618 PLANTJE .TAVANIC7E RARI0RES.
the character of the genus. Besides these improvements in the account of the genus, he has given the distinguishing characters of several new species.
In 1805 M. Palisot de Beauvois, in his Tlore d'Oware et de Benin/ 1 in describing his Sterculia acuminata, in- troduces several alterations into his character of the genus, most of which belong only to the species he has there described, and some of which may be considered of generic importance, especially the remarkable structure of a nth eras.
In 1806 Salisbury proposes a new genus, Southwellia, formed of all such species of Sterculia as have the seg- ments of the calyx connivent with cohering apices; this being the only character distinguishing it from the rest of the genus.
In the same year Poiret, in the article Sterculia of the Botanical Dictionary of the 'Encyclopedic Methodique/ 2 223] considers the flowers as hermaphrodite, and adopts the explanation given by Forskael, Cavanilles and de Jussieu. In his generic character there is no improvement; but he suggests the probable expediency of afterwards dividing the genus, when the structure of the various plants referred to it is better known.
In 1819 Sir James (then Dr.) Smith, in Rees's 'Cyclo- pedia/ in his account of Sterculia, takes no notice of the structure of seed, and is inclined to agree with Schreber and Willclenow in referring it to Dodecandria. If the genus should be hereafter subdivided, he seems more dis- posed to trust to differences in the styles and stigmata than to those very remarkable modifications of fruit, with some of which he was acquainted, believing them not to be sup- ported by other characters, and in proof of this instancing St. platanifolia and colorata, so similar in fruit, and so unlike each other in the form of the flower.
In 1824 DeCandolle, in the first volume of his 'Prodro- mus/ forms a tribe which he calls Sterculiea, consisting of Sterculia and Heritiera, referring it to Buttneriacea. The
1 Vol, i. p. 40. 2 Vol. vii. p. 428.
�� �
