210 Proceedings of the Royal Irish Academy. 



On repeating the experiments of Harries and Miiller, we were unable to 

 obtain the compound melting at 68 - 5° C, and obtained instead a substance 

 with the same empirical formula, Ci 8 H 1E 0-, but which melted at Sl c -83° 0. 



The compound C^HisOi may exist in the two structurally isomeric forms 

 II and II T. It is possible that the compound prepared by Harries and Miiller 

 may have the formula III, and that isolated by us the formula II, but since 

 the oximes of the two substances have the same melting-point, 195 r -196°C, 

 it is also possible that our substance was a purer preparation of the substance 

 previously described by Harries and Miiller. 



We should expect that the oxime of a compound which has the formula II 

 would readily change into an isoxazol 



C 6 H 5 < 'H : CH . C CH . CH 3 C 6 H,CH : CH . C CH . CH, 



I! \ II I 



NOH HO.CH.C 6 H 5 N— 0— CH.C 6 H S 



but the analysis of the hydroxylamine derivative of the compound melting 

 at 81°-83° C. showed that it was not an isoxazol, and for this reason we were 

 at first inclined to regard the substance as a tetrahydropyrone derivative. 



However, as the compound gives an orange colour with concentrated 

 sulphuric acid, it seems more likely to be an unsaturated ketone than a 

 tetrahydropyrone, and this view is to some extent confirmed by the behaviour 

 of the substance on further condensation with benzaldehyde. 



The compound C^HieOj, melting at 81°-83° C, condensed readily with 

 benzaldehyde in the presence of alcoholic hydrochloric acid to form 

 not benzylidene-diphenyl-methyl-tetrahydropyrone, GsHmOj, which was 

 expected, but another well crystallized compound melting at 156° C, which 

 has the empirical formula CVHjoO. 



The latter compound was also obtained by condensing benzaldehyde with 

 a-benzylidene-methylethylketone in the presence of alcoholic hydrochloric 

 acid. 



Using the same condensing agent d-benzylidene-methylethylketone 

 interacted with anisaldehyde and with piperonal to form analogously con- 

 stituted compounds having the formulae C'kH^Oj and (J 27 H M O s respectively. 



If the statement of Harries and Miiller, that a-benzylidene-methyl- 

 ethylketone is formed from benzaldehyde and methylethylketone in the 

 presence of alkali, and is not formed from the same components in the 

 presence of acids, were rigidly true, we should expect that the compound 

 '-'liHjoO could not be formed directly from methylethylketone and benzalde- 

 hyde in the presence of alcoholic hydrochloric acid. 



Direct experiment, however, showed that the contrary is true — the 



