2 
Kew Herbarium. D. os ssleleaiolid extends the range and brings the 
number of known cere up tos 
I have not able to examine specimens of D, latifoliwm, A. 
eight, more commonly about three, and rarely absent. This last-named 
condition occurs in thetwo specimens of D. longissimum, Seem., at Kew, 
and it is possible that the sexes may be more widely separated here. 
In the male flower the inflated tube points to the presence of perfect 
anthers, the style and stigma are small, and the ovary, if present, is not 
to be distinguished from the pedicel. In the female flower r, which opens 
D. acwminatum before the males, the straight stoutish uniform tube 
aba the eye, the style and stigma are large, and the ovary is 
obviou 
It is no surprise that this genus should prove to be diclinous ; for 
as the corolla-lobes— 5 in the male, in the Panett flower. Here, 
n some genera of Rubiacee, e.g. Canthiu um, Burck remarks that a 
s character “flores . 4-meri” for D, latifolium, the species 
which I have not seen, really fdtrannen that in his specimen the corolla 
of the female flower = its en and that those remaining were 4-merous 
male flowers.—I. H. Burx11t. 
Fig. 1,ovary, style, and stigma of male flower ; 2, ied of corolla of male fluwer laid 
open ; 38, ovary, style, and stigma of female flower ; 4, part of corolla of female flower, 
laid open; 5, seed (immature). All enlarged. 
