Jtondelelia. FENTANBRIA Kckogynia. 135 



lar, entire, smooth above, a few hairs, particularly on the nerve and 

 veins underneath : from four to six inches long, and from one to two and 

 half broad.— Petioles short, and somewhat winged, arising from the 

 decurrency of the leaves, hairy — Stipules triangular, with ensiform api- 

 ces, hairy eta both sides. — Panicles terminal, larg^, composed of many, 

 opposite, compound, four-sided, hairy branches and hranchlets. — 

 Bractes numerous, hairy ; those of the ultimate divisions of the p;mic!e 

 ensiform, with a hastate base, and embracing three flowers; besides 

 which each flower is furnished with a small proper bracte. — Flowers 

 very numerous, small, \\ hite, sub-sessile, fasicled, and on the extremities 

 of the ramifications, always in opposite fascicles of three. — Calyx five- 

 toothed, hairy, permanent. — Carol one-petalled. Tube slender, 

 many times longer than the calyx. — Segments ovate, spreading. — 

 Filaments very short, inserted immediately under the fissures of the 

 division of the border of the corol, in the mouth of the tube. — An- 

 thers incumbent. — Germ inferior, turbinate, two-celled, each con- 

 taining many ovula attached to the elevated centre of the partition. 

 — Style as long as the tube. Stigma Iwo-lobed ; lobes oblong.— 

 Capsule spherical, size of a small grain of millet, dry, brown, crown- 

 ed with the remaining calyx, two-celled, two«valved, opening across 

 the apex contary to the partition. — Seeds few, very minute, and im- 

 perfect, so much so that I have not been able to ascertain their inter- 

 nal structure. 



Obs. Its bark is employed by the natives as a mordant in some 

 of their dyes, hence the specific name. 



S. R exserta, R* 



Arboreous. Leaves petioled, broad-Ianceolar. Stipules semilu- 



• This is a rather common tree in various parts of Nipal, where it blossoms din ins* 

 the summer. It varies >o much in the form and villosity of the leaves, and the more 

 or less crowded fascicles of flowers, that I suspect at least two species are con* 

 founded under it, which I am uuable at present to distinguish satisfactorily. Tha 

 recurved, reniform, terminal appendix of the stipules is sometimes so lar^e, as to 

 give the petioles the appearance of being aurictilate. — R. orissensis, Roth, rue syst. 

 veg. v. 234, and probably also R. Heynii (thyrsiftora, Roth- MS.; ibid, belong to (hit 

 epecies.— N. W. 



