114 XVI1.-Researches on the Constitution of Azo- and Diazo-derivatives. V. Compounds of the Naphthalene P-Xeries-continued. By RAPHAEL MELDOLA F.R.S. and GILBERT T. MORGAN. IN the last paper referring t o the present branch of investigation (Trans. 1888 460; Meldola and East) it was shown that the azo-derivatives of &naphthol contain an atom of hydrogen capable of beiiig displaced by acid radicles a property which they share in common with their a-analogues (Zincke and Bindewald Ber. 1884,3030). Of the alkyl-derivatives of the azo-P-naphthol compounds the only re-presentative at present known is the benzeneazo-P-naphthyl ether of Weinberg (Ber. 1887 3171). The importance from a theoretical point of view of ascertaining whether this displaceable hydrogen-atom of the azo-@-naphthol compounds is attached to the oxygen o r to the nitrogen of the azo-group has from time to time been pointed out in the course of the present series of researches and has recently been emphasized in a paper read by one of the authors at the Bath meeting of the British Association (Phil.Nag. 1888 403). The present paper contains the results of experiments undertaken w itli the object of throwing further light on this question. I. Benzoyl-derivatives of Azo-/I-Naphthol Compounds. Acetyl-derivatives of azo-P-napht,hol compounds can be prepared, as pointed out in the last paper by two distinct methods. If the radicle attached to the azo-group is more o r less of an acid character, such as nitrophenyl CsH4.N02 ( p or rn) the corresponding fi-naph-thylamine azo-compound is attacked by nitrous acid in the presence of glacial acetic acid with the formation of the azo-naphthyl acetate (Zoc.cit. p. 465).* This method does not appear t o be applicable * The formation of phenylic acetates from the corresponding amines by means of the diazo-reactinn in the presence of glacial acetic acid was made known in it paper communicated to the Chemical Society of Germany on January 24th 1888 (Be?, 1888 Sol) and in a subsequent paper read to this Society (Trans. Nay 1888, 465) i t was stated that " as the reaction promises to be widely applicable for the synthesis of phenylic acetates and other et,hereal salts it is proposed to extend the investigation in this direction and to make B special study of the conditions which determine the displacement of amidogen by the C2H30.0-group by means of the diazo-reaction." Notwit.hstanding the publicity thus given to the method and the explicit claim to continue its further investigation a paper appeared in the Amer.Chem. Jour. of last September (10 368) by W. R. Orndorff containing the results of the application of this reaction for the formation of phenyl cresyl an THE CONSTITUTION OF AZO- AND DIAZO-DERIVATNES. 11 5 when the radicle attached to the azo-group is not acid such as phenyl or naphthyl C6H5 CloH7 &c. in the azo-P-naphthylamine com-pounds. I n such cases however azo-p-naphthyl acetates can be prepared by the direct acetylation of the azo-/I-naphthol compounds by means of sodium acetate and acetic anhydride and the same method is generally applicable whatever may be the nature of the radicle associated with the azo-group.The same remarks apply t o the benzoyl-derivatives of azo-/I-naphthol compounds some of which have been prepared by both methods in the manner described below. Berzzeneaxo-p-Na~hthyz Benzoate C6H,*N,*CI~H~-OC7H50 (p). This compound cannot be prepared by fusing benzeneazo-6-naph-thylamine with excess of benzoic acid and adding the calculated quantity of sodium nitrite to the mixture complete decomposition takes place under these circumstances with the formation of tarry products. A mixture of benzeneazo-p-naphthol with dry sodium benzoate and henzoic anhydride if heated for some hours to the fusing point of the anhydride gives only a small yield of the azo-p-naphthyl benzoate owing to the difficulty of keeping the mixture in a homo-geneous fluid condition and to the continuous loss of anhydride by sublimation.After many partially successful attempts with benzoyl chloride the following method was found to give perfectly satisfactory results :-Benzeneazo-&naphthol is mixed with about an equal weight of dry and finely powdered sodium benzoate and the mixture is then made just pasty with benzoyl chloride. The flask containing the mixture is fitted with a cork perforated for the reception of a tube drawn out to a fine point and heat is applied by immersing the flask in boiling water.* The reaction commences immediately on the application of heat and is complete in about an hour when the product is repeatedly extracted with hot water as long as benzoic acid is dissolved out.The contents of the flask fuse to a dark reddish oil under boiling water a property which facilitates the washing operations as the oil and hot water can be thoroughly mixed by agitation and the former rapidly subsides t o the bottom of the flask on allowing the contents to remain a t rest f o r a few minutes. The benzoyl-derivative is easily decomposed by other acetates from aniline toluidine &c. The investigation of the method has been going on in my laboratory since the beginning of last year and these and other acetates and benzoates had been prepared from the diazo-salts of the corre-sponding amines before the appeai'ance of the paper by the author refevred to. * At higher temperatures the reaction takes a ditferent course ; tarry products -R.M. are formed which have not been further investigated 116 MELDOLA AND MORGAN RESEARCHES ON THE ammonia even when the latter is very dilute so that it is unsafe to attempt to shorten the washing by the use of ammonia. In some of our first preparations unaltered heuzenenzo-@naphthol was always found, and this was afterwards traced to the use of ammonia in the washing waters. Sodium or ammonium carbonate also readily hydrolyse the bekzoyl-derivative and it is evident that the benzoyl-group is much more readily removed from this compound than is the acetyl from benzeneazo-p-naphthyl acetate. When completely washed the sub-stance is cry st.allised from boiling alcohol in which it dissolves readily. A specimen was further purified for analysis by dissolving in chloro-form in which it freely dissolves in the cold and allowing the solvent t o evaporate spontaneously in an open dish.The substance forms dense orange-red crystals melting at 125". 0.1025 gram gave 0.2951 gram COz and 0*0438 gram H,O. 0.0894 gram burnt in a vacuum with CuO gave 5.9 C.C. moist N at Calculated for HBH,,N202. Found. 13" C. and 770.1 mm. bar. C . 78.41 78-52 H . 4.55 4.74 N . 7.95 7.90 Metanitrobenzeneazo-P- Napht hy 1 Benzoate, (m)NOZ*CBH**Nz*C10H6*0 C,HbO (p) . Mehnitrobenzeneazo-/bnaphthylamine (m. p. 182') is fused with excess of benzoic acid and the calculated quantity of dry sodium nitrite added in small portions to the fused mixture. Nitrogen is freely given off and when all the nitrite has been added the product is washed with hot water till free from benzoic acid.The washing may in this case be accelerated by the use of dilute ammonia without fear of decomposition. The washed product is crystallised from hot glacial acetic acid in which it is not very soluble and finally from boiliiig alcohol in which it dissolves only very sparingly separating on cooling in the form of dull reddish-orange silky needles. The melting point is 171". The following results were obtained on analysis :-0.16'76 gram gave 0.4280 gram GO2 and 0.0612 gram HzO 0.2196 gram burnt in a vacuum with CuO gave 19.8 C.C. moist N Calculated for at 17" C. and 761.7 mm. bar. C23H,,N304. Found. C . 69.52 69.64 H . 3-77 4.05 K 10.57 10.4 CONSTITUTION OF AZO- AND DIAZO-DERIVATIVES.11 7 The substance is formed in this process according to an action perfectly similar to that which gives rise to the production of the acetate. C,0H,<~&~"H"N02 + NaNO + 2C7H50*OH = CIOH6<N2'C6H4"02 + C7H50*ONa + N2 + 2H,O. 0.C7H,0 The benzoate may also be prepared by direct benzoylation and this method is much more convenient when the compound is required in quantity. For this purpose metanitrobenzeneazo- - naphthol (m. p. 194") is mixed with about its own weight of dry powdered sodium benzoate and enough benzoyl chloride added to make the mixture int,o a paste. The contents of the flask are then kept just at the boil-ing point of the benzoyl chloride for 4-5 hours. The product is washed with dilute ammonia and purified by crystallisation in the manner already described.11. Reduction of Axo-P-Naphthol-derivatives containing Acid Radicles. I n order to see whether any information concerning the position of the displaceable hydrogen-atom could be derived from a study of the products of reduction of the acetyl and benzoyl-derivatives these compounds were reduced with tin and hydrochloric acid and with stannous chloride under various conditions. Reduction of Benxenenzo-P-Na~hth~l Acetate. An alcoholic solution of the above compound is soon decolorised if boiled with tin and hydrochloric acid. On distilling off the alcohol and filtering t o remove tarry impurities the filtrate on standing gradually deposits a white curdy substance which under the microscope is seen to consist of agglomerations of white needles.The addition of strong hydrochloric acid to the solution promotes the deposition of the white compound. The latter was therefore purified by being collected dissolved in hot water filtered and 1-eprecipitated by the addition of strong hydrochloric acid this treatment being repeated till the substance dissolved without leaving any residue. The white substance thus obtained proved to be the hydrochloride of a mixture of bases. Many specimens mere prepared and analped without any concordant results being obtained ; all attempts t o separate the mixture into its constituents by fractional crystallisation or by con-version into other salts were unsuccessful. The free bases could not be isolated owing to their extremely oxidisable character.This pro-duct or -3uction contains a considerable quantity of x-amido-p-xiaph 118 MELDOLA AND MORGAN RESEARCHES ON THE fhol the presence of the latter compound being shown by the formation of P-naphthaquinone on oxidation by ferric chloride. I n spite of the unfavourable results obtained in our first experiments, the reduction was repeated an alcoholic solution containing the theoretical qiiantity of stannous chloride being gradually mixed with a slightly warm alcoholic solution of the substance After the mixture had stood for some hours the alcohol was distilled off and t,he product treated as above. The same white hydrochlorides were obtained but the siibstance was evidently a mixture and a-amido-p-naphthol was identified as one of its constituents as before.The analyses showed that in this case the mixture contained different proportions of the bases from those present in the product formed by violent reduction with tin and hydrochloric acid. 'In both series of experiments the filtrate after the removal of the white hrdrochl orides first deposited, was treated with sulphurettcd hydrogen in order to remove tin and after the removal of the stannous sulphide the filtrate was evaporated to a small bulk made alkaline with caustic soda and submitted to steam distillation when a considerable quantity of aniline was obtained, showing that this base is one of the products of reduction. According to the current hypotheses of the constitution of the azo-p-naphthol compounds the acetyl-derivative of benzeneazo-&naphthol may have one of the following formula :-I.11. 111. The first of these is what may be called the old formula the second is that proposed by Liebermann and the third is Zincke's that is the hydrazone formula. Supposing the compound to be capable of complete reduction without the elimination of acetyl these three compounds might be expected to yield the following products :-I. Aniline and a-amido-p-naphthyl acetate. IT. Aniline and a-acetamido-p-naphthol. 111. Acetanilide and a-amido-p-naphthol. We may state a t once that a most careful search for acetanilide in the product of reduction has led to negative results. If the acetyl were attached to tlie nitrogen-atom it is improbable that t h i s group would be removed under the conditions of reduction since acetanilide is well known to be a stable compound nnder such circumstances.The third formula appears therefore to be inadmissible and th CONSTITUTION OF AZO- AND DIAZO-DERIVATIVES. 119 decision rests between Nos. I and II.* A complication of the question however here arises from the possibility of intramolecular migration of the acetyl. Thus according to the formule the naphthalene portion of the molecule ought to give on complete reduction-I. 11. The investigations of Bottcher (Bey. 1883 1933) have shown that such a compound as No. I is incapable of existence as it immediately becomes transformed into No. 11. This was proved by preparing the acetyl-derivative of a-nitro-P-naphthol and reducing the latter when a phenolic substance viz.acetamido-P-naphthol was obtained instead of the ba.sic amido-P-naphthyl acetate. The acetyl-group accordingly becomes transferred from the oxygen to the nitrogen probably through the intermediate formation of the anhydro-base and hydra-tion of the latter :-Whichever of the formula (I or 11) be the correct one it might therefore be expected that extreme reduction would give rise to the same acetamido-&naphthol. The latter is a perfectly stable phenolic substance melting according to Bottcher at 225". It does not appear to be present however aniong the products of reduction of benzene-azo-6-naphthyl acetate obtained by us in accordance with the method previously described. Caust.ic soda does not dissolve any stable sub-stance out of the mixed hydrochlorides but simply gives a mixture of highly oxidisable bases the chief constituent of which is undoubtedly a-amido-/%naphthol.The extreme reduction of the acetate therefore gives rise to the formation of aniline and a-amido-@-naphthol as chief products a fact which is in itself of considerable importance in connection with the present line of investigation since it shows that t,he acetyl-group is split off during reduction arid this lends support to the view that the said radicle is attached to oxygen and not to nitrogen in the original compound. * The formula proposed by one of the authors (Meldola Phil. Jfug. Nor. 1889, p. 4.11) is for the present left out of consideration because i t indicates the formation of products of reduhon identical with those indicated by KO.I 120 MELDOLA AND MORGAN RESEARCHES ON THE IdentiJcation of a- Amid+ Naphthol. At this stage of the work seeing that the products of reduction of the acetyl-derivative contained some other base in addition to amido-p-naphthol it became necessary to discover a method by which the mixed bases could be converted into stable derivatives capable of being separated and characterised. The identification of the amido-naphthol itself presents but little di%culty since it gives /3-nayhtha-quinone on oxidation with ferric chloride. It is well known how-ever that the formation of this quinone is somewhat capricious and when formed it is not easy to identify unless in a condition of tolerable purity. The quinone could not be obtained pure from our mixed hydrochlorides became the oxidising agent also attacked the other base with which the amidonaphthol was mixed and so gave rise to the formation of impurities of a resinous character.The method of oxidation being thus dest,ructive to the other base had to be abandoned. It was then found that the purified hydrochlorides when dissolved in water and treated in the cold with dilute hydro-chloric acid and sodium nitrite gave a crystalline yellow substance which after being collected and washed was identified as nitroso-p-naphthol (melting point about 1 1 0 O ) . This last substance is formed by the action of nitrous acid on a-amido-&naphthol according to a process of decomposition which has not hitherto been studied but which has also been observed by Dr. H. E. Armstrong." The action of nitrous acid thus confirms the presence of a-amido-@-naphthol but as this reagent did not appear t o give a satisfactory product with the other base its use was also abandoned.Benzoyl and acetyl-deriva-tives of the mixed bases were next prepared and these promised to answer our requirements so that in the first place the benzoyl and acetyl-derivatives of pure a-amido-&naphthol were prepared aud studied in order that we might familiarise ourselves with their properties. As these derivatives have not hitherto been described, we think it may be of use to give further particulars as they are well characterised and stable crystalline substances very easy of prepma-tion and thus offering many advantages over the quinone method foy the identification of a-amido-@-naphthol when the latter is mixed with other oxidisabIe bases.Benzeneazo-/l-naphthol was reduced in alcoholic solution with tin and hydrochloric acid the alcohol distilled off and the amid+ naphthol hydrochloride allowed t o crystallise out of the solution in the usual way. This salt was collected dissolved in water and again crystallised with the addition of strong hjdrochloric acid this process * Private communication to one of the authors COSSTITUTIOS OF AZO- ASD DIAZO-DERIVATIVES. 1 ',I being repeated till the substance dissolved completely in water. The aniline arid tin salts are thus got rid of aud the pure amidonaphthol hydrochloride is left as a white crystalline compound. The latter was collected drained completely from adhering mother-liquor and dried iii the water-oven.The dry salt was then just covered with glacial acetic acid and powdered anhydrous sodium acetate and acetic anhydride added. The mixture was boiled for 2-3 hours and the acetyl-derivative thrown o u t by dilution with water. The substance was collected washed and purified by crystallisation from alcohol in which it readily dissolves when hot and separates very slowly on cooling in the form of dense colonrless transparent rhombic prisms, melting sharply a t 206". hnaljsis showed thnt the compound was a d iacetyl-derivative :-0.1081 gram gave 0.2742 gram GO and 0.0502 gram H,O. 0.1460 , , 7.6 C.C. moist N at 17' C. and 738.4 mm. bar. 0.3458 , , . 18.2 , , 14.5" C. and 731.5 ,, Calculated for Found. NI1.C H 0 +-.-) c~oH6<*.~,~32,03 * I.11. - c 69.13 69.18 H 5.35 5.16 N 5-76 5.84 5.94 -The substance is accordingly acetamido-p-nnphthyl acetate. It is of interest to note that t.he acetylation in this case leads to the forma-tion of a diacetyl-derivative and not of an anhydro-base although the amidonayhthol is an ortho-derivative. As a parallel instance we may mention a-p-naphthylenediamine which although an ortho-compound also gives a diacetyl-derivative and nut an anhydro-base, on acetylation (Lawson Iuuug. Diss. 1885 p. 26). By heating dry amido-B-naph tho1 hydrochloride with anhydrous sodium benzoate and benzoyl chloride to the boiling point of the latter for about two hours a dibenzoyl-derivative is formed. The product of the reaction is washed repeatedly with hot water and dilute ammonia and finally pnrified by two or three crystallisations from boiling alcohol in which the substance dissolves somewhat sparingly.The pure compound forms whitish needles melting at 226.5". 0.1186 gram gave 0.3420 gram GO and 0.0544 gram H,O. 0.1508 gram burnt in a vacuum with CuO gave 5 C.C. moist N at 24.5" and 765.3 mm. bar. VOTI. LV. 122 MELDOLA AND JIORGAN RESEARCHES ON THE Calculated for NH*C H 0 Cl0%<O.C7&(j Found. C . 78-47 78.66 H . 4-63 5.09 N . 3.81 3.73 These results show that the compound is benzamido-5-naphthyl benzoate and t,hat as in the case of acetylat,ion both the hydroxyl and nniidogen of the amidonaph thol are simultaneously a t tacked. 2lfild RelEzmSon of Benzeneazo-p-Naphtliol nnd its Acetgl-derivative, 111 accordance with the well-known fact that aromatic azo-corn-pounds by mild reduction yield hydrazo-compounds which nrider tbe influence of acids become transformed into bases of the diphenyf or riapht,hylphenyl series it became of interest to investigate the action of mild reducing agents on the azo-compounds of p-naphthol.Corn-mencing with the simplest case benzeneazo-/3-nnph thol was dissolved in a sufficient quantity of alcohol to retain the whole of the substance in solution when cold and stannous chloride mixed with alcohol was added t o the solution till reduction had been effected. The alcohol Was distilled off nnd the residual solution after filtration to remove a small quantity of tarry matter was mixed with dilute sulphuric acid and allowed to stand for 24 hours.A small qnantity of a crystaJline substance consisting of slender white needles had sepa-rnted but the yield was too small to enable analyses to be made. The presence of sulphuric acid was proved qualitatively in the compound after the latter had been purified by crystallisation from boiling water in which it dissolves with areat difficulty. The sub-stance is in all probability the sulphate of dinmido-oxynaphtliylphengl, aud we hope to return to its investigation on a future occasion but the small yield has deterred us from making further experiments in this direction as the preparation of this compound has only an indirect bearing on the present line of work. The free base is very unstable forming dark products of oxidation with extreme rapidity on exposure to the air.During the boiling of the solution of the sulphate for the purpose of purifying this salt a conside~able loss was incurred by the formation of a resinous product. If the sulphate is basified by ammonia a greenish colouring matter is produced by the action of atmospheric oxygen. A small quantity of the sulpliate heated with dry sodium acetate and acetic anhydride gave an acetJ1-derivative which after crystallisation from alcohol formed small, white needles melting at 130-131". Benzeneazo-P-naphthyl acetate when reduced in a similar manne COXSTITUT[OY OF AZO- AND DI.\ZO-DERIVATIVES. 123 with stannous chloride in cold alcoholic solution gives a mixtupe of bases from which nothing definite could be isolated. Amido-/j-naphthol is undoubtedly present but not in sufficient quantity to separate out as a hydrochloride in the presence of excess of acid.The presence of amido-@-naphthol was proved by the formation of nitroso-/%naphthol by the action of sodium nitrite on the acid solu-tion. The other base which is present is so rea lily decomposable that a dark resinous scum is continually formed during the evaporation of the solution. The addition of sulphuric acid to the solution of the hydro-chlorides does nof cause the sepiiration of an insoluble sulphate. Nevertheless the absence of a naphthylphenyl base cannot be inferred from this circumstance because it is highly probable that such a base, if formed would contain the acetyl-group in one of the amidogens and would thus lose the pyoperty of fonning an insoluble sulphate.We have proved that such a naphthylphenyl-derivative is formed by a method which is described further on so that the uiistable base pre-sent in solution with t'he amidonaphthol is in all probability the ncetyl-derivative of diarnidoh,ydroxynaphtliylphenyl with the acetyl in one of the amidogen-groups by intramolecular transference i.~., The proof that such a base is formed is furnished by the isolation of the diacetyl-derivative. The mixtiire of dry hydrochlorides. obtained by the reduction of benzeneazo-p-naphthyl acetate in sZigh/l.i/ wurm alcoholic solution with Ftannous chloride was boiled for aboiit two hours with glacial acetic acid dry sodium acetate and acet.ic+ anhydride. The product was diluted with water collected washed, and crjstallised two or three times from hot alcohol.Microscopic needles were obtained melting sharply a t 252". From the first mother-liquor there slowly separated the familiar rhombic prisms of' acetamido-p-naphthyl acetate (m. p. 206"). The substance meltiu:.; a t 252" was analysed with the following results :-I. 0.1513 gram gave 0.3892 gram GO and 0.0724 gram H,O. 11. 0-1310 , 0.3368 ,, 111. 0.1194 , 0.3076 , , 0-0581 ,, IV. 0,2154 , 13.1 C.C. N at 15" C. and 745 mm. bar, Found. Calculated for -7 C22H20N204. I. 11. 111. IV. C 70.21 70.15 '70.12 70.26 -H 5-31 5.31 - 5.40 -N . . - 7.44 - - 6-9s K 124 MELDOLA AND MORGAN RESEARCHES ON THE These numbers indicate that. the compoiind is the triacetyl-derira-tive of diamido-hydroxynaphthylphenyl the constitutional formula being-\-/ The other base mixed with the amido-&naphthol is accordingly, the monacetyl-derivative of diamido-hydroxynaphthylphenyl the acetyl being most probably present in the a-amidogen-group throngh intramolecular migration in the manner explained.The absence of two arnidogen-groups in the monacety 1-derivative appears as we have already stated from the fact that it does not form an insoluble sul-phate a property which is so eminently characteristic of all the bases of the benzidine series and which as we have previously shown, pertains t o diamido-hydroxynaphthylphenyl itself. Summing up the results given by the present investigation of the products of reduction of benzeneazo-p-nnphthyl acetate the following conclusions have been ai-rived at :-1.By violent reduction with tin and hydrochloric acid the chief product is a-amido-/I-naphthol a small quantity of the monacctyl-derivative of diamidohjdroxynaphthylphenyl being a t the same time produced. The splitting off of acetyl f:bvours the view that this radicle is attached t o the oxygen atom and not to the nitrogen atom. 2. Reduction in cold alcoholic solution with stannous chloride gives monacetyldiamido- h ydroxyna p ht h yl phenyl as the chief product, a-amido-p-naphthol being a t the kame time formed in smaller quantity. 3. Moderate reduction with stannous chloride in warm alcoholic solution gives a mixture of the two bases from which the triacetjl-derivative of the hydroxynaphthylpheiiy 1 base can be isolated by acetylation and crystallisation of the product.I n all three cases a,niline is also one of the products of reduction. Reduction of Benzeizeaxo-p-Naphthy 1 Benzoate, When this compound is reduced by boiling its alcoholic solution with ttin arid hydrochloric acid the proclucts are aniline and a-amido-&naphthol the benzoyl-group being split off as in the case of the acetate when it is submitted to violent retluction. A cold alcoholic solution of the benzoate when mixed with a solution of st,annous chloride also gives a considerable quantity of amido-/3-napht hol together with a benzoj 1-derivative of a naph1hJllphenyl base. Th COXSTITUTION OF AZO - AND DIAZO-DERIVATIVES. 1 S 5 separation of the two products in this case offers no difficulty as the hasic properties of the latter are neutralised by the benzoyl-group and the substance is insoluble in dilute acids.After the reduction is complete the alcohol is distilled off the residue diluted with water and filtered. The filtrate contains the amido-P-napht<hol hydrochlo-ride which can be identified in the usual manner. The dark tarry, insoluble residue left after this treatment is extracted two or three times with hot dilute hydrochloric acid then washed with water and finally extracted with cold alcohol which dissolves o u t a tarry impurity and leaves the crude naphthylphenyl-derivative in the form of a whitish crystalline powder. The latter af'ter being twice crystallised from boiling alcohol in which it is rather sparingly soluble forms white silky needles melting a t 172-173" :-0*10@4 gram gave 0.2860 gram C02 and 0.0416 gram H,O.0.0813 gram burnt in a vacuum with CuO gave 3.5 C.U. moist N a t 14" C and 750.9 mm. bar. Calculated for C& 0 2 . Found c . 77.96 7 7-69 H . 5.08 4-60 N . 7-91 7-86 The formula deducible from these analyses indicates that the com-pound is a monobenzoyl-derivative of diamidohydroxynaphthlyphenyl. The yield is n u t very great as tshe chief products of reductioii are, even under the described conditions of working chiefly amido-p-naphthol and aniline. The fact that the substance is not basic tells against the view that two amido-groups are present and points LO the tyansference of the benzoyl from the oxygen to the nitrogen i n the naphthalene nucleus by the same process as that which takes place with ihe corresponding acetate.We are therefore led to the view that the constitutional formula of the compound is-The proposed LJ formula certainly indicates a phenolic substance, whereas our compoiind is insoluble i n aqueous alkali. The disguisiug of the phenolic character of the molecule by the proximity of the C,H,O*NH-gronp seems however more probable than that two free nmido-groups should be present when the distinctly non-basic character of the substance is borne in mind. For t h i s reason we attach the greater weight to the above f.mniilit 126 MASSON AND KIRKLAKD ACTION OF BROJlINE AND The general conclusion to which these experiments point is that tho acid radicle acetyl or benzoyl introduced into benzeneazo-B-naphthol or its derivatives displaces an atom of hydrogen which is attached to oxygen and not to nitrogen. This is shown by the facility with which flie radicle is split off during reduction and by the absence of acetanilide or benzanilide among the products of reduction. The transference of the radicle from the oxygen to the nitrogen in the naphthylphenyl base is effected by intramoiecular change in the ortho-position through the intermediate formation of an nnhydro-base. An investigation of the products of reduction of the alkyl-derivatives of azo-/?-naphthol compounds has been carried on concurrently with the foregoing experiments and the results w-ill be made known in a subsequent, communication. In the early part of tbe present research we received much assistance from Mr. E. H. R. Salmon and more recently we have had the beiiefit of the co-operation of Mr. J . H. Coste. It gives us mnch pleasure to acknowledge the services rendered by these gentlemen. Finsbury Tech n ical Co 1 leg e