Page:Miscellaneousbot02brow.djvu/636

 620 PLANTS JAVANICE RARIORKS.

the direction of the embryo dependent on that of the seed, even in the carpella of all the species then published.

In 1831, in the ' Flore de Senegainbie/ the joint work of MM. Guillemin, Perrottet and Richard, a new point is introduced into the character of Sterculia} namely an incomplete arillus, which, however, if it really exists in any case, is probably to be found in one species only, namely, Sterculia cordifolia : in all the other species which I have examined, there is either only a minute carimcula umbi- lical is or strqphiola, as in St.fcetida, or more generally no trace whatever of this appendage.

In 1832, in the ' Meletemata Botanica ' of Schott and Endlicher, the Natural Order Sterculiacece is divided into three principal tribes — Pombacea, llelicterece, and Ster- culiece. This last tribe, as in DeCandolle's ' Prodromus/ is limited to the genera Sterculia and Heritiera. But Sterculia is divided into twelve genera, chiefly from modi- fications of the flower, or from the texture and period of dehiscence of the folliculi, and in one case from the seed being winged ; no modification of internal structure of seed being introduced into any of the characters.

In the same year, the third volume of Dr. Roxburgh's work such a generic character of Sterculia" is given as to comprehend all the Indian species, and indeed so con- sul structed as to include all that are now known, except Courtenia, a new genus of the present essay; and even that would be excluded only from its generally having double the usual number of ovaria. Several new species are well described in the work, and the direction of embryo noticed in most of them ; the only species in which the radicle is described as pointing to the umbilicus being his Sterculia alata.
 * Flora Indica ' was printed at Calcutta. In this valuable

In 1840 Professor Endlicher, in the 13th part of his ' Genera Plantarum,' modifies the arrangement of the Lin- nean genus Sterculia given in the ' Meletemata,' all the genera there established, except his Pterygoid (the St. dldtci

1 Vol. i. p. 79. 2 p. 141.

�� �