Annotinine, Cl61121031\T, reacted with phenyl lithium to yield diphenylallnoti~line, C28H3J03N, containing two hydrosyl functions. Oxidation of diphenylannotinine by the Oppenauer procedure yielded a monohydroxyketone, C28Ha~03X, while oxidation with chrolnic acid yielded a dihydroxyketone, CPSH~IO.~N. A rnechanis~n for the formation of these cornpounds is discussed in relation to the chemistry of annotinine.Annotinine, C I~H~~O~N , is known to contain an epoxide function, a lactone ring, and a tertiary nitrogen atom (I). The relationship oE these groups to one another has been demonstrated by Wiesner et a/. (2) and by MacLean and Prime (3). This relationship is represented in the partial formula below.The structure of annotinine is now known (4, 5 ) but for the purposes of this paper the partial structure above suffices.Some years ago we found that annotinine reacted readily with phenyl lithium to form diphenylannotinine,, C28H330aN (I), which was isolated a t that time in an amorphous state. Diphenylannotinine did not dehydrate readily as would be expected of a diphenplcarbill01 nor did it show any reactions characteristic of a compound containing an epoxide function. Our formula also differed from that of WIeier and Marion (6), who proposed the formula CZ8Ha5O4N for the product of reaction of annotinine with phenpl lithium. Because of the anomalous behavior of diphe~lylan~loti~li~le and because of the difference between our results and those of NIeier and h/irarion, we have investigated the reaction more thoroughly.We have now obtained dipheng7Iannotini1le in crystalline form. This compound has a pronounced tendency to solvate, which may explain the formula of NIeier and Marion and the difficulty in obtaining good analytical results for the base. Drastic oxidation of diphe~lyla~lnoti~li~le yielded traces of benzophenone, which indicated that both phenpl residues were attached to the same carbon atom. Oxidation by the Oppenauer procedure using potassium tertiary butoxide or alumi~zum isopropoxide yielded the monohydroxyketone, C28H3103N (11). Both I and I1 were oxidized by chromic acid in acetic acid to the dihydroxyketone, C28H3104N (111), which is apparently identical with the product iso-'Manuscript received i n original form Septenzber 12, 1857, and, as revised, Aprzl 24, 1958. Conlribzltion front the Burke Chentical Laboratories, Hanzzlton College, 11cc1Waster University, Hamilton, Ontario, and the Research Laborator~es, Dominion Rubber Co. Lld., Guelph, Ontario. Part of this work was take?tfrom a thesis subtnttted by G.