CONTEIBUTIONS TO BOTANY. 
99 
the peduncle and its widely spreading branchlets being long, 
slender, and nearly glabrous i the persistent calyx is smooth, 
with five small teeth ; the petals are linear and thin in texture ; 
these in the greater number of instances, together with the sta- 
mens, are wanting, having fallen away, as almost universally oc- 
curs in the female flowers of Stemonurus found in herbaria •, and 
it is probably owing to this circumstance, that Prof. Bliune, in 
his generic character of Platea, states that the female flowers are 
deficient of corolla and stamens. The stamens are the length 
of the petals, the filaments being quite free, very compressed 
and broad at the base, tapering above, thin, and almost mem- 
branaceous in texture, somewhat indexed at their summit, where 
they are terete and affixed near the dorsal sinus of the anthers, 
which are oblong, 2-lobed, bifid and sagittate at base, and emar- 
ginated at the apex ; the lobes are membranaceous, opened by a 
longitudinal fissure, the cells being quite void. The ovarium is 
cylindrical, as long as the stamens, and crowned with a sessile 
5-lobed pulvinated disk, which is slightly umbilicated in the 
centre, where a short prominence is seen, this being the withered 
style and stigma : its single cell contains two large suspended 
o\^es. It is worthy of remark, that in all the flowers retaining 
the corolla, I could find no instance in which the petals presented 
any appearance of opening, so that it is very probable that these, 
together with the stamens, in falling away retain the cylindrico- 
cupular form they present in the bud. 
Sarcostigma. 
The following observations on the structure and affinities of 
Sarcostigma were completed in readiness for the press, when the 
last part of the ‘ Plantse Javanicse Rariores ’ made its appearance : 
in that important work we are favoured with an interesting ac- 
count and an excellent figure of a new species of this genus 
from Java. The remarks there offered, in regard to the affinities 
of Sarcostigma, will be seen to be greatly at variance with my 
own deductions ; and hence it becomes necessai’y that I should 
offer a few explanatory words on the subject. It would be pre- 
sumptuous in me to attempt to contravene the inferences there 
deduced by the most profound botanist of our time, showing the 
relation which that genus bears to Phytocrene, Nansiatum, and 
lodes ; but fully acknowledging all that is there affirmed, I may 
ventm’e to show, that a yet stronger and much closer extent 
of analogy will be found to exist in the structure and deve- 
lopment of the floral parts, as well as a greater approximation 
in habit, to what we find in Stemonurus and Pennantia. From 
the facts shown below, it will be seen that Sarcostigma accords 
o 2 
