Dr. Hooker {Brit. FI.) to be the E. hexandra of DeCandolle,
a fact of which Sir J. E. Smith was probably not aware
when he published it in his English Flora under the name
of E. tripetala. The true E. Hydropiper does not appear
to have been observed in Britain till it was detected by
the writer of this, in August last, on the S.E. side of Llyn
Coron, near AbberfFraw, Anglesey, forming matted tufts
4 to 6 inches wide, both on the sandy margin of the lake,
and in the shallow water immediately contiguous. The
flowers and stoloniferous stems were generally covered by
the sand and mud, the petals being in that case white; and
it would appear they are tinged with pink only when exposed
to the air as well as thé light. The accompanying
drawings are from sketches taken in Anglesey by Mr. Wilson
of Warrington and myself, from fresh specimens.
E. hexandra was also found growing with it, but the two
species are never intermixed in the same tuft.
In general habit and appearance, our plant approaches
closely to E. hexandra, having a round creeping stem, internally
divided by radiating septa into eight longitudinal cavities,
and throwing out at each joint several white simple
roots, accompanied by two, sometimes four erect opposite
spathulate leaves, united in pairs. Each leaf has a central
nerve, dilated at the summit, and occasionally furnished
with lateral veins, with two membranous ovate toothed stipules
at the base, one on each side of the stem. Flowers axillary,
solitary, nearly sessile, alternate. Calyx-segments 4,
nearly ligulate, obtuse, shorter than the petals and fruit.
Petals 4, ovate, white or very pale red, closed over the
stamens and germen, in which state the impregnation in
both species is effected under water, as in Subularia aqua-
tica, the anthers lying in close contact with the stigmas.
Stamens 8, fixed opposite to the petals and calycine segments;
filaments flattened, somewhat dilated at the base,
anthers 2-celled. Styles none; stigmas 4, very minute, persistent,
forming the extreme points of the valves of the
capsule. Capsules 4-valved and 4-celled, roundish, with
slight depressions at the junction of the valves, convex at
the summit, becoming at length obscurely angular by the
swelling of the seeds: septa alternate with the valves
adhering to a central axis. Seeds about 4 in each cell,
pendulously attached to placentae which crown the intersection
of the dissepiments, curved almost double ; embryo
of the same shape, dicotyledonous. The base oFtlie seeds
is strongly umbonated, exactly resembling the operculum
of a moss: this is the chalaza of Gærtner; and the semitransparent
membrane occupying the space between the
inner sides of the bend, like a ligature to retain the seed
in its curved form, is the raphis, or bundle of vessels connected
with the chalaza. The outer coat is furrowed and
transversely striated, very elegant under the microscope.
The strongly hooked shape of the seeds, and their attachment
at the top of the dissepiments, are both very important
characters, sufficient, independently of others, to establish
the specific difference of our plant from E. hexandra,
w'hose seeds are nearly straight, and have an upward direction
from the point of their attachment at the base of the central
axis. Dr. Hooker, who has kindly furnished the characters
and synonymes, fully concurs in this opinion ; and remarks,
“ Cafnbassèdes, in the Mém. du Mus. d' Hist. Nat.
9me année, has, with much judgement, raised Elatine and
two allied genera to the rank of an order, Elatinece. That
author seems to consider Elatine to be exclusively an European
genus. I have just been describing one from the
East Indies.”
I subjoin an amended specific character of E. hexandra,
as drawn up by my acute and indefatigable friend Wilson ;
and I may add, that I have examined many specimens, both
from the original station near Shrewsbury and from Anglesey,
wdiich have invariably 3 petals and sepals, 6 stamens,
and a 3-valved capsule.
E. hexandra. Leaves opposite, spathulate. Flowers
alternate, pedicellate, erect, hexandrous, tripeta-
lous. Calyx longer than the fruit, segments
spreading, orbicular. Petals obovate. Capsules
turbinate, concave at the summit, 3-celled. Seeds
nearly straight, about 12 in each cell, (I find about
8.—J. E. B.) ascending, the end where they are
attached to the receptacle being the lowest.