80 d TILIACER. - Grewia. 
obtuse or acutish, sometimes unequal at the base, unequally serrated ; upper 
side at length nearly glabrous, under glabrous pubescent or hoary: stipules lan- 
ceolate-subulate : peduncles axillary, 2—4, twice or thrice as long as the petiole, 
3-flowered, and as well as the pedicels divaricating: sepals oblong-lanceo- 
late, twice the length of the narrow-obovate petals: torus scarcely longer 
than the glands: style about equal to the stamens ; stigma 4-lobed : drupes 
globose, with 1-2 1-celled nuts.— DC. prod. 1. p. 511; Spr. syst. 2. p. 581; 
Roth. nov. sp. p. 251; Rowxb.! fl. Ind. 2. p. 586; in E. I. C. mus. tab. 223; 
Wall.! L. n. 1089 ; Wight! cat. n. 255.—0G. subineequalis, DC. prod. 1. p. 511. 
There are before us so many variations in the form of the leaves and pu- 
bescence, that we cannot point out any constant characters but in the fruit. 
290. (14) G. tiliefolia (Vahl:) arborescent: leaves 5-nerved, roundish- 
cordate, rounded or acuminated at the apex, usually oblique at the base, 
equally or unequally bluntly toothed or serrated ; upper side at length gla- 
brous, under glabrous or sometimes densely pubescent : stipules lanceolate, 
auricled on one side at the base: peduncles axillary, numerous, from half as 
long as the petiole to sometimes nearly equal to or even slightly exceeding it, 
3-5-flowered, straightish as well as the pedicels: sepals linear-oblong, twice 
as long as the oblong petals: torus scarcely exceeding the glands: style 
about the length of the stamens; stigma 4-lobed : drupes 2-lobed ; lobes 
lobose, with two hemispherical 3-celled nuts, or one spherical 6-celled nut,. 
in each lobe.—Vahl symb. 1. p. 85 ; Roth. nov. 8p. p. 251 ; DC. prod. 1. p. 511; 
Spr. syst. 2. p. 5815 Roxb. fl. Ind. 2. p. 587 ; Wall.! L. n. 1094 ; Wight ! cat. 
n. 266.—G. arborea, Roxb. in E. I. C. mus. tab. 227 ; Roth. nov. sp. p. 241; 
Spr. syst. 2. p. 581.—G. variabilis, Wall. ! L. n. 1087. c. 
We have not seen Wall. L. n. 1087. a or b, and do not know whether they 
be referable to this or to the last species. Most authors proceed on the sup- 
position that Roxburgh's G. arborea is distinct from G. tiliefolia ; but Rox- 
burgh ed called it so before Vahl's description reached him, when he 
adopted ahls name both in his Hort. Bengh. and Fl. Ind. It is strange 
that G. Don (in Mill. dict.) should refer to the Hort. Bengh. p. 92, for G. 
arborea, no such occurring, the name tiliefolia being even there (at p. 93) 
printed by mistake trifolia. 
291. (15) G. rotundifolia (Juss. :) leaves 5-6-nerved, cordate-roundish or. 
ovate, equal or oblique at the base, crenulated, hoary and softly velvety on 
both sides: peduncles axillary, several together, several times longer than 
petioles, 3-flowered ; pedicels divaricating: sepals linear, 2-3 times as long 
as the oblong slightly bidentate petals : torus searcely exceeding the glands: 
style longer than the stamens ; stigma 4-lobed and crenulated : drupes 
slightly 2-lobed, with 1-2 usually 2-celled nuts in each.—Juss. ann. mus. 4- 
p. 92. t. 50. f. 35 DC. prod. 1. p. 511; Spr. syst. 2. p. 581; Wall.! L. m 
1085; Wight ! cat. n. 263, 268.—G. orbiculata, Herb. Miss.! ; Roth. nov. 8p. 
p. 246.—G. orbicularis, Herb. Lamb. + G. Don in Mill. dict. 1. p. 550. 
.292. (16) G. populifolia (Vahl:) shrubby: leaves 5-nerved, roundish, 
slightly cordate at the base, unequally crenated ; upper side nearly glabrous, 
under pubescent: petioles slender: peduncles axillary, solitary, 1-flowered, 
shorter than the leaf, with two opposite bracteoles above the middle: se 
linear, a little longer than the stamens and narrow linear bifid petals: torus 
scarcely exceeding the glands: ovarium glabrous: style a little longer than 
the stamens ; stigma 4-lobed: drupes deeply 2-lobed ; lobes globose with à 
shining, glabrous crustaceous epicarp, and 1 1-celled nut in each.—/* 
symb. 1. p. 33; DC. prod. 1. p. 511; Spr. syst. 2. p. 581 ; Wight ! cat. n. 267. 
E eis Roth. nov. sp. p. 249; Wall.! L. n. 1088.—Chadara tenax» 
orsk. ! 
Roth states that the petals are obtuse; our specimens, from the same — 
source, have the points bifid. We had no opportunity of dissecting Forskahl's 
plant, but it appeared quite the same as Ours, as did also G. populifolia of 
