TRANSACTIONS OF SECTION B, 703- 



production of the corresponding' carboxylic acid in small quantity. The substitut- ■ 

 ing (groups appear to occupy the 1, 2 positions in the ring. 



_ The monohydroxy furfural is identified, hy its very characteristic relations 

 with phlorogluciiiol and resorcinol, as a constituent of the liguocelluloses. 



Caro's reagent, prepared from potassium persulphate and sulphuric acid, 

 according to the directions of Baeyer and Yilliger,' reacts under similar conditions 

 in a different manner. In the first place the reaction appears to be quantitative, and 

 when the furfural has taken up 0., a trace of either reagent in excess persists. 

 The temperature remaining at 15'^-20°, and there being no evolution of gas, we may 

 expect to find a product of empirical formula C-H^,, and as the aldehydic group 

 disappears this should be a hydropyromucic acid. From tlie isolation and 

 analysis of a crystalline methylphenyl-hydrazide we confirm the product as a 

 monocarboxylic acid. On reduction with sodium amalgam an aldehydic product 

 is obtained with the brilliant colour reactions of the monobydroxyfurfurals. 

 Control observations on pyromucic acid proved that this acid is reduced under 

 similar conditions to furfural. 



The reactions of this hydroxyfurfural, though similar to those described in our 

 previous paper, are sufficiently diiierentiated to indicate that we have obtained a 

 second isomeride. Moreover, the corresponding acids are differentiated, the one 

 giving Pb and Ba, salts, insoluble in acetic acid ; the salts of the new acid are 

 soluble, and, moreover, the acid undergoes hydrolysis with such ease as to make its 

 isolation in the pure state a matter of great difficulty. In the course of the usual 

 processes for isolating the Ba and Ca salts, decomposition occurs, and the crystalline 

 salts isolated are those of a dibasic acid. On boiling the original product in 

 solution at constant volume, formic acid distils continuously and with traces only 

 of ether acids. The yield of formic acid amounted in one experiment to 07 grm. 

 per 1 grm. of original furfural. All these observations indicate that the oriinnal 

 product of oxidation is the acid C'^H^O . (COOH) (OH) 1-4 .... A characteristic 

 reaction of this original acid is the production of a yellowish-red precipitate with 

 ferric chloride, similar to that obtained with pyromucic acid. 



Jt is to be noted that the Garo reagent oxidises the constituent of the 

 lig^iocelluloses, which gives the brilliant colour reactions with phenols character- 

 istic of these natural products, which we know to be a hydroxyfurfural. In this 

 respect the reagent differs from the oxidants ordinarily used for bleaching purposes, 

 e.^., hypochlorites and permanganates. 



A quantitative experiment with a typical aldose (dextrose) and the Caro reagent 

 gave a somewhat unexpected, entirely negative result. The cupric reduction 

 (Fehling solution) was unaffected. 



The typical ketose laovulose, on the other hand, is slowly oxidised. 



All of these matters are under investigation. 



13. On the Synthesis of Benzo-y-pyronc. 

 By Dr. S. Ruhemann and H. E. Stapleton, B.A. (fixon.) 



The important group of yellow vegetable dyesi the chief of which are chrysin, 

 fisetin, and morin, are derived primarily from a phenyl benzo-y-pyrone— 



" 0^ — . Ph 



/ II 



C,H, CH 



This was synthetised in 1898 by Kostanecki, but the mother substance itself, 

 benzo-y-pyrone, had not, up to the middle of 1900, been isolated. The authors 

 succeeded in preparing this compound from pheuoxy-fumaric acid— 



COOH . n = CH . COOH 



I - - 



O.C.H, 

 « Ber., 1899. 



