72 NEILGHERRY PLANTS. 
STROBILANTHES. 
Calyx about equal, 5-parted to the base; lacinez linear, somewhat broader towards the apex. Corolla 
funnel-shaped, the tube not passing abruptly into the limb; lobes equal or nearly so, rarely sub-bilabiate. 
Stamens four, didynamous, inserted on the middle of the throat, usually within the filaments, united at the 
base by membrane ; anthers oblong, muticous, cells parallel, equal, contiguous, or in some diverging at the 
base, whence the anthers are sagittate. Stigma subulate, incurved or involute, spongiolose on the back. 
Capsule columnar, 4-sided, 2-celled almost to the base 4-seeded about the middle ; partition thin, incom- 
plete towards the apex, adnate, or sometimes separating from the valves. Seed discoid, angular, with an 
areola on both sides, the angles more prominent towards the hilum, attached to hooked retinacula. Herbs 
and shrubs, spikes more or less dense, axillary and terminal, erect, cernuous, or drooping. Bracts foliaceous 
or foliaceo-membranous, persistent or caducous, exposing the flowers, bracteoles small or sometimes want- 
ing. Flowers, in most, delicate blue or white. 
fuse, which however is corrected by an abreviated essential character in these terms, “ Calyx 5-parted, sub- 
equal. Corolla funnel-shaped subequal. Stamens incluse, sometimes monadelphous ; anthers straight, muti- 
cous. Capsule 4-sided not unguiculate, 4-seeded in the middle. Spikes axillary and terminal, usually com- 
pact, broadly bracteate and cone-like, in some loose.” To illustrate this genus three plates are given. 
The first, S. sessilis, is I believe a true representative—the second, S. Wightiana, ought in my opinion to 
form the type of a new genus which might include my S. duridus, and S. micranthus, all readily distinguish- 
able by their peculiar stamens, which are totally unlike those of all the other species I have examined. 
So long as S. luridus was the only one I knew having that peculiarity, I considered it a new genus, but 
on finding it correspond with one of Nees’ species I submitted to his authority, and placed it in this genus, 
I now think, erroneously. As it now stands, in De Candolle’s Prodromus, this is a very difficult genus as 
regards the determination of the species, which I think might have been to some degree obviated by greater 
care in their subdivision and grouping, hence the desirableness of having the whole genus recast. 
STROBILANTHES sEsstxis (Nees), suffruticose, very serrated; spikes ascending, one or two together, op- 
hairy, stem erect, 4-angle : leaves sessile, ovate, posite: bracts large, orbicular, emarginate or slightly 
acuminate, crenate, spikes axillary, opposite and ter- retuse at the apex, dark livid brown; bracteol: 
crenate: spikes about an inch: calyx about 3 an the back: corolla 5-lobed, lateral lobes reflexed, 
inch long, the lobes lanceolate, the two lower ones hence apparently bilabiate: stamens united by pairs 
. a ¥ e , all equal, exerted; anthers oblong: cap- 
pale blue up to purplish. Stamens shortly monadel- sule about the length of the calyx, somewhat com- 
phous at the base, jo yamembrane. Capsule pressed, 4-seeded: seed orbicular, glabrous. 
oblong, smoo' obtuse or somewhat attenuated at Neilgherries, in wood near Nedawuttim. Flower- 
the base, 4-seeded in the middle. ing January and February 1846. When I visited 
STROBILANTHES PERROTTETIANUS (Nees), shrub- the station in 1847 and 1848, I did not find it in 
by, branches reddish, hairy: leaves ovate, caudato- flower though I found the plant in abundance, hence 
cuspidate, undulato-crenate, hairy, very rough above: it seems only to flower once in several years. Cor- 
spikes axillary, opposite, secund, ov nodding, olla deep purplish brown, scarcely exceeding the 
dense, hairy ; bracts ovate, acute, the interior ones large dull lurid bracts, 
a : : us. —_ | at first considered this the type of a genus near 
—An erect shrub, 3-8 feet high, thickly covered Strobilanthes and proposed calling it, with reference 
with brownish-purple bristles, becoming smoother by to the stamens, Didyplosandra, thinking that they, 
: 8 inches long, 1 to 2 broad, hairy; added to the peculiar habit, were sufficient to con- 
peduncles 1-3 inches long, simple or bifid or trifid, stitute this a distinct genus. But on comparing them 
naked. Spikes about an inch long, compact, hairy. with those of S. Wightiana, a species (of which I had 
membrane, capsule contracted at the base, ferenc f ic value, or such as to call 
compressed, 2-seeded in the middle. : iia ia eo le 
e 
unfrequent on ” ays outskirts of My own impression, however, is, that they, and some 
und, ers to i ; 
: W.), a large, ra- the genera Goldfussia an Strobilanthes, either united 
mous shrub ; branches virgate, bearing the inflores- or recast on amended characters. As they now stand 
: 1 
cence on the lower naked portions: leaves i istinguishable. 
oblong, acuminate, pubescent on both sides, Snely os a gp ea —— 
os 
- 
