416 



MEMOIRS OF THE NEW YORK BOTANICAL GARDEN [Vol. 8, No. 5 



3) Also from Meerburgh's herbarium, bearing a label saying "Dolichos kolii 

 [or kolii] Linn Sp Plukn Aim: 290 pi 10." I cannot make, much of this, except that 

 the Plukenet reference is probably to "Phaseolus hirsutus flexicaulis Mungo af- 

 finis e Maderaspatan caule tereti; sub nomine Colue recepi" (Pluk. Almag. Bot. 

 290). This is said to be a synonym of Phaseolus mungo L. Mant. PI. 101 (1767). 



Sheets 1) and 2) are unquestionably Vigna unguiculata (L.) Walp.; 3) is Do- 

 lichos uniflorus Lam. 



In the descriptive sentence used by van Royen and Linnaeus for D. biflorus, 

 taking into account the * Erecti in which Linnaeus placed the species, the erect 

 smooth stems and the pedunculate flowers do not apply to any plant which later 

 authors have put under Dolichos biflorus in its widest sense. On the other hand 

 the descriptive sentence referred to fits well Vigna unguiculata and the specimens 

 of van Royen and Meerburgh. 



David van Royen died in 1799, and must have written the label on sheet 1) 

 sometime after 1763. Meerburgh's works were published between 1775 and 1798, 

 and the labels on his specimens appear contemporary. There is thus strong evi- 

 dence that at that time at Leiden the name Dolichos biflorus was being applied 

 to Vigna unguiculata, and that the plant which subsequently became known as 

 D. biflorus was called a different name, "Dolichos kolii." The name D. biflorus 

 was only written on specimen 3) quite recently. 



So far there is no evidence at all that Dolichos biflorus (i.e. Vigna unguicu- 

 lata) and D. uniflorus were anything but clearly distinguished from each other. 

 The trouble seems to date from Murray's thirteenth edition of the Systema naturae 

 (1774), for on p. 548 he adds as a synonym of Dolichos biflorus "Phaseolus vul- 

 garis lablab effigie, flore parvo ochroleuco, siliquis falcatis gemellis. Pluk. aim. 

 291. t.213. f.4." The plant described and figured by Plukenet is clearly Dolichos 

 uniflorus and not Vigna unguiculata. Since then Dolichos biflorus has been gen- 

 erally applied to D. uniflorus or related species of Dolichos, and not at all to 

 Vigna unguiculata. 



Fortunately Vigna unguiculata (L.) Walp. is based on Dolichos unguiculatus , 

 published in 1753, at the same time as D. biflorus, so that the latter merely be- 

 comes a synonym of V. unguiculata. 



I propose that the specimen in van Royen's herbarium, labelled Dolichos bi- 

 florus by David van Royen, the first of the three specimens referred to already, 

 be taken as the type of Dolichos biflorus L. 



The authorities of the Museum National d'Histoire Naturelle at Paris very 

 kindly sent on loan a fragment of the type of D. uniflorus Lam. ("De M. Sonnerat 

 au Jardin R."), showing stems, leaves and flowers; also a tracing of the pods, 

 showing them clearly falcate and 6 mm. wide. There can I think be no reasonable 

 doubf that D. uniflorus Lam. has been correctly interpreted as the commonly cul- 

 tivated horse gram of India, from whose axils single flowers often arise as they 

 do in Lamarck's type. The calyx-segments are shorter than those of D. daltoni. 



The only African gathering that agrees with the Indian plant is Snowden 1826 

 from Kampala in Uganda, where it was said to be an introduction cultivated as a 

 cover-crop. 



The following variety is however no doubt native in Africa.] 



[2b Dolichos uniflorus Lam. var. stenocarpus Brenan, var. nov. 



Leguminibus angustioribus 4-5.5 mm. (nec. 6-8 mm.) latis differt. 

 Dolichos benadirianus Chiov. Ann. Bot. Roma 13:385. 1915; Bak. f. Leg. Trop. Afr. 

 448. 1929. 



INDIA: Punjab: Simla, Hiya Khud, 5th waterfall, 1520 m., Sept. 1, 1885, H. Collett 

 596. Central Provinces: Pachmarki, wild in woods remote from cultivation, Oct. 9, 1901, 



