1953] 



PLANTS COLLECTED IN NYAS ALAND 



255 



tain; Luchenya Plateau, very abundant in forest-re growths, shrub 2-3 m. high, 

 branched into a very dense flattish crown, flowers yellow, 2000 m., June 27, 1946, 

 16470. Confined to Nyasaland. 



Geissaspis drepanocephala Bak. Kew Bull. 1897: 260. 1897; Bak. f. Leg. Trop. 

 Afr. 313. 1929. 



Kota-kota District: Nchisi Mountain, sporadic in Brachystegia woodland, per- 

 ennial herb, 1400 m., July 26, 1946, 16951*; sporadic in Brachystegia woodlands, 

 perennial herb, bracts green, 1400 m., Aug. 2, 1946, 17110*. Belgian Congo, Tan- 

 ganyika Territory, Nyasaland, and N. Rhodesia. 



Geissaspis ? descampsii De Wild. & Dur. Bull. Soc. Bot. Beige 39(2): 65. 1900; 

 Bak. f. Leg. Trop. Afr. 317. 1929. 



? District: North Road, between Mzimba and Kasungu, stony soil in Brachy- 

 stegia woodlands, shrub 60 cm. high, 1400 m., Aug. 23, 1946, 17387. 



Two specimens in the Kew Herbarium — St. Clair Thompson 1123, Pole Evans 

 & Erens 638 — both certainly conspecific with Brass 17387, have been kindly com- 

 pared with the holotype (Belgian Congo: Samba, Mar. 1891, Descamps s.n.) of G. 

 descampsii by Dr. G. Troupin at Brussels. He writes that the two specimens sent 

 "Different de G. descampsii De Wild, surtout par les folioles plus grandes, plus 

 longuement et densement serrees aux bords; quant aux glandes (ou poils glandu- 

 laires), elles n'apparaissent pas sur les jeunes feuilles de Pholotype, mais bien 

 sur un autre specimen (Vanden Brande 31, Marungu) determine provisoirement. Vu 

 le materiel peu abondant, il est quasi impossible de se rendre compte de l'eventu- 

 elle variation de cette espece." He also very kindly sent some leaflets from the 

 holotype of G. descampsii. In view ->f all this I feel that further gatherings of G. 

 descampsii in the Belgian Congo may bridge the apparent difference in leaflet- 

 size. I do not therefore feel prepared at present to separate Brass 17387 from G. 

 descampsii. I have seen plants conspecific with Brass 17387 from Tanganyika 

 Territory, Nyasaland, and N. Rhodesia. 



Desmodium repandum (Vahl) DC. Prodr. 2: 334. 1825. 

 Hedysarum repandum Vahl, Symb. Bot. 2: 82. 1791. 



Desmodium scalpe DC. Prodr. 2: 334. 1825; Bak. f. Leg. Trop. Afr. 328. 1929. 



Zomba District: Zomba Plateau, frequent in weedy growths on rain-forest 

 paths, herb about 80 cm. high, standard and wings red, keel greenish, tipped with 

 red, 1450 m., June 3, 1946, 16176. Mlanje District: Upper Ruo River, shrub, flow- 

 ers red, about 850 m., July 4, 1946, Vernay 16658. Mlanje Mountain; Luchenya 

 Plateau, occasional in moist openings in forest, shrub 50-70 cm. high, flowers 

 orange-red, 1850 m., July 8, 1946, 16732. Widespread in tropical Africa, also in 

 the Transvaal, South Africa, and Madagascar. 



I am greatly indebted to Dr. Bernice G. Schubert, who has most kindly con- 

 firmed my supposition that the name Desmodium scalpe must be replaced by Des- 

 modium repandum. She writes: "I have on loan from Copenhagen what Schindler 

 considered the type of D. repandum from Forskal's herbarium and shall send you 

 a photograph later on. It is a miserably scrappy specimen but I don't think there 

 is any doubt about the identity." 



Desmodium salicifolium (Poir.) DC. Prodr. 2: 337. 1825; Bak. f. Leg. Trop. Afr. 

 330. 1929. 



Hedysarum salicifolium Poir. in Lam. Encyc. 6: 422. 1804. 



Kota-kota District: Chia area, in semi-shade on bank of a water-hole, flowers 

 purple and white, 480 m., Sept. 1, 1946, 17462. idespread in tropical Africa and 

 on Madagascar and the Mascarenes. 



