332 MR. J. MIERS ON THE HIPPOCRATEACEX OF SOUTH AMERICA. 
which period little was known concerning it. Linnæus, in 1753 *, and Jacquin, in 1760 +, 
established 2 species; and Decandolle, in 1824 $, enumerated 23 species, of which 15 
were of American origin; but some of these do not belong to the genus. Since then 
several others have been described. "They are all suffruticose plants, with a climbing 
habit, mostly scandent, with opposite branches, the younger ones generally covered with 
ferruginous tomentum. Their leaves, also opposite, are usually glabrous, either entire, 
or subserrated on the margins, and upon rather short petioles. The inflorescence is 
axillary or terminal, in corymbose panicles, generally dichotomously divided, with a very 
short branch in each dichotomy, more or less tomentous, each branchlet bracteolated at 
its base, and bearing numerous shortly pedicellated flowers, generally of small size, and 
subagglomerated at the ultimate divisions. The flower consists of five small, fleshy, 
suborbicular sepals, five alternate petals, much longer, oblong, somewhat fleshy, more 
or less tomentous on both sides, and invariably furnished on the inside above the middle 
with a transverse tuft of hairs (a feature peculiar to this genus alone); they are inserted 
beneath the margin of the disk, become patent or reflected on the expansion of the 
flower, and are quincuncially imbricated in æstivation, more especially at the apex. 
The disk is rather large, very fleshy, in the form of a truncated cone, somewhat 5- 
lobed at its base, wbere it is partially free from the torus, is always puberulous, often 
thiekened around the truncated apex, where it is more densely tomentose, and hollowed 
in the centre for half its length, in which cavity the ovary and stamens are seated. So 
constant are all these characters that any species of Hippocratea may be surely recog- 
nized from one of any other genus by a hasty inspection of the flower. "The stamens are 
invariably three in number, inserted at the bottom of the cavity within the disk, around 
the ovary ; the filaments are greatly compressed, much dilated towards the base, where 
their margins almost meet, are slightly tomentous, generally twice the length of the disk, 
and are somewhat reflected at the apex ; the anther is rather large, of a deep orange colour, 
quadrately orbicular or slightly 4-lobed, the lobes broader than long, subreniform, upon a 
small dorsal connective, attached, in an inclined direction, at their middle, to the broadish 
apex of the filament, becoming by its sudden reflexion quite extrorse; and it opens a 
little above and across its middle by a transverse suture, at first gaping bivalvately, and 
then patently, in the form of a flattened rosette, showing, in an almost unilocular cell, 
four round patches of pollen intersepted by a cruciform obsolete partition. The ovary,. 
which rises but little above the disk, is somewhat conical and trigonoid, is 3-locular, the 
cells within its rounded angles being alternate with the stamens, and opposite the stig- 
mata, each cell containing from four to eight crowded divaricated ovules, collaterally affixed 
in pairs to the internal angle of the cell, upon its very short axis. The peculiar mode of 
growth of the ovary is deserving of attention: its axis, crowned by the persistent style, 
from first to last never lengthens, appearing atrophied; but an increment is first per- 
ceived soon after the expansion of the flower at the top of each cell in the form of three 
protuberanees, which continue to grow upward and transversely till they severally attain 
a size of more than fifty times the length of the ovary, resolving themselves finally into 
* Linn. Sp. Pl. 1st edit. 1191. T Stirp. Amer. p. 9, tab. 9. , + Prodr. i. 567. 
