1328 NIEEENSTEIN THE COLOURINQ MATTER OF CXXII1.-The Colouring Matter of the Red Pea GaZl. By MAXIMILIAN NIERENSTEIN. THE colours of oak galls are very varied and rich. They range from white and cream through all tints of yellow to deep orange, from pale green t o a rich dark hue and through almost every shade of red some being very beautiful and attractive. These red colours are generally ascribed to the presence of anthocyanins, which are supposed t o be derived from the tannins present in galls (compare Gertz “ Studien ofver Anthocyans,” 1906 ; Connold, “ British Oak Galls,” 1908 ; Eiister “ Die Gallen der Pflanzen,” 1911 ; Magnus “ Die Entstehung der Pflanzengallen,” 1914). Our knowledge of the anthocyanins has been fundamentally increase THE RED PEA GALL. 1329 by the recent investigations of Wheldale Willstiitter Everest and others (compare Petkin and Everest ‘‘ The Natural Organic Colouring Matters,” 1918).Their researches have conclusively proved that the anthocyanins are derived from the different flavones present in plants. This suggested an inquiry into the colouring matter of the so-called anthocyanin of the “red pea gall,” fre-quently found on the leaves of different British oak trees especially Quercus pedunculata when galled by Dryophanta diuisa Adler. It seemed reasonable to expect that the anthocyanin of this gall would in all probability be derived from cyanidin the anthocyanin of quercitin and if so it might furnish some evidence regarding the much discussed question as to the relationship between the pathological products produced by the gall and those normally present in the plant (compare Dekker ‘ I Die Gerbstoffe,” 1913).It was incidentally also thought possible that an anthocyanin derived from a gall might prove t o be closely allied to quercetone or isoquercetone both anthocyanin-like oxidation products of quercetin described by Nierenstein and Wheldale (Ber. 1911 44, 3487) and Nierenstein (T. 1915 107 869; 1917 111 4) as it was probable that the accelerated oxidative processes common to larva and imagines (compare Krogh “ The Respiratory Exchanges of Animals and Man,” 1916) which in addition to numerou~ inquilines are present in large numbers in galls (compare Connold, Zoc. cit.; Kuster Zoc. cit.) would favour the production of an oxidation product such as quercetone and not that of a reduction product such as cyanidin (compare Everest Proc.Roy. SOC. 1914, [B] 87 444). The investigation of the red colouring matter derived from the “red pea gall ” has however t o some extent proved disappointing. It was found that dryophantin the name suggested for this pig-ment was in no way allied either t o the flavones or to the anthocyanins but that it consisted of purpurogallin and two mole-cules of dextrose. On the other hand it must be mentioned that purpurogallin has not previously been found in nature. Dryo-phantin is derived from pyrogallol like gallotannin and is there-fore of pathological origin like the latter. Dryophantin how-ever cannot be regarded as an anthocyanin and probably the same can be said of the other so-called anthocyanins derived from plant galls.It is therefore proposed t o classify these red pigments in a new group of natural organic colouring matters to which the name gallorubrones is assigned 1330 NIERENSTEIN THE COLOURJNCI MATTER OF Preparation of Dryophantin. The galls used in this investigation were collected in the vicinity of Bristol and East London during the months of August and September 1913 1915 1917 and 1918 and care was taken to avoid admixtures with the different galls of the Neuroterus species frequently met with on the same leaves as the galls of Dryophanta diuksa. I n all 94 grams of the galls were collected and the dried material was powdered and extracted in a Soxhlet apparatus a t first with ether and subsequently with chloroform so as to remove W ~ X chlorophyll and the so-called gall-fats.The carefully dried powder was then again extracted in a Soxhlet apparatus with alcohol which dissolved both the colouring matter and the tannins. The alcohol was distilled off in a vacuum and the viscid residue redissolved in water. The cold aqueous solution made up to 150 c.c. was shaken with 5 grams of fat-free caseinogen to remove the tannins (compare Korner and Nierenstein Chem. Z e i t . 1911, 36 31) filtered and extracted with ether. The ether left on evaporation only traces of a tarry substance apparently a by-product. The aqueous solution was evaporated under diminished pressure a t about 5 5 O (water-bath temperature) and the residue dissolved in boiling alcohol and filtered.The red alcoholic extract, after being evaporated to a small bulk was poured into water the mixture extracted several times with ether and the small quanti-ties of alcohol and ether present were removed from the aqueous liquid by prolonged heating on a boiling-water bath. The solu-tion on cooling became semi-solid owing to the separation of crystals ; these were collected and washed repeatedly with ether and dilute alcohol. The deep red product obtained in this way was purified by several crystallisations from dilute and finally absolute alcohol. The air-dry substance was dried a t 130° for analysis without apparent loss of weight. The total amount of dryophantin thus obtained corresponded with about 4 grams and there was no apparent difference if fresh or old material (about six months old) was used which showed that there was apparently no deterioration on keeping.Found C=50*2 50.4; H=5*5 5.4. Dryophcamtin was obtained in deep red glistening needles with a bronzy lustre. It was almost insoluble in cold alcohol sparingly soluble in hot water but fairly readily so in boiling methyl and ethyl alcohol and in larger quantities of boiling acetone. It sinbred a t 216O and melted a t 219-220° to a viscous liquid. The C2sH,0, requires C = 50.6 ; H = 5.2 per cent THE RED PEA GALL. 1331 addition of ferric chloride to its alcoholic solution produced a brick-red precipitate and a similar precipitate but slightly darker in colour was also obtained by the addition of lead acetate. A trace of ammonia turned the alcoholic solution deep blue which became red on acidification.These colour changes could be pro-duced in an unlimited number of times in the same solution with-out affecting its sensitiveness to these reagents. Similar blue solu-tions were also obtained by the addition of sodium potassium or barium hydroxides to alcoholic solutions of dryophantin. In this connexion it must be mentioned that similar colour changes are also given by purpurogallin itself (compare Wichelhaus Ber., 1872 5 848; Struve Annalen 1872 163 164; Hooker Ber., 1887 20 3259). On repeating these observations it was found, however that the colour changes are not so permanent in the case of purpurogallin as in the case of dryophantin. Hydrolysis of Dryophan t i i z . Experiments having shown that dryophantin was a glucoside, its decomposition with acid was studied in the following manner.0.5246 Gram dissolved in 550 C.C. of boiling water was digested with 5 C.C. of sulphuric acid for two hours. A deep red crystal-line product commenced to separate and more of it was deposited on cooling. This was collected in a Gooch crucible washed with cold water so as to remove all traces of sulphuric acid and dried a t 160O. I n this way 0.1928 gram of purpurogallin was obtained. Found Purpurogallin = 36.7. C,H,,O, requires purpurogallin = 40.4 per cent. The low value obtained for purpurogallin is due to its sparing solubility in water and i t was found that the filtrate recovered on hydrolysis of dryophantin to which had been added the washings of purpurogallin contained 2.2 per cent.of purpurogallin when determined colorimetrically by Willstiitter and Stoll’s method for the estimation of small amounts of purpurogallin (Annulem 1918, 416 46). The total amount of purpurogallin from dryophantin corresponded therefore with 38.9 per cent. which is 1.5 per cent. below the theoretical if the hydrolysis of dryophantin is expressed as : C,,H,801 + 2H,O = C,,H,O + 2C6Hlz06. A second experiment gave 37.1 per cent. of purpurogallin grnvimetrically and 1 - 9 per cent. colorimetrically corresponding with 39.0 per cent. of purpurogallin which is 1.4 per cent. below the theoretical 1332 MCBAIN AND KAM: The purpurogallin recovered from dryophantin was recrystallised from glacial acetic acid and had the correct melting point of 274-275O generally given for purpurogallin (Found C = 59.8 ; H=3*7.Calc. C=60.0; R=3*6 per cent.). The acetyl deriv-ative which had been prepared by digesting with acetic anhydride, cryatalhed from alcohol in orange-yellow ’ needles melting at 179-180° and the melting point was not depressed after mixing with the tetra-acetyl derivative of purpurogallin (Found : C=59*2; H=4.7. The filtrate from the first hydrolysis was quantitatively tested for dextrose by Fischer and Freudenberg’s method (Ber. 1912, 46 915) and the dextrose estimated volumetrically in several portions of the hydrolysate by Bertrand’s method (Bull. SOC. chim., 1906 [iii] 35 1286) as used by Geake and Nierenstein (Ber., 1914 47 893) for the estimation of dextrose in gallotannin. Calc. C=58.7; H=4*1 per cent.). Found Dextrose=62.8 63.1 63.0. C,,H,80, requires dextrose = 63.3 per cent. The filtrate of the second hydrolysis was prepared as in the experiment for the quantitative estimation of dextrose and then concentrahd to a small bulk. It was subsequently converted into dextrosazone which crystallised from dilute alcohol in glistening, yellow needles melting a t 203-204O (Found N = 15.8. Calc. : N=15*6 per cent.). The author begs to acknowledge his indebtedness to the Govern-ment Grant Committee of the Royal Society for a grant from which much of the cost of the investigation was defrayed. BI o- CHEMICAL LABORATORY, CHEMIOAL DEPARTMENT, UNIVERSITY OF BRISTOL. [Received October 6th 1919.