110 MR. BENTHAM ON MALVACEE 
its fruit and general habit. It is indeed so closely allied to Berrya 
among Tiliacez as only to be distinguishable from that genus by 
the singular, almost petaloid sessile stigmas, and we have accord- 
ingly now removed it to the latter family. Among the Kew- 
Garden drawings is one of a plant of unknown origin, which is 
evidently the same Carpodiptera, although there are three stigmas, 
instead of two as in the Cuban specimens. 
QUARARIBEA, Aubl. 
This genus is generally referred as a section to JMyrodia, 
which it resembles in its fruit, although the flowers are very 
different. The andrecium, with its one-celled anthers, is truly 
Bombaceous, near that of Matisia ; whilst in Myrodia the anthers, 
in their two parallel or diverging cells, and in their usually defi- 
nite number and arrangement, are decidedly Sterculiaceous, closely 
resembling those of several genera of the Helicterez. 
Subtribe DvR10NEx. 
Of the five genera forming this subtribe three are monotypic, 
and the two others have only two species each. They have, more- 
over, so much general similarity in their habit, in their scaly in- 
dumentum, in their involucre and fruit, that they might have been 
considered as constituting a single genus. Yet there is so much 
diversity in their calyx, in the presence and absence of petals, in 
their style, in the number of ovules, and especially in their andree- 
cium, that we have, for the present, thought it better to preserve 
the five genera as usually adopted. Two of them (Neesia and 
Boschia) have been occasionally placed among Tiliacee, from 
which they are readily known by their anthers. 
The distinction between SrERCULIACEX and BUETTNERIACES, 
taken each in their general sense, although adopted by most bota- 
nists, rests on no one tangible character. In both, the number of 
stamens usually bears some definite ratio to that of the sepals. The 
supposed introrse anthers of Buettneriacee originated in a mis- 
take. The “sterile stamens” of most tribes of Buettneriacex 
are the same thing as the *teeth of the staminal column" in 
Helicteres and its allies, and the degree of connation of the sta- 
mens varies in both supposed orders. In both also we meet with 
great diversity in the dehiscence of the fruit and in the embryo 
and albumen. If, however, we unite the two, rejecting only the 
