CONTRIBUTIONS TO BOTANY. 
3 
pedicels are sometimes branched, when the inflorescence becomes 
somewhat paniculate ; at other times the flowers are condensed 
into globular heads upon the peduncle ; sometimes the axillary 
flowers appear in fascicles of pedicillated single flowers, or are 
simply umbellate, or in umbels compounded to the second or 
third degree. I have frequently observed the racemes growing 
abundantly on the stems devoid of leaves. The flowers are ge- 
nerally furnished with bracts; they are extremely minute, and, 
though often hairy, are sometimes destitute of pubescence : they 
are, with very rare exceptions, universally unisexual and dioe- 
cious. They are said to be sometimes monoecious ; but this ap- 
pears doubtful. In the two instances recorded by DeCandolle, 
I found, by an examination of the original specimens, that they 
were decidedly dioecious. St. Hilaire records the existence of a 
monoecious species of Cissampelos ( C. monoica) : this has not been 
confirmed by any other observer, and is the only instance on 
record. I have, however, seen two cases where the flowers are 
distinctly hermaphrodite, or, rather, polygamous. I have ob- 
served, in Anomospermum, a solitary ovary in the male flowers 
in a few instances ; and I found it a general feature in a speci- 
men of Tiliacora from the island of Ceylon. 
The arrangement of the floral envelopes (sepals) is usually in 
several ternary imbricated series, gradually decreasing outwards, 
the two internal whorls being in most instances considerably 
larger than the others; and they probably constitute the true 
normal number of six sepals ; and all the outer ones, frequently 
very minute in size, may be considered as bracts. These six 
sepals, though in aestivation generally in two imbricate series, 
are fixed in a nearly circular whorl around a small central torus; 
but sometimes as many as five ternary whorls are seen arranged, 
one above another, upon a cylindrical gynaecium, as in the Mag- 
noliacece. The number in each series is generally three, though 
sometimes four, five, or six occur ; in Anamirta and Quinium we 
have a pentamerous arrangement ; in Antitaxis the floral parts 
are disposed in opposite pairs, while in Antizoma we have the 
remarkable instance of two opposite sepals hooding two petals 
plaeed before them : rarely, as in Rhaptomeris, owing to the 
confluence of the margins of its six sepals, the calyx is gamo- 
phyllous, being quite tubular and campanulate. In Synclisia, 
according to Mr. Bentham, the sepals are somewhat united at 
base into a very short tube; while in Stephania and Cyclea, 
although the sepals remain distinct, they assume, by their erect 
position and approximated margins, the semblance of a tube. 
The aestivation of the sepals, although in most cases imbricate 
is sometimes valvate, as in the cases last mentioned : this occur? 
in Tiliacora, Abuta, and Limacia. The symmetry in the arrange- 
• B 2 
