426 D. Prain — Some additional Leguminosse. [No. 2, 



as a section of Phaseolus. In the Genera Plantarum (i. 539) Bentham and Hooker, 

 while still recognising the group, doubt whether it constitutes a section of Phaseolus, 

 and suggest that it may be found preferable to refer it to Vigna. The natural 

 character of the group is, however, somewhat marred in the Genera Plantarum by 

 the inclusion of a species figured by Wallich as a Phaseolus (PL As. Bar. i. 6, t. 6) 

 which Kurz has clearly shown to be a Dunbaria (Journ. As. Soc. Beng. xliii. 2, 186 j 

 xlv. 2. 255). Kurz, who treated the group in the sense originally understood by 

 Bentham, recognised qxiite clearly that it can by no possibility be included in Phaseo- 

 lus ; he has consequently adopted a suggestion made in a MSS. note that Wallich has 

 left in Herb. Calcutta, and referred all the Dysolobia to Canavalia. For this, at first 

 sight, there is soinething to be said ; the structure of the pod in all the species is very 

 much that of Canavalia. When, however, it is considered that the calyx differs 

 altogether from the calyx of Canavalia, that the style is bearded, and that the seeds 

 are hirsute, it seems less convenient to adopt Wallich's suggestions than to adopt 

 Bentham's. Baker has attempted a compromise ; in the Flora of British India he 

 still treats Dysolobium as a section of Phaseohis ; he leaves in it, however, only two 

 forms, viz. : — the species of the group that has the longest beak to its keel, and 

 the Dunbaria that has, by inadavertence, been cited as a Dysolobium in the Genera 

 Plantarum ; the other two he has referred to Vigna. The last species of the group 

 he has, in the absence of flowers, dealt with tentatively as a Psophocarpus. Taubert 

 (in Engler's Natilrlichen PJianzenfam. iii. 3, 380) has thrown no new light on the affi- 

 nities of the group ; on the contrary he has accorded it, without qualification of any 

 kind, the treatment and the position regarding which the authors of the Genera 

 Plantarum have so expressly enjoined caution. 



That the group as originally recognised by Bentham forms, in consequence of 

 its firm, septate pods and its hirsute seeds one of the most natural and definite 

 genera in the whole of the Phaseolidee does not, the writer thinks, admit of ques- 

 tion ; to settle the dubiety that has prevailed as regards its proper position, it 

 seems to the writer most convenient to adopt Mr. Bentham's name in a generic sense 

 and to treat the forms it covers as a group apart alike from Canavalia, Phaseolus 

 and Vigna. 



Key to the Species. 



Racemes lax long-peduncled, flowers large ; pods closely vel- 

 vety-villous, seeds sparsely velvety ; (pods keeled along 

 suture but not winged) : — 

 Leaflets rounded cuspidate, chartaceous, hirsute on nerves 

 beneath ; flowers 1'75 in. long, keel with long laterally 

 deflexed beak, style bearded down the face ... 1. D. grande. 



Leaflets narrowed to a point, membranous, glabrescent ; 

 flowers only '6 in., long, beak of keel not deflexed, style 

 penicillate round stigma ... ... ... 2. D. lucens. 



Racemes dense short-peduncled ; flowers small ("3 in. long or 

 less) ; pods softly hirsute with long hairs, seeds densely 

 velvety ; (beak of keel not deflexed) : — 

 Leaflets roundish cuspidate ; pod neither keeled nor winged 3. D. dolichoides. 

 Leaflets lanceolate; pod subquadrangular, prominently 



winged along the angles ... ... ... 4. D. tctragonum: 



