SOUTH AMERICAN PLANTS. 79 



4 



the two posterior lobes more erect ; it is nearly smooth outside, 

 and very pilose within. Another characteristic feature is, that 

 the upper moiety of the ovarium, and the lower portion of the 

 style, are densely covered with white hairs, the basal gland being 

 smooth ; it has constantly five stamens, of which the three anterior 

 are somewhat shorter. The hairs of the corolla and pistillum 



X J 



as in the rest of the plant*. 



calyx 



Habrothamnus. 



I take this opportunity of cancelling the suggestion made on a 

 former occasion, in regard to the validity of this genus (Lond. 

 Jo. Bot. V. 151 j and 111. So. Am. PL i. 75). From an examina- 

 tion of dried specimens, I could detect no difference in its floral 

 structure from that of Oestrum^ upon which a generic distinction 

 could be drawn, and there seemed no other alternative, but to 

 unite the whole group, as a separate section of Cestrum. I have, 

 however, lately had an opportunity of examining a plant of this 

 genus in a living state, and can here detect some slight dif- 

 ferences, which are not distinguishable in dried specimens. In 

 CestruiUj the aestivation of the corolla is induplicato-valvate, the 

 edges of each lobe being partly turned in upon both margins, 

 and closely applied and adherent to those of the contiguous lobes 

 (see Lond. Journ. Bot. vii. 58; and 111. South Amer. Plants, i. 

 126) : but in Habrothamnus ^ each lobe has its margins completely 

 turned in, so that they adhere, in a somewhat conduplicate form, 

 firmly to one another, and are only connected with those of the 

 adjoining lobes by apposition, not by adhesion; although the 

 margins of the several lobes thus all converge towards the axis, 

 each lobe is respectively free, and not valvately or induplicately 

 connected with the adjoining lobes, as in Cestrum-, this peculiar 

 mode of aestivation, which is only a modification of the plicative or 

 valvate, so peculiar a feature among the Solanacea, I propose to 

 distinguish by the name of implicative; it is somewhat analogous to 

 the volutive form oi Anthocercis, a figure of which is shown in ' 111. 

 South Amer. Plants, i. 170,' but there the margins are respectively 

 imbricated or overlapped, which is one of the principal distin- 

 guishing features of the Atropacece in the Solanal alliance. 

 Another difference is observable in the structure of the stigma, 

 which is not exactly that of Oestrum. In Habrothamnus^ the 

 style is a little thickened at its summit, and slightly infundibidi- 

 form, being terminated by a thin and distinct, almost entire 

 margin, shghtly bilobed ; this orifice is closed by a large, sphe- 



A draiving of this plant with sectional details is shown in plate 58- 



