252 DiADELPHiA DECANDRiA. Erijtliiina. 



be a tree of very considerable magnitmle. Leaves tcniate. 

 Leaflets the pair iiiipqually broad-cordate ; the terminal one 

 reniform Cordate, all are smooth, entire, and pointed, from 

 three to five inches each way. Petioles round, smooth, very 

 rarely prickly, about six inches long. Stipules narrow-fdcate, 

 more permanent than the leaves. Racemes terminal, project- 

 ing- horizontally at right angles from the apex of the new haf- 

 less branchlets. Bractes minute, caducous; one under each 

 pedicel, and two near their apices. Plowers tern, numerous 

 and crowded, large, as in E. indica, and the colour the i?ame 

 bright dazzling scarlet. Calyx minute, not above one-tenth 

 part of the length of the banner, spathiform, entire, acute, 

 opening on the upper side, withering. Banner oblong, erect. 

 Wincjs ininute, even shorter than the calyx, semi obovate. 

 Keel half the length of the banner, and four or five times 

 longer than the wings, two-petalled, but firmly united, and in 

 that state, when expanded ovate; colour pale greenish yel- 

 low, which is also the colour of the wings ; with the addition 

 of a slight tinge of pink. Filaments seem one and nine, but 

 the ten are united into one, a little above the base, alternitely 

 short. Germ pedicelled, villous, containing from four to six 

 ovula attached to the upper margins. Style tapering. Stiyma 

 acute. Legume pedicelled, lanceolar, subulate, pointed, from 

 three to five inches lone, smooth, light brown. Seeds owe^ 

 two or three, rarely more come to maturity, lodged equally 

 distinct from the ends, oval, smooth, light brown, of the size 

 of a French bean. 



Note. Notwithstanding the great similarity of this tree, to 

 E. indica, JMoitricou of Van Ilheede, there can be no doubt 

 of its being specifically distinct ; the small, entire, spathiform, 

 calyx and minute triangular m ings, are sufficient nmrks by 

 which to distinguish it from all our other huVxan E ryihrinaSy 

 exclusive of the colour of the prickles, which is, no doubt, the 

 most obvious criterion. The contents of the germ furnish an- 

 other proof; in that containing from ten to fifteen ovula, in 

 this only from four to eight, generally about six. 



