and the distillate fractionally crystallized from ethyl acetate, when it yielded cerotic acid, a mixture of acids melting between 62 and 66° C., and a small amount of an unsaturated oily acid, which was apparently oleic acid. The total amount of fatty acids obtained from the original petroleum extract was about 10 Gm.

II. Ether Extract of the Resins.

This was a nearly hard, dark green resin, and amounted to 472 Gm. It was dissolved in about 2 liters of ether, and the solution extracted with several successive portions of a solution of sodium carbonate, followed by a 10 per cent. solution of potassium hydroxide, when the following products were obtained.

The first few extractions with sodium carbonate afforded a rather thick, dark green liquid (a), containing a large amount of a crystalline substance in suspension, which proved to be mono-sodium homo-eriodictyol. The next portions of sodium carbonate afforded a further quantity of monosodium homo-eriodictyol, together with an intensely yellow liquid (b). The treatment with sodium carbonate was then continued until nothing further was removed from the liquids, when no more mono-sodium homoeriodictyol separated, but a clear, greenish-brown liquid (c) was obtained. The ethereal liquid was then washed with water, which removed a further quantity of a substance that imparted a dark greenish-brown color to the aqueous liquid (d). As the ethereal liquid still possessed a very dark green color, although now free from substances of an acidic nature, it was extracted with a 10 per cent. aqueous solution of potassium hydroxide which afforded a deep green, alkaline liquid (e). After washing the ethereal liquid, which was still colored, and which contained nothing soluble in 20 per cent. sulphuric acid, it was dried, and the ether removed. It then finally yielded a quantity of a soft, green solid (ƒ).

The above-described treatment of the ether extract of the resins showed it to be an extremely complex mixture, and as all the products thus obtained evidently differed in character, they were separately examined.

(a) This liquid, from which the crystalline mono-sodium homo-eriodictyol held in suspension had been removed by filtration, was acidified with sulphuric acid, and extracted with ether in the usual manner. By this means the following products were obtained: (1) A small amount of a black resin, insoluble in ether; and (2) a large amount of a soft, green, resin, readily soluble in ether, which evidently represented the principal constituent of the ether extract of the resins (C). As there was no evidence of any crystalline substance in this soft, green resin, it seemed of interest to ascertain whether it was of a glucosidal nature. A portion of it was accordingly dissolved in alcohol, a little of water added, and subsequently an amount of sulphuric acid corresponding to about 5 per cent. of the weight of liquid employed. This mixture was boiled for five hours,

after which the alcohol was removed, and the residue submitted to steam distillation, but no volatile products were obtained. The liquid contained in the steam distillation flask was then filtered from the apparantly unchanged resin, and extracted with ether, but this removed only a trace of substance. The aqueous liquid was finally freed from sulphuric acid by means of barium hydroxide, filtered and concentrated, when it yielded nothing but a small amount of crystalline barium ethyl sulphate, which was doubtless produced by the action of the sulphuric acid on the alcohol employed. The chief constituent of the ether extract of the resins was therefore not a glucoside.

Isolation of a new yellow, crystalline substance, C16H12O6.

(b). This intensely yellow liquid was acidified with sulphuric acid, and extracted with ether in the manner previously described. The ethereal liquid afforded a small amount of a yellow varnish, which was dissolved in ethyl acetate and allowed to stand. In the course of a few days, bright yellow, crystalline crusts were deposited, which were removed by filtration, and washed with ethyl acetate. The substance thus obtained was dissolved in a large quantity of boiling alcohol, the solution filtered, and concentrated to a small bulk, when minute, glistening spangles of a deep golden-yellow color separated from the boiling liquid. This substance, which appeared to be homogeneous, was collected on a filter, washed with alcohol, dried, and analyzed.

4.08.

0.0895 gave 0.2092 CO, and 0.0330 H.O. C = 63.75; H The remainder of the compound was then recrystallized from alcohol, and again analyzed.

0.0873 gave 0.2050 CO, and 0.0328 H2O. C= 64.04; H = 4.17. C16H12O, requires C 64.00; H 4.00 per cent.

=

This substance did not melt when heated to 337° C., but from its appearance and the analytical results it was evidently pure. As it is not identical with any compound hitherto described, it must be regarded as a new substance. It was very sparingly soluble in alcohol and in ethyl acetate, and apparently insoluble in the other ordinary solvents, but it dissolved in alkalies and in concentrated sulphuric acid, yielding solutions of a bright yellow color.

Acetyl Derivative of the Yellow Substance.-A small quantity of the yellow, crystalline substance was dissolved in hot acetic anhydride, and the solution boiled for one hour. After cooling, a considerable volume of water was added, and the mixture allowed to stand for several hours, when a colorless, crystalline substance separated in tufts of needles. This was collected on a filter, and, after recrystallization from a mixture of ethyl acetate and alcohol, it melted at 213-215° C. As the total amount of the yellow substance obtained was only about 0.3 Gm., it could not be subjected to further study.

It has already been noted that a considerable amount of mono-sodium homo-eriodictyol separated from the liquids (a) and (b), and it was from this source that the greater portion of the homo-eriodictyol was obtained.

(c), (d), and (e). These liquids were acidified with sulphuric acid ⚫ and extracted with ether, as before described. (c) and (d) each afforded about 10 Gm. of a greenish-brown resin, while (e) yielded about 20 Gm. of a bright green resin, but from none of these products could anything crystalline be obtained.

(ƒ). This substance, as previously indicated, represented that portion of the ether extract of the resins which was not dissolved by shaking the ethereal solution successively with solutions of sodium carbonate and potassium hydroxide. In appearance it resembled the petroleum extract of the resins, and was found to contain some of the hydrocarbons and other constituents of that extract which had not been completely removed by the treatment with petroleum, but no other crystalline substance could be isolated from it.

III. Chloroform Extract of the Resins.

This was a hard, greenish resin, and amounted to about 39 Gm. A portion of it was dissolved in amyl alcohol, and the liquid extracted with sodium carbonate solution, when the greater part of the resin was taken up by the alkali, yielding a solution of a bright yellow color and a small amount of a brown, sparingly soluble sodium compound. Nothing crystalline could be obtained from these products.

IV. Ethyl Acetate Extract of the Resins.

This was a very soft, dark brown resin, and amounted to about 49 Gm. It was not entirely soluble in sodium carbonate solution, and was evidently a complex mixture of substances. The only crystalline product that could be obtained from it was a small amount of homo-eriodictyol.

V. Alcohol Extract of the Resins.

This was a thick, dark brown, tarry liquid, and amounted to about 24 Gm. It was entirely soluble in sodium carbonate solution, but was evidently a mixture of substances, and contained nothing crystalline.

Summary and Conclusions.

The results of this investigation have shown the leaves of Eriodictyon californicum to contain, in addition to some essential oil, resins, and other amorphous bodies, the following substances, in the amounts approximately indicated:

Triacontane, C3H62 (m. p. 65.2° C.)

Pentatriacontane, CH (m. p. 74-5-75° C.)]

72

about 0.7 per cent.

Formic, acetic, cerotic, and other acids, in a free state.

Glycerides of formic, butyric, cerotic, and other acids.

The total amount of acids, free and combined, amounted to about 0.5 per cent.

A Phytosterol (m. p. 136-137° C.), in very small amount.

12

Eriodictyol, CH1O。 (m. p. 267° C.), a new, crystalline substance of phenolic nature, about 0.23 per cent.

14

Homo eriodictyol, C16H1О6 (m. p. 223° C.), another new, crystalline substance of phenolic nature, about 3 per cent.

A new, yellow, crystalline substance, C16H12O6, also of a phenolic nature, about 0.014 per cent.

Glucose (phenylglucosazone, m. p. 205° C.), a considerable amount. The amount of essential oil obtained corresponded to about o.1 per per cent. of the weight of the leaves. It was a yellow liquid, which possessed the characteristic odor of the drug, and had the following constants : d=0.9372 at 15°/15° C.; a„—0° 24′ in a 25 Mm. tube. It was readily soluble in 70 per cent. alcohol.

The total amount of crude resinous substances was 29.2 per cent. of the weight of the leaves, and about 75 per cent. of these resins was soluble in ether. They represented an exceedingly complex mixture, from which some of the above-mentioned crystalline substances were isolated.

The statememt of Thal (loc. cit.) that Eriodictyon leaves contain the so-called "ericolin," we are not able to confirm. The peculiar odor which he observed on heating an extract of the leaves with dilute sulphuric acid, and which he regarded as characteristic of "ericinol"-the assumed hydrolytic product of ericolin-we believe to have been due to the formation of a small amount of furfural. The latter compound has now been definitely identified as the chief volatile product resulting from the action. of a dilute mineral acid on a purified extract of Eriodictyon leaves, and is doubtless produced from the carbohydrates contained therein.

The so-called eriodictyonic acid, which was described some years ago by Quirini (loc. cit.), was evidently not a pure substance. The method by which it was obtained indicates that it must have consisted, to a large extent, of the new crystalline substance of a phenolic nature which we have designated homo-eriodictvol.

In conclusion it may be noted that our investigation of Eriodictyon leaves has disclosed the occurrence, among other compounds, of four substances which have not hitherto been known. It is our intention to continue the chemical study of two of these newly isolated substances, namely, eriodictyol and homo-eriodictyol.

24

THE BOTANICAL CHARACTERS OF SOME CALIFORNIAN SPECIES OF

GRINDELIA.

BY PIERRE ÉLIE FÉLIX PERRÉDÈS, B. SC., F. L. S.

Pharmaceutical Chemist.

[A Contribution from the Wellcome Research Laboratories, London.]

In a paper presented to this Association last year by F. B. Power and F. Tutin, on the "Chemical Examination of Grindelia," the authors stated that the material employed by them, having been obtained "directly from California," 99 66 was evidently Grindelia robusta, one of the varieties of the latter, or a closely related species. The material in question was, in fact, found to conform most closely in its characters to the description of Grindelia camporum, Greene."

In a discussion which followed the reading of the paper it was remarked by one of the speakers (Professor Rusby) that the authors had failed "to state the exact species of the grindelia used, and that in California one would meet Grindelia robusta or Grindelia squarrosa, and more likely the latter" (loc. cit., p. 201).

Since the presentation of the above-mentioned paper, I have had an opportunity of submitting a sample of the material used by Power and Tutin to Dr. Willis L. Jepson, author of the "Flora of Western Middle California," and he states that the plant is undoubtedly G. camporum as defined by him in the aforesaid work. With regard to Professor Rusby's statement, the fact may be noted that Grindelia squarrosa, Dunal (Donia squarrosa, Pursh) can hardly be said to occur in California at all. Its distribution is given by Gray in the "Synoptical Flora of North America (Vol. I, Part II, p. 118) as " Plains and prairies, Minnesota and Saskatchewan to Montana and south to Missouri and Texas, west to Nevada, Arizona, and borders of California (Mex.)," and by Britton and Brown in their "Illustrated Flora of the Northern States and Canada" (Vol. III, p. 321) as "In dry soil; Illinois and Minnesota to Manitoba, south to Missouri, Texas, Nevada and Mexico. Adventive in southern New Jersey." Grindelia squarrosa, Dunal, is not mentioned in the "Botany of California," and I, myself, have not seen any undoubted specimen of this species from California, either at the British Museum or at Kew. Dr. Jepson, who has made a life-long study of the Flora of California, also informs me that he has never found it in that State.

The Grindelia camporum of Greene, as extended by Jepson in the "Flora of Western Middle California," is the common "gum plant" of California, and I am told by Dr. Jepson that the frequency of its occurrence as compared with that of the next most abundant species, viz., Grindelia cuneifolia, Nuttall, is at least as a thousand to one.

Both G. cuneifolia and G. camporum are now generally acknowledged

*Proc. A. Ph. A., 1905, 53, p. 193.