CONSTITUENTS OF THE ESSENTIAL OIL OF ASARUM CANADENSE. 59 VL-The Constituents of the Esseizticcl Oil qf Asurum Canadense . By FREDERICK B. POWER and FREDERIC! H. LEES. THE aromatic essential oil distilled from the underground portion, rhizome, and rootlets of Asarunz canadense, comtnonly known as Canada Snake-root, was first investigated by one of us a number of years ago (Power, Inaug. Diss., Strassburg, 1880 ; PYOC. Amev. Pharm. ASSOC., 1880,28, 464). I n that investigation, the following substances were isolated. (1) A terpene, C,OHIB, b. p. 163-166'; (2) two fragrant alcohols, differing considerably in their boiling points and also in their odour, but both possessing the same empirical formula, CloHl,O. The alcohol of lower boiling point, 196-199', was termed nsurol, and had an odour which recalled that of coriander, but was also somewhat camphoraceoua, whilst the alcohol of higher boiling point, 222-226', had a rose-like fragrance ; (3) a fraction, possessing but little odour, b.p. 254-257", representing the largest single con- stituent of the oil, which, upon oxidation with chromic acid, afforded an acid of the composition C,H,,O,. This acid mas subsequently showu by Petersen (Beis., 1888, 21, 1062) to be veratric acid and was obtained by him by the oxidation of an analogous substance contained in the oil of Asarurn europaeum, boiling a t about 250", which he proved to be eugenol methyl ether ; (4) a fraction collected a t 275-350°, which contains a deep blue oil of undetermined composition ; (5) a large amount of acetic acid, combined with the above-mentioned alcohols in the form of acetic esters, together with a very small amount of a less soluble, oily acid, which appeared t o consist of, or at least to contain, valeric acid.I n consideration of the advance in knowledge of the constituents of essential oils since the period of the first investigation, and the means which are now available for the more positive identification and classification of these constituents by the preparation of well-defined and mostly crystallisable derivatives, it has seemed desirable again to subject the oil in question to a careful chemical examination. EX P E R I N E N T A L . The oil employed for this research, about 2 kilos. in amount, was distilled by Messrs. Schimmel & Co. of Leipzig. Its density a t 15'/15" was 0.952, and its rotation aD= - 3'24' in a 100 mm.tube. The oil was first shaken with a 5 per cent. solution of sodium carbonate in order to remove the free acids, which were examined in connection with the acids obtained by the subsequent hydrolysis of the60 POWER AND LEES: THE CONSTITUENTS OF THE oil. It was then shaken three times successively with a 5 per cent. solution of sodium hydroxide, and afterwards with water until the washings were neutral. The combined alkaline liquids were shaken twice with ether to remove any adhering oil, then acidified with sulphuric acid, completely extracted with ether, and the ethereal liquid dried with calcium chloride. After distilling off the ether, the residual liquid was distilled in a vacuum.Under 10 mm. pressure, it began to distil at 155O, rising rapidly to 250', and the last portion was observed to solidify in the condenser. When refractionated, there mere obtained : I. A light coloured oil boiling at 172' under 35 mm. pressure. 11. A dark oil which boiled somewhat below 200' under 10 mm. pressure and solidified on cooling. The Phenol, C,H,,O,. The first of the preceding fractions was distilled under the ordinary (762 mm.) pressure and afforded : (1) A light greenish liquid, becoming brown on standing, and boiling below 245'. (2) A light yellow liquid, boiling at 245-260°, which did not solidify a t - 169 (3) A small residue, from which R little crystalline substance separated on cooling. Of these fractions, (I) and (2) were analysed. (1) 0.1402 gave 0.3606 CO, and 0.1008 H,O.C = 70.1 ; H = 8.0. (2) 0.1493 ,, 0.3923 CO, ,, 0,1066 H,O. C=71.*6 ; H=7*9. These two fractions were then subjected to a final fractionation (a;) A few drops only distilled below 220'. ( b ) From 220' the mercury rose rapidly to 2.15". (c) The chief portion distilled between 245' and 255', and was (d) Only a few drops distilled above 255'. Eraction (c) was then analysed. 0.1523 gave 0.3992 CO, and 0.1073 H,O. These results indicate that the phenol contained in asarum oil has the empirical formula C9H,,0,. It is a nearly colourless, oily liquid, having an odour recalling, but more agreeable than, that of creosote. In the process of liberating the phenol from its alkaline solution a somewhat clove-like odour was developed, and this at first led us to suspect that the phenol contained some eugenol.This, however, is under the ordinary pressure, with the following result : fairly constant a t 248-252". C = 71.5 ; H = 7.8. C,H,,O, requires C=71*1 ; H=7*9 per cent.ESSENTIAL OIL OF ASARUM CANADENSE. 61 rendered highly improbable, both on account of the constancy of the analytical results and the characteristic colour reaction which was afforded by all the fractions, but most strikingly by the principal fraction (c). Thus a very small quantity of the phenol, when dissolved in 90 per cent. alcobol, gives, with a trace of very dilute ferric chloride, a beautiful violet colour which only gradua.lly fades, whilst eugenol, under the same conditions, gives a green. If, however, the phenol from asarum oil be dissolved in absolute alcohol, and a trace of a stronger solution of ferric chloride added, it affords a green colour, whilst eugenol, under these conditions, gives a deep blue.The amount of the phenol available did not permit of the formation of any deriva- tives, but as it is not identical with any of the known phenols of the formula indicated, it is evidently a new substance, and we shall endea- vour to determine its constitution by some synthetical experiments. IdeiLtiJication of Palmitic Acid. It was noted that in the first distillation of the phenol, a solid sub- stance separated in the condenser, and that in a subsequent fractiona- tion the higher fraction, designated as I1 (p. 60), solidified on cooling. This was, therefore, brought upon a porous tile, and the substance subsequently dissolved in hot light petroleum, from which, on cooling i t crystallised in colourless, pearly plates.I t s melting point was 60-61", and this remained unchanged on recrystallisation. On analysis, i t was identified as palmitic acid : 0.0874 gave 0.2382 00, and 0.0982 H,O. C: = 74.3 ; H = 12.5. C,,H,,02 requires C = '75.0 ; H = 12.5 per cent. Ident$catim of Finene. X e p a t i m of the Terpene.-The oil, which had been shaken with a dilute solution of sodium hydroxide as previously described, was washed with small, successive portions of water until the washings were neutral, and dried with anhydrous sodium sulphate. It was then distilled under diminished pressure and the portion collected which boiled below 100' under 10 mm.pressure. After several frac- tionations of this portion and drying with potassium carbonate, the lowest fraction, which distilled below 85' under 10-15 mm. pressure, was collected. Its density was 0.8566 at 18'/16', which proved the absence of any olefinic terpene. These liquids were then further fractionated under the ordinary pressure, when the greater portion finally distilled below 165', (a), chiefly at 159-161", and only exceedingly small fractions were collected between 165' and 170" (/I) and from 170-180° (7). These were analysed, with the following results :62 POWER AND LEES: THE CONSTITUENTS OF THE (a) 0.1252 gave 0,4003 CO, and 0.1320 H,O. C = 87.2; H = 11.7. (/I) 0.1526 ,, 0.4873 CO, ,, 0,1606 H,O. C=87.1 ; H = l l * $ . (7) 0.1558 ,, 0.4920 CO, ,, 0.162s H,O.C = 86.1 ; I1 = 11.6. CloHl, requires C = 85.2 ; H = 11 *8 per cent. Ymction below 165' -This fraction, boiling chiefly between 159O and 161°, which is seen to consist of a nearly pure hydrocarbon, amounted to about 2 per cent. of the original oil. Its physical con- stants were as follows : d 16"/16" = 0,8563. a r = + 1'36'. It readily formed a crystallisable nitrosochloride melting sharply at 103-104°. From the latter, the nitrolpiperidide was prepared, which, after re- crystallisation from methyl alcohol, melted sharply a t 118 -1 19'. This fraction thus consisted of pinene, and its low rotation indicates it to be a mixture of the d- and I-forms. Petersen (Ber., 1888, 21, 1059) has previously recognised the terpene existing in the oil of both the European and American species of Asccruma as pinene, in the former as the Lvariety, but identified it only by the formation of an oily monobromide and by its conversion into dipentene.As it was possible that the very small fraction of our oil collected between 170" and 180' might contain dipentene or limonene, it was treated with bromine, but only an uncrysfallisable, oily product was obtained. After being carefully dried, a bromine determination was made of this, with the following result : 0.2915 gave 0.3'762 AgBr. Br = 54.9. C,,H,,Br, requires Br = 54.1 per cent. This result serves to prove the absence of either dipenteno or limonene, both of which form crystallisable tetrabromides, CloHl,Br,. By a careful examination of all the fractions, no terpene other than pinene could be detected in the oil.Hydrolysis of the Oil. For further examination, all the oil boiling above the tergene frac- tion was now hydrolysed by boiling with alcoholic potassium hydr- oxide for about 2 hours in a flask provided with a reflux condenser. After distilling off the greater portion of the alcohol from a water- bath, the liquid was brought into a separating funnel and sufficient water added to effect the separation of the oil. The latter was then drawn off, the aqueous alkaline liquid shaken with successive portions of ether, and the ether extracts mixed with the separated oil. The latter was then mashed several times with water, and these washings added to the aqueous alkaline liquid. The ethereal solution of the oil was quickly dried with calcium chloride, filtered, the ether distilled off, and the residue finally subjected t o fractional distillation, firstESSENTIAL OIL OF ASARUM CANADENSE.63 under diminished pressure, and then in part under the ordinary pres- sure. The following fractions were eventually obtained : 195--203', 203-208", 208-216', 216-222", 222-235", 235--?145',and 245-2660". Identijication of I;inccbool. Pkaction 195-203'.-This was a large fraction which, when re- distilled under the ordinary pressure, passed over mainly a t 199' and almost entirely at 198-202' under 768 mm. pressure. It is a colour- less, fragrant liquid. It was analysed and its physical constants were determined with the following results : 0,1372 gave 0.3910 CO, and 0.1436 H,O. C = 77.7 ; H = 11.6.C,,H,,O requires C = 77.9 ; H = 11.7 per cent. d 15*5°/150=0*S711. aD= + 10'48' in a 100 mm. tube ; La]D= + 12.4". When oxidised with chromic acid, it afforded citral, which was ob- tained as a pale yellow liquid of strong, lemon-like odour, distilling at 110-115° under a pressure of 10-12 mm. The latter, by condensa- tion with pyruvic acid and P-naphthylamine, was converted into the crystalline a-citryl-/?-naphthacinchoninic acid, melting a t 1cj5-19S0. The identity of this fraction with d-ZinaZoob is therefore definitely established. It corresponds to the substance C,,H,,O (b. p. 196-199"), which in the first investigation of the oil was designated asarol. Fraction 203-2085.-Thia fraction was too small for further ex- amination, and evidently consisted simply of a mixture of the preced- ing and the following fractions.Fraction 20S-216°.-This was a small fraction, which distilled mostly between 208' and 212'. It was analysed, and its physical constants were determined, with the following results : 0.1368 gave 0,3866 CO, and 0.1424 H20. C = 77.1 ; H= 11.6. 0.14'72 ,, 0.4154 CO, ,, 0.1524 H,O. C=77.0; H=11*5. C,,H,,O requires C = 77.9 ; H = 11 -7 per cent. d 15~5°/150=0~911; aD= -0'24' in a 100 mm. tube. IdentiJicatiorn of Borneo 2. The liquid had a camphoraceous and also somewhat rose-like odour. When subjected to a temperature of - 10" for an hour, no crystalline substance separated. As this fraction of the oil was relatively small, and as its constituents were evidently contained to some extent in the next higher fraction, the two fractions were mixed.A portion, how- ever, of the higher fraction was reserved for special examination. This mixture of the two fractions was now gently oxidised with Fittig's oxidation mixture (Ber., 1885, 18, 3207) in the following64 POWER AND LEES: THE CONSTITUENTS OF THE proportions: 10 parts of oil, 80 parts of potassium dichromate, and 120 parts of sulphuric acid, the latter diluted with three times its volume of water. The oxidising mixture was added in small amounts a t a time to the oil, which was kept cool by immersion of the containing flask in water, After all the chromic acid solution had been added, the mixture was heated on a water-bath for about 20-30 minntes. It was then distilled from a water-bath under diminished pressure and the camphor which separated in the condenser and dis- tillate was collected by filtration, and dried on a porous tile.A little of the sulnlimed product was found to meIt sharply at 175O. A deter- mination of its specific rotation in 90 per cent. alcohol gave the follow- result : uD = - 1'45' ; I = 0.5 dcm. ; c = 8.684 ; [aID = - 40.3'. For further identification of the camphor, the oxime was prepared, and found to melt at 115-116O. As camphor could not be detected in the fraction of the original oil, its formation by the above method of oxidation is conclusive proof of the presence of 1-bonzeol in the oil. The chromic acid liquor remaining from the distillation of the cam- phor was subsequently shaken out several times with ether, the ethereal solution washed with a little water, dried with calcium chloride, and the ether removed by distillation.The residual light yellow oil, which had a strong odour of acetic acid, was found to be not entirely soluble in cold sodium carbonate solution. It was consequently redissolved in ether and the ethereal solution shaken out several times with a dilute solution of sodium carbonate in order to remove the acids, The ethereal solution was then washed with a little water, dried with calcium chloride, and the ether removed by distillation. The residue was a light yellow oil possessing a coumarin-like odour, and on standing a short time became a crystalline paste. This was drained on a porous tile, when the substance was obtained quite white. After recrystallisa- tion from dry ether it melted at 62O, and was insoluble in sodium carbonate solution.On analysis : 0.1148 gave 0.2731 CO, and 0*0902 H20. C = 64.9 ; H = 8.7. C,,H,,O, requires C = 65.2 ; H = 8.7 per cent. This substance is undoubtedly identical with the ketokactons, C,oH,,O, (m. p. 62-63'), which was isolated as a product of the oxidation of terpineol by chromic acid by Wallach, and has been further studied by him, as also by Tiemann and others (AnnaZen, 1893, 275, 153; 277, 118; Ber., 1895, 28, 1773, 1781). The sodium carbonate solution from which the ethereal solution of the above ketolactone had been separated was acidified with hydrochloric acid, and shaken out several times with ether. TheESSENTIAL OIL OF ASARUM CANADENSE. 65 ethereal solution was washed once with water, dried with calcium chloride, and the ether removed by distillation.The residue was a light yellow syrup, whicb, on standing, deposited a crystalline acid. The syrup was consequently diluted with ether, in which the crys- tals appeared to be sparingly soluble, and from which they were easily separated by filtration. After washing with dry ether, the substance was finally recrystallised from boiling ether. It melted at 173-174" and dissolved with effervescence i n a cold solution of sodium carbonate. 0.1 16s gave 0.2266 CO, and 0.0695 H,O. C7Hl0O4 requires C = 53.2 ; H = 6.3 per cent8. This acid is evidently identical with terebic acid, CTHIo04 (m. p. 175O), which has been found as a direct oxid+tion product of terpineol, as also of the ketolactone, CloH,,O, (Tiemann and Mahla, Ber., 1896, 29, 2621).The syrup from which the terebic acid crystallised was not examined but probably contained terpenylic acid, which always accompanies terebic acid when terpineol or the ketolactone is oxidised with chromic acid mixture. I t is thus shown that the fractions of the oil which served for the identification of borneol also contained a con- siderable amount of terpineol. I n the first investigation of asarum oil, a considerable fraction was collected at 222-226', and as a portion of this was still available it WJS thought of interest to examine i t again. It was therefore oxidised with a chromic acid mixture in the manner just described, and among the products of oxidation there were isolated and identified : camphor (m. p. 175") ; the ketolactone, C,,H,,O, (m, p.62') ; and terebic acid (m. p. 173-114"). It therefore contained borneol and terpineol, and, apparently, a small amount of geraniol, as it had the characteristic rose-like odour. C = 52.9 ; H = 6.6. .Tdentijcation of Teqirzeol. phoraceous and also a somewhat rose-like odour. its physical constants were determined, with the following results : C = 77.5 ; H = 11.4. Praction 216-222°.-This was a small fraction. I t had a cam- I t was analysed and 0.1647 gave 0.4682 CO, and 0.1698 H,O. 0.1633 ,, 0.4610 CO, ,, 0.1680 H,O. C = 77.0 ; H= 11-4. C,,H,,O requires C = 77.9 ; H = 11.7 per cent. d 15-5°/15"-0*9267 ; a,= - S"26' i n a 100 mm. tube. The portion of this fraction which had not been used in connection with t h e preceding one, as described under the latter, was employed for the direct identification of terpineol.I n view of the presence of small amounts of other alcohols, the following method was employed. VOL. LXXXI. F66 POWER ,4ND LEES: THE CONSTITUENTS OF THE The liquid was shaken with a concentrated solution of hydriodic acid (sp. gr. Z*O), when a heavy, dark oil was formed. This was separated from the aqueous layer, and shaken with a dilute solution of sodium bisulphite to remove any free iodine. The oil was then washed with water and allowed to stand, when after a short time crystals began to form, and finally the whole became a crystalline paste. This was spread on a porous tile, when a small quantity of nearly white needles was obtained, which, after recrystallisation from light petro1,eum (b.p. 30-40°), melted a t 80'. This melting point was identical with that of dipentene dihydriodide, C,,,HI8T2, which, for the purpose of comparison, we had also prepared from pure crystallised terpineol, and when the two hydriodides were intimately mixed, the melting point remained unchanged. The formation of t h i s derivative, and of the products of oxidation described in the preceding section, proves conclusively the presence of terpineol in the oil. The optical rotation of the fraction from which it was obtained indicates i t to be the I-form. Identijcation, of Gevanio 1. Fruction 223--3359-This fraction was collected within the above limits, in view of the possible presence of both citronellol and geraniol. It was relatively small in amount, and was analysed, and its physical constants were determined, with the following results : 0.1544 gave 0,4370 CO, and 0,1549 H,O.C = 77.2 ; H = 11.1. 0.1420 ,, 0.3999 CO, ,, 0.1433 H,O. c! = 76.8 ; H= 11.2. C,,,H,,O requires C= 77.9 ; H = 11.7 per cent. I t possessed a camphoraceous and also a fragrant, rose-like odour. Although its high density and rotation indicated that i t contained a considerable amount of terpineol, and its analysis also showed an admixture with some of the next higher fraction, the small amount of liquid precluded its further purification by simple distillation. The odour of this fraction afforded such convincing evidence of the presence of geraniol t h a t Erdmann's method, which depends on the formation of the crystalline geranioldiphenylurethane (m.p. 82O), was resorted to for the identification of the substance (J. pr. Chem., 1897, [ii], 56, 8). The oil was treated with diphenylcarbamic chloride in presence of pyridine, as described by Erdmann ; the syrupy residue left after distilling the product with steam was then purified by extraction with ether, and the ethereal solution emporated after extraztlon with dilute hydrochloric acid. The residual light brown oil was mixed with a little alcohol, when it soon formed a crystalline paste, which was drained on a porous tile. The substance was finally recrystallised d 15*5'/15O = 0.9340 ; uD = - 9'8' in a 100 mm. tube.ESSENTIAL OIL OF ASARUM CANADENSE. 67 from a little alcohol, from which it separated in fine, glistening needles melting sharply at 81-82O. On analysis : 0,1444 gave 0.4163 CO, and 0.1015 H,O.C,,H,70,N requires C = 79.1 ; I€ = 7.7 per cent. The fraction by gentle oxidation with chromic acid afforded a little citral, but although the amount of the latter was too small for con- version into the naphthacinchoninic acid derivative, the evidence was already sufficiently conclusive of the presence of geradol in this fraction of the oil. There was, on the other hand, no indication of the presence of citronellol. It may be noted that in the first investigation of asarum oil by one of us, a fraction was obtained corresponding approximately in boiling point (222-226O) to that just described, and that this, on more energetic oxidation with chromic acid, afforded, besides acetic acid, a small amount of a crystallisable acid.As a specimen of the latter had been preserved, it has been re-examined and shown to be a mixture of terebic and terpenylic acids. Frcbction 235-245O.-This was very small in amount, and was evidently a mixture of the preceding and following fractions ; a little of the crystallised geranioldiphenylurethane mas obtained from it by thc method previously described. C = 78.6 ; H = 7%. Identijcation of Eugenol Methy? Ethes-. flraction 245-260°.-This constitutes the largest fraction of the oil. On redistillation under the ordinary pressure it was easily resolved into a large fraction, which was collected between 250° and 256O, but distilled for the most part between 252O and 254'. It is a colourless, nearly odourless liquid, and was analysed, and its physical constants were determined, with the following results : 0 1648 gave 0.4522 CO, and 0.1238 H,O.C = 74.8 ; H = 8.3. CllH1,O, requires C = 74.2 ; H = 7.9 per cent. d 15'/16"= 1.0239 ; uD= - 2'44' in a 100 mm. tube. I t has been shown by Petersen (Ber., 1888, 21, 1064) that the oil obtained from the allied European species of Asuvum contains a sub- stance of the same composition, boiling at about 250', which on oxidation affords veratric acid, and was Fully identified as eugenol methyl ether. I n the first investigation of the oil of Asarum canadense by one of us, a fraction was collected a t 254-257',which on oxidation with chromic acid afforded a small amount of a crystalline acid, COH1004, and this Petersen has likewise found to be identical with veratric acid, The same specimen of acid, after recrystallisation from water, me now find to soften at 1 7 2 O , and to melt compIet,ely at 177-178".F 268 POWER AND LFES: THE CONSTI'I'UENTS OF THE The confirmation of the identity of this fraction with eugenol methyl ether has now been effected by the preparation of the crystalline bromoeugenol methyl ether dibromide, C,H,Br( OCH,)-C,H,Br,, which melts at 78-79O (Wasserman, Contpt. vend., 1879, 88, 1206). This was accomplished as follows : To the liquid dissolved in dry chloroform, and cooled in a mixture of ice and salt, the requisite quantity of bromine, also dissolved in chloroform, was added, drop by drop, and any slight excess of bromine removed afterwards by shaking the solution with a little sulphurous acid.The chloroform solution was separated, dried, and filtered, and the chloroform removed by rapidly drawing dry air through the solution. The residue was a thick syrup, which, when dissolved in alcohol, deposited a quantity of glistening crystals. These, on recrystallisation from absolute alcohol, separated in glistening, felt-like needles, which melted at 78-79". The optical activity of the fraction is due to admixture with a small amount of a higher fraction, which it is difficult to separate completely by fractional distillation. 8earch for isoEugeno2 Methyl Ethr.-As it has been assumed by Mittmann (A?& Pharm., 1889, 22'7, 543) that the substance con- tained in asarum oil is not eugenol methyl ether but the isomeride, we have thought i t desirable to ascertain the correctness of this opinion. For this purpose, a portion of the original oil which had been deprived of terpene was fractionated under diminished pressure before being subjected to hydrolysis. As eugenol methyl ether boils at 128-130' (10 mm.) and isoeugenol methyl ether at 142' (10 mm.), fractions were first collected at 130-140' and a t 140-155O under a pressure of about 10 mm.Further fractionation mas conducted under 60 mm. pressnre, at which eugenol methyl ether was found to boil a t 16Cj0, and isoeugenol methyl ether a t 179". A large fraction was thus collected at 163-16'7' (60 mm.), and also a fraction at 175-185" (60 mm.). For the differentiation of these two substances recourse was had to bromination, as eugenol methyl ether in the cold yields the bromo bromide, whereas isoeugenol methyl ether under tlie same conditions yields only a dibromide melting a t 99--l0lo (Ber., 1890,23, 11 67).On applying this test to the two fractions, only the cryst illine derivative melting a t 78-79' was obtained, which proves that the original oil does not contain isoeugenol methyl ether. Fraction boiling above 260'. This fraction was distilled under reduced pressure, and after a large number of distillations under 60 mm. pressure the following fractions were obtained : Below 175', 175-2 9 5', 195 -2 1 O', 210 --220°, and 220-2 30'.ESSEKTIAL OIL OF ASARUM CANADENSE. The characters of these fractions are shown in the following table : 69 Boiling point (60 nim.). Below 175" 175-195 195-210 210-220 220-230 Analysis.- C=76*0; H= 9.3 C=78'4 ; H=10.3 C=81*1 ; H=10'5 - Rotation in 100 mm. tube. Solubility in 70 per cent. alcohol. Very freely soluble Very freely soluble Very freely soluble Less freely soluble Turbid Colour. Slight Light yellow Bluish Bluish Greenish The fraction collected below 175" consisted chiefly of eugenol methyl ether. The three subsequent fractions had a n odour resembling that of cedar wood, and when a few drops were dissolved in glacial acetic acid and a drop of concentrated hydrochloric or sulphuric acid added, a n intense reddish-violet colour was produced. The fraction 220-230' was very small in amount, The analysis OF the principal fractions, and particularly their ready solubility in dilute alcohol, proved that they consisted of oxygenated compounds, and did not contain a sesqui- terpene." As the fraction 21C!-220° was the largest, this was again carefully distilled, and the following fairly constant fraction obtained, which was more fully examined.Fq*action 212-217' (60 mm.).-This is a thick, viscid liquid, having R fine blue colour and an odour recalling that of cedar wood. It does not solidify when exposed for some time to a temperature of - IS". It is very freely soluble in '70 per cent, alcohol and affords the same colour reaction as the fraction from which it was obtained. It was analjsed, and its physical constants mere determined, with the followkg results : 0.1069 gave 0.3133 CO, and 0*1009 H,O. A molecular weight determination gave the following result : 0 4184 gram depressed the freezing point of 30.17 grams of phenol by 0.48', whence mol.wt. = 214. This result mould agree very well with a sesquiterpene alcohol of the formula C,,H,,O (mol. wt. = 222), but the analytical figures do not accord with those required for this substance (C = 81.1 ; H = 11.7 per cent.). It is probable, therefore, that the fraction snalyscd still consisted of more than one substance. The statcinent in " The Volatile Oils," by Gildemeister and Hoffniaiin (p. 123) hat " Seiiimler, in 1889, obtained from asartun oil a hydrocarbon, C16H2,, boiling a t about 255" " is an error of translation. It properly refers to the oil of Cccrlij~cz acaulis or Carline thistle (German, Eberwzcrz), which is described on p. 690 of the same work. C = 79.9 ; H = 10 5. d15'/16'=1*0063~ a D = -3'; 2=100mm;C=3'678;[a]~= -Sl.5'.70 POWER AND JJEES: THE CONSTITTJENTS OF THE Treatment with Ph.osp?twic Oxide.-In order to obtain further evidence respecting the character of these bluish fractions, an attempt was made to dehydrate them, The remainder of the fractions 195-210Oand 210-220' (60 mm.), about 5 grams of each, was separately dissolved in dry benzene, phosphoric oxide added, and the liquids boiled for about an hour, when they acquired a deep purple-red colour.After distilling off the benzene, the residues were distilled under diminished pressure, Fraction, 195-210' (60 mm.) afforded a liquid which distilled between 175' and 210' under 60 mm. pressure. It had a bright green colour, a cedar-like odour, and was insoluble in 70 per cent. alcohol.0-106s gave 0.3198 CO, and 0.1000 H,O. C = 81.7; H = 10.4 per cent. Fraction 21 0-220° (60 mm.) afforded a liquid which distilled chiefly It had an olive-green d 15O/15O = 0,975 ; [ a ] D = - 37'. between 200' and 220' under 60 mm. pressure. colour, a cedar-like odour, and mas insoluble in 70 per cent. alcohol. 0.1073 gave 0.3300 CO, and 0.1006 H,O. C = 83.9; H = 10.4 per cent- d 15'/15O =: 0.985 ; [ a ] D = - 35.5'. Both these liquids, when dissolved in glacial acetic acid and treated with a drop of hydrochloric acid, afforded a purplish or red colour. Although the insolubility of these products in alcohol and the increase in the per- centage of carbon by the above treatment was evidence of the forma- tion of a hydrocarbon, the substances themselves were not sufficiently pure to admit of further identification.They mere finally dissolved in dry ether, the solutions saturated with dry hydrogen chloride, and allowed to stand for several days, but from the very dark, oily residues no crystallisable hydrochloride could be obtained. Although several essential oils are known to afford high boiling fractions of a deep blue colour, which have been designated as ccewlein by Gladstone, and as ccxulene by Piesse, no properly charac- terised compound has as yet been isolated from any one oE them. Acids obtained by the Hydrolysis of the Oil. The strongly alkaline, aqueous liquid, separated from the hydrolysed oil and completely extracted with ether, as previously described, was evaporated to a small bulk, then acidified with sulphuric acid and distilled with steam.The first portion of the distillate was slightly turbid, but it soon became clear. The entire acid liquid was then made alkaline with sodium carbonate, and extracted several times with ether. After distilling off the ether there remained a small amount of a dark coloured, highly aromatic oil. This was insolubleESSEKTIAL OIL OF AYARC'M CANADENSE. 71 in a cold solution of sodium hydroxide, but dissolved readily on warming. The alkaline solution of the substance was shaken with ether t o remove any impurity, and then acidified with sulphuric acid, which liberated the original oil. This was again taken up with ether, the ethereal solution washed with a little water, dried, evaporated, and the slightly coloured, oily residue finally placed in a V~CLIOUS desiccator over paraffin to remove the last traces of ether, and then analysed.0,0463 gave 0.1298 CO, and 0 0403 H,O. This substance, to which we assign the provisional formula C,,H,,O,, is evidently a Zactone. Although existing in extremely small amount, so that we have not been able to examine it further, its powerful odour indicates that it must have considerable influence on the odour of the original oil. To it is also possibly due the somewhat clove-like odour which was observed in the isolation of the phenol (p. 60). After the lactone had been separated from the liquid which had been made alkaline with sodium carbonate, this liquid was concentrated, strongly acidified with sulphuric acid, and shaken four times successiveiy with ether.The ethereal solution was washed twice with water, dried, and distilled. On fractionating the residue, nearly all distilled between 110' and 120'. A portion of this was converted into the barium sa,lt, and from the latter the silver salt mas prepared which gave the following figures on analysis : C,H,O,Ag requires Ag = 64.7 per cent. This served to confirm the presence of acetic acid, the previous invest- igation having shown that esters of this acid were present in consider- able amount in the oil. Acids of Higher Boiling Point.-The residue from the distillation of the acetic acid was very small in amount, and was therefore mixed with a larger portion of acids obtained by shaking the original oil with a solution of sodium carbonate, The whole of the acids of higher boiling point, after standing over potash in a vacuous desiccator, was first fractionated under 10 mm., and then under the ordinary pressure, when the following three fractions were obtained.C = 76.1 ; H= 9.7. C,,H,,O, requires C = 76.4 ; H = 9.1 per cent. 0.085 gave on ignition 0.0550 Ag. Ag = 64.7. (1) Below 240'; (2) 240-280'; (3) 280-300'. The last fraction became solid on standing, and from the residue in the flask crystals were separated which, after recrystallisation from hot light petroleum, melted at 57-58O ; these consisted apparently of pal- mitic acid, which had been extracted by means of caustic alkali from the original oil. The first two fractions were redistilled and the follow-72 ESSENTIAT, OIL OF ASARUM CANADENSE. ing fractions collected : A, 140 -200 ; B, 200--230' ; C, 230-270°.D was fraction (3) from the first distillation (b. p. 280--300'). They were yellowish, oily liquids, nearly equal in amount, and were present altogether to the extent of about 2 grams in a kilo. of the original oil. They were first converted into ammonium salts, and then fractionally precipitated by silver nitrate. The various silver salts were spread on porous plates, and then dried at 80' for half-an-hour. On analysis, they gave the following results : A,. B,. Ag= 45.1 ,, 13,. Ag=47*1 ,, C,. Ag=36*7 , C,. Ag=39.5 ,, D,. Ag=34.5 ,, D,. Ag=39*2 ,, Ag = 46.1 per cent. ; A,. Ag = 48.4 per cent. C,H,,O, A g requires Ag = 48.4. CI,H,,O, Ag requires A g = 35- 1 per cent. It is thus seen that these acids of high boiling point constitute a n exceedingly complicated mixture, the amounts of eilver found corre- sponding t o those required for salts of acids ranging from C,H,,O, t o C,,H,,O,. A further separation and identification of them would require a very much larger quantity of material than was available for the purpose, It may also be noted that from the method by which the chief portion of these acids was obtained, it is evident that they exist in the oil in a free state, and not in the form of esters. Xurnmar y. From the results of this investigation, the oil of Asarum canadense 1. A phenol, C,H,,02, 2. 3. d-Linalool, 4. I-Borneol, 5. I-Terpineol, 6. Gersniol, 7. Eugenol methyl ether, 8. 9. A lactone, C14H2002, is seen to contain the followiiig substances : Pinene, apparently a mixture of the d- and I forms, A blue oil, of undetermined composition, consisting of oxygen- ated substances of alcoholic nature, 10. Palmitic acid, 11. Acetic acid, and 12. A mixture of fatty acids intermediate between acetic and palm- itic acids. I n order to ascertain approximately the amounts of the principal constituents, the following determinations were made with the original oil :POWER AND SHEDDEN: DERIVATIVE8 OF GALLIC ACID. 73 1. The eugenol methyl ether was determined by Zeisel's method. 0.1898 gram of oil gave 0.1S46 gram AgI, whence eugenol methyl ether = 36.9 per cent. The amount of esters,calculatedasC,,K,7'C,H,O,, is27.5 per cent. The total amount of alcohols, C,,H,,O, free and as esters, after acetylating the hydrolysed oil, was found to be 34.9 per cent., hence the amount of free alcohols is 13.3 per cent. I n reality, the amount of free alcohols is somewhat larger than this, as i t is known that lin- alool and terpineol cannot be quantitatively acetylated. As about 2 per cent. of pinene was found in the oil, the con- stituents of high boiling point, blue oil, &c , would amount to some- what less than 20 per cent. 2. 3. 4. THE WELLCOME CIIEMICAL RESEARCH LABORATORIES.