AND STERCULIACES. 107 
it as distinct on numerous grounds, and pointed out the truly 
Malvaceous character of its anthers. As the structure of the 
staminal column is also Malvaceous (except that the stamens appear 
to be limited to ten), as the pollen is remarkably muricate, and 
as the shape of the corolla with the dark spot at the base of the 
petals is so much like that of Hibiscus, there appears no reason 
against removing it to the tribe Hibisceew of Malvaceæ, of which 
it has all the technical characters. The original species of Kydia 
must also be removed to Malvaces, as having truly one-celled an- 
thers; but their shape, as well as the general structure of the sta- 
minal column, places the genus in Abutilex rather than in Hibiscee. 
Hisiscus, Linn. 
This, the largest genus among Malvacew, comprising about 150 
known species, varies more than any other in the calyx and bracts, 
in the woolly or glabrous seeds, &c. ; but the characters appear to 
us to be too much blended together, or to pass too much one into 
the other in many instances, to be considered as more than sec- 
 tionl We would therefore restore to Hibiscus the proposed 
genera Bombycodendron, Zoll. (sect. Bombycella, DC.), Lagunea, 
Cav., Paritiwn, A. de St. Hil, and even Abelmoschus, Medik. 
On the other hand, Thespesia, Corr., appears to be sufficiently di- 
stinct in the calyx, in the clavate style, and in the hard, almost 
woody fruit, although not always indehiscent even in T. populnea, 
as well as in the apparently constant character of the obovoid, 
not reniform, seeds: the genus should, however, include the H. 
Lampas and its allies, forming Garcke's subsection Tiparium of 
DeCandolle’s section Azanza. 
Tribe BoMBACER. 
T have already given the principal reasons for which I should 
consider the Bombacew as a tribe or suborder rather of Malvacew 
than of Sterculiacez, and have observed that it is chiefly with the 
arborescent Hibiscew that they stand in close connexion. Ham- 
pea, indeed, and some allied genera are scarcely separated from 
them, except by the filaments all terminating the staminal column 
without any barren truncate or 5-toothed edge; and the latter 
character is not even quite constant in Hibisces, for in some 
species of Lagunaria and Gossypium the column is divided to the 
summit into antheriferous filaments. Some genera of Bombacex 
present indeed exceptional characters, never or seldom observed 
