172 Proceedings of the Royal Irish Academy. 



We tried to increase the yield of the compound by allowing the 

 condensation of the diacetoresorcinol with the henzaldehyde to continue for 

 a much longer time, and obtained a quite different product. 



A mixture of diacetoresorcinol and benzaldehyde with dilute aqueous 

 sodium hydroxide, to which some alcohol was added, was allowed to stand for 

 three months at the laboratory temperature. 



The yellow solid, which separated, was filtered and washed with water, 

 acid, and alcohol. It was recrystallized a few times from chloroform and 

 alcohoL When dried at 110 = C, it melted at 218 a C, and gave on analysis 

 the following results : — 



01655 substance gave 04744 CO, and 00722 H-O, 

 corresponding to C 73 1, H 485. 

 C^H, A requires C 77-34, H 486. 



- -!>'.'.■: : l>:ne-diacetwrtiorcinol crystallizes in light yellow equilateral 

 triangular plates, with some diamond-shaped crystals formed by the 

 juxtaposition of pairs of triangular plates. The crystals had hexagonal 

 symmetry. The substance is quite different in crystalline form and 

 melting-point from the isomeric £-dibenzyhdene-diacetoresoreinol got by 

 the demethylation of dihydroxy-dichalkone-dimethylether. A mixture of 

 the two compounds softens at about 1S5 = — a temperature much lower than 

 the melting-point of either substance. This modification of dibenzylidene- 

 diacetoresoreinol is nearly insoluble in dilute alkali, alcohol, or ether, readily 

 soluble in warm chloroform, and sparingly soluble in boiling benzene. The 

 crystals turn a transient orange colour on contact with concentrated sulphuric 

 acid, in which they dissolve to a yellow solution, which has a green fluorescence. 

 When the sulphuric acid solution is diluted by addition of water, only a faint 

 white precipitate is obtained. 



Scarcely any coloration is produced by adding ferric chloride to boiling 

 alcohol saturated with the compound. 



The amount of y-dibenzylidene-diacetoresorcinol at our disposal being 

 small, we were unable to examine it so fully as seemed desirable. 



When dissolved in dry chloroform, 7-dibenzylidene-diaeetoresorcinol 

 interacts very slowly with bromine to form a crystalline derivative, which, 

 however, on warming with alcoholic potash, does not form diflavone. 



9 :e it seemed probable, therefore, that the substance was not diflavanone, 

 we heated 05 gramme of it with 05 gramme of anhydrous sodium acetate and 

 ."> ccs. of acetic anhydride to gentle boding for a few minutes. The mixture 

 was cooled, water was added, and the oily layer was extracted with chloro- 

 form. The chloroform layer was washed with ddute sodium bicarbonate, and 



