aurin"

or "peonine" is therefore nothing but aurin containing more or less pararosaniline.

As already stated (this Journal, 1879 [i], 562), methylamine acts more readily on aurin than ammonia, only trimethylpararosaniline being formed, and no intermediate products, which is shown by dyeing silk in an ammoniacal solution, the characteristic shade of Hofman's violet being produced, while the product containing aurin dyes in a neutral or faintly acid solution, a claret colour.

The most remarkable fact is that it is the most difficult to separate the bases from the aurin by means of dilute acids, from which we came to the conclusion that they formed with it definite compounds.

We heated, therefore, equal molecules of aurin and common rosaniline with alcohol. They soon dissolved, and on concentration a crystalline powder having a dark green metallic lustre separated out, which, by recrystallising from hot dilute alcohol, is obtained in well defined crystals.

Rosaniline aurinate is more freely soluble in alcohol than its constituents. It dyes silks exactly like other rosaniline salts, the colour of aurin not interfering. That the compounds go on the fibre as a whole, is shown by the fact that the dye-baths may be completely exhausted without obtaining any other shade.

Aurin being a phenol and rosaniline an amido-base, we tried next to obtain a compound of the two most simple members of both groups, and heated a mixture of equal molecules of aniline and phenol for some hours to the boiling point. After cooling and standing for some time, large thin plates crystallised out, and finally the whole solidified. Aniline phenate, CH,N,CHO, forms also slowly in the cold; it crystallises from alcohol or petroleum-naphtha in beautiful glistening plates, melting at 29.5°. Its uncorrected boiling point is 184.5°, and its smell resembles that of phenol, but is much fainter. As it has no irritating action on the skin, and is undoubtedly a powerful antiseptic, it may in many case be advantageously used in place of phenols.

Since these observations were made, we have found that Hofmann's statement (Annalen, 74, 126) as to phenol not acting on aniline, even at 250°, is incorrect. His endeavour was to obtain diphenylamine. The reason why he did not observe the formation of aniline phenate, we cannot explain.

The further investigation of these and other compounds of phenols with amido-bases has been undertaken by Mr. G. Dyson, who, as a mean of three analyses, found in aniline phenate 7.57 per cent. of nitrogen, the calculated percentage being 7:49. He has also already prepared rosaniline phenate.

We have already stated that the red alkaline solutions of aurin sometimes turn brown. This change takes place sometimes quickly,

sometimes very slowly; presence of air is not required as we formerly supposed (this Journal, 1879 [i], 562). On adding an acid to the brown solution, a smeary mass is precipitated, from which we could isolate a body forming colourless crystals. This compound is dioxyphenolketone, CO(CH4.OH)2, which Cars and Graebe obtained by heating aurin with water to 220-250° (Ber., 11, 1348).

XXVII.-On some Derivatives of Diphenylene Ketone Oxide.
By A. G. PERKIN.

THE diphenylene ketone oxide employed in the following experiments was prepared by the process recently described by my Father (page 35 of his volume), namely, by distilling a mixture of salicylic acid and acic anhydride. On concentrating the alcoholic mother-liquors left from the purification of this body, they were found to contain small quantities of a second product. It was afterwards observed that when the mixture of salicylic acid and acetic anhydride was very rapidly di tilled, the amount of the new compound produced was larger than when the operation was conducted slowly. At first it was separated from the diphenylene ketone oxide simply by repeated fractional crystallisation, but afterwards the crude diphenylene ketone oxide was macerated with a small quantity of cold alcohol, and, after filtration, the alcohol was removed from the filtrate by boiling with water. Caustic potash was then added to dissolve oily matter present, and the whole was allowed to stand for about an hour, after which the solid matter in suspension was thrown on a filter and well washed. The product thus obtained was crystallised from alcohol, the first crop of crystals separating out being collected apart. By repeated treatment in this way, the new product was obtained in long thin transparent plates, having a beautiful satiny lustre. On analysis, it gave the following numbers :

:

I. 01391 gram substance gave 0.3870 of CO, and 0.0497 of OH,.

These numbers give the following percentages agreeing with the formula C1H2O2 :

:

This substance is only slightly soluble in boiling alcohol, but more so in coal-tar naphtha. It melts at 192°, and distils with partial decomposition. When boiled with alcoholic potash, it is converted into the potassium salt of an acid, and on adding water and boiling off the alcohol, the acid is precipitated by hydrochloric acid as a somewhat gelatinous product if the solution is cold, but as a bulky precipitate on boiling.

This acid may be purified by two or three crystallisations from alcohol, from which it separates in minute needles. Specimens dried at 100° gave on analysis the following numbers :—

I. 0.1265 gram of substance gave 0·3246 of CO2 and 0-0431 of
OH2.

II. 0·097 gram of substance gave 0·2488 of CO2 and 0·0328 of OH2.
III. 0-1345

0.3457
0.3838

0-0416

These give percentages agreeing with the formula CH,O

This acid melts at 275°. It sublimes in brilliant needles, and can be distilled apparently with but little decomposition; it is difficultly soluble in alcohol. When gently fused with caustic alkali, it produces salicylic acid, judging from the iron reaction.

Silver Salt. This was prepared by dissolving the acid in dilute ammonia and boiling the solution until it was free from excess of that substance. Silver nitrate was then added by small quantities at a time, the mixture being well agitated. After some quantity of precipitate had been formed, the solution was filtered and an excess of silver nitrate added, which caused the liquid to coagulate into a gelatinous mass, somewhat like precipitated alumina, but, contracting, and becoming crystalline when heated in a water-bath; it was then collected and well washed with cold water. It gave on analysis the following numbers, the different preparations being examined:

I. 01172 gram of substance gave 0·0360 gram Ag.

II. 01483 gram of substance gave 0.2642 gram CO, and 0.0275 gram OH2.

III. 01881 gram of substance gave 0.0582 gram Ag.

These give percentages agreeing with the formula C1H,AgO

This salt is slightly soluble in cold water, more easily in boiling water. From its hot aqueous solution, it is deposited in groups of small transparent needles, which appear very definite under the microscope. It is not coloured by boiling with water.

From the foregoing results, it appears that this acid has the formula CHO. Its constitution may perhaps be represented thus:

Its formation from the compound C1,H,O, is, however, difficult to understand, because if it be an anhydride of the acid, as its properties would indicate, its formula should be C1HO; but this would require 75.6 per cent. carbon and 2.7 of hydrogen. Unfortunately, this matter cannot at present be further investigated, owing to the want of product, which is obtainable from salicylic acid only in small quantities, and then is difficult to purify. I hope, however, to return to this subject.

On the Action of Nitric Acid on Diphenylene Ketone Oxide.

When diphenylene ketone oxide was added to a cold mixture of one part of nitric acid (sp. gr. 1:45) and one part of sulphuric acid, in small quantities at a time, and with constant stirring, it gradually dissolved without evolution of nitrous fumes. As the solution became more and more concentrated, crystals began to separate out, and the operation was continued until the whole had a semi-solid appearance.

The resulting mass was then thrown into a funnel provided with an asbestos plug, and the residue left after the acid had drained off, was washed, first with a little nitric acid (sp. gr. 1.45), then with weaker nitric acid, and finally with water, and dried. The mother-liquor and washings, after standing for some hours, deposited a further crop of these crystals, which were treated in a similar manner.

The product, which consisted of a mass of fine needles having a satiny lustre, was then further purified by crystallisation from alcohol or coal-tar naphtha. On analysis, it gave the following numbers :

-

I. 0-1600 gram of substance gave 0.3200 gram of CO, and 0-0301 gram of OH2.

II. 01544 gram of substance gave 0.3103 gram of CO2 and 0.0300 gram of OH2.

These numbers agree with those required by the formula C13H6(NO2)2O2; thus:

Dinitrodiphenylene ketone oxide is very slightly soluble in boiling alcohol, from which it separates on cooling in the form of fine needles, having a beautiful satiny lustre and electrical properties, and which, when viewed under the microscope, affect polarised light. It is more soluble in boiling coal-tar naphtha, and, when heated it sublimes in the form of needles. It melts at 262°.

The acid mother-liquors and washings obtained during the preparation and subsequent purification of dinitrodiphenylene ketone oxide were found to contain a small quantity of a much more soluble compound, which is probably a "mononitro" derivative, but which, up to the present, has not been obtained in a sufficiently pure state to be examined.

Fuming nitric acid (sp. gr. 1-54), well cooled with ice, was found to act on diphenylene ketone oxide in a similar manner to a mixture of nitric and sulphuric acids, but the yield of dinitro-product was not so good, there being produced at the same time a much larger quantity of the more soluble body above referred to.

The Action of Reducing Agents on Dinitrodiphenylene Ketone Oxide.

When dinitrodiphenylene ketone oxide was suspended in alcohol, and digested with tin and hydrochloric acid, it gradually dissolved, and the solution acquired a slightly yellow tint. When the reduction was completed, which was after the lapse of three or four hours, a little water was added and the alcohol boiled off. The solution was then poured into an evaporating dish and set aside to cool. While cooling, it deposited a quantity of semi-crystalline matter, which evidently consisted of the hydrochloride of a new base. This was thrown on a filter, and washed two or three times with dilute hydro

chloric acid.

The mother-liquors and washings, when further evaporated, deposited on cooling another crop of crystals, which were treated in a similar manner.