ME. D. OLIVEIt ON AUBANTIACE.S. 9 



above, while in three at least of the four remaining species the 

 petals are greatly imbricate. This may be an exceptional state of 

 the one gathering -which I have seen, which affords very few buds 

 to examine ; if constant, it could not be regarded as of higher than 

 specific importance, so closely is this species allied to the rest of 

 the genus. For the present 1 regard this plant as a variety of 

 P. citrifolia. The stamens when definite are clearly normally 

 symmetrical with the calyx and corolla, almost invariably, exclusive 

 of Citrus and ^Egle, being the double number of the petals*. 



In Micromelum, Glycosmis, Clausena, Murraya, the filaments are 

 always alternately shorter, the longer filaments alternating with 

 the petals. In Atalantia this alternation in length is more or less 

 apparent, while in the remaining genera the stamens are nearly or 

 quite equal at flowering. 



The ovarium, in some genera sessile, resting in or surrounded 

 at the base by a fleshy, free, or accrete, often cupulate disk, in 

 others is very shortly but distinctly stipitate," especially in Para- 

 mignya and Clausena ; in some species of which latter genus the 

 short gynophore is considerably narrower than the ovary; yet the 

 base of the latter is apparently encircled by a connate " disk," 

 which in species with a pilose ovary is conspicuous, owing to its 

 usually remaining nearly smooth. The contorted dissepiments of 

 the ovary in one or two species of Micromelum, after or indeed at 

 the time of flowering, form a curious feature to which I cannot 

 recollect a precise parallel f (vide infra, p. 19). The style varies 

 in length, but we do not find the stigma truly sessile in any 

 ■Aurantiaccce, unless we include the American genus Casimaroa. 

 Often considerably elongated and conspicuously narrower than the 

 ovary, which it may exceed in length, as in Murraya, Citrus, Para- 

 mignya, in other genera it is much shorter; in Clausena either 

 long or short, while in Glycosmis we find the extreme in its reduc- 

 tion ; the entire pistil being in one or two species shortly conical 

 or almost cylindrical. The passage of the ovary into the style is 

 in certain genera very gradual, in others very abrupt, the base of 

 the style being also sometimes slightly constricted. There may be 

 m a few species a true articulation at their union. In nearly, if 

 not quite, all of the species, with the exception of Glycosmis, the 

 style after flowering separates at or towards the base, either irre- 



■Limonia oligandra, Dalz., ' Journ. Bot. and Kew Misc.' ii. 258, is a species 

 of Toddalia. 



T I think in a young berry of Clausena Wallichii I have observed a similar 

 twisting. 



