The most important differences shown by the various celluloses are in relation to the solvent action of the alkalis upon them, and the degree of this action is usually shown by the condition of the cellulose fibres after washing and drying. The products of the action are in the first instance gelatinous, and those fibres which undergo degradation thereby, show an agglutination of the cellulose fibrils on drying. (3.) The loss of weight sustained by boiling with an alkaline solution of arbitrary strength has been observed in two stages for the purpose of separating its more purely solvent action (continued for five minutes) from what may be termed its degrading action (continued one hour subsequently). These observations throw a certain light on the nature and order of stability of the bodies of which the fibre is composed; and attention will be drawn to this point in regard to the distinctive character of the jute fibre.

(5.) A high percentage of ash-constituents is usually, in plant structures, associated with the presence of gummy or pectic substances; and the relatively small distribution of the latter throughout the wood and bast of plants accords with their low percentage of inorganic constituents. It is to be noted that the Boehmeria fibre stands conspicuously high in regard to its ash, and the presence of pectous substances thus indicated is confirmed by the large loss in weight sustained in the boiling alkaline solution.

(6.) Cellulose structures which differ from pure cellulose may be regarded as containing, in addition, (a) bodies of the pectic group; (b) substances connected with the trihydric phenols; (c) substances containing furfural,—the union of these with cellulose being probably such as is known as combination by residues, i.e., to form with the cellulose residue a chemical whole. The groups of compounds under (a) (b) and (c) differ from one another and from cellulose in respect of elementary composition, and its determination is a certain measure of the quantitative relations of these groups to one another. It may be remarked, that (b) and (c) are in all cases yet observed co-incidental, and agree also in respect of high carbon percentage: consequently as factors in the mean carbon percentage of a fibre they cannot as yet be separated.

(7.) The chemical evidence of lignification is the formation of substitution-derivatives on exposure to the action of chlorine gas, and the proof of the formation of these is afforded by their characteristic colour-reaction with solutions of the neutral sulphites.

To sum up these results and bring out more clearly the distinctive character of the jute fibre, I may recapitulate its more striking points of differentiation from the other fibres included in this investigation. (1.) High carbon percentage. (2.) Power of resisting the continued action of boiling alkaline solutions, from which, together with the re

sults of an examination of the substances dissolved, it is to be inferred that the pure fibre contains no constituents of the pectic group. (3.) Such uniformity.in composition and properties as to permit us to regard it as a chemical whole. (4.) Its comparative simple microscopic features.

Some of these characteristics are represented amongst the other fibres, but are never united as in the case of jute, which is therefore to be preferred as a simple type of lignification; and over biologically complicated structures, such as wood, its superiority is still more manifest.

IV.-On a Condensation-product of Phenanthraquinone with Ethylic Aceto-acetate.

By FRANCIS R. JAPP, M.A., Ph.D., Assistant Professor of Chemistry in the Normal School of Science, South Kensington, and FREDK. W. STREATFEILD.

IN a former communication (this Journal, Trans., 1882, 270) we described the acetonquinimide of phenanthrene, C17H15NO2, obtained by the interaction of phenanthraquinone, acetone, and ammonia. In endeavouring to extend this reaction to other ketones, we substituted ethylic aceto-acetate for acetone. No reaction took place at ordinary temperatures, but on heating phenanthraquinone, ethylic aceto-acetate, and concentrated aqueous ammonia for a short time under pressure at 100°, a dark-coloured mass was obtained, from which, by an appropriate process of purification, a compound was isolated, crystallising in needles, and fusing at 184.5-185.5°. (A description of this process. of purification is superfluous, as a much better method of preparing the substance is given further on.) The compound did not contain nitrogen. On analysis it yielded figures agreeing with the formula C20H1604:

A compound of this formula would be formed from one molecule of phenanthraquinone and one of ethylic aceto-acetate, by the elimination of one molecule of water.

The reactions of this compound show that in its formation one atom of oxygen from the quinone is eliminated along with two atoms of hydrogen from the ethylic aceto-acetate, the two resulting dyad groups then uniting by means of the free affinities. From the fact that no such condensation occurs when ethylic diethaceto-acetate is substituted for ethylic aceto-acetate, we may conclude with a high degree of probability that the two hydrogen-atoms thus eliminated are furnished by the methylene group in aceto-acetic acid. The compound would thus possess the formula

CH–CO

.CO–CH,

and might be termed ethylic phenanthroxylene-aceto-acetate.*

The reaction is analogous to the condensations of aldehydes with ethylic aceto-acetate described by Claisen (Ber., 14, 345), in which, however, gaseous hydrochloric acid was employed as a dehydrating agent.

The dehydrating action of aqueous ammonia has not, so far as we are aware, been previously observed. It resembles, however, the dehydrating action of aqueous caustic potash upon acetone described by Heintz (Annalen, 196, 118). Condensations between aldehydes and ketones have also been effected by means of dilute caustic soda (Schmidt, Ber., 14, 1459; Claisen, ibid., 14, 2468).

The foregoing analogy led us to examine whether a caustic alkali could not be substituted for ammonia in the above reaction. Not only did this prove to be the case, but the yield by the new method was fully twice as great as when ammonia was employed; whilst, owing to the almost total absence of resinous bye-products, the process of purification was materially shortened.

After several trials the following mode of conducting the experiment was adopted, as yielding the best result:-100 grams of phenanthraquinone, ground to an impalpable powder (this is essential, as larger particles escape conversion), are introdnced into a flask with

* The dyad radical (C ̧‚H)” is phenanthrylene; the dyad radical (CHO)" may be styled phenanthroxylene.

90 grams (an excess) of ethylic aceto-acetate; 150 c.c. of dilute potash (1 part of solid caustic potash to 6 of water) are now added, and the mixture is gently warmed, agitating all the time. The reaction takes place quickly with considerable rise of temperature, and the orange colour of the quinone disappears, giving place to the light grey of the crude condensation-product. The product is boiled with water, washed with alcohol, and crystallised from boiling benzene till the fusing point remains constant,

From 100 grams of quinone over 100 grams of a product, once crystallised from benzene and practically pure, were obtained.

Gaseous hydrochloric acid does not effect the condensation of phenanthraquinone with ethylic aceto-acetate.

Properties.-Ethylic phenanthroxylene-aceto-acetate is deposited from its hot benzene solution in tufts of fine white silky needles. It fuses with blackening and evolution of gas at 184.5—185.5°. It is soluble also in alcohol and in glacial acetic acid. On oxidation with a chromic mixture it yields phenanthraquinone.

Hot caustic potash decomposes it, yielding a purple or a green solution, according to the concentration of the potash. Dilute potash appears to saponify it slowly in the cold. These reactions have yet to be studied.

With bromine in acetic acid solution it appears to form, after long standing, an addition-product. This product, which is much less soluble in acetic acid than the original compound, is slowly deposited from the solution. By recrystallisation from hot glacial acetic acid it was obtained in flat yellow needles. The fusing point could not be determined, as the substance, without previously fusing, became quite black at about 150°. A bromine determination gave figures agreeing with the formula CH16O.Br. (Br calculated, 3333; found, 33.84 per cent.). Of course analysis is incompetent to decide between this formula and the formula of a substitution compound C2H1O,Br2; but judging from the analogy of the compounds discovered by Claisen, the probability is greatly in favour of the first formula. The usual method of deciding this question by determining the quantity of bromine requisite for the formation of the compound is scarcely applicable in the present case, owing to the extreme slowness with which the compound is formed. In the case of the condensation-product of chloral with ethylic aceto-acetate, Matthews (Dissertation, Bonn, 1882, p. 28) observed a similar sluggishness in the way in which this compound combined with bromine.

14

Action of Hydriodic Acid upon Ethylic Phenanthroxylene-aceto-acetate. -A quantity of the above compound was mixed with amorphous phosphorus in a flask, and an excess of faming hydriodic acid was added. A reaction took place, accompanied by a rise of temperature, and the

substance fused to a black pitchy mass. This product, which became semi-solid on cooling, was washed successively with water, with cold alcohol, and with small quantities of ether. The brownish substance which now remained was dissolved in boiling alcohol, and the solution filtered from unchanged amorphous phosphorus. The alcoholic solution deposited on cooling a granular substance, which by repeated crystallisation was obtained in star-shaped aggregations of a pink colour. This colour is due to an impurity, and is best got rid of by dissolving the crystals in boiling light petroleum. On allowing the petroleum solution partially to cool, the colouring matter separates out first on the sides of the vessel, and the solution, when poured off at the proper moment, deposits almost colourless crystals. A final crystallisation from benzene removes the last traces of colour. The pure substance fused at 124°.

We found that the pink-coloured substance could be bleached by exposing it in solution to the action of daylight.

The above somewhat complicated process of purification so diminished the quantity of substance, that from 60 grams of ethylic phenanthroxylene-aceto-acetate only 4 grams of the pure reductionproduct were obtained.

The compound contained no iodine. Analysis of different preparations yielded numbers corresponding with the formula C20H16O3:

The hydriodic acid had therefore removed one atom of oxygen from the compound C20H1O. This process may be most readily explained by supposing that the acetyl-group of the ketonic acid is first reduced to the group CH-CH(OH), which then parts with water, and is converted into the vinyl-group CH, CH-. This hypothetical intermediate compound would be a derivative of B-hydroxybutyric acid, and the ease with which this acid parts with the elements of water, and is converted into crotonic acid is well known. According to this view the reduction-compound would possess the constitution