1897.] D. Praia — Some additional Leguminosa?. 381 



there are some rhomboid leaflets on nearly every plant. This is Dr. Wallich's 

 Uraria retusa in its entirety; Mr. Baker says that it forms part of Dr. Wallich's 

 U. hamosa also, which may well be the case, for evidently Dr. Wallich misunder- 

 stood those Indian species, as elaborated by Dr. Roxburgh, of Iledijsarwn (or 

 Doodia, as Roxburgh afterwards called them), that constitute the genus Uraria. 

 The writer has not, however, seen an instance of this particular confusion among 

 the Wallichian specimens. 



To this species the most nearly related Uraria is U. alopecuroides Wight, (Doodia 

 alopecuroides Roxb.), which differs in having a stouter habit, rather larger racemes 

 with broader more shortly tailed bracts, and larger differently shaped clouded leaves. 

 Wight and Arnott (Prodr. 222) have suggested that U. alopecuroides may be the 

 same as U. lagopoides Wall., which is impossible, because U. lagopoides Wall, is cer- 

 tainly U. lagopus DC. ; in this Wight and Arnott have been followed by the F. B. I. 



4. Uraria lagopus DC. 



This species has been confused with the preceding by Dr. Wallich who issued 

 it as U. lagopoides Wall. Cat. n. 5676 E. from Nepal and n. 5676 F. from Silhet, and 

 still more hopelessly with the next two species by practically every Indian author who 

 has dealt with the genus. The confusion is, as it chances, devoid of excuse since 

 this is the most easily characterised species of its own group, owing to its having hir- 

 sute pods, those of U. lagopoides, U. alopecuroides and U. neglecta being qnite glabrous. 



From the synonyms given in the Flora of British India must be excluded U. 

 alopecuroides Wight, (7c. t. 290) regarding which no mistake is possible because it is 

 a copy of Roxburgh's manuscript drawing; also Doodia alopecuroides Roxb. (Fl. lnd. 

 iii. 368) which is the description of the plant there delineated, and Hedysarum 

 alopecuroides Roxb. (Hort. Beng. 57), the earliest reference to the plant in question. 

 The F. B. I. has referred here also a part of Uraria hamosa Wall., since it has 

 identified Wall. Cat. n. 5681 C. with U. lagopus. This is not, however, possible ; 

 Wall. Cat. 5681 C. has glabrous pods with 3-6 segments, the segments being 

 much smaller than those of U. lagopus. 



The F. B. I. locates the species in Burma ; no specimens from Burma have ever 

 been received at Calcutta. Specimens have, however, been sent from the Chumbi 

 Valley, the plant consequently occurs at a considerably higher level than the F. B. J. 

 indicates. 



5. Uraria alopecuroides Wight, Ic. t. 290. U. repanda Wall. Gat. 

 5677. 



This is Dr. Roxburgh's Doodia alopecuroides. Just as Dr. Wallich failed to 

 recognise Roxburgh's D. lagopodioides, to which he gave the name U. retusa, so he 

 failed to recognise D. alopecuroides and issued it as U. repanda. The plant is per- 

 haps not specifically distinct from U. lagopoides ; it certainly can never ba referred 

 to U. lagopus, for though it has the erect habit of the latter instead of the trailing 

 habit of U. lagopoides, its bracts differ but slightly and its calyx, corolla, and pods do 

 not differ at all from those of U. lagopoides. 



The localities of the species are : — 



Dehra Dun ; Vicary ! Behar ; Kurz ! Anderson ! Wood ! Clarice ! Khasia Hills ; 

 G. Mann! Burma; Wallich ! King's Collector! It probably came originally from 

 Dehra Dun to Dr. Roxburgh; it was one of the species of which he received seeds 

 from Genl. Hardwicke — these, as we know now, came mostly from the Sub-Hima= 

 hiyau tract. This species has not been collected in the Himalaya. 



