296 ; LEGUMINOS X. BaUHINIA. 
c. Calyx spathaceous : anther-bearing stamens 5, with or without fertile filaments 
between them. 
913. (5) B. variegata (Linn. :) arboreous, unarmed : leaves roundish, cor- 
date or rounded at the base, upper side glabrous, under when young shortly 
and softly villous afterwards glabrous ; leaflets oval, obtuse, united far above 
the middle, nearly parallel, 5-nerved: racemes axillary and terminal: calyx 
spathaceous, ovate, slightly acuminated, 5-toothed at the apex: petals ob- 
long, nearly sessile, the upper somewhat larger and on a rather longer claw 
than the others: fertile stamens 5, with sometimes the rudiments of 1-5 abor- 
tive filaments, all shortly united at the base; “ inner verticil wanting " (Grah.); 
legume straight, linear, acuminated, compressed, gabrous, 5—12-seeded ; seeds 
approximated.— Wight ! cat. n. 630.—2; four petals reddish and veined with 
purple ; the fifth variegated with purple, brown, and yellow.—Ham. in Linn. 
soc. trans. 13. p. 496.— B. variegata, Linn. sp. p.535; DC. prod. 2. p. 514 ; Spr. 
syst. 2. p. 333 ; Roxb. fl. Ind. 2. p. 319 ; in E. I. C. mus. tab. 1238 ; Wall.! L. 
n. 5795.—B. purpurea, Wall. L. n. 5797, e, i? (partly), k (partly), m? (ac- 
cording to Graham).—Rheed. Mal. 1. t. 32.—8 ; four petals whitish ; the fifth 
variegated on the inner side with yellow and green.— Ham. l. c. p. 497.—B. 
. candida, Roab. fl. Ind. 2. p. 318 ; in E. I. C. mus. tab. 634 ; Wall. L. n. 5796. 
“ This tree can only be said to differ from B. candida in the colour of the 
flowers : had I met with this first, I should certainly have considered the other 
asa variety only."—Roxs. We follow Hamilton in uniting them; in the 
herbarium it is impossible to distinguish the two. 
$2. Anther-bearing stamens 3 (or occasionally 4), or rarely 5, with or without 
2-7 sterile filaments, all connected at the very base ; ovary-stalk cohering 
on one side with the tube of the calyv.—Symphyopoda, DC. 
. 914. (6) B. phenicea (Heyne:) arboreous?, unarmed, with a dark mi- 
nutely warted bark : leaves ovate, cordate at the base, upper side glabrous, 
. under minutely pubescent; leaflets united for about two-thirds of their 
. length, obtuse, diverging, 3-4-nerved: panicles terminal, clothed with short 
dense tomentum ; pedicels short with 3 bracteas at their base: calyx spatha- 
ceous : petals a little longer than the calyx, oblong-lanceolate, acute, nearly 
equal, glabrous, with claws about as long as the limb: stamens 5, all fertile; 
inner verticil eut into numerous bristle-shaped threads: ovary shortly tomen- 
foto Heus ie Wall. L. n. 5800. : 
e have taken the above character from a description of this plant sent us 
by Dr Graham. Dr G. remarks that the dower tudsars cylindric-clavate ; the 
calyx 5-parted, the segments 1 inch long, linear, and generally united at the 
apex; whence we have considered it as a modification of spathaceous. . t 
appears closely allied to the next, and principally differs by having 5 fertile . 
stamens. May it not be a hybrid between B. purpurea and. B. variegata ? 
915. (7) B. purpurea (Linn.:) arboreous, unarmed: leaves roundish, cor- 
date at the base, coriaceous, glabrous ; leaflets oval, obtuse, parallel, unit 
to the middle or a little above it, the free portions sometimes overlapping — 
each other, 4-5-nerved : racemes terminal, panicled: calyx usually cleft to 
the base of the limb into 2 reflexed segments, the one emarginate, the other 
3-toothed, sometimes all cohering at the apex and spathaceous: petals ob- 
. long-lanceolate, attenuated at the base: fertile stamens 3 (or occasionally 4), 
long, ascending; “ inner verticil large, thick, and slightly fimbriated on the 
ba 3 (Grah.) : legume long, linear, pointed, with 6-12 rather distant seeds. 
— Linn. sp. p. 536 ; Spr. syst. 2. p. 383 ; Roxb. fl. Ind. 2. p. 320; in E. T. C. 
net "s = H Ham. m EE soc. trans. 13. p. 497: Wall.! L. " i 
rtly) ; t! cat. n. 629.—B. Coromandeli . prod. 2. p. 919— 
Peed, Mal. 7 e deliana, DC. p p is 
De Candolle's B. purpurea, probably described from a specimen from the — 
Island of Timor, is certainly different : we have never jen in our plant the 
