242 



MEMOIRS OF THE NEW YORK BOTANICAL GARDEN [Vo/. 8, No. 3 



Rhus longipes Engl. yar. grandifolia (Oliv.) Meikle, comb. nov. 



Rhus villosa L. f. var. grandifolia Oliv. Fl. Trop. Afr. 1: 439. 1868; Engl, in DC. 

 Monogr. Phan. 4: 425. 1883; Pflanzenw. Afr. 3 2 : 209. 1921. 



Kota-kota District: Nchisi Mountain, common amongst rocks in Brachystegia 

 woodland, shrub 3-5 m. high, weak, subscandent habit, flowers greenish, fruit red 

 (immature), 1400 m., July 24, 1946, 16888, Distribution seems to be more or less 

 the same as that of typical R. longipes. 



This species has been confounded and obscured in a most remarkable manner. 

 True Rhus villosa L.f. (R. incana Mill.) from S. Africa is quite distinct, with 

 a different habit and indumentum, and with thick, coriaceous, blunt, obovate or 

 spathulate leaflets. Moreover, so far as tropical African material is concerned, 

 it can be shown that the epithet villosa (or incana) has been applied indis- 

 criminately to several perfectly distinct species. Engler (in DC. Monogr. Phan. 

 4: 431. 1883) first correctly identified our plant as a distinct species, under 

 the name Rhus longipes, but he seems to have overlooked the importance of 

 the distinction by continuing to cite tropical African specimens, conspecific 

 with his R. longipes, under the epithet villosa. Furthermore (Bot. Jahrb. 24: 501. 

 1898) he adds to the confusion by publishing a new name, R. huillensis, and re- 

 ferring to it two Welwitsch specimens, one of which (Welwitsch 4412) is identical 

 with R. longipes, the other (Welwitsch 4415) is quite a different species. This 

 unfortunate error has been repeated in the Gossweiler Catalogue (Jour. Bot. 

 66 suppl. 1: 92. 1928). Engler may have realized his mistake, for later (Pflanzenw. 

 Afr. 3 2 : 1921) he appears to restrict the epithet huillensis to the small-leaved 

 species represented by Welwitsch 4415 (a variety of R. quartiniana A. Rich.), 

 though he continues to apply the epithet villosa to tropical African material. 



R. ruzizensis Engl, is, I think, identical with R. longipes Engl., and the 

 name should be rejected as a later synonym. 



The name Rhus inamoena Standi, has crept into several check-lists of African 

 trees and shrubs, but I have not been able to trace any description and suspect 

 that it is merely a manuscript name copied from an herbarium label; many of the 

 specimens so named are referable to R. longipes. 



R. longipes Engl, is undoubtedly a very variable species, though it is likely 

 that many of the variants will prove to be states or growth-phases rather than 

 varieties in the proper taxonomic sense; var. grandifolia with its large, broad 

 leaflets and softly villose shoots is, however, fairly readily distinguished from 

 the type, though it must be admitted that intermediates do occur and that the 

 villosity of the stems is a great deal more obvious in young flowering specimens 

 than in mature fruiting material. 



Rhus nfonticola Meikle, sp. nov. 



R. chirindensi Bak. f. affinis, sed statura humiliore, ramis mox glabrescenti- 

 bus, foliolis eleganter reticulato-venulosis valde differt. 



Frutex parvus circiter 1 m. altus; rami flexuosi, juventute pills albis ca- 

 ducis dense obtecti, mox glabrescentes, leviter lenticellati, cortice brunneo- 

 rufescenti. Petioli validi usque 6 cm. longi, parce pilosi, supra valde applanati 

 vel leviter canaliculati. Foliolum terminale obovatum vel ellipticum, usque 8 cm. 

 longum et 3.5 cm. latum, ad basin in petiolulum ± alatum usque 8 mm. longum 

 sensim coarctatum, ad apicem in cuspidem acutam usque 8 mm. longam attenua- 

 tum; lamina siccitate brunnea vel olivacea, margine arte recurvata, integerrima; 

 costa utrinque prominens, nervi laterales 12-20, prominentes, adscendentes, ad 

 marginem conjuncti, venae et venulae eleganter reticulatae; foliola lateralia simi- 

 lia, breviora, usque 7 cm. longa et 2.8 cm. lata. Inflorescentia terminalis, albo- 

 pilosa, multiramulosa, usque 10 cm. longa. Flos <$ virescens, glaber, breviter 



