallel between colour and fluorescence must not be pressedtoo far.Simple ring compounds, such as benzene, are fluorescent,although the emitted light lies in the ultra-violet portion of thespectrum, and by substitution, or by the juxtaposition of rings, theoscillations causing fluorescence may be so far retarded as t o enter thevisible region. Thus anthracene has a distinct fluorescence. Theproduction of colour occurs in a quite different way. The replacementof hydrogen in benzene by substituents, however heavy, is quiteincapable of retarding the oscillations so as to bring the ultra-violetabsorption of benzene into the visible region. The colour of benzenederivatives is brought about, not by the shifting of a previouslyexisting ultra-violet band towards the red, but by the production of anew band or bands, and for this a definite arrangement of conjugateddouble linkings is required.The author considers the key to theproblem to be given by a comparison of benzene with its isomeride,fulvene. In both compounds, the number of carbon and hydrog6n atoms,and of double linkings is the same, and only the disposition of thelatter is altered.FH:CHCH:CH >C:CH,.Benzene (colourless). Fulvene (coloured).From a comparison of a large number of coloured substances, theauthor concludes that there is only one true organic chromophore,R R RA-A-Anamely, the grouping \I II It, with three double linkings. Thisgrouping is not, however, sufficient in itself to bring about colour, butrequires some further condition (m&t frequently the closing of a ring)to produce visible colour. The reason for this is that the double link-ings must first be brought into a definite relative position before therhythmic oscillations which cause colour can b9 set up.It will beobservedgrouping,that nitroquinol dimethyl ether contains the requiredtogether with the ring :C OMeaa H. von Liebig, Anzalen, 1908, 360, 128 ; A., i, 445.K 132 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Coloured solutions of such simple coloured substances as diacetylmust contain polymerised molecules in order to render the abovegrouping possible, an assumption for which there is good evidence.Fluorescence is explained as due to a pulsating interchange ofdouble linkings in rings, brought about by impinging light-waves.The tritane derivatives examined ranged from the colourless triphenyl-methanet through compounds exhibiting both colour and fluorescence,to the penta- and hepta-acetyl derivatives of the complicated hydroxytritanol ether,OH OH OH OH <>-ox I I (/OHA- /\ICPh, CPh, YPh, QPh,co co-o-OH\/ I \/ I YI OH OHwhich exhibit colour, fluorescence, and pleochroism, a property hithertoonly possessed by chlorophyll amongst colouring matters.The picryl ethers of certain amidines have been noted as showing aremarkably strong fluorescence, a1 though nitro-groups generallyhinder the appearance of this property.89 The non-fluorescent1 : 2-diphenyl-3- benzyloxyamidine yields a highly fluorescent ether,and the same is found to be true of other non-fluorescent imide bases.The essential grouping appears to beIR C N RI .-----IN*O*CaH,(NO,),R'and that the residual affinity of the nitro-groups, acting in the direc-tion of the imino-nitrogen atom, is connected with the property issuggested by the fact that the salts of the same imides with heavymetals, which are certainly internally complex, are highly coloured .The phenyl groups are not necessary, since picryl diguanide,NH,*Q :NHNH,*C:NHy*O.C,H,(NO,), ?is also fluorescent.The effect is attributed to internal oscillations oflinking in the molecule.Pulsations of the ring-system of benzene were invoked to explainthe absorption of benzene in the ultra-~iolet,~0 it being shown that the89 H.Ley, Ber., 1908, 41, 1637; A., i, 570.90 E. C. C. Baly, W. H. Edwards, and A. W. Stewart, Trans., 1906, 89, 514ORGANIC CHEMISTRY. 133number of absorption bands corresponded with the number of distinctmodes of deformation of the ring. This hypothesis has now been putto a severe test by the examination of compounds containing morethan one ring.g1 Naphthalene itself has three bands in the ultra-violet, one of which is attributed to the benzenoid ring, and the othertwo to the conjugation of this with the ethylenic linkings of thesecond ring. When this second ring is entirely reduced in 1 : 2 : 3 : 4-tetrahydronaphthalene, only the one benzenoid band is observed.I n1 : 4 : 5 : 8-tetrahydronaphthalene, the two rings are exactly alike, andthe symmetrical arrangement resembles that of p-xylem, and the twospectra are consequently almost identical.A similar spectrum is given by acenaphthene (I), whilst acenaph-thylene (11) has, in :addition, bands due to the influence of the doublelinking outside the ring, producing colour.CHZCHThe complete absorption spectra of a number of compounds contain-ing nitro- and nitroso-groups have been examined, and the position of theband is found to depend on the nature of the atom to which the groupis atta~hed.~2 Dinitro-paraffins form both coloured and colourlesssolutions, the spectra OF the former, and of the alkaline salts of thesecompounds, being unlike those of aci-mononitro-salts, R*CK*NO,M.They are therefore assumed to contain an isomeric modification withquinonoid grouping, R* C< NO(0M) No->O, producing colour, in equi-librium with the colourless f0rm.~3In continuation of the work on the strongly-coloured aci-ethers ofnitrophenols, a dark violet aci-ether of hexanitrodiphenylamine hasnow been prepared94 from the violet silver salt, It is thereforeassumed to have the quinonoid constitution,C6H2(N02)s*N:C6H2(N02)~:No2Me.The fact that halogen-phenols, in which it is difficult to assume aquinonoid rearrangement, form both coloured and colourless silversalts 95 points to the necessity of further investigation of this kind ofisomerism, and an interesting case of the same kind has been recentlydiscovered.9691 E. C. C!, Baly and W. B. Tuck, Trans., 1908, 93, 1902.92 E. C. C. Baly and C. H. Desch, ibid., 1747.9s E. P. Hedley, Ber., 1908, 41, 1195; A., i, 382.94 A. Hantzsch and S. Opoloski, ibid., 1745 ; A . , i, 526.g5 Ann. Report, 1907, 114.96 0. Dimroth and 0. Dienstbach, Ber., 1908, 41, 4055 ; A., 1909, i, 62134 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.-4-Oximino- 1 -phenyl-5- triazolone, N€’h<N -r forms three CO*C:N*OH’series of salts, yellow, red, and green, of which the red is usually t hstable form, although the red silver salt is labile, and passes into thestable green modification. Further, the benzoyl and acetyl deriv-atives, which no longer contain a labile atom, also occur in red andyellow modifications, of which the latter is the stable form.That thecase is one of chemical isomerism, and not merely of polymorphism, isshown by the fact that the red benzoyl derivative dissolves inchloroform to a deep red solution, which, however, immediatelybecomes yellow. Concentrated solutions are orange, and contain thetwo modifications in equilibrium.Many aminoazo-compounds are known to form two series ofdifferently coloured salts. The orange salts 97 resemble azobenzene incolour and must have the azo-constitution,Ph*N:N*C,H,*NR,HX,whilst the violet salts are quinonoid, as, for example,Ph*N H*N: C6H,: NMe,CI.The solid orange salts are usually labile, and are converted into theviolet salts on heating, or, in some cases, on rubbing.The violet salts were proved to be unimolecular in solution.Indifferent solvents favour the violet modification, whilst the orangemodification persists in alcohol, ether, or acetone.The solutions inconcentrated sulphuric acid are all yellow, and resemble azobenzene.The same colour relationships appear in the aminoazobenzene-sulphonic acid series, and the colour-changes in helianthin and methyl-orange have therefore received a similar e ~ p l a n a t i o n , ~ ~ the inter-mediate formation of an internal azo-salt being assumed, as representedin the following scheme :Helianthin.1/ \ Methyl-orange.I n aqueous solution. Solid andL Solid. r > in solution.Violet. Orange. Orange. Orange.C,H4* so, C6H,* so, C H *SO,H C,H,*SO,NaI S * -+ N NaOH iJ 1 + - ~ - 3 M ‘H I HzO -+ #l l I I IC,H,: NMe,Quinonoid.Azo.C6H,*NMe2H C6H4*NMe, C,H,*NMe,Azomethines are of interest from their analogy to azo-compounds.Since the simplest nzomethines and their ethers are colourless, whilsttheir hydrochlorides show an absorption band in the blue region of the97 A. Hantzsch and F. Hilscher, Ber., 1908, 41, 1171 ; A., i, 484.93 A. Hantzsch, ibid., 1187 ; A., i, 469.& ORGANIC CHEMISTRY. 135spectrum, it has been concluded 99 that the latter are quinonoid oxoniumsalts, the simplest ethyl ether hydrochloride, for instance, having theformula C6H,*CH2*N:C6H,:o<cl. EtA. recent observation 1 is of interest as showing that the possessionof residual affinity may suffice to give chromophoric properties toa group of compounds, even when no ethylenic linking is present.Iodochlorides, such asOMeare strongly coloured red.Hitherto all chromophores attached to thebenzene ring have contained a double bond.The investigations into the constitution of triphenylmethyl, whichderive their interest mainly from their bearing on such problems asthose just discussed, have added little to the knowledge summarisedin last year’s Report (p. 118). The hexaphenylethane formula seemswell established for the solid substance. An attempt to preparehexaphenylethane by heating triphenylmethyl triphenylacetate,CPh,*C02*CPh,,2yielded only an amorphous product, together with triphenylmethane.The product obtained from Gomberg’s triphenylmethyl, and formerlysupposed to be hexaphenylethane,3 is now4 proved to be p-benz-hydryltetraphenylethane, CHPh2*C,H4*CPh,.by its synthesis fromp-benzoyltriphenylmethane, which reacts with magnesium phenylbromide to form p-benzhydryltriphenylcarbinol ; condensation withaniline hydrochloride and elimination of the amino-grmp then givesGomberg’s compound, which is thus an isomeride, and not a polymeride,of triphenylmethyl. The preparation of the latter compound fromthe magnesium chloride has been improved,5 and the existence ofisomeric modifications of magnesium triphenylmethyl chloride hasbeen defended against the criticisms of Tschitschibabin. The existenceof both a coloured and a colourless modification in solutions of tri-phenylmethyl is inferred from the disappearance of the yellow colouron shaking with air, the yellow modification being the more readilyoxidised.6 On removing the peroxide by filtration, the colourless99 F.G. Pope, Trans., 1908, 93, 532, 1914.H. Kauffmann, Ber., 1908, 41, 4413; A., 1909, i, 95.R. Anschiitz, Annalen, 1908, 359, 196 ; A., i, 331.M. Gomberg, ibid., 1903, 36, 370 ; A., 1903, i, 244,A. E. Tschitschibabin, Ber., 1908, 41, 2421 ; A., i, 624.J. Schmidlin, ibid., 1908, 41, 423, 426 ; A . , i, 150.Ibid., 2471 ; A., i, 623136 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.filtrate again becomes yellow, an ethereal solution in equilibrium atthe ordinary temperature containing about ten times as much of thecolourless as of the coloured form. Lowering the temperature favoursthe production of the colourless modification.The theoreticalexplanation of these facts has made little progress since last year.Aromatic Diaxo-compounds.I n last year’s Report reference was made to a new formula (11) fordiazonium salts advocated by Cain,7 which was put forward to account2(1.) (11.) (111.)for the following facts : (1) The property of giving rise to diazoniumderivatives is confined to the aromatic primary amines, and is notpossessed by the bases of the fatty series or fully saturated ringseries; (2) the facility with which nitrogen is eliminated fromdiazonium salts suggests an unstable double linking \GIN, ratherthan the single linking \CON, assumed to be present in theKekuld (IV) and Blomstrand (V) formulae?///C6H,*N:N*C1 C,H,*NCli N.(IV.) (V.1An ortho-quinonoid configuration would, however, afford an equallysimple explanation of these facts, and accordingly i t has recently beensuggested that diazonium salts might appropriately be represented bya dynamic formula, in which the valency of the triad nitrogen atomin the foregoing static ortho- and para-quinonoid formulae (I, 11, 111)is directed sucessively to the carbon atoms marked 1, 2, and 3, but isnot held continuously by any one of them. This extremely labilecondition of the oscillating linking would account for the productionof hydrazines rather than diamines on reduction.8These views on the structure of diazonium salts have beenvigorously attacked by Hantzsch,g who contends that if diazoniumsalts are hemi-p-quinoids, as represented by Cain’s formula, then onreduction they should yield p-diamines and not hydrazines.He alsorepudiates the suggestion of a. dynamic formulation, contending that7 Ann. Report, 1907, 120.8 G. T. Morgan and F. M. G. Micklethwait, Trans., 1908, 93, 617.Ber., 1908, 41, 3532ORa ANlC CHEMISTRY. 13’7this hypothesis, instead of obviating the difficulty of explaining theproduction of hydrazines rather than diamines, increases it twofoldinasmuch as a mixture of ortho- and para-diamines should result fromthe reduction of such a compound. This view of the matter, however,involves the assumption that the properties of a substance with adynamic structure are simply those of the tautomerides representingthe extreme phases of the molecular oscillation. It is at leastconceivable that the predominant properties of the compound mightbe those of the intermediate phase or phases.This conception, whenapplied to diazonium salts, furnishes a reason for the formation ofhydrazines as the main products of reduction.Cain,lo who has replied to Hantzsch’s criticisms of his formula,points out that there is experimental evidence for the view thatreducing agents would break the unstable linking, >CH*N, ratherthan the more stable azo-bond, -N:N-. He also calls attention toHantzsch’s admission that the existence of aromatic diazonium saltsand the non-formation of such derivatives of the aliphatic aminespoint to some interaction of the residual affinities of the aromaticnucleus with the unsaturated diazonium complex. He maintains thatthis connexion is more definitely expressed by his quinonoid formulathan by the vague addition made by Hantzsch to the Blomstrandconfiguration formula and expressed by the dotted line in thefollowing formula, C,H,*N2X.Euler,ll who advocatesanother para-quinonoid formula for diazoniumsalts,.-. .also lays stress on Hantzsch’s admission of the imperfection of theBlomstrand formula (V.).Although these controversial matters require further experimentalevidence for their complete elucidation, yet this’discussion has servedthe useful purpose of recalling attention to the important fact, sofrequently overlooked, that the residual affinity of the aromaticnucleus is the determining factor both in the production of diazoniumsalts, and also in many other characteristic properties of aromaticcompounds.The study of the decomposition of diazonium salts in solutions oflo Ber., 1908, 41, 4189 ; A ., 1909, i, 70.ll. Ibid., 3979 ; A , , 1909, i, 70138 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.varying strengths and at different temperatures has led to thefollowing results.121. Dry:diazonium chlorides give the maximum value for the velocityof decomposition in aqueous solutions. Freshly prepared specimensdecompose more slowly, and contain apparently traces of a protectingsubstance, the nature of which has not been elucidated.2. Diazonium chlorides and bromides in solutions of the gameconcentration decompose at the same rate.3.Increase in the concentration results in a small rise in thevelocity of decomposition, which is somewhat greater with thebromides than with the chlorides. The velocity increases as thedecomposition progresses, probably owing to the formation of halo-genated hydrocarbons by the action of the liberated hydrogen halide.4. Diazonium iodides decompose more rapidly than the chloridesor bromides, even in very dilute solutions.5. Increase of concentration has but little effect on the velocity ofdecomposition of p-nitrobenzenediazonium chloride, and only in veryconcentrated solutions is any acceleration noticed. Neither strongnor weak acids have any protecting influence.6. The less basic is the diazonium hydroxide set free by hydrolysisthe more rapidly is the diazonium acetate decomposed in aqueoussolution, Sodium chloride has a protecting influence on p-nitro-benzenediazonium acetate.7.Nitrous acid decomposes diazonium salts, not catalytically,but by actual participation in the decomposition.The experimental data on which this last generalisation is basedagree with those formerly obtained by Cain,13 and confirm hisopinion that a solution of pure p-nitrobenzenediazonium chloride andanother containing the same concentration of this salt with a traceof nitrous acid sufficient to give the starch-iodide test decompose withequal velocities under similar conditions of temperat~re.1~A study of the oxidation of primary aromatic hydrazines has led toa method of converting these substances quantitatively into thediazonium salts, from which they were obtained by reduction.In the first place, the oxidation of these hydrazines with copper,silver, and mercuric oxides leads to the deposition of the correspond-ing metal, with the liberation of nitrogen and the formation of ahydrocarbon :R*NH R*NH R N H + H N b H 4 H ~ O H -+ H ~ Hl2 A.Hantzsch and It J. Thompson, Bey., 1908, 41, 3519 ; A., i, 1021.l3 Ibid., 1905, 38, 2511; A., 1905, i, 724.l4 Ibid., 1908, 41, 4186 ; A , , 1909, i, 70ORGANIC CHEMISTRY. 139Manganese and lead ,dioxides give similar results, and potassiumpermanganate and hydroxide furnish benzene, azobenzene, anddiphenyl.The oxidation proceeds most smoothly with alkaline potassiumchromate, when a practically quantitative yield of nitrogen andhydrocarbon is produced.15Althongh the foregoing experiments give no indication of theformation of diazonium salts, yet it has been found possible to obtainthese compounds in excellent yield by introducing chlorine or bromineintolan alcoholic solution of the hydrazine a t temperatures below - 20°.This mode of procedure gives the solid diazonium salt, although asolution of the same can be obtained in glacial acetic acid by addingchlorine or bromine a t OO.16A more convenient method of obtaining the solid diazoniumbromide is first to prepare its perbromide by adding bromide to theaqueous solution of the diazonium salt.The dry perbromide, whenmixed with the corresponding hydrazine in cold alcohol, undergoes thefollowing change : 17R*NBr R*NH R-NBr R*NBrBrNBr= 3 = 3 111 + 3HBr.H I ~ H H h r N 2 1 4-The ortho- and para-aminophenols can be diazotised to diazo-oxides,but the ortho-aminonaphthols, on treatment with nitrous acid in thepresence of mineral acids, undergo oxidation to P-naphthaquinone.It has been found that these aminonaphthols and their sulphonicacids can be diazotised readily, providing that mineral acids areabsent.A solution of 1 -amino-P-naphthol-&sulphonic acid, sodiumnitrite, and sodium chloride~slowly deposits the cyclic diazo-derivative,NaS03*C,,H,<~, and this result is also obtained by treating thesulphonic acid a t 40--50° with sodium nitrite, zinc sulphate, and zinchydroxide; other metallic salts may be used instead of zincsulphate.l*Although diazo-oxides can be obtained from ortho-, para-, and peri-aminophenols, diazoimines can be prepared only from ortho- and peri-ls F.D. Chattaway, Trans., 1908, 93, 270, ls ]bid., 852. l’ Ibid., 958.D.R.-P. 189179, 195228, and 195322 ; A., i, 231, 842140 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.diamines. The unsubstituted para-diamines do not give rise to diazo-imines, although certain of their arylsulphonyl and aryl derivativeshave been found to yield p-diazoimides and p-diaxoimines of varyingdegrees of stability.The simplest coml3pund of this group is the explosive phenyl-p-phenylenediazoimine l9 (I) obtained by adding ammonia to thediazonium salt of p-aminodiphenylamine, NH,*C,H,*NH*C,H,.Aeeries of increasingly stable p-diazoimines has been prodnced bygradually introducing nitro-groups into the phenyl nucleus of this base,so that when the trinitrophenyl (picryl) derivative is reached ap-diazoimine (IV) is produced, which is as stable as the p-diazoimides(V) obtained from the arylsulphonyl-p-diamines and mentioned inearlier reports.20(IV. 1 w. )I n the foregoing diagram these derivatives of the still unknownp-diazoiminobenzene are arranged in the order of their stability, whichincreases as the acidity of the substituent group becomes morepronounced. I n spiteof the great difference in stability, the membersof this series all exhibit the two properties which are characteristic ofp-diazoimino-derivatives. They combine additively with P-naphthol,and change quantitatively into the corresponding diazonium salt ontreatment with cold concentrated mineral acids.z1Wydrazones and Hyd~*oxyaxo-compounds.I n considering the diazonium salts, special stress was laid on thefact that these are known only in the aromatic series.The compoundsnow under discussion may be either entirely aromatic or mixedaliphatic-aromatic derivatives.Several important investigations on these substances have beenpublished during the last year, and a t last i t seems likely that someagreement will be reached in regard to their constitution.l9 Annalen, 1888, 243, 282, and Ber., 1902, 35, 895.2o Ann. Report, 1906, 124 and 151.21 G. T. Morgan and F. M. G. Micklethwait, Trans., 1908, 93, 602ORGANIC CHEMISTRY.141The coupling of a diazonium salt with a phenol gives rise to anaromatic hydroxyazo-compound, whilst an aliphatic-aromatic azo-derivative results from the condensation of the diazonium salt withan aliphatic substance containing the group *CH,*CO=.It is in the first place necessary to consider the constitution of thealiphatic and aromatic compounds concerned in these condensations.Aliphatic compounds containing the above group tend to conservethis ketonic configuration, even although in certain reactions thedynamic enolic form comes into play.X*QH, A X*GHY-co YOOH 7Stable ketonic form, Labile enolic form.In the aliphatic series the ketonic form is the more stable.The converse holds with the phenols. Although i t is frequentlynecessary to assume the intervention of the dynamic ketonic(quinonoid) form, this in the aromatic series is the labile condition,and whenever possible the phenol or its condensation product revertsto the stable hydroxylic (benzenoid) configuration.7H:C H 'f?H2CH: C: H *C: 0Labile ortho-$!H:CH*GH ' quinonoid form.CH:CH-&OH ,, CH:CHStable hydroxylic ' cH2<cH:cH>~obenzenoid form.Labile para-quinonoid form.Recent experimental evidence is almost unanimous in demonstratingthat these properties of the aliphatic diketones and aromatic hydroxy-derivatives are shared by the mixed and aromatic azo-compoundsrespectively, the aliphatic-aromatic derivatives tending always toacquire the hydrazone configuration, whilst the purely aromaticderivatives assume almost invariably the hydroxyazo-structure.Aliphatic series :d X*JC;'*N:NR7X*$XN*NHRy*c:o Y*C*OHStable hydrazo-form.Labile azo-enolic form.Aromatic series :$?H:CH*E*N:NR .-) QH:CH*$XN*NHRCH:CH*C*OH CH: CH*C:O -Stable o-hydroxyazo-form, Labile o-hydrazone.Stable p-hydroxyazo-form. Labile p-hydrazone142 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRYAn investigation of the mechanism of the coupling of diazoniumsalts and aliphatic ketones has been carried out with tribenzoyl-methane, which reacts in its enolic form, since the first product is adiazo-oxide (I) containing the diazo-complex attached to the enolicoxygen. This is proved by the ease with which the diazo-group canbe removed merely by treatment with @naphthol or a-naphthylamine,This diazooxide passes successively into two isomerides, the first, ared azo-compound (11), and the last and most stable, a colourlesshydrazone (111).a76H5 FdH5 72%fi*OXN*C,H, 7:0 Yo70 70C*N:N*C6H5 + C : N N*C6H5I /\ $0 yo-+ C/\70 70C6H5 '6*5 C6H5 C6H511.Red azo-compound. 111. White acyl hydrazone.On hydrolysis, these three isomerides give the same products, namely,'GH5 C6H5I. Yellow diazo-oxide,ethyl benzoate and the hydrazone (IV).C6H5-y0C6H5 * COY:N*NH*C,H,IV. Yellow hydrazone.These results show that in the aliphatic series the final stable pro-duct is a hydrazone rather than an azo-compound.When a very reactive group, such as nitroxyl, is present, the trane-formation of diazo-oxide into azo-compound and hydrazine derivativemay occur simultaneously,o-Dinitrotoluene, C6H5*CH(N0,)2, and its homologues couple withdiazonium salt in alkaline solutions, the first product being a diazo-oxide (I).This substance readily changes into two isomerides, a redazo-compound (11) and a white hydrazine derivative (111). In theinitial coupling, the dinitro-compound is used in the form of its alkalisalt, c6H5*C(N0,):NO*OK.W>C<N?2 ,+ C6H5 N.N*C6H,No2>C:NO*O*N,*C,€€5 11. Red azo-compound.'+ C,H,*CO*~-?f* C6H5 C6H5 I. Yellow diazo-oxide.NO, NO111. White hydrazine.The constitution of the hydrazine derivative is shown by the actionof water, which hydrolyses it first into nitric acid and nitrosobenzoyl-a2 0. Dimroth and M.Hartmann, Ber., 1908, 41, 4012, A., 1909, i, 66ORGANIC CHEMISTRY. 143phenylhydrazine, C,H,*CO*NH*N(NO)*C,H,, this compound beingfinally decomposed into nitrous acid and s-ben~oylphenylhydrazine.~~In the aromatic series the first product, the O-azo-derivative, ordiazo-oxide, has also been isolated in a few cases where the velocity oftransformation into the C-azo-derivative has been lessened by thepresence of substituents in the reactive para-position. p-Bromobenzene-diazonium chloride and p-nitrophenol couple to give p-bromobenzene-diazo-4-oxynitrobenzene, C6H4Br*N: N*O*C,H,*NO,, which at 80"becomes transformed into its isomeride, the red p-bromobenzene-2-azo-4-nitrophenol, C6H4Br0hT:N*C,H:,(NO2)*OH.Auwers 24 is inclined to regard these intermediate O-azo-derivativesas diazonium oxides, but inasmuch as they are produced only in theabsence of acids stronger than acetic acid, it seems preferable toregard them as diazo-oxides, R*N:N*OY, analogous to the diazo-amines,R*N:N *NHY, which are formed under similar conditions and undergothe same change into C-azo-derivatives.In 1907 W.Borsche succeeded in condensing o-nitrophenylhydrazineand 2 : 4-dinitrophenylhydrazine with p-benzoquinone and its homo-logues ; the products were identical with the hy droxyazo-derivativesprepared by coupling o-nitrobenzenediazonium and 2 : 4-dinitrobenzene-diazonium chlorides with phenol and its homologues. 25-----.-.-.......;O H2iN*NH*C,H4*N02 N* NH* c,H,*N02( 0 ):. . /A+ II II\/,\ _..____..\/\ A0 u 04 Unstable hydrazone(not i solnted).N:N* C,H,*NO,(o)11 I + ~O-N,*C,H4*N02 -3- I] I\/ \/633 6HStable p-hydroxyazo-derivative.These condensations are of great interest, because the products areadmittedly p-hydroxyazo-derivatives, formed by a process which oughtto give rise to p-quinonehydrazones were it not for this tendency ofthe aromatic hydroxyl derivatives to conserve their hydroxylicstructure.23 G.Ponzio and G. Charrier, Atti R. Accad. Sci. Torino, 1908, 43, 303 ;24 Ber., 1908, 41, 4304 ; A., 1909, i, 67.25 Borsche, Annalen, 1907, 367, 171 ; A., i, 66.Gazzetla, 1908, 38, i, 526 ; A., i, 482144 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.When a p-quinoneoxime is employed instead of the p-quinone itselfin the foregoing condensation, then the primary product can beisolated.:.~'~..."H~~N.NH.C,H,.NO, . N*NH*C6H4*N0,i\-+ v,, . . . . . . . .It I'\/II IIgOH NOHStable p-quinoneoximehydrazone.I n this case the phenolic OH is replaced by NOH, so that the strivingis\ /""\C:O to become -(J/.-*-\ of thearomatic residue Cnon-existen t.To the foregoing evidence may be added the closely relatedtransformation of p-benzoquinonebenzoylphenylhydrazone into benzene-azo-p-phenyl benzoate under the influence of cold potassium hydroxide,already referred to in last year's Report (p. 126).The production of P-benzeneazo-a-naphthyl benzoate from as-benzoyl-phenylhydrazine and P-benzoquinone simply means that the ortho-quinonehydrazones are even less stable than the paraquinone-hydrazones and pass spontaneously into ortho-azo-derivatives.The reduction of this /3- benzeneazo-a-naphthyl benzoate has beencompared with that of its isomeride, a-benzeneazo-P-naphthylbenzoate ; the former gave benzanilide, aniline,, and N-benzoyl-2-amino-a-naphthol, whilst the latter yielded benzanilide, aniline, andN-benzoyl-1 -amino-P-naphthol.27 The corresponding acetates behavesimilarly on reduction.The reduction of benzene-o-azo-p-tolyl benzoate leads to the hydrazo-derivative, which, on heating in acetic acid, regenerates the azo-derivative and yields simultaneously N-benzoyl-o-amino-p-cresol andaniline : 28/ \...../aH3*C6E4<OH NHBz + Ph*NH2.Benzanilide was never obtained in this reduction.Bearing in view the fact that acyl groups readily shift from oxygent o nitrogen in aminophenols, Auwers considers that the foregoingThere is evidence that this compound can assume the dynamic azo-form,N0,*C,H4*N:N*C,H4'NH*OH, in certain of its reactions, as, for example, oxidation(Zoc.eit., p. 148).27 Be?-.$ 1908, 41, 403 ; A., i, 228.28 K. Auwers and M. Eckardt, Annulen, 1908, 359, 336 ; A , , i, 480ORGANIC CHEMISTRY. 145results favour the assumption that these esters are really azo-derivatives and not o-quinoneacy 1 h ydrazones.Mercuric acetate was shown by Dimroth to condense with phenolsand aromatic amines, entering the unsubstituted para- and ortho-positions with reference to the oxygen and nitrogen atoms respectively.On this account it has been employed in the study of hydroxyazo-compounds, for, as will be seen from the following formulae, thenumber of mercuriacetate groups introduced into the azo-compoundshould be a criterion of its constitution :X Xp - and o-Hydroxyazo-derivatives. p - and o-Quinonehydrazones.Considered as an azo-derivative, the para-compound (I) shouldcondense with not more than two mercuriacetate groups, whilstas hydrazone it should form a tri-substit uted mercuriacetate.Theortho-compound (11), i n its hydrazo-form, should also take up threemercuriacetate groups, but as an azo-derivative it should onlycondense with one molecule of mercuric acetate. I n every caseexamined, the result corresponded with the hydroxyazo-structure,both in the ortho- and para-~eries.2~These results, like the other chemical evidence already cited, are allin favour of the view that the constitution of both ortho- and para-hydroxyazo-derivatives is what their usual designations imply, andthat these substances have not in ordinary circumstances a hydrazonestructure.The spectroscopic evidence, as interpreted by Tuck, is, to acertain extent, against this view and in favour of the assumption thatthe ortho-hydroxyazo-compounds and their acyl derivatives have thequinonehydrazone constitution.Both Auwers and C. Smith, however,take exception to the comparisons instituted between the absorptioncurves, the former referring to the diEerence between the curve forbenzoquiaonebenzoylphenylhydrazone and those for a set of azo-derivatives, whilst the latter asserts that the curve for benzeneazo-p-tolyl benzoate, presumably a hydrazone, resembles that of benzeneazo-phenol, admittedly an azo-compound, more than t h a t of the foregoingquinonehydrazone.From the chemical evidence now available, it is permissible todeduce the general rule that a hydroxyazo-compound will be an azo-15) C.Smith and A. D. Mitchell, Trans., 1908, 93, 842.KEP.-VOL. V. 146 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.derivative or a hydrazone according as to whether its oxygenatedgenerator is an enol or a ketone respectively.The case of 2-pyridone (IV) is of interest in this connexion, forthis substance, which differs from the phenols in not giving the ferricchloride and nitrosoamine colorations and in not yielding acetyl andnitroso-derivatives, nevertheless forms an azo-compound, 5-benzene-azo-2-pyridone (I1 or 111), with benzenediazonium chloride.Theorientation of the substituents in this compound is established by thefollowing series of operations, starting with 6-hydroxynicotinic acid(I), a substance of known constitution : 30C,Hb*N:N(\ /\:o f- I/:.. c6H5*N20 NH-+ ,)OH orN NH(11.) (111.) (IT. 1This azo-derivative and the isomeric benzeneazo-3-hydroxypyridinemerit further examination from the point of view of the constitutionof hydroxyazo-compounds.Although the balance of chemical evidence is in favour of the viewthat all aromatic hydroxyazo-derivatives have the azo-structure, thequestion of the nature of their salts with the mineral acids next arises.Some years ago Hewitt showed that, on nitration, benzeneazophenolitself behaves in accordance with its hydroxyazo-c0nstitution,3~ and inthis respect differed from its sulphate, which behaves as if it were aquinonehydrazone.More recently it has been found that benzeneazophenol andits ethoxy-derivative, benzeneazophenetole, both admittedly azo-compounds, have similar absorption spectra in concentrated hydro-chloric acid.32 Accordingly the formula for the salts of benzeneazo-phenol must be applicable to the salts of benzeneazophenetole, and atthe same time should account for the chemical properties ofbenzeneazophenol sulphate.After discussing several formula?, Foxand Hewitt, who have prepared and examined spectroscopically thesalts of a series of azophenols and their ethyl ethers, decide onthe following structure : /-)NHqq: /=>:0<8(4H,)L \= c130 W.H. Mills and S. T. Widdows, Trans., 1908, 93, 1373.31 Trans., 1900, 77, 99. Tuck, ibid., 1907, 91, 450ORGANIC CHEMISTRY. 147The compounds having this constitution are characterised spectro-scopically by strong and persistent absorption in the yellow and greenand chemically by behaving towards substituting agents as quinone-hydrazones.33Heterocyclic Rings contain,ing Oxygen.The Cotmarin, Group-The coumarin condensation with malic acidor ethyl acetoacetate and the substituted phenols has been studied, theresults showing that phenols containing alkyl, hydrbxyl, or dialkyl-amino-groups in the positions indicated below give good yields of thecorresponding coumarins.XX()OH\//)OHXI vChlorine, as a substituent in these positions, has a similar effect,but to a less appreciable extent. The introduction of nitro-, carboxy-,or carbethoxy-groups prevents conde~sation.3~The condensation is affected considerably by the strength of thesulphuric acid employed; when 73 per cent.acid is used, a satisfactoryyield of 4-methylcoumarin may be obtained from phenol and ethylacetoacetate, otherwise the amount produced is very ~ m a 1 1 . ~ ~One of the most interesting, and still only partly explained, reactionsof coumarin is its hydrolysis into coumarinic and o-coumaric acids.When treated with sodium ethoxide, coumarin and 7-methylcoumnrinyield respectively ethyl o-coumarate and ethyl 4-methyl-o-coumarate.But under these conditions, 4 : 7-dimethylcoumarin, which containsone methyl group in the lactonic ring gives rise to more complexproducts, namely, 3 - [ 2 : 5-dimethylhydrocoumarilyl] - 4 : 7 - dimethyl-coumarin and 1-[2 : 5-dimethylhydrocoumarilyl]-2 : 5-dimethyldihydro-coumarone (11) respectively.A molecular proportion of the p-4-di-methylcoumaric acid produced by hydrolysis condenses with unaltered4 : 7-dimethylcoumarin, and the product then undergoes rearrange-ment 36 and a coumaryl ring is produced (I). The second product isformed from the first by hydrolysis and loss of carbon dioxide.The interaction of 4 : 6-dimethylcoumarin and sodium ethoxideleads t o the production of similar products differing from compounds(I) and (11) only in the position of the methyl substituent in thebenzene ring.The reduction of coumarin with zinc dust and alkali hydroxide33 Fox and Hewitt, Trans., 1908, 93, 333.34 A.Clayton, ibid., 2018.36 F. Peters and H. Simonis, Ber., 1908, 41, 830 ; 8.) i, 339.36 K. Fries and W. Klostermann, Annalen, 1908, 362, 1 ; A., i, 820.L 148 ANNUAL REPORTS ON THE PBOGRESS OF CHEMISTRY.CMeCMeleads to melilotic acid as the main product, and, in addition, to twoby-products, which are stereoisomeric a- and P-tetrahydrodicoumaricacids, the isomerism of which resembles that of the symmetricallydi-substituted succinic acids, and persists even when they are condensedinto tetrahydrodicoumarins.f\iCH2--- ~H-~H--cH,/\ I I\/OH HO*OC CO*OH HO\/a- and &Isomeric acids.Ia- and B-Tetrahydrodicoumarins.4 : 7-Dimethylcoumarin behaves differently on reduction, and givesrise to three products : 2-hydroxy-4-methylphenyldimethylcarbinol (I)(hydroxythymol), 2-hydroxy-a-4-dimethylstyrene (Il), and thymol(111)./\CM~,*OH /\C M 8: CH, /\CHMe2MP()OH Me(,)OH ~ e j / o a(1.1 (11.1 (111.ORGANIC CHEMISTRY.149The second product also exists in a polymeric bimolecular form.These vinylphenols and their polymerides may be produced bydistilling o-coumaric acid and its homologues under reduced pressure.The simplest member of the series, o-vinylphenol (o-hydroxystyrene),prepared from o-coumaric acid itself, readily polymerises.The bimole-cular form is insoluble in alkali hydroxides, and becomes depolpmerisedwhen distilled under the ordinary pressure, although under 15 mm.it passes over unchanged.37As the coumarins aro colourless, and do not condense with eitherhydroxylamino or phenylhydrazine, whilst the thiocoumarins areyellow and yield oximes and phenylhydrazones, it has been suggestedthat coumarin has the modified structure (I), whilst thiocoumarinretains the configuration (11) generally attributed to coumarin : 3876H4-0 CH:CH*C ' '> 7GH4-O>cs.GKCHThe FZavone Gv-oup.The residue *C:C*CO* is often associated with the development of . .colour, and i t may occur in compounds in four different ways; both*C:C* and GO* may be present in a ring, or both may occur in achain, or one may be in the ring and the other in the chain.As thecase of the ethyl linking in the ring (" cyclostatic ") and the carboxylgroup in the chain ("streptostatic") has not been studied,l-hydroxybenzoylcoumarone, a compound having the requiredstructure, has been synthesised from coumarilic chloride and phenolby the action of aluminium chloride,. .The colour is absent from the alkyl derivatives, The methylcompound is synthesised in the following manner : salicylaldehydeand p-methoxybenzophenone give 2-hydroxy-4'-methoxychalkone,OH*C,H,*CH:CH*CO.C,H,.OMe, a yellow substance containing bothchromophores in the streptostatic condition.The acetate of this compound yields a dibromide, which on heatingwith alcoholic potash furnishes the required compound, l-methoxy-benzoylcoumarone,3937 K.Fries and G. Fickemirth, Ber., 1908, 41, 367 ; A., i, 160.38 A. Clayton, Trans., 1908, 93, 524.39 F. Zwayer and 8. Jon Hostanecki, B e r . , 1908, 41, 1335 ; A., i, 443150 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY,The two chromophores are cyclostatic in the allied group of flavones,the hydroxy-derivatives of which exhibit colour, whereas their alkyland acyl derivatives are colourless.A series of flavones have been synthesised from orcacetophenonedimethyl ether and the isomeric isoorcacetophenone dimethyl ether,these ethers being produced by condensing acetyl chloride and orcinoldimethyl ether. One example of this synthesis will suffice to indicatethe general method.Orcacetophenone is condensed with methylo-methoxybenzoate, yielding 2 : 6 : 2’-trimethoxy-4-methylbenzoylaceto-phenone (I) ; this intermediate product, when boiled with concentratedhydriodic acid, condenses into 1 : 2’-dihydroxy-3-methylflavone (11)(greeni~h-yellow).4~A 0OH COBradin, Haematoxylin, and their Derivatives.The culminating point in a long series of researches on brazilin andhaematoxylin has been reached this year with the demonstration of theconstitution of these two substances and of the colouring matters,brazilein and hamatein, into which they are converted by oxidisingagents.Brazilein and hamatein are respectively the colouring matters ofbrazil-wood and logwood, both of which find extensive application indyeing; they stand in the same relationship to brazilin and haematoxylinas p-benzoquinone does to quinol.A s in both cases the quinol derivative is more amenable to chemicaltreatment than the quinonoid substance, the problem of ascertaining theconstitution of these substances has mainly been worked out withbrazilin and haematoxylin, The f ormer of these substances containsthree, and the latter four, hydroxyl groups, and it was found advisableto protect these by methylation before systematically breaking downthe compounds by oxidation.Trimethyl brazilin yields on treatment with potassium permanganatea number of acidic substances, all of which are of great importance intracing out a skeleton formula for the parent substance.The follow-ing are two pairs of these oxidation products.MeO/\O* CH,* CO,H CO,H/\OMeCO,Hi ’OMe2-Carboxy-5-methoxyphenoxy- sn-Hemipinic acid.40 J.Tambor, Ber., 1908, 41, 787, 793; A , , i, 349, 358.{,!CO,H \/acetic acid151 ORGANIC CHEMISTRY.0 CO,H/\OMe M ~ o ( \ ~ \ c H , CO,HCH,!,)OMe\,/\/(&OH)'CH2'CO2H 2-Carboxy-4 : B-dimethoay- co phenylacetic acid.Brazilic acid.Bearing in mind that the empirical formula of trimethylbrazilin isC,,H&,, a careful consideration of the oxidation products led to thegraphical formula for trimethylbrazilin (I) :0 0MeO/\/\ CH, HO/\/\CH, I I 'C(OH), \/\/I IC(OH)\CH CH2\-// \\.-/ \-./\A/ CHMe0 OMe HO OHI. 11. Brazilin.It will be seen that this constitution agrees with the formation ofthe foregoing oxidation products.I n confirming this formula, all themore important degradation products of trimethylbrazilin have beensynthesised with the exception of brazilic acid.O*CH,*CO,Et 0. CH2*C0,HMeof\(\/\ co co --+ co\-// \\-/Me0 OMeEthyl methoxyphenoxyacetateand m-hemipinic anhydride.\ CO2H/-\.\-/Me0 OMeI.M~o{) 0. CH, CO,H MeO(\OH1 \)\ /O-\CH CO\ / <I? Me0 OMe I-\ \-/Me0 OMe11152 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Of these syntheses, one of the most important is that of brazilinicacid (I), an oxidation product of trimet hylbrrtzilin, which on reductionchanges into the lactone (11) of dihydrobrazilinic acid. Both thesesubstances have been synthesised in the manner indicated on p. 151.A similiar synthesis, using m-hemipinic anhydride with pyrogalloltrimethyl ether instead of the foregoing resorcinol dimethyl ether, ledby a precisely comparable series of reactions to the lactone (111) ofdihydrohamatoxylinic acid :Me0MeO{)O*CH2*C02H\/O-\\ACH COMe0 OMe111.a result which proves that hamatoxylinic acid and haematoxylin mustbe represented respectively by formuls IV and V.41Me0 HO 0MeOAO*CH2*C02H H O / y \ C H 2L \ l C ( O R ) \I t\/\ 'CH \CH, co\ C O P \-//-\ / \\-/ \-/Me0 OMe HU OHIV.V. Hcematoxylin.Brazilein has the empirical formula C,4H,205,2H20, losing water a t130-140O. On methylation it yields trimethylbrazilein (11) andtetramethyldihydrobrazileinol, the former of these on treatment withdilute aqueous potassium hydroxide becomes hydrated to trimethyl-brazileinol (I) ; the tetramethyl derivative has the configuration (I)with methoxyl in place of the lower hydroxyl :0 0MeO/\/\QH, I MeOf\f\FH,\/\/C(oMei \\\ \\2\)\,C(OM")\?(OH) CH, %-- c: CH2/ >- / >- \--> Me0 a MeU OH1.11.41 W. H. Ytdtin, jun., and R. Robinson, Tmws., 1908, 93, 489ORGANIC CHEMISTRY. 153The trimethyldihydrobrazileinol is reconverted into trimethylbrnzileinon heating, this reversible change being quite comparable with thechange of trihydroxytriphenylcarbinol (111) into aurin (IV), andvice versa.111.-+ f-H O f \ r ()OHC'\/\/\/\/-\\-/ ..0IV.The close relationship existing between brazilin and hsmatoxylinindicates that hamatein is hydroxybrazilein. The behaviour ofhamatein on methylation fully justifies this hypothesis. The colouringmatter is converted into tetramethylhaematein (11) and pentamethyl-dihydrohsemateinol ; the former of these, when digested with dilutepotassium hydroxide, is hydrated to tetramethyldihydrohzmateinol (I),a change which is reversed on heating :Me0 0 Me0 0M ~ o < ) / \ ~ H ~ M e O ( \ / \ C H ,\/\/c(oMe)\\--/u CH2\/ \ /C(OMe)\C(0H) OH, 2\-.-.-)- /\ tjMe0 OH Me0 0I.11.These reactions, together with more confirmatory evidence, lead tothe following formulz for brazilein (I) and htlematein (11) : 420 HO 0I. 11.The synthesis of brazan, a substance obtained from brazilin, hasP. Engels, IT. 1%. Pcrkin, jun., and R. Robinson, Trans., 19OS, 93) 1.115154 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.been effected in the following manner from resorcinol and 2 : S-dichloro-naphthaquinone : 430 0 00 0 00.--with zincBrazan (o-phenylene-BB-naphthylens oxide).Pyyanol Salts.I n connexion with the experiments on brazilin and hzematoxylin,the synthesis and coustitution of certain pyranol salts have beenstudied. As a typical example of these compounds, one may take theproduct obtained by condensing either resorcinol and benzoylacet-aldehyde or P-resorcylaldehyde and acetophenone.The hydrochlorideof this pyranol contains an additional molecule of water, and may berepresented by the following formulze :H C1\/0 oc1I. 11.The balance of evidence is in favour of the second formula, whichrepresents the salt as 7-hydroxy-2-phenylbenzopyranol( 1 : 4)anhydro-hydrochloride with one molecule of water of crystallisation.All thehydrochlorides examined could be formulated in this way, exceptingthat some contained even more water of crystallisation.The platinichloride of the above pyranol was obtained in bothC3 S. yon Kostanecki and V. Lampe, Ber., 1908, 41, 2373 ; A., i, 671ORGANIC CHEMISTRY. 155hydrated and anhydrous (111) forms, and all the ferrichlorides describedwere free from wafer.c1 FeCl,0 0/ \ A C - A I I I ICH CH,111. IV.Formula (IV), for example, represents the anhydroferrichlorideof 2 : 3-indenobenzopyranol(l : a), the free base having the structureindicated by (V) :FeC14/\0 1 1/\0 0 1 1\A/\/I l l\/\/\/CHf /\/\c /\ /\A/\/I I I 1 I I I I ICH*OH CH2 CH*OHV.VI. VII.The ortho-quinonoid formula is adopted for these salts because theyare coloured, and in every way comparable with those of naphtha-xanthhydrol. Formulae (VI) and (VII) represent respectively this sub-stance and its anhydroferrichloride. It will be seen that the hetero-cyclic nuclei of 2 : 3-indenobenzopyranol( 1 : 4) and naphthaxanthhydrolare similarly constituted (V and VI), and their salts, for example,the anhydroferrichlorides (IV and VII), are formed in a similarmanner with elimination of water. Now the 8 naphthaxanthhydrolanhydroferrichloride must be formulated as an ortho-quinoid whetherthe double linkings are represented as in (VII) or whether they areturned towards the naphthalene nucleus.Accordingly, it may fairlybe assumed that the same ortho-quinonoid structure exists in thesalts of 2 : 3-indenobenzopyranol(l : 4).oc1I i I (yellow)156 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The action of hydrogen chloride on 2-hydroxybenzylideneaceto-phenone (I) under different conditions affords a noteworthy example ofring formation with production of oxonium salts. Hot concentratedhydrochloric acid acting on the ketone in glacial acetic acid produces2-phenylbenzopyranol anhydrohydrochloride (II), a very soluble saltisolated in the form of its platinichloride, (C,,H,,OCl),PtCl,.Dry hydrogen chloride and the ketone interacting in etherealsolution give rise to the dichloride (111) (the hydrochloride of2 -p henyl benzopyranol anhydrohydrochloride).The original memoir should be consulted for further examples ofthese interesting pyranol salts, and for a fuller exposition of thearguments adduced in favour of their constitutional f o r m u l ~ .~ ~A closely allied series of oxonium salts has been produced fromdisdicylideneacetone (I) ; this substance when treated with alcoholichydrochloric acid undergoes condensation to a red oxonium chloride(II), which furnishes a brick-red ferrichloride, Cl7H,,O2*FeCI4.Dilute aqueous sodium hydroxide converts the Z-o-hydroxystyryl-benzopyrylium chloride (11) into the sodium salt, Z-o-hydroxystyryl-benzopyranol-2 (III), from which, however, the free carbinol couldnot be isolated, as diluto acids induce further condensation with theformation of a substance containing two heterocyclic nuclei.Thisso-called dibenzospiropyran (IT) is hydrolysed by strong acids to theoxonium salt, and by alcoholic soda to the original disalicylidene-acetone (J).45_/cH:cH \ c o * a ~ : CH-C,H,*OH -+ / \\=/I (yellow).71-/--\c*cH: CH.C,H,*OH // \-O(-jl/\=’ \=/ I1 (red).i? ,CH:CH,111 (yellow solution).Indigotin ccnd Indigoid Dyes.The investigations recently carried out on the dyes of the indigogroup fall chiefly under four headings : (1) improvements in thephenylglycine synthesis of indigotin, (2) the production of halogenated44 W. H. Perkin, jun., R. Robinson, and M. R. Turner, Trans., 1908, 93, 1085.45 11. Deckcr and H.Felser, Ber., 1908, 41, 2997 ; A . , i, 906OKGANlC CHEMISTRY. 157indigotins, (3) the synthesis of indigoid dyes containing sulphur,(4) the synthesis of more complex dyes, chiefly of the anthraceneseries, which are capable of employment in the hyposulphite vat.1. Modifications in the process of synthesising indoxyl by heatingphenylglycine with alkalis have formed the subject of many recentpatents. The addition of magnesium powder t o the fused mixture ofphenylglycine with sodium and potassium hydroxides and barium oxidemore than doubles the yield of indigotin subsequently precipitated byaerial oxidation from aqueous solutions of the melt. E'usion undergreatly reduced pressure at 200--230° gives a n 80 to 90 per cent.yield of indigotin.462.A well-defined chloroindigotin, C16H902N2Cl, has been preparedby direct chlorination of indigotin suspended in hot nitrobenzene.Bromination in the same medium has led to the formation of a tri- andtwo tetra-bromoindigotins, and more highly halogenated indigotinshave been produced, containing both chlorine and bromine. Thesubstitution of hydrogen by halogen in indigotin increases thebrilliancy and fastness of the dye, and it is noteworthy that theleuco-derivatives of these halogenated indigotins differ from indigo-white in being coloured, the shade varying from yellow to brown.47The sulphonic acids of the halogenated indigotins have also beenprepared ; their tinctorial properties differ considerably from those ofindigo carmine.483. 2-Hydroxythionaphthe11, an important compound in the synthesisof thioindigoid dyes, is obtained, together with its carboxylic acid,C,H,<;(O~)>CH or c,H~<~-?>cH,,by condensing chloroacetic and thiosalicylic acids or their esters inalkaline solutions and then heating the intermediate product, carb-methoxy-o-thiobenzoic acid, C02H*C6H4*S*CH2*C02H, with sodiumhydroxide a t 180'.Oxidation of 2-hydroxythionaphthen, or itscarboxylic acid, results in the formation of a red colouring matter,thioindigotin or 2 : 2'-bi~thionaphthenindigotin,~~Thionaphthen (I), the parent substance of this group of dyes,is now readily obtained by reducing the synthetical 2-hydroxythio-naphtha with zinc and glacial acetic acid :I. 11. 111.4(i L. Lilienfeld, D.B.-P.189021, 195352 ; A., i, 371, 797.47 D.11.-P. 193438, 193970, 193971, 195085, 195291 ; A, i, 468, 695, 798.43 C. G. Scliwalbe and H. Jochlieim, Ber., 1908, 41, 3798 ; A . , i, 1019.D.R.-P. 192075, 194237, 194254 ; A . , i, 451, 672158 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.2-Hydroxythionaphthen yields 1 : l-dibromo-2-ketodihydrothio-naphthen (11), and this, on treatment with lead acetate, gives rise tothe intensely yellow thionaphthenquinone (111). Like hydroxythio-naphthen, the dibromide and the quinone are employed in thesynthesis of thioindigoid d ~ e s . 5 ~The interaction of thionyl choride, and various styrene derivatives,at 180-270°, leads to the formation of thionaphthen compounds,up-dibromostyrene giving rise to hexachlorothionaphthen,a substance containing only carbon, chlorine, and sulphur.51furnishes thioindirubin, 2 : 3-bi~thionaphthenindigotin,~~The condensation of 2-hydroxynaphthen and thionaphthenquinoneDyes containing indigotin and thionaphthen nuclei have beenIsatin and hydroxythionaphthen give rise to thioindigo- synthesised.scarlet, 2 - t hionaph then-3-indole-indigotin,The isomerides, 2-thionaphthen-2-indole-indigotin and 3-thionaph-then-2-indole-indigotin, are obtained from indoxyl with thionaphthendibromide and thionaphthenquinone respectively.The application of peri-carboxymethylthionaphthoic acid,1 8C0,H *CH,*S*C,,H,- CO,H,in the preparation of blue thioindigoid dyes has been patented.534.Certain complex anthracene derivatives containing nitrogen havethe property of yielding colouring matters suitable for the hyposulyhitedye vat.54Flavanthren, which possesses this property, may be convenientlyconsidered a t this stage.This yellow colouring matter is remark-able because it yields a series of reduction products, most ofwhich are more intensely coloured than the original substance. Theprogressive reduction of flavanthren indicates the formation of sevenreduction products, five of which have actually been isolated.A. Eezdzik, P. Friedlander, and P. Koeniger, Ber., 1908, 41, 227 ; A , , i, 200.51 G. Barger and A. J. Ewins, Trans., 1908, 93, 2086.6a P. Friedlander, Monatsh., 1908, 29, 359, 375 ; A., i, 673.53 D.R.-P. 198050 ; A , , i, 797.D.R.-P. 197554 ; A., i, 807ORGANIC CHEMISTRY.0159OR0Flavanthren (yellow).OHa-Hexahydroflavanthren hydrate(blue).IJ/OHFlavanthrinol hydrate (blue).tiDihydroflavanthren hydrate (blue).OH/\A/\I I I I+--OHa-Te t rah ydroflavan thren hydrate(not isolated, brown solution).'xHFlavanthrine hydrate (greenish-brown).The foregoing diagram represents, according to Scholl, the successivestages in the reduction of flavanthren.The products thus isolatedcontain water of hydration, which is assumed to be attached to twocarbons united by a double linking in one of heterocyclic rings.Dihydroflavanthren (green), a-hexahydroflavanthren (blue), flavan-thrinol (blue), and flavanthrine (brown) have, however, all beenobtained in the anhydrous condition by expelling the water a t150--2OOO. This dehydration is not accompanied by any markedchange in colour. The di-, a-tetra-,and a-hexa-hydroflavanthren hydratesare obtained by alkaline reducing agents, flavanthrinol is formed byheating a-hexahydroflavantbren hydrate, and flavanthrine hydrate isproduced by reducing flavanthren or the foregoing hexahydro-hydrat160 ANNUAL ItEPOltrIS ON THE PlWGEESS OF CHEMISTRY.with red phosphorus and hydriodic acid a t 210'.A t 170°, flavan-thren, when reduced, gives rise to another tetrahydroflavanthren, towhich in solution the following constitution (I) is ascribed, althoughwhen precipitated it probably exists in the ketonic form (11). ThisP-tetrahydroflavanthron cannot be hydrated, for it already containstwo hydrogen atoms in the position occupied by water in the otherreduction products.0 OH ..0I.B-Tetraliydroflavanthrenred in solutioii (e~iol).Acids. s t--organic solvents011. Green precipitated ketonicform.This production of intensely coloured .reduction products from afaintly coloured substance is a striking example of the rule that thepartial conversion of the chromophores of a colouring matter intoauxochromic groups brings about an intensification of the colour. Thepartial reduction of yellow picric acid to dark brown picramic acid isa case in point. This generalisation accounts for the colour of theintermediate reduction products OE flavanthren. The colour of thefinal product, flavtznthrine, like that of meso- and P-anthramines, maybe referred to the fact that they are all derivatives of anthracene, ahydrocarbon which must be regarded as having a chromophoric ortho-quinonoid constitution.55Oxazine Group.The oxazine dye produced by the condensation of nitrosodimethyl-aniline and methyl gallate is termed '' prune )' (I), and an interestingseries of derivatives may be obtained from i t by the action of anilineand similar aromatic amines. Pruneanilide was formerly supposed tobe an additive compound of the oxazine and aniline, but it has now beenshown to be a substituted derivative, the dye and aniline condensing inmolecular proportions while the hydrogen which should be displacedis employed in reducing another molecule of prune to its leuco-deriv-ative.Actually, the yield of pruneanilide was doubled by blowingair through the mixture, A similar result was obtained with ccelestinblue B [the oxazine (11) from nitrosodimethylaniliue and gallamide],55 R.Scholl, Bcr., 1908, 41, 2304, 2534 ; A,, i, 696, 740ORGANIC CHEMISTRY 161the yield of anilide being considerably increased by the introductionof air.0 O H 0 OHThe constitution of pruneanilide has been demonstrated by con-densing nitrosodimethylaniline and methyl dibromogallate,C, Br,(OH),* CO,Me,when bromoprune is produced, which must have the following con-stitution (111). When this bromoprune is heated with an alcoholicsolution of aniline, pruneanilide results. Accordingly the anilino-residue occupies the position adjacent to the quinonic oxygen (IV).The anilide of ccelestin blue 13 has a similar constitution.660 OH.2 111.2% Axine Group.The relationship between azines and quinoxalines ( ' I ethopyrazines ")is well illustrated by the conversion of the two isomeric a/?-dinaphth-azines into bases of the quinoxaline series.s-ap-Dinaphthazine is first/-\ /-\\-/\NNI II>-< --3 \ - A N >-<N/ II\/\/\ \/\/\I l l 01 I 1\/\/ \/\/0co, €€"\/56 E. Grandmougin and E. Uodmer, Ber., 1908, 41, 604 ; A.. i, 289.REP.-VOL. V. 162 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.oxidised by chromium trioxide into diketo-s-ap-dinaphthazine. Thissubstance, when heated with concentrated sodium hydroxide, gives riseto 2-phenylnaphthaquinoxalinecarboxylic acid, which, on heating, losescarbon dioxide, and yields 2-phen ylnaphthaquinoxaline (" 2-phenyl-naphthapyrazine ").as-ap-Dinaphthazine is converted into 3-phenylnaphthaquinoxaline by a precisely similar series of changes.57The prasindones are a group of hydroxyazonium bases containing ahydroxyl group in the para-position with respect to the azoniumnitrogen. In some cases, water is eliminated from the azonium andphenolic hydroxyls, so that the base exists in the form of an anhydride.An attempt to prepare the simplest member of this series has not beensuccessful. o- Aminodiphenylamine has been condensed with 3-amino-4-hydroxy-o-benzoquinone, and the following series oE changes hasbeen realised, but it was not found possible to isolate the basecorresponding with the final product : the prasindone nitrate.NI I'6*5 N/\C P , NO,When 2-anilino-l-aminonaphthalene was substituted for o-amino-diphenylamine in the foregoing condensation, the synthesis was carrieda stage further, but in this case the prasindone hydrate, althoughactually isolated, could not be dehydrated.5s/\ C6H5 OHFurther evidence in favour of Kehrmann's betaine formula forisorosindone (I) has been obtained by converting it into the chloride of57 0.Fischer and E. Schindler, Bar., 1908, 41, 390 ; A., i, 221.62 F. Kehrmann and R. Schwarzenbach, ibid., 472 ; A., i, 297ORGANIC CHEMISTRY. 163its acetyl derivative (IV) by two distinct processes. In one thecompound is simultaneously reduced and acetylated by zinc dust andacetic anhydride, the product, a diacetyl leuco-derivative (II), beingthen oxidised to the chloride.I n the other process the isorosindone istreated with acetic anhydride alone, when the acetate (111) of itsacetyl derivative is directly produced, and can be converted into thecorresponding chl0ride.5~/\C,HS 0-CO-CH,(111.)Carboxonium dyes with a structure somewhat similar to thepreceding substances have quite recently been obtained by condensingm-acetylaminophenol and benzotrichloride in nitrobenzene a t 1 Go.One of these is acetylaminophenylfluorone (I), to which an ortho-quinonoid structure is ascribed. The non-acetylated base is of greatCPh CPhinterest, as on eliminating the amino-group, phenylfluorone (11) isobtained, which is the chromogen of fluorescein.Hydroxyphenylfluorone(111), produced by replacing the amino-group by hydroxyl, is identicalwith resorcinolbenzein, and is remarkably like fluorescein, which, as its69 F. Kehrmann and K. L. Stern, Beer., 1908, 41, 12; A . , i, 220.Y 164 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY,carboxylic acid, should, if Kehrmann's formulation be accepted, havethe formula (IV).C,H,*CO,HCPh 6(111.) (IV, 1The other condensation product is diacetylaminophenorosaminechloride (V), which yields an unstable colour base changing rapidlyinto a colourless, stable form. Hydrolysis leads to the simplest rosamine,tho salt of which is represented by (V.I).,OCPh CPh/\A/\ /\A/\I l l\/\//\/ NHAJ I I JNHA~ NH,(/\//\/NH2OCl oc1The Boaaniline Group.The researches of Baeyer and Villiger have shown that many basesof the triphenylmethane group exist iu two forms, one colourless andthe other coloured. There is little doubt that the colourless baseshave the carbinol formula, HO*C(C,H,NR,),, but some uncertaintystill exists as to the nature of the coloured amines.I n many casesthe coloured amine is the dehydrated imino-base, which has only beenobtained pure when the amino-groups are phenylated, as inC,H5N :C6H4:C( C6H,-NH*C,H5),.The metliylated rosanilines have as yet only been isolated in thecolourless carbinol form, and although the observations of Hantzschand Osswald indicate that isomeric quinoneimonium bases of thisseries may exist, these substances have not been obtained in a state ofpurity.Naphtho-blue, NMe,CI:C,,H,:C(C,H,*NMe,),, has yielded its basein two isomeric forms, the colourless cnrbinol (I) (m.p. 155') andthe dark green, quaternary ammonium hydroxide (11) (m. p.26 I"), which may possibly have the alternative pseudo-ammoniumformula (111).76H4*WH8)2 y,H,*N(CH3)2H 0.7 C,,H,*N (C HJ2(1.1~:C,,H,:N(CH,),*OHC,H, *N (CHJ 2(11.)CGH4*N(CH3)260 F. Kehrmann and 0. Dengler, Ber., 1908, 41, 3440 ; A., i, 1002ORGANIC CHEMIS'I'RY. 165(111.)Victoria-blue R, NHEtC1:C,oH6:C(CsH4*NMe2)2, has also given riseto a coloured as well as a colourless base, but these substances are notisomeric. The colourless compound is the carbinol (I) (m. p. 170°),but the coloured substance is the anhydrous imino-base (11) (m.p.1 9 2 O ) .7GH4*N(CH3)2 VBH, * N ( C H 3 ) 2HO.Q*CI,H6*NH*C9H, 7 : C IoHG: N C 2 H,C6H4*N(CH3)2 C6H4*N(CH3)2(1.1 (11.1The composition of these two pairs of diphenylnaphthylmethanebases has been confirmed in each case by complete analyses,6l whichshow that the substances were obtained in a fairly pure state.The corresponding diphenylnaphthylmethane colouring matter,containing methyl instead of ethyl in the foregoing formulz, givesrise also to the corresponding colourless carbinol and the colouredim ino-base.All attempts to isolate the quinonoid form of the bases from crystal-violet, malachite green, and o-chloromalachite green were unsuccessful ;although the coloured bases could be seen in solution, they were toounstable to be anaiysed.G2One of the nitration products of thiodiphenylamine is the dinitro-sulphoxide (I), which has been taken as the starting point in thesynthesis of 8-phenylphenazothionium hydroxide and its salts.Whencondensed with phenol or phenetole in concentrated sulphuric acid, thesulphoxide gives rise t o the sulphate of a phenazothionium base, thechloride of which is represented by formula (11).8-Phenetyl-3 : 3'-dinitrophenazothionium chloride undergoes hydro-lysis to the base 8-pbenetyl-3 : 3'-di&trophenazothionium hydr-oxide (111), but when treated with aqueous alkali hydroxide theanhydride (TV) (anhydro-S-phenetyl-3 : 3'-dinitrophenazothionium) isset free.til I<. Noelting aiid K. Yhilil)p, Bcr., 1908, 41, 579 ; A., i, 295.y2 Bid., 3908 ; A., 1909, i, 61166 ANNUBL REPORTS ON THE PROGRESS OF CHEMISTRY.NH NH/\C1 C,H,*OEt ''/ 11.Green salt.1L *7 $HydrolysisN N,A/\/\ /A//\/\HO,N:I I I I\ / \ / V N 0 2S S6,H,*OEt /\OH C,H,*OEtIV. Reddish-brown anhydride. 111. Crimson hydroxide.The chloride (11) has been reduced to a diamine, which, on oxidation,gives rise to X-phenetyl-3 : 3'-diaminophenazothionium chloride (V).N/\//\AI I I I/ \ / \ / V N H 2 HN SCorresponding phenazothionium salts have been produced containinghydroxyphenyl i n the place of ~henetyl.6~Thiopyrine, the sulphur analogue of antipyrine, is regarded byMichaelis as having the formula (I). It has now for the first timebeen converted into l-phenyl-3-methyl-5-thiopyrazolone (IV) in themanner indicated in the following diagram :N*C,H, N*C,H,/\ /\Me# -S -2 (C&5'COC1) MeRC1 E.S*Co*C6H5MeC--CH --- +- MeC-CH(I.)$0 heleg;gN N C,H,/\G*S*CO*C,H,/\ 8 g*SH HydrolysisMeC -CH f-- MsC-CH(117.)6s S, Smiles and T.P. Hilditch, Trans., 1908, 93, 145, 1687ORGANIC CHEMISTRY. 167The presence of the thiol group SH is indicated by the mode inwhich the thiopyrazolone undergoes oxidation. Alkaline hydrogenperoxide gives rise t o the sulphonic acid, whilst nitrous acid or iodinein potassium iodide solution leads to the disulphide. Nevertheless,I-phenyl-3-methyl-5-thiopyrazolone also reacts in its thiocarboxyl form,and, like phenylmethylpyrazolone, it condenses with aldehydes andketones.64AZJcaloicls.Damascenine, the alkaloid from Xigella, has been investigated, andits constitution established.Boiling with alkalis converts it into anisomeride, damascenic acid, which by a series of steps yields %amino-3-hydroxybenzoic acid, proving it to be 2-methylamino-3 -methoxy-benzoic acid (I). The alkaloid, being devoid of acid properties,appears to be a betaine (11).C0,H co.0/\NHMe /\&H,Me I lOMe (,!OM.(11.1\/(1.1Methyldamascenine, present in the same plant, proves to be themethyl ester of damascenic acid. An attempt to synthesise thealkaloid, starting from methylanthranilic acid, failed owing to theimpossibility of introducing methoxyl 01’ hydroxyl in place of theamino-group in position 3, the diazonium compound changing into astable azimino-~ompound.~~Atropine may be synthesised by a method giving a much better yieldthan does the condensation of tropine with tropic acid, by acetylationof tropic acid, conversion into the chloride, condensation with tropinehydrochloride, and elimination of the acetyl group, the last processtaking place spontaneously when the acetyltropine is allowed t o remaina short time in aqueous solution.On the other hand, when an attemptis made to replace the chlorine in P-chlorohydratropyltropine byh ydroxyl, an intramolecular change takes place, hydrogen chloridebeing transferred, and ccpoatropine hydrochloride is obtained :CH2*CH-CH2 CH,C1 CH,--CH-CR, yH2 I &Me dH*O*CO*bHPh -+ 1 :>$Me C!lH*O*CO*C!PhCH2*CH--UH, UH,--CH--UH,This apoatropine synthesis is a general one for tropeines, startingfrom a-, p-, or y-halogenated propionic or m-butyric acids.66I I 1A.Micliaclis, AnmZcii, 1908, 361, 251 ; A . , i, 685.m 0. Keller, Arc7~. Phrtrnz., 1908, 246, 1 ; A . , i, 283.fi6 R. Wolffenstein and I,. Mamlock, Ber., 1908, 41, i 2 3 ; R. Wolffenstein and,J. Rolle, iEid., 733 ; A., i, 281, 282168 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.A contribution t o the study of cinchonine has been made in theform of an examination of cinchoninone, the ketone obtained byoxidation of the alkaloid. This ketone is amphoteric, and also exhibitsenol-keto-tautomerism, yielding both an oxime and a n 0-benzoylderivative. By the action of nitrous acid, cinchonic acid and an oximeare obtained, the latter yielding Koenig’s meroquinenine on hydrolysis.This leads t o formulse (I and 11) for cinchoninone and cinchoninerespectively : 67CH,* CH--CH C H : C H, CH,*CH-CH*CH:CIT,I II l l I II IUH--N-CH CH-N-CH,CO* C,NH, OH* CH C,NH,(1.) (11.1Although little progress has been made towards establishing theconstitution of the strychnos alkaloids, nevertheless crystalline acidshave now been obtained by oxidation, the composition of whichindicates that both alkaloids contain a secondary alcohol grouping66When papaverinium alkyl salts are treated with very dilute alkalis,the liberated papaverinium hydroxides undergo condensation to phenol-betaines.Thus N-methylpapaveriniurn methoaulphate yields thephenolbetaine,C,H2(0Me)*CH“Me-CHC,H,(OMe),=CH,*C/ >O )I .69I n the morphine group of alkaloids, a synthesis of a compoundobtained in the partial breaking down of morphine has been effectedfor the first time.Morpholquinone is known t o be 3 : 4-dihydroxy-phenanthraquinone, and this may be prepared from 3-nitrophenan thra-quinone by reduction, diazotisation, and conversion into the 3-hydroxy-compound, nitration in the 4-position, a repetition of the reduction anddiazotisation giving morpholq~inone.~~q-apocodeine is the 3-methyl ether of apomorphine, and thusstands in the same relation to apomorphine as codeine does t omorphine. 71Much work has also been carried out on the constitution of65 P. Rabe, BcT., 1908, 41, 62 ; A . , i, 100.LM H. Lenchs, ibid., 1711 ; A., i, 563.‘jg H. Decker and G.Dunnnt, Amalcn, 1908, 358, 288 ; A . , i, 204.7o ,J. Schmidt and J. Still, Bey., 1908, 41, 3696 ; A . , i, 995.7l L. Know and F. I:nxlw, ibid., 3050 ; .4., i, 908ORGA4NIC CHEMISTRY. 169sparteine T2 by means of a study of the reactions of a-methylsparteine.When degraded by Hof mann’s reaction, methylsparteine yields methyl-hemispar teilene,CH CHThe methylsparteines behave as unsaturated bases, having a methylgroup attached to nitrogen. The isosparteine into which they may beconverted is a ditertiary base, which does not reduce acid per-manganate. The conversion may be carried out by heating a-methyl-sparteine di-iodide with water at 1 2 5 O , when isomeric change toisosparteine methiodide takes place./CH(C,HJ,N)*CHCH.CH, /c (cSH14N)0CH2\ CH ~ - CH,*CH,-- ,NMe*OH -+ CH-\CH2--- CH,’ \CH,--- CH2-+ CH ~ CHM e/Na-i\lethylsparteinium a-Methylsparteine.hydroxide./CH( CsH,,N)*C‘H,\\CH2--- C*2isoSpar te i 11 e .The reverse change may be carried out by heating a-methyliso-sparteinium hydroxide in a vacuum, when it gives a-methylsparteine.The latter compound takes up two atoms of iodine to form themethiodide of an iodo-base, which behaves as if it were an iodoiso-sparteine,CH- /CJww14PJ)*C CH(CH,I)--H~=~;~Y~I,\CH2---- CH2but it is possible that its formation is due to the isomerisation ofthe a-methylsparteine under the influence of hydrogen iodide.Jateorrhizine and columbamine, alkaloids from columba root, havebeen further in~estigated,~3 and the monomethyl ether of columbamineis found to be identical with the dimetbyl ether of jateorrhizine.The acid obtained on oxidation of this ether is a trimethoxy-o-71 C.hlourco a i d A. Valenr, Compt. wnd., 1907, 145, 815, 929, 1184, 1343 ;1908, 146, 7 9 ; 147, 127 ; BdI. Soc. chim., 1903, [iv]: 3, 674 ; A . , i, 43, 44, 103,206, 563, 736, 1006.7s K. Feist, A 7 d ~ . Phnivn., 1907, 245, 586 ; d., i, 100170 ANNUAL REPORTS ON THE PROGKESS OF CHEMISTRY.phthalic acid, but it has not yet been determined whether themethoxyl groups have the 3 : 4 : 5- or 3 : 4 : 6-position. A thirdalkaloid, palmatine, closely resembling berberine, is also presentin the root, but its relation to the other two is unknown.Polypeptides.The progress in the building up of polypeptide niolecules has beenexceedingly rapid during the past year.Perhaps the most importantstep that has been made is that of introducing tyrosine groups intothe molecule. The tyrosine group is of such frequent occurrence innatural proteins that this advance was necessary before compoundsshowing the reactions of the true proteins could be prepared, especiallyas the difference between the naturally occurring and the syntheticpolypeptides is now recognised as consisting less i n the number ofamino-acid groups present than in the conjunction, in natural sub-stances, of groups of several different kinds.The usual method of synthesis is not applicable when an amino-hydroxy-acid, such as tyrosine, is employed, because of the actionof phosphorus pentachloride on the hydroxyl group.This may beprotected by the introduction of the carboinethoxyl group, whichresists the action of phosphorus pentachloride and*acyl chlorides, andis readily removed afterwards by hydrolysis. The protecting group isintroduced by means of methyl chlorocarbonate. The products are,however, inactive.74A tetrapeptide was isolated last year from silk, and was shown toconsist of two glycine groups and one group each of d-alanine andZ-tyrosine. Several attempts to synthesise this substance have beenmade, but in every case the product was found to differ from thenatural polypeptide in not being precipitated by ammonium sulphate,or in only being precipitated from very concentrated solutions. Suchcompounds, isomeric with the tetrapeptide sought after, have beenprepared by the action of Z-tyrosine ester on chloroacetyl-d-alanyl-glycine, and by coupling glycine-d-alanine anhydride with chloro-acetyltyrosyl chloride methyl carbonate, followed by hydrolysis.Theglycyltyrosylglycyl-d-alanine obtained in the second case is probably amixture of stereoisomerides.75 Various tyrosine polypeptides, derivedfrom glycine, d-alanine, and Z-leucine, were also not precipitatedby ammonium ~ u l p h a t e . ~ ~The 3 : 5-di-iodo-Ltyrosine group has also been introduced, onaccount of its occurrence in the proteins of coral. Glycyldi-iodotyro-i4 E. Fischer, Sitzuizgsber. K. Akad. Wiss. Berlin, 190s) 542 ; A., i, 544.75 E. Fischer, Ber., 1908, 41, 850, 2860 ; A , , i, 324, 887.76 E.hbderlialden and A. Hirszowski, ibid., 2840 ; A . , i, 8S7ORGANIC CHEMISTRY. 171sine has been ~ynthesised,7~ starting from the action of iodine ontyrosine.Various dipeptides have been prepared containing the d-valinegroup, which are of interest, not only in connexion with polypeptidesynthesis, but also for the stuiy of the Walden inversion.78 Morecomplex groups have also been introduced, notably the a-aminostearylgroup 79 and P-amino-acid residues, such as those of P-aminobutyricacid and a-methylisoserine.80It is found that glycine ethyl ester, like ethyl oxalste, may bereduced with sodium amalgam, the final product, after treatment withalcoholic hydrogen chloride, being aminoacetal.81 The appiication of asimilar method of reduction to polypeptides gives unsatisfactory yields,but the corresponding aminoacetals are readily prepared by the con-densation of aminoacetal with chloroacyl chlorides and treatment ofthe products with ammonia.The presence of histidine in many natural proteins, arid the exist-ence of proline (pyrrolidine-2-carboxylic acid) in the product,s of thehydrolysis of’ gelatin, have led Fischer and his pupils to include thesegroups also in the synthetic scheme, and methods of synthesis havetherefore been devised for the purpose of preparing the materials.82Z-Leucyl-Z-histidine is stable towards concentrated hydrochloric acid,and since both Z-leucine and Z-histidine are present in oxyhzemoglobia,the dipeptide was sought for, but unsuccessfully, in the products ofhydrolysis of that substance.The use of hydrofluoric acid has been recommended for the hydrolysisof proteins, as causing less secondary reactions than any other acidBs3Synthetical Thevapeutic Agents.Considerable activity is being shown in the production of localThis substance itself is the anaesthetics of the “novocaine” type.hydrochloride of diethylaminoethyl p-aminobenzoate,NH,*C,H4*C0,*C2H4*N( C2H5),,HC1.A series of similarly constituted compounds has been prepared, andtheir physiological action has been ascertained.Some of these sub-stances have pronounced local anaesthetic properties, but are not77 E. Abderhalden and hf. Guggenheim, Ber., 1908, 41, 1237 ; A . , i, 420.78 E. Fischer and H. Scheibler, Annalen, 1908, 363, 136 ; A., i, 957.79 E.Fischer and W. Iiropp, ibid., 362, 338 ; A., i, 773.s1 E. Fischer, Ber., 1508, 41, 1019 ; d., i, 323.82 E. Fischer and A. Iirkimer, ibid., 2728 ; E. Fischer and I,. H. Cone, Annalen,1908, 363, 107 ; E. Fischer and G. Reif, ihid., 118 ; A., i, 858, 1004, 1007.83 L. Hugounenq and A. Morel, Co?npt. rend., 1908, 146, 1291 ; A., i, 706.F. W. Kay, ibid., 348 ; A . , i, 773172 ANNUAL mpoms ON THE PROGRESS OF CHEMISTRY.suitable for use in medicine, owing either to their high general toxicityor to the local irritation produced by their i n j e ~ t i o n . ~ ~I n view of these results, it is open to doubt whether the many com-pounds recently described in patents are, in general, of much therapeuticvalue.The hydrochlorides of the clialkylaminoalkyl benzoates, such asC,H4-C02*CH2*CH2*NEt,,HCl, are stated to be useful anasthetics,*5whilst the alkylaminoalkyl aminocinnamates (for example,NH,*C,H,*CH: CH*CO;CH,* CH,*N Ett2)are said to surpass the alkylaminoalkyl benzoates in this respect.86It is claimed that the alkylaminoalkyl salicylates unite theseanasthetic properties with those of salicylic a ~ i d .~ 7Salicylic acid and its acetyl derivative (“ aspirin ”) have certainundesirable physiological properties, which are said to be favour ablymodified in the recently-described :anhydrides of acylsalicylic acids,Of these products, acetylsalicylic anhydride, O(CO*C6H,*O*CO*CH,),,and cinnamoylsalicylic anhydride, O(CO*C6H,*O*CO*CH:CH*C6H5)2,appear to be the most promising.88The recent commercially successful synthesis of the powerfulhaemostatic adrenaline by the Farbwerke vorm.Meister, Lucius, andBriining has stimulated research in this direction. An interestingseries of observations on the conversion of catechol methylene ethersinto cyclic carbonates by the action of thionyl chloride may be men-tioned at this stage.89The beazoylaminoacetylcatechol ethers,. C,H4*CO*NH*CH,*CO*C,H,(OR),,when hydrolysed with aqueous acids under pressure, yield arnino-OH()OHi/ CO*CH,*N H,acetylcatechol, which is stated to have powerful hamostatic propertieslike the active principles of the suprarenal capsules.g0The root of Canadian hemp (Apocynum cannmbinum) retards thebeart in systole. Apocynin, one of its physiologically active con-84 F.1,. Pyman, TI-am., 1908, 93, 1799.D.R.-P. 187209 ; A . , i, 167.8: D.R.-P. 188671 ; A., i, 176.h8 D,R.-P. 201325 and 201326 ; A . , i, 954.89 G. Earger and A .J. Ewins, ! L ’ r c ( ~ i ~ . , 1908, 93, 563, 735, 2081.yo D.E.-1’. 189483 ; A., i, 262,*‘j D.R.-P. 157593 ; A., i, 169ORGANIC CHEMISTRY. 173stituents, has been examined, and shown to be identical with aceto-vanillone. This conclusion has been confirmed by the synthesis ofapocynin from vanillin.g10 Bz()OM,CHO4BIg M e I \/ --+OH{)OM. -+\/C'HOVanillin.OBz 0132//)OM8 {)OM,CrOs \/ ++ \/VH*OH YoCH3 CH,Benzoylapocy nin. Benzoylapocynol.I I .i. OH+(HydrolysisOHf)OMe ()OM.\/ Yo \/ f-NafEtOHTH*OH.CHsApocynol.CH3Apocynin.5 : 5-Diethylbnrbituric acid (I' veronal") and its homologues stillreceive much attention, and new methods of preparation form thesubjects of many patents.Practically quantitative yields of theseacids may be obtained by treating with nitrous acid the dialkgl-malonylguanidines produced by condensing ethyl dialkylmalonateswith guanidine.92 A detailed study of the dialkylmalonic acids hasbrought to light the interesting fact that complex anhydrides of thesesubstances may be produced by treating the acid chlorides withaqueous pyridine. The duodecimolecular anhydride of diethylmalonicacid has in ethylene dibromide or benzene a complexity correspondingwith the formula [~$>C<~~>O],,. In nitrobenzene the degreeof association is diminished to that of an octamolecular anhydride.93Orgcmic Derivatives of Arsenic.The last few years have witnessed a great revival of interest in thestudy of organic derivatives of arsenic, owing to the circumstance thatcertain of these substances have proved efficacious in the therapeuticsof diseases of protozoic origin.Sodium p-aminophenylarsonate ('' atoxyl ") and its acetyl derivativehave been successfully employed in the treatment of trypanosomiasis(sleeping sickness).The former of these substances was discovered byq1 H. Finnemore, Trans., 1908, 93, 1513, 1520,B2 D.R.-P. 189076 ; A., i, 370.98 A. Einhorn, Annalen, 1908, 359, 145 ; A . , i, 312174 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.BBchamp in 1863, but its true nature was discovered only in 1907 byEhrlich and Bertheim.p-Aminophenylarsonic acid, NH,*C,H,*AsO(OH),, is the chiefproduct of the interaction of aniline and aniline arsenate a t 180°, butPyman and Reynolds have recently demonstrated that a small amount(2 or 3 per cent,) of bis-p-aminophenylarsinic acid (I) is simultaneouslypr0duced.~40These investigators have also prepared the corresponding bis-2-aminotolyl-5-arsinic acid, so that the reaction is probably a general one.These compounds are the aromatic analogues of cacod ylic acid,AsMe,O*OH.The camphor analogue, dicamphorylarsinic acid (11),has recently been isolated from the products of the action of arseniouschloride on sodium camph0r.~5The use of arsenious chloride for the introduction of arsenicinto aromatic compounds was first studied by Michaelis andhis collaborators.The dimethylaminophenylarsine oxide,NMe,*C,H4*As0,obtained from the chloride and dimethylaniline, has now been oxidisedt o dimethyl-p-aminophenylarsonic acid (dimethylatoxyl) by alkalinehydrogen peroxide.96Aminoarsonic acids can be obtained from all primary aromaticamines having an unsubstituted para-position, and by means of thediazo-reaction the corresponding hydroxy-arsonic acids have beenprepared.97Phenylarsonic acid gives a nitrophenylarsonic acid, which, onreduction, yields an aminophenylarsonic acid isomeric with the atoxylacid. The isomeride is probably a meta-derivative; unlike atoxyl, itretains its arsenic on boiling with hydriodic acid. I n the para-seriesthe arsenic is replaced by iodine.98The arsenic in these compounds is not removed by boiling withaqueous alkali hydroxides, but when fused with these reagents it iseliminated as alkali arsenate.The alkyl esters of arsenious acid, As(OEt),, etc., may be obtainedby heating arsenious oxide with the primary alcohols in the presenceB4 Trans., 1908, 93, 1180.y5 G.T. Morgan and F. M. G. Micklethwait, ibid., 2144,96 A. Michaelis, Ber., 1908, 41, 1514 ; A, i, 590.y7 L. Benda and R. Kahn, ibid., 1672, 3859 ; A., i, 591.g8 A. Bertheim, ibid., 1455 ; A., i, 590ORGANIC CHEMISTRY. 175of an insoluble dehydrating ngent (for example, dry copper sul-hate).^^When arseniousoxide is added to an ethereal solution of magnesiumphenyl bromide, it is found to undergo the Grignard reaction, with theproduction, after about thirty minutes, of diphenylarsine oxide,[ As(C6H,),],0.Prolonged treatment leads to the formation oftriphenylarsiue, As( C,H,),. The tri-p-tolylarsine, As(C,~;CH,),, issimilarly prepared, but in this case is unaccompanied by the 0xide.lMagnesium benzyl bromide gives rise to dibenzylarsine hydroxide,As(CH,*C6H,),*OH,H20.Silicolz Compounds.The chemistry of silicon derivatives offers many points forcomparison with that of carbon compounds of similar structure,remarkable similarities and equally striking diEerences of behaviourbeing observed. Thus the silicones, the silicon analogues of theketones, are not only compounds of considerable molecular com-plexity, tending to form molecules of the form (R2SiO), and boiling atvery high temperatures, but they are not reduced by the usualreagents for reducing ketones, and do not form oximes or phenyl-hydrazones.The latter circumstance may be due to the readinesswith which the Si-K linking is broken by water. Benzylethyl-silicone, 6 5 c H2>Si:0, is prepared with ease by the action ofwater on benzylethylsilicon dichloride. Two isomeric dibenzylsilicols,Si( CH2* C6H5)2(OH)2, were obtained, differing considerably in stability,but both yielding the silicone by loss of water. The nature of theisomerism remains unexplained, and other results which have beenobtained suggest that the isomerism of silicon-oxygen compoundsmay prove to be of a nature not hitherto encountered amongst carboncompounds.Stereoisomeric silicon compounds of such optical activity as to leaveno doubt as to the occurrence of optical antipodes in this series, havenow been prepared by sulphonating benzylet hylpropylsilicyl oxide andthe corresponding compound containing isobutyl in place of propyl,and crystallising the salts of the sulphonic acids with active methyl-hydrindamines. The sodium salts of the active isobutyl compoundshave a molecular rotation nearly twice as great as that of the propylcon~pounds.~C H *CH2 5SJ W.R. Lang, J. F. Mackey, and R. A. Gortner, Trans., 1908, 93, 1364.F. Sachs and H. Kantorowicz, Ber., 1908, 44, 1031 ; A,, i, 1031.R. Robisoii and F. S. Kipping, Trans., 1908, 93, 439.F. S. Kipping, ibid., 457 ; B.D. W. Luff and-F. S. Kipping, ibid., 2004, 2090176 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The utility of the Grignard reaction is very apparent in this work.I n the researches quoted above, for instance, benzylethylsilicondichloride, Si(C7H7)EtC12, was caused to react with magnesium iso-butyl bromide, giving benzylethylisobutylsilicyl chloride, which yieldsthe required silicol or oxide on treatment with water. The siliconicacids, the analogues of the carboxylic acids, may be prepared by theaction of magnesium alkyl or aryl haloids on silicon tetrachloride, anddecomposition of the resulting compound with water. Thus silico-butyric acid (propylsiliconic acid) is prepared by the followingreactions :PraMgBr + SiCI, = PraSiC1, + MgClBr.P&3iC1, + 2H20 = PFSi0,H + 3HC1.The trichlorides form ortho siliconic esters with alcoh01.~These esters may also be prepared by the action of magnesiumorganic haloids on ethyl orthosilicate, Si(OEt),, but it is only possibleto replace one ethoxyl group by this process. I n this way, variousaryl-substituted orthosiliconic esters, such as the xylyl and a-naphthylderivatives, have been prepared, yielding siliconic acids whendecomposed with water.5Silicochloroform reacts with aniline in benzene solution to formtrianilinosilicon hydride, SiH(NHPh),. This compound is useful as asource of silicoiodoform, which is obtained from it in good yield bythe action of hydrogen iodide in benzene solntion.GSulphur Compounds.The properties of the organic derivatives of sulphur present manypoints of interest, and it is not surprising that the number ofresearches dealing with these substances is on the increase. Theisomerism and tautomerism of many of them has an importantbearing on a department of chemistry which still remains veryobscure, namely, the structural arrangement of inorganic com-pounds.The action of alkalis on sodium alkyl thiosulphates has previouslybeen little investigated. The principal action appears to be theformation of the disulphide: 2R*S20,Na -+ R,S,. It is notnecessary, for the preparation of the disulphide, that the thiosulphateshould be isolated. Thus the product of the action of p-nitrobsnzylchloride on sodium thiosulphate, when treated with sodium carbonate,yields di-p-nitrobenzyl disulphide directly.7 The same is true ofW. Melzer, Ber., 1908, 41, 3390 ; A , , i, 967.E. Khotinsky and B. Seregenkoff, ibid., 2946 ; A,, i, 1032.0. Ruff, @id., 3738; A., i, 966.T. S. Price and D. F. Twiss, Tram., 1908, 93, 1395, 1401ORGANIC CHEMISTRY. 1 7 7alkyl compounds, and the substance obtained in solutioii byGutmann 8 from sodium ethyl thiosulphate, and supposed by him tobe EtSOH, is diethyl disulphide. Other disulphides may be betterprepared by the electrolytic oxidation of thiosulphates, dithio-diglycollic esters, for instance, being prepared in this way.gCertain disulphides, such as benzyl disulphide and 4 : 4'-dithio-acetanilide, are found t o occur in two isomeric modifications, oE whichone is converted into the other by the action of light.1° The natureof this isomerism remains unexplained.The oxidation of sulphides to sulphoxides is most satisfactorilyperformed by means of hydrogen peroxide, both for aromatic l1 and fattycompounds. Thus thionyldiglycollic acid, SO(C H,*CO,H),, is preparedin this way from thiodiglycollic acid.1,The constitution of thianthren (diphenylene disulphide) has beendefinitely proved to be that of an ortho-compound, by oxidation t o thedisulphone and treatment with phosphorus pentachloride, the productsbeing benzene-o-disulphonyl chloride and o-dichlorobenzene :Organic polysulphides, uplike their inorganic analogues, are notreadily prepared by the action of sulphur on alkyl mercaptides,oxidation taking place. They are obtained in certain cases by theaction of sulphur on disulphides in absolute alcoholic solution,saturated with anhydrous ammonia, but quantitative yields of tri-sulphides are obtained by the action of thionyl chloride on mercaptansaccording to the equation4R*SH + SOCI, = R,S, + R2S, + H20 + 2HC1.14The di- and tri-sulphides are separated by fractional distillation. Thetetrasulphide of acetic acid is obtaineJ by the action of sulphur chlorideon thiolacetic acid in ethereal solution.Sulphination, or the introduction of sulphinic groups into aromaticcompounds, is carried out by passing sulphur dioxide into a mixtureof the compound to be sulphinated with aluminium chloride. Thismethod has been described independently by two groups of workera.15The process is accelerated by passing in hydrogen chloride simul-A. Gutmanu, Bes.., 1908, 41, 1650 ; A., i, 497.T. S. Price and D. F. Twiss, Trans., 1908, 93, 1645.Ibid., 2836 ; A., i, 875.l o 0. Hinsberg, Ber., 1908, 41, 626 ; A., i, 25T.I p 11. Gazdar arid Y. Smiles, FrcCIw., 1908, 93, 1833.l3 J. J. B. Dews, Ber., 1908, 41, 2329 ; A . , i, 635.R. Holniberg, A m a h i , 1908, 359, 81 ; if., i, 308.l5 S. Smiles and R. Le Rossignol, Tra?~s., 1908, 93, 745 ; E. I<iioevcna,a~l mdJ . Kenner. RcT., 1908, 41, 3315 ; if., i, 970.REP.-VOL. V. 178 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.taneously with the sulphur dioxide, The work of Smiles and LeRossignol has led to the recognition of interesting steric influences inthis reaction. The sulphination will continue beyond the stage of thesulphinic acid to those of the sulphoxide and the sulphonium compound,the process being, in the case of phenetole :OEt*Ph -+ OEt*C6H4*SO2H -+( OEt*CGH4),S0 -+ (OEt*C6H4),S*OH.The group enters the same position as the sulphonic group insulphonation. How far the reaction will proceed is dependent ontwo factors, the directive influence of such groups as ethoxyl, andsteric hindrance by accumulation of ortho-substituents. Thus, withphenetole, the sulphinic group enters the para-position, and therebeing no ortho-substituents, the reaction proceeds t o the formationof the sulphonium compound. In the ethers of quinol, sulphinationtakes place in the ortho-position, and progress beyond the snlphinicacid stage is rendered difficult, and almost impossible, by accumulationof ortho-substituents.x X<7)SO,H -\ so/---/- \-X X X OH XTho relative influence of the two factors has been investigated inseveral cases.The constitution of the aromatic sulphinic acids has been furtherstudied by the examination of their oxidation products.16 Potassiumpermanganste, in glacial acetic acid, converts them into a-disulphones,which are identical with those obtained by the combination of sulphonylchlorides with sodium arylsulphinates. This indicates that the acidsreact as R*SV102*H, and not as R*SIVO*OH. The U-disulphones, suchas C,H,*SO,*SO,*C,H,, are stable, rather inert substances, onlydecomposed by hot concentrated alkalis.Aldehydes and ketones react with alkaline sodium hyposulphite,forming compounds which appear to be esters of sulphoxylic acid,RRCO + Na2S204 + NaOH = REC(OH)-SO,Na + Na,S0,.17The sodium formaldehydesulphoxylate, known commercially asrongalite, probably has one of the two following structures :CH2<g>SNa*OH or 0H.C H,* O*SO,Na.16 T. P. Hilditch, Trans., 1908, 93, 1524.l7 E. Fronun, Ber., 1908, 41, 3397 ; A . , i, 968ORGANIC CHEMISTRY. 179It forms a dibenzyl derivative,CH,<g>S\C711,) 0. C,H, or C7H70*C H, O*SO,-C,H,,with benzyl chloride.Sodium thiosulphate and formaldehyde react to form thioform-aldehyde, which assumes the termolecular form, ( CH,S),.lSA few organic selenium and tellurium compounds have beenprepared. A similar method to that employed for the preparationof sulphonium compounds leads t o the formation of selenoaiumderivatives, namely, the condensation of phenolic ethers with seleniumdioxide in presence of aluminium chloride. The selenonium basesare very stable, their salts are hardly acted on by sodium hydroxide,but yield hydroxides with silver 0xide.1~Di-a-naphthyl selenide and telluride, ( CloH,),Se and ( Cl,H7),Te,and similar compounds have been prepared by the action of seleniumor tellurium on mercury dinaphthyl and its analogues.20Many of the researches not recorded in this Report, although ofgreat importance, are omitted because they represent intermediatestages in the att'ack on one of the outstanding problems of this branchof chemistry. At this point it is often impossible to correlate andexplain concisely the different lines along which the attack is beingdirected. But when success has been. attained, it will then be foundthat these investigations will fall naturally into their places in therecords of future reporters as necessary steps in tho execution ofanother noteworthy achievement of organic analysis or synthesis.CECIL H. DESCH.GILBERT T. MORGAN.L. Vanino, J. pr. Chin., 1908, [ii], 77, 367 ; A . , i, 318,l9 T. P. Hilditch and S. Smiles, Tmns., 1908, 93, 1384.2o R. E. Lyons and G. C. Bush, J. Amer. Chrn. Soc., 1908, 30, 831 ; A., i, 417