rent line of hairs from the insertions of the leaves,
leaves scarcely sheathing, sessile, ovato-lanceolate,
sub-acute: flowers axillary, two or three from each
axile, filaments glabrous, capsule ovate, cells about 8-
seeded, in 2 rows.
Quilon, October to December, in low wet ground.
2079. D ithyrocarpus petiolatds (R . W .), ascending,
sparingly ramous: leaves sheathing, elliptico-
lanceolate acute, tapering at the base into a longish
petiol; sheath inflated, ciliate on the margin: panicle
terminal; branches racemose, flowers secund and,
with the rachis, villous.
Neilgherries. I am still uncertain whether I ought
to consider this a distinct species or a mere form of
D. Rothii. All the three species here represented .are
very like, and if really species prove this to be a
very natural genus, but still the differences seem such
as to preclude their being united, certainly not until
we have had opportunities of studying them better
than I have had it in my power to do. The Anei-
lema hispida of Wallich’s list certainly belongs to
this genus.
2080. D ithyrocarpus R othii (R. W., Trades-
cantia paniculata, Roth, not Roxb.), stem creeping at
the base, erect at the ap e x : leaves sheathing, lanceolate,
^ acuminate; sheaths ciliately woolly: panicle
terminal, somewhat globose, compact; branches racemose,
many-flowered, densely villous, viscid, anterior
petal much narrower, sub-spathulate: filaments glabrous
: stigma obtuse: capsule 2-celled, with a single
sub-lenticular seed in each. /
Neilgherries, Ceylon ? Roth remarks that his plant
does not correspond with Roxburgh’s figure, but I
think his description corresponds with mine; which
is certainly not Roxburgh’s plant, so far at least as
can be made out from his. figure and description.
The figure differs in the form of the leaves and
sheath (which is woolly on the margin), in the composition
of the panicle, which as shown by him is
distinctly compound, each branch panicled, while in
mine they are racemose. In his the calyx is said
to be simply hairy while here it is shaggy and viscid.
I cannot so well compare the flowers as my drawing
is made from a dried plant, and may not be so correctly
represented as in his. Roth describes the capsule
as 3-celled, perhaps a typographical error.
2080. D ithyrocarpus undueatus (R. W.), ascending
: leaves ovato-lanceolate, acuminate, waved
on the margin, sheathing: sheaths large inflated, the
throat thickly beset with coarse bristly h a irs: panicles
terminal, branches racemose: calyx shaggy, viscid,
lobes obovate obtuse: odd petal narrow obtuse
sub-cuniate: style filiform, curved: stigma simple :
capsule 2-celled, 2-sceded.
Station. ^ I am uncertain whence I obtained this
plant. I t is nearly allied to the preceding, but I think
certainly distinct, its whole aspect being so different.
The leaves and sheaths externally are glabrous, but
a line of hairs extends down the stem from the
woolly margins of the sheaths.
2081. Streptolirion voebuile (Edgeworth, Lin-
nean Trans.)
I am uncertain now whence I obtained the plant
from which the drawing was taken, but I think from
Assam, about 15 years ago, at which time the drawing
was made. I shall somewhat abridge Mr. Edge-
worth’s description of the plant which is very full.
Glabrous, twining; stems rooting at the base: leaves,
cordate acuminate, long petioled; petiols sheathing
at the base; sheaths truncated, ciliate: racemules
axillary and terminal, 2- 6-flowered: floral leaves
becoming modified, losing their sheaths, the petiols
shortening or disappearing and the limb changing to
cordato-ovate, acute or folded: upper flower of the
raceme often sterile: bracts lanceolate, delicately
membranous: three exterior lobes of the perianth
elliptic acutish; interior ones linear, a little dilated at
the a p e x : stamens six, filaments bearded, with yellow
hairs above the middle; anthers versatile, cells horizontally
divaricated: ovary tapering into the sty le ;
stigma capitate, puberulous: capsule ovate, 3-celled,
3-valved; cells 2-seeded: seed slightly angular, ru-
gosely furrowed.
This genus differs but little except in habit from
Tradescantia. The perianth is the same with the
exception of the petals being smaller than the sepals
and the filaments in both are bearded and all the
anthers polleniferous. The form of the anthers however
is _ peculiar in so far as they resemble in form
the sterile anthers of Aneilemas. The habit is very
distinct, and, added to the above differences, well
entitles this plant to form the type of a distinct
genus. I t ranks between Aneilema and Tradescantia
rather than between Tradescantia and Cyanotis
on account of the anthers forming an easy transition
from the one to the other.
2082. Cyanotis cristata (Rsem. and Sell., Com-
melyna cristata, Lin., not Burm. FI. Ind. tab. 7. f. 4.
Tradescantia imbricata? Roxb.), lower part of the
stem diffuse, creeping; floriferous extremities ascend-
mg or erect, marked with attenuate pubescent lines
decurrent from the sheaths of the leaves and sprinkled
with long h a irs: leaves sessile, succulent, ovato-lanceolate,
glabrous, slightly ciliate: spikes terminal,
secund, progressively lengthening from 2 to 12 -15
pairs of bra cts: bracts lanceolato-falcate, imbricate,
each supporting a flower: flowers small, scarcely ex-
serted, sepals lanceolate acute, pubescent, petals connate
to near the apex, limb obtuse : stamens scarcely
exserted, filaments simple, bearded: style glabrous:
stigma capitate : capsule ovate, cells 2-seeded.
Bolamputty Hills, frequent in woods, flowering
November and ^ December. I have extended the
character of this plant under the impression that
more than one species is confused under this name.
My plant seems to correspond sufficiently well with
Linnams’ figure in the Flora Zeylanica, but not with
Burmann’s, in the Flora Indica, of which also, I think,
I have specimens, a figure of which is given in plate
No. 2088.
2083. Cyanotis pilosa (Rami, and Sell., Tradescantia
pilosa, Willd. Herb.), stems scapose, procumbent,
spreading, somewhat branched and, with sheaths
and under surface of the leaves, more or less floccose:
radical leaves long linear, obtuse, villoso-ciliate: stem
leaves like the radical ones, but smaller: spikes terminal,
secund, aggregated, few-flowered: bracts falcate,
calyx woolly, lobes lanceolate acute, filaments
densely bearded, not tumid at the apex: ovary
pilose; style bearded: stigma clavate: capsule small,
cells 2-seeded.