quantity for a detailed examination; but, judging from its mode of generation, it will probably be found to be tolyldiphenylrosaniline,

A few additional experiments performed with phenyltolylamine may still briefly be mentioned.

Chloride of benzoyl attacks this substance, especially on application of heat. The product of the reaction remains liquid for a long time, but, when appropriately treated with water, alkali, and spirit, it ultimately solidifies, and separates from boiling alcohol in well-formed crystals,

which are more soluble than the corresponding diphenyl-compound. The benzoyl derivative is very readily converted into nitro substitutes. In contact with ordinary concentrated nitric acid the crystals are at once liquified, and ultimately entirely dissolve. Addition of water to this solution precipitates a yellow crystalline dinitro-compound,

which is deposited from boiling alcohol in reddish-yellow needles. fectly similar treatment converts the diphenyl body into a mononitro substi tute. Cold fuming nitric acid, which when acting upon the diphenyl body gives rise to the formation of a dinitro-substitute, transforms the benzoylated phenyltolylamine into a nitro-derivative containing, according to an approximate analysis, not less than 5 atoms of NO,.

Dinitro-phenyltolylbenzoylamide dissolves in alcoholic soda with a feebly crimson colour. Ebullition of the solution eliminates the benzoyl atom in the form of benzoate, and on cooling small yellowish red crystals, C. H. (NO2)

4

C13 H12 N, O=C, H. (NO2) › N,

H

are deposited, which are easily purified by crystallization from alcohol. Lastly, when treated with reducing agents, the dinitronated phenyltolyl-benzoyl compound is converted into fine white needles of a new base, to which I hope to return as soon as I shall have procured a somewhat larger supply of phenyltolylamine.

It scarcely requires to be mentioned that it is not necessary to prepare the pure toluidine-blue for the purpose of obtaining phenyltolylamine. It suffices to maintain for some hours a solution of ordinary but dried acetate of rosaniline, in its double weight of toluidine, in a flask provided with an upright condensing tube, at a boiling temperature, and to submit the violet

blue mass produced to destructive distillation over a naked burner. The distillate is treated with hydrochloric acid, and subsequently with water, when aniline and toluidine, together with several other basic substances accompanying the phenyltolylamine, remain as hydrochlorates in solution. The oily layer which separates generally solidifies, or may be purified by rectification. The resulting crystals are crystallized from alcohol.

The same method is also adapted for the preparation of diphenylamine, aniline being substituted for toluidine.

If I have bestowed upon diphenylamine and phenyltolylamine rather more attention than these substances at the first glance appear to claim, I have done so in the hope of gaining additional data for the investigation of the remarkable colouring matters from which these bases are derived. Both constitution and mode of formation of these colouring matters are still involved in darkness. Theory, as it often happens, has not kept pace with practice. The anticipation I expressed in a former note, that the study of the behaviour of the colouring matters under the influence of chemical agents might disclose their true nature, has only very partially been realized. Up to the present moment, chemists have not succeeded in giving a satisfactory account either of the atomic construction of these compounds, or of the mechanism of their formation; and it would therefore scarcely be worth while to return to this question before its definite solution, unless the publication of erroneous statements by M. Schiff had threatened to divert the researches of chemists from this subject.

According to M. Schiff *, the transformation of aniline into aniline-red by means of stannic chloride is represented by the equation

20C, HI, N+10 Sn Cl=3(C2, H1, Ng, HCl)+6(C, H, N, HCl)+

20

19

H, NC1+10 Sn Cl+4 C, H, N.

The formation by means of mercuric nitrate by the equation 20C, H, N+20 HgNO3=3(C20 H1, N ̧, HNO3)+6(C ̧ H, N, HNO2) + H1 N, NO,+10Hg, NO,+4C, H, N.

4

20

The latter process is accomplished at as low a temperature as 80° C., and, according to M. Schiff, is so elegant that he was enabled to make quantitative experiments. "Within a few hundredths," he says, we have obtained the requisite quantities of the sought-for materials."

M. Schiff's equations are not conspicuous for elegance and simplicity, but they are absolutely inadmissible for other reasons. These equations utterly ignore the very essence of the process. I have pointed out, some time ago, that the formation of rosaniline involves the presence of both aniline and toluidine. Pure aniline furnishes no rosaniline, nor can this body be procured from pure toluidine. This fact I have since further established by many varied experiments, both on the small and on the large scale. The formation of rosaniline thus becomes the means of ascertaining rapidly the presence of toluidine. The amount of the latter Compt. Rend. vol. lvi. p. 271. + Ibid. p. 545. 20

VOL. XIII.

490

Dr. Hofmann on the Colouring Matters

[Recess,

in crude aniline* may become so minute that its presence can no longer be traced by distillation, or by conversion into oxalates. It may be recognized, however, with the utmost facility by submitting the mixture to the action of either corrosive sublimate or arsenic acid; on application of a gentle heat the crimson colour is immediately produced.

In the equations proposed by M. Schiff there figures, moreover, ammonia as an essential term. The existence of ammoniacal salts in the crude rosaniline was pointed out some time ago by Prof. Bolley. But this ammonia (which, as I have satisfied myself, is scarcely ever absent) is, according to my opinion, not an essential product of the reaction that gives rise to the formation of aniline-red. I have established by special and careful experiments that appropriate treatment of a mixture of aniline and toluidine with chloride of mercury at a moderate temperature is capable of producing very considerable quantities of rosaniline without elimination of more than a trace of ammonia. The ammonia generally observed belongs to a different phase of the reaction, being more especially due to the almost invariable production of a small quantity of aniline-blue.

If we wished, even now, to represent in formulæ the relation between rosaniline and the substances which give rise to its formation, the equation C, H, N+2C, H, N=C,, H1, N2+6H

6

20 19 3

might be looked upon as an expression closely approaching truth. The hydrogen figuring in this equation is eliminated in the form of water, hydrochloric, hydrobromic, hydriodic acids, &c.

But even this equation gives no account of the mechanism of this remarkable process; indeed we cannot hope for the solution of this chemical problem before we shall have succeeded in splitting up the molecule of rosaniline into the atomic groups which enter into its composition. It is true its transformation into aniline- or toluidine-blue, as well as into the several violets which are generated by the substitution of alcohol radicals, prove even now that the rosaniline-molecule still contains three atoms of typical hydrogen, and hence that the complex atom C2, H1 functions in this triamine with the value of six atoms of hydrogen; but this indeed is the limit of the experimental evidence as yet obtained.

20

16

16

With regard to the number and nature of the simpler radicals into which the carbon and hydrogen atoms of the complex atom C20 H are grouped, we can only speculate. Derived from the radicals phenyl, C, H,, and tolyl, C, H., under the influence of dehydrogenating agents, this complex atom

6

* Aniline obtained by distillation with potash from certain varieties of indigo, is apt when treated with corrosive sublimate, to furnish traces of rosaniline. I infer from this result that aniline thus produced contains a small proportion of toluidine. The formation of this substance from indigo would be as readily intelligible as the conversion under certain conditions of indigo into salicylic acid, a fact established by Cahours's observations.

Aniline prepared from crystallized isatin does not yield a trace of rosaniline.

may possibly contain the bivalent radicals phenylene, C, H,, and tolylene, C, H.,

when the molecular construction of the three colouring matters might be represented by the formulæ

16

We must not, however, forget that this is simply an hypothesis, and that the elements in the complex atom C20 H1 may be associated in a great variety of other groups. An interesting observation quite recently made by Dr. Hugo Müller, and communicated to me by my friend while these pages are passing through the press, may possibly assist in further elucidating the nature of this class of bodies. Dr. Müller has found that rosaniline and its coloured derivatives are instantaneously decolorized by cyanide of potassium, a series of splendidly crystallized, perfectly colourless bases being produced. The composition of these bodies, which will probably be found analogous to a substance similarly obtained from harmaline by Fritzsche, remains to be established.

November 17, 1864.

Major-General SABINE, President, in the Chair.

In accordance with the Statutes, notice of the ensuing Anniversary Meeting for the election of Council and Officers was given from the Chair.

Mr. Gassiot, Dr. J. E. Gray, Dr. Hirst, Mr. Lubbock, and Dr. Odling, having been nominated by the President, were elected by ballot Auditors of the Treasurer's accounts on the part of the Society.

Among the presents announced was a Photograph of the Moon, from Mr. Warren De la Rue, respecting which an extract of a letter from the Donor was read as follows:-" Except to remove white and black spots, the photograph is untouched. The size of the original negative is about one inch, and from this was taken, in the first instance, a positive on glass nine inches in diameter. The glass positive was used for the production of four negatives, each containing a quarter of the disk. The proper distance for viewing the picture is about six feet, or two diameters."

VOL. XIII.

2 P

The following communications were read:

I. "Comparison of Mr. De la Rue's and Padre Secchi's EclipsePhotographs." By WARREN DE LA RUE, F.R.S. (See page 442.)

II. "On Drops."-Parts I. and II. By FREDERICK GUTHRIE, Esq. Communicated by Professor Stokes, Sec. R.S. (See pages 444 and 457.)

III. "On the Chemical Constitution of Reichenbach's Creosote." By HUGO MÜLLER, Ph.D. Communicated by Warren De la Rue, F.R.S. (See page 484.)

IV. "Researches on the Colouring Matters derived from Coal-tar.— No. IV. Phenyltolylamine." By A. W. HOFMANN, LL.D., F.R.S. (See page 485.)

V. "On the Spectra of some of the Nebulæ." By W. HUGGINS, Esq., F.R.A.S. ;—a Supplement to the Paper "On the Spectra of some of the Fixed Stars," by W. HUGGINS, Esq., and W. A. MILLER, M.D., Treas. and V.P.R.S. Communicated by the Treasurer. Received September 8, 1864.

(Abstract.)

The author commences by showing the importance of bringing analysis by the prism to bear upon the remarkable class of bodies known as nebulæ, especially since the results obtained by the largest telescopes hitherto constructed appear to show that increase of optical power alone would probably fail to determine the question whether all the nebulæ are clusters of stars too remote to be separately visible.

The little indication of resolvability, the absence of central condensation, the greenish-blue colour, and the intrinsic brightness characterizing many of the nebulæ classed by Sir W. Herschel as planetary, induced the author to select chiefly nebule of this class for prismatic observation.

The apparatus employed is that of which a description is given in the paper, "On the Spectra of some of the Fixed Stars," by the author and Dr. W. A. Miller, to which this is a supplement.

No. 4373*, 37 H. IV. Draconis. A bright planetary nebula, with a very small nucleus. The light from this nebula is not composed of light of different refrangibilities, and does not therefore form a continuous spectrum. It consists of light of three definite refrangibilities only, and, after passing through the prisms, remains concentrated in three bright lines. The strongest of these occupies a position in the spectrum about mid

*These numbers refer to the last catalogue of Sir J. F. W. Herschel, Phil. Trans. Part I. 1864, pp. 1-138.

t