to stand 24 hours, collected and purified by crystallisation from alcohol.

0.1108 gave 0.2465 CO2 and 0·0399 H2O. C = 60·67; H = 4·00. C15H6O6(C2H3O), requires C = 60-79; H= 3.96 per cent.

It forms a silky mass of colourless needles melting at 213-215°, very sparingly soluble in alcohol. It is insoluble in cold alkaline solutions.

In order to determine the number of acetyl groups present, a slight modification of Liebermann's method was adopted. To a solution of the substance in boiling acetic acid, a few drops of sulphuric acid were added and the whole boiled for about a minute; a considerable quantity of boiling water was then added, and the crystals of luteolin which separated on cooling, were collected and weighed.

0.5283 gave 0.3310 luteolin. C15H10O6 = 62·65.

C15H7O6(C2H3O), requires C15H10O. 69-41 per cent.

=

C15H10O62.99

It was, therefore, a tetracetyl derivative.

Dibromotetracetylluteolin, prepared from dibromoluteolin in a similar way to the above compound from luteolin, was obtained in the form of fine colourless needles melting at 218-220°; it is very sparingly soluble in alcohol.

0.1359 gave 0.2220 CO2 and 0·0353 H2O. C = 44·62; H = 2.88. C15H OBr2(C2H3O), requires C 45'09; H= 2.61 per cent. Tetrabenzoylluteolin was prepared from luteolin by the method of Baumann and Schotten, using a ten per cent. solution of caustic soda. The colourless sticky product thus formed became solid after some hours; it was then ground into a paste with water, well washed with dilute alkali, and purified by crystallisation from benzene.

C= 73.33; H=3·95.
H= 3.65 per cent.

0.1114 gave 0.2996 CO, and 0.0397 H2O.
C15H7O6(CHO), requires C = 72.24;
C1sH6O6(CH2O).
4
C = 73.50; H = 3.70

3

It was, therefore, a tetrabenzoyl derivative. From benzene, in which it is sparingly soluble, it was obtained as a spongy mass of colourless needles melting at 200-2010.

Action of Fused Alkali on Luteolin.-Rochleder (loc. cit.) has stated that luteolin, when fused with alkali, yields protocatechuic acid, and probably phloroglucin, but as judging from the formula of luteolin the production of the latter seemed somewhat unlikely, experiments were made to determine this point. Unfortunately, however, but a small quantity of pure luteolin was available for this purpose, and the results obtained, although sufficient to prove that phloro

glucin is not produced at the temperature employed, must be considered as but preliminary to an exhaustive study of this reaction.

Luteolin was heated at 170-200° for half an hour with 10 times its weight of potassium hydroxide dissolved in a little water; the melt was dissolved in water, the solution neutralised with acid, extracted with ether, the extract evaporated, and the crystalline residue dissolved in a little caustic potash. After saturation with carbonic anhydride, the alkaline liquid was extracted with ether, and the residue left on evaporating the ethereal extract was purified by crystallisation from water. The mass of almost colourless needles thus obtained, melted at 210°, and when dissolved in water, the solution gave no coloration with ferric chloride. It could not, therefore, be phloroglucin, which gives such a characteristic reaction with this reagent. The further examination of this substance will no doubt throw considerable light on the constitution of luteolin.

To isolate the second product of the reaction, the remaining alkaline solution was neutralised with acid and extracted with ether; on evaporating the ethereal extract, a crystalline residue was obtained which after crystallisation from water, formed colourless needles melting at 195°; these, when dissolved in water, gave a strong green coloration with ferric chloride. It had all the properties of protocatechuic acid and was found to be identical with it.

Methylation of Luteolin.-One part of luteolin dissolved in a solution of 10 parts of caustic potash in methylic alcohol, was treated with excess of methylic iodide, and digested at the boiling heat for 24 hours. After removal of the excess of methylic iodide, and the greater portion of the alcohol by distillation, the residue was poured into water, the precipitated product dissolved in ether, and the resulting solution after being washed with dilute alkali was evaporated to a small bulk. The crystalline mass which separated on cooling, was collected, rinsed with a little ether, and purified by crystallisation from alcohol.

0.1078 gave 0.2643 CO2 and 0·0555 H2O. C = 66.86; H = 5·71. 0.1264 0.3102 CO2 and 0·0635 H2O. C = 66·92; H = 5·58. C1H2O. (CH,), requires C = 66-66; H = 5.55 per cent.

This compound is deposited from alcohol as a spongy mass of needles, of a faintly yellowish tint, insoluble in alkalis, and melting at 191-192°. A determination of the methoxy-groups present indicated that the compound contained but three of them, in which case it would be a derivative of methylluteolin. Owing, however, to lack of material, this result cannot at present be confirmed, but it is hoped soon to be able to thoroughly investigate this reaction.

If the properties of luteolin be considered, one cannot but be.

struck with their close similarity to those of fisetin, C15H10O6, the colouring matter of "Young Fustic" (Rhus Cotinus., L.). This substance according to the investigations of Schmid (Ber., 1886, 19, 1739) and Herzig (Ber., 1895, 28, 293) contains four hydroxyl groups, yields a dibromo-derivative, is readily decomposed into resorcinol and protocatechuic acid, and, as I have shown (loc. cit.), also unites with mineral acids. Fisetin (Herzig, loc. cit., and Kostanecki, Ber., 1895, 28, 2302) is most probably a tetrahydroxy-ẞ-phenylpheno7-pyrone, and its constitution is as follows.

Though it is at present too early to speak with certainty as to the constitution of luteolin, it is most probable that its further examination will show it to differ only from fisetin in the position of the hydroxyl group in the pheno-7-pyrone ring.

The study of luteolin will be continued by the author in conjunction with Mr. G. Y. Allen.

Clothworkers' Research Laboratory,
Dyeing Department,
Yorkshire College.

XXIII.-Lead Tetracetate and the Plumbic Salts. By A. HUTCHINSON, M.A., Ph.D., and W. POLLARD, B.A., Ph.D.

Introduction,

IN a note published in this Journal in September, 1893 (Trans., 1893, 63, 1136), we pointed out that the crystals obtained when minium is dissolved in glacial acetic acid were to be regarded as lead tetracetate, a salt of lead dioxide, and that it would in all probability be possible to prepare other salts of quadrivalent lead from this substance. During the past two years we have, as opportunity offered, attempted a füller study of the properties of this compound, and, although some points are still under investigation, we venture now to lay before the Society the results we have so far obtained.

Lead Tetracetate.

Historical. Since the time of Berzelius, chemists have been aware that minium is soluble in acetic acid, and a method of detecting and estimating certain impurities found in the commercial article has

been based on this fact. Little, however, was known of the properties of this solution till Jacquelain noticed (Comptes rendus mensuels des Travaux Chimiques, 1851, 1; Abstr., J. pr. Chem., 1851, 53, 151), as Dumas had done before him, that a solution of minium in aqueous acetic acid soon decomposed and deposited lead dioxide; he found, also, that this decomposition was greatly accelerated by heat or by the addition of water, and further observed that when he employed glacial acetic acid at 40° as the solvent, the solution, on cooling, deposited a crop of slender, colourless, oblique prisms. The formation of these crystals had, Jacquelain tells us, been previously noted by Balard, who did not, however, study them.

A few years later, Schönbein (J. pr. Chem., 1858, 74, 315) made similar observations on the behaviour of the solution of minium in acetic acid, and found that sulphuric acid precipitated only a part of the lead from this liquid, leaving in solution the "acetate of lead peroxide;" Schönbein does not appear to have been acquainted with Jacquelain's work, nor to have obtained any crystals from his solutions. On filtering off the crystals of "acétate de bioxide de plomb," and attempting to dry them between filter paper, Jacquelain found that they quickly turned brown, decomposing into lead peroxide and acetic acid. On the addition of water, this decomposition became complete, and he was therefore able to determine the percentage of acetic anhydride in the substance by titrating the aqueous solution. with standard alkali. The lead was estimated as chloride in another portion.

The results led him to adopt the improbable formula

=

PbO2,3(C1H303) [O = 100, Pb 1294, H = 12·5, C = 75], which requires lead dioxide = 43.86 and acetic anhydride 5614 per cent. Finding that Jacquelain's improbable formula was based on insufficient data, we determined to submit the substance anew to investigation.

Preparation.-Commercial red lead was added little by little to hot glacial acetic acid till no more dissolved and lead peroxide began to separate; the solution was then either filtered hot, or the crystals deposited on cooling were subsequently freed from peroxide by collecting them in a funnel on a porcelain filtering plate, and washing away the finely divided peroxide by cold acetic acid. The crystals were purified by recrystallisation from hot glacial acetic acid, and dried over sulphuric acid in a vacuum.

As regards the interaction which takes place, Jacquelain seems to have held the view that a compound of minium with acetic acid is first formed, for he speaks of an "acetate of minium" which, as the solution crystallises, splits up, yielding crystals of acetate of

dioxide of lead, and ordinary lead acetate which remains dissolved. Schönbein, on the other hand, believed that when minium dissolved in acetic acid, the two oxides contained in the former separated, and that both lead acetate and acetate of the peroxide were present in solution. As, however, lead peroxide is insoluble in acetic acid, Schönbein adopted the hypothesis that lead peroxide can exist in two conditions, in one of which it is capable of union with acetic acid and in the other incapable of such combination.

We are inclined to think that Schönbein's view is in the main the correct one, for it is conceivable that the peroxide set free from minium by abstraction of lead monoxide should, at the moment of its formation, be capable of combining with acetic acid, although it is insoluble when once formed. We have not, however, so far been able to obtain fresh direct experimental evidence bearing on this point.

The substance can be readily analysed by taking advantage of the extraordinary ease with which water decomposes it; a weighed portion was treated with hot water, the PbO2 collected on a tared filter, dried at 110°, and weighed, the acetic acid being estimated in the filtrate by titration with a standard alkali. The lead determinations were checked by direct conversion of other portions into chloride and sulphate by evaporation with the respective acids.

Analyses I, II, and IV were made on different samples, II and III on the same.

I. 0.8757 gave 0.4735 PbO2, and required 984 c.c. of potash solution (1 c.c. = 0·00448 KOH).

II. 0.8770 gave 0·4755 PbO2, and required 799 c.c. of soda solution (1 c.c. 0·00395 NaOH).

The pure crystals begin to melt at 175°, and decompose at a temperature a few degrees higher. The only substance known to us. which dissolves the tetracetate without change is glacial acetic acid, in which it is readily soluble when hot, crystallising out again on cooling. Before using this solvent for the molecular weight determinations, we estimated the amount of tetracetate contained in a solution saturated at 17°. For this purpose, two portions of the solution were weighed, and the lead determined in the one case as chloride, and in the other as sulphate.