THE GLUCOSIDES OF CASCARA SAGRADA.

We next desire to speak in detail of the work we have done upon the glucosides of cascara sagrada. The literature on the glucosides of buckthorn and cascara sagrada is quite extensive, and we will give a general account of what has been done upon them by other investigators. Buckthorn bark (Rhamnus frangula) had been investigated quite considerably before cascara sagrada was known and studied. Casselmann,' and later Enz,' worked on buckthorn bark and obtained therefrom citron-yellow silky crystals, which were tasteless and odorless, and melted at 226° C. Their composition was for a long time a matter of dispute among chemists, Hesse3 claiming that the formula was C,,H,O, which was also verified by Faust; but Casselmann set up the formula C,H,O, as the result of his analyses. Faust then decomposed the substance by treating it with alcoholic hydrochloric acid and discovered that it was a glucoside, as it yielded him sugar and an acid which he named frangulinic acid. This he obtained in golden yellow crystals, melting at 248°-250° C., difficultly soluble in water, chloroform, and benzene, but easily soluble in ether and alcohol, as well as in alkalies, in which latter it dissolved with formation of a purple-red color. Later investigations showed that this frangulinic acid was the same as emodin, which is trioxymethylanthraquinone. According to Baeumker' it is quite an active laxative. Thorpe and Robinson' and Thorpe and Miller went into the matter more closely and

1 Casselmann: Ann. Chem. (Liebig), 104, 77.

2 Enz: Vierteljahr. d. prakt. Pharm., 16, 106.

8 Hesse: Ann. Chem. (Liebig), 117, 349.

4 Baeumker Exper. Beitr. zur Kenntniss der pharm. Wirkung von Frangulasäure, Göttingen, 1880.

5 Thorpe and Robinson: J. Chem. Soc., 57, 38.

6 Thorpe and Miller: Ibid, 61, 1.

21

determined that the glucoside had the formula C,,H,,O, and was split up by acids into emodin and a dextrorotatory sugar, which is not glucose, however, but was identified by them as Rhamnose. Schwabe's' work on cascara sagrada is quite extensive, and he concluded that the active principle is emodin, which melts at 254° C., and whose formula he determined to be C1,H,,O,+H,O. The cascarin of Le Prince' is certainly not a pure substance, to judge by the description given.

15

10

Summed up, the work done indicates that buckthorn bark contains a glucoside frangulin and that this is split up into emodin, which is trioxymethylanthraquinone, and a sugar, Rhamnose or isodulcite, and further, that cascara sagrada contains emodin but not frangulin. We proceeded as follows in obtaining the glucoside of cascara sagrada which had, up to the time of our work, not been obtained. The drug was extracted with chloroform to remove fats, etc., and the residue extracted with eighty per cent. alcohol, and the resulting extract dried and dissolved in hot water. On cooling, some resinous, waxy substance separated and was filtered off. The filtrate was treated with lead acetate, which produced a yellow precipitate. This was filtered off and stirred with hot water on a water-bath. As the lead tannates are difficultly decomposable by hydrogen sulphide it is advisable to pass this gas through the suspended precipitate of lead salts at a temperature of about 100° C. This was done until on shaking the flask, whose mouth was closed by the thumb, the latter was raised by the pressure of the gas. The lead sulphide was filtered off and the filtrate evaporated to dryness, resulting in a dark brown substance which consisted mainly of tannins, as portions of them dissolved in water gave good inks on treatment with ferric salts. This tannate mass appears to be composed of several tannins which we did not undertake to investigate, reserving that for a later time, should Prof. Trimble not find time to undertake the same. The filtrate from the lead tannin was treated with basic lead acetate and gave a dark red-brown precipitate of lead glucosides. It is not by any means pure, or it would be colored more nearly cinnabar-red. The precipitate is stirred with hot water in a flask and treated with hydrogen

1 Schwabe: Archiv. d. Pharm., 226, 569.

2 Le Prince: Compt. rend., 115, 286.

sulphide as before. The filtrate from the lead sulphide on evaporation yields a hard, brown-red substance which is very difficult to obtain in a crystalline form, as efforts to crystallize it from acetone and ethyl acetate resulted only in our obtaining a few dark brown-red needles melting at 237° C., the most of it separating out in an amorphous condition. Not sufficient of it was obtained to make an analysis, but we could confirm that it was not emodin, as it gave no purple color on being treated with caustic, potash. It is the glucoside of cascara sagrada which has so far eluded capture, and we have named it purshianin as already explained, being analogous to the frangulin of buckthorn bark. On heating it with alcoholic hydrochloric acid we obtained a sugar and a product which proved to be emodin. The product obtained by heating the purshianin with alcoholic hydrochloric acid was poured into cold water, when a yellow crystalline substance separated out. This was recrystallized from ethyl acetate, which proved the best solvent for the purpose, and separated therefrom in reddish yellow needles melting at 254° C. (one lot melted at 265° C.) and producing a blood-red color on treatment with caustic alkalies. The crystals were dried at 110° C. and appeared to lose some moisture during the process, which was not, however, determined. The analysis of these reddish yellow needles, dried at 100° C., resulted as follows:

I. 0.257 gram gave 0.6304 gram carbon dioxide and 0.0972 gram water or 66.9 per cent. carbon and 4.2 per cent. hydrogen. II. 0.091 gram gave 0.02330 gram carbon dioxide and 0.032 gram water or 66.83 per cent. carbon and 3.9 per cent. hydrogen.

These figures, together with the melting-point 254° C., leave little doubt but that the substance in hand was emodin, which has also been found to be the active principle of buckthorn bark and likely in a measure of rhubarb. The sugar that is formed when purshianin is saponified appears to be dextrorotatory and non-fermentable, but we have not yet examined it with sufficient care to arrive at a definite conclusion as to what it is. It will

be necessary to obtain a quantity of it, make its osazone, and endeavor to recognize it by the properties and analysis thereof. It appears that the glucoside purshianin is certainly one of the active principles of the drug as in doses of one-fifth of a grain it produces the effects of the drug as far as these affect the bowels. Purshianin is tasteless and odorless, and soluble in alcohol, ethyl acetate, acetone, alkalies, and hot water. We wish to reserve for ourselves its further study, which we hope will bring to light wherein the difference between frangulin and purshianin lies. Frangulin is an orange-yellow powder melting at 225° C., according to Thorpe and Robinson, while we find that purshianin is a dark brown-red crystalline substance melting at 237° C. Curiously enough, both are glucosides and yield the same substance on being saponified; viz., emodin. The difference cannot hence lie in anything but the sugars that are combined with the emodin to form the glucosides, or perhaps in the way in which these are combined. A fact it is, nevertheless, and notwithstanding the above apparent marked similarity, that cascara sagrada acts more agreeably and effectively than buckthorn bark, which it has practically supplanted. It may be possible that both of these drugs do actually contain as their active principle the identical glucoside, and that the cascara either contains more of it or produces a happier result on the patient by virtue of its other accompanying constituents. We hope to be able to solve this question this autumn, and also to isolate the bitter principle of the bark and determine in what way magnesia or lime removes or alters the same so as to render the preparation free from bitterness.

LITERATURE UPON RHAMNUS FRANGULA AND RHAMNUS PURSHIANA. 1875, Liebermann: Ber. d. chem. Ges., 8, 970; J. Chem. Soc., 29, 251. 1876, Liebermann and Waldstein: Ber. d. chem. Ges., 9, 1775; J. Chem. Soc., 31, 477.

1879, Prescott: Am. J. Pharm., 51, 165.

1885, Kennedy and Wiltrout: Proc. Am. Pharm. Ass., 33, 499.

Limousin: J. Pharm. Chim., (5), 6, 80; Pharm. J. Trans., 15, 615.

1885, Wise: Western Druggist, 7, 89.

1886, Wenzell: Pharm. Rundschau, 4, 79; Pharm. J. Trans., 16, 918. 1887, Butterfield: Pharm. J. Trans., 18, 473.

1888, Meier and Webber: Am. J. Pharm., 60, 87.

1888, Eccles: Drug. Circ., 32, 54; Proc. Am. Pharm. Ass., 36, 402.

1888, Schwabe: Archiv. Pharm., 226, 569; Proc. Am. Pharm. Ass.,

37, 491.

1889, Findlay: Pharm. J. Trans., 20, 491.

1889, Fuge: Pharm. J. Trans., 19, 736.

1889, Zeig: Proc. Am. Pharm. Ass., 37, 261.

1890, Rusby: Proc. Am. Pharm. Ass., 38, 203; Bulletin Pharm., 5,

115.

1892, Blanc Revue Therapeutique, 326.

1892, Le Prince: Compt. Rend., 115, 286; Pharm.J. Trans., 23, 183; Proc. Am. Pharm. Ass., 41, 691.

1892, Phipson: Compt. Rend., 115, 474; Repert. d. Pharm., October. 1893, Braeutigam: Pharm. Centralh., 34, 383.

1890, Thorpe and Robinson: J. Chem. Soc., 57, 38. 1892, Thorpe and Miller: J. Chem. Soc., 61, 1.

BALTIMORE, AUGUST 1, 1897.

MOLECULAR WEIGHTS OF SOME CARBON COMPOUNDS; A FEW WORDS MORE.

BY C. L. SPEYERS.
Received May 26, 1898.

Na short review by Noyes' of an article by me,' Noyes writes: "No discussion of the results is given, however, nor is any evidence of their accuracy presented, which is particularly unfortunate, since, according to the reviewer's experience, a Beckmann thermometer, which is subjected to considerable variations of pressure, may give quite unreliable readings, owing to the imperfect elasticity of the bulb."

In regard to the discussion of the results, I would state that the measurements made at the boiling-points under ordinary pressures need no discussion; such measurements have been thoroughly discussed by others. In regard to the measurements made under reduced pressure, I would state that I am altogether at a loss to account for the peculiar results obtained. The values given under "cor." in my article show that the rise in the temperature is, in most cases, probably the rise actually caused by the substance dissolved and that the rise is not appreciably influenced by irregular boiling nor by the very slight change in pressure during a set of determinations.

In regard to an error that might seem to come from the imperfect elasticity of the thermometer bulb, I would state that

1 Rev. Am. Chem. Research, 4, 55 (1898).

21. Phys. Chem., 1, 766 (1897).