disposal was too small to admit of a satisfactory fractional distillation, but an attempt was made to effect at least a partial separation of the various substances whose formation might be expected.

The first fraction we obtained boiled at about 200°, and had a specific rotation of [a] +20.9°. The highest fraction had a rotation of [a] -10, and an intermediate fraction a rotation of [a] +8·5°. The impurities in the fraction of lowest boiling point have therefore a small positive or even a negative rotation, so that the value of the positive rotation actually obtained for the first fraction is undoubtedly too low.

When the potassium ethyl salt of ordinary camphoric acid was electrolysed under as nearly as possible the same conditions as we here employed, the fraction of lowest boiling point had a specific rotation of [a] +29.8°. There is thus a distinct approximation of the corresponding fractions from ordinary d-camphoric acid and l-isocamphoric acid, and we think ourselves justified in concluding that the allo-asymmetric carbon atom has the same configuration in both compounds, since if it had the opposite configuration, the rotation of the product would have a value of -30°, instead of a positive rotation of nearly that magnitude.

The fraction of lowest boiling point here obtained is of course not pure ethyl allo-campholytate, but a mixture of this with ethyl camphononate from which separation by distillation was rendered impossible by the small quantity at our disposal. Combustion analyses yielded numbers intermediate between those required by the two esters, and the other properties of the liquid were such as would be shown by a mixture of the two substances.

UNIVERSITY COLLEGE,

DUNDEE.

XXXVIII.-The Constitution of Camphoric Acid. By JAMES WAlker.

IN 1893 (Trans., 63, 495), I showed that the campholytic acid derived from sodium ortho-ethyl camphorate by electrolysis was an aß-unsaturated acid, this conclusion being based on the fact that its dibromide decomposed on neutralisation according to the following equation:

CgHngBr, CO,Na=CgHngBr+CO,+NaBr,

13

a behaviour which Fittig had proved to be characteristic of the dibromides of such acids. Now, as appears from the researches of

Kolbe and of Crum Brown and myself, when an unsaturated compound is formed on the electrolysis of a salt of a carboxylic acid, the acid loses the groups CO2H and H from neighbouring carbon atoms, the double bond in the unsaturated compound being thus between the a- and ẞ-carbon atoms, as reckoned from the carboxyl group which has disappeared. The double bond of campholytic acid then is in the aß-position with regard to the carboxyl group of the camphoric acid which has disappeared on electrolysis; but it is also in the aß-position with regard to the carboxyl group, which remains after electrolysis. This gives us only two possibilities for the relative position of the carboxyl groups in the camphoric acid molecule. Either the molecule contains the group

These groupings, and no others, fulfil the required conditions, but the first is inadmissable, since it represents camphoric acid as a malonic acid derivative, which is altogether at variance with the known facts. There remains therefore only the conclusion (loc. cit., 509) that the camphoric acid molecule contains the complex

•C CO,H
•CH•CO,H

This conclusion, although it received immediate confirmation from the work of Noyes in 1894, was not generally accepted at the time, chiefly, no doubt, on account of the apparent difficulty in explaining the formation of camphanic acid from camphoric acid (compare Trans., 1895, 67, 345).

If exception is taken to the reasoning given above, it can only be on the ground that campholytic acid is not an aß-unsaturated acid, that it is not a direct product of the electrolysis, or that the carboxyl group and hydrogen atom which are removed on electrolysis do not leave the molecule from neighbouring carbon atoms. The second and third of these points will be discussed later in connection with Perkin's formula for isolauronolic acid. With regard to the first, reference need only be made to the decomposition of the hydrobromide of campholytic acid on neutralisation (this vol., p. 380), and to the nature of the isomerism with isolauronolic acid (p. 382), which is universally admitted to be an aẞ-unsaturated acid. The reasoning is further

strengthened by the proof that isolauronolic acid itself is, along with campholytic acid, a direct product of the electrolysis (this vol., p. 376).

An investigation as to the products of electrolysis of potassium alloethyl camphorate (Trans., 1895, 67, 337; 1896, 69, 248), permitted a further conclusion to be drawn, namely, that camphoric acid contained the complex

•C CO,H HČCH CO,H

Now Perkin and Thorpe (Trans., 1897, 71, 1169), by their synthesis of camphoronic acid, a product of the oxidation of camphoric acid, proved conclusively that this acid has the constitution which had been suggested by Bredt, namely:

which, when combined with the complex derived from the behaviour of the electrolytic products, yields the Perkin-Bouveault formula,

This formula was first suggested by Perkin, and afterwards, apparently independently, by Bouveault. Perkin, however, was led by his own experiments (Trans., 1898, 73, 796) to favour another formula which was proposed by him at the same time, namely,

The behaviour of dihydro-isolauronic acid chiefly weighed with Perkin in preferring this to the Perkin-Bouveault formula, but, as Blanc has shown (Bull. Soc. Chim., 1899, [iii], 21, 830), Perkin's observations are open to a totally different interpretation from that put upon them by the author, and on the whole tell in favour of the Perkin-Bouveault formula.

That Perkin's second formula is irreconcilable with the results of electrolysis is plain from the fact that it does not contain a complex with the carboxyl groups in the ortho-position relatively to each other. If the electrolysis of the sodium ethyl salt of an acid having Perkin's formula (II) occurred in the ordinary way, the ethyl ester of the unsaturated acid thereby produced should have formula I. Now this acid is a ẞy-unsaturated acid, and Perkin himself attributes the formula marked III to isolauronolic acid,

which, along with its stereoisomeride, campholytic acid, is the substance actually produced. That an ester having formula III should be formed directly from a substance having formula II is in the highest degree improbable, and that it should be formed by the isomeric transformation of the acid having formula I is almost equally unlikely, for not only is the electrolysis conducted at the ordinary temperature and in a solution which is, at most, feebly alkaline with sodium hydrogen carbonate, but the ester produced immediately separates from the aqueous liquid, and is thus exposed to the action of no chemical agents whatever. The improbability of an isomeric transformation of a primary By-product into an aẞ-acid is still further increased by the fact that the ẞy-acids obtained from the electrolysis of potassium allo-ethyl camphorate show no tendency whatever, under the experimental conditions, to pass into aẞ-acids. The formula for isolauronolic acid (and campholytic acid) derived from the Bouveault-Perkin formula for camphoric acid is

As Blanc has pointed out, this formula has no asymmetric carbon atom, and the acid corresponding to it must therefore be inactive. The isolauronolic acid formula of Perkin, on the other hand, has one asymmetric carbon atom, so that the corresponding acid should exhibit optical activity, unless it be a racemic mixture. Now, the isolauronolic acid prepared from camphoric anhydride and aluminium chloride by Blanc's method is optically inactive, and so are its salts and esters. The acid prepared by electrolysis is also inactive, and this inactivity is likewise characteristic of isolauronolic acid prepared

from the campholytic acid of electrolysis, as well as of the campholytic acid prepared from isolauronolic acid (this vol., pp. 378, 380). The hydrocarbon isolaurolene, CH14, derived from the hydrobromides of campholytic and isolauronolic acids (this vol., p. 380), is also inactive, which is in accordance with formula IV derived from the Perkin-Bouveault formula of camphoric acid, and in disagreement with formula V, derived from Perkin's later formula for this substance, which still contains an asymmetric carbon atom.

There is absolutely no reason to suppose that racemisation of active compounds takes place during electrolysis, for the allocampholytic acid produced by the electrolysis of potassium orthoethyl camphorate is strongly active, in accordance with the formula deduced from either the Perkin-Bouveault or the Perkin formula for camphoric acid. An attempt, too, to resolve isolauronolic acid into active constituents by means of cinchonine was made by Blanc, and failed. It must be conceded, therefore, that the inactivity of isolauronolic acid, from whatever source derived, as well as of the related substances, campholytic acid and isolaurolene, affords very strong evidence in favour of the Perkin-Bouveault formula for camphoric acid. It is impossible, either from Perkin's second formula or from Bredt's formula, to derive a formula for isolauronolic acid which shows this substance to be at once an aß-unsaturated acid, and inherently inactive.

The Perkin-Bouveault formula for camphoric acid has recently received synthetic confirmation from Noyes (Ber., 1900, 33, 54), who, by heating calcium aßß-trimethyladipate, prepared the corresponding ketone, thus:

CMe2

CMea

Now, just as diethyl ketone can be produced from diethylhydroxyacetic