1954] 



PLANTS COLLECTED IN NYASALAND 



69 



inclusis 1 mm. longi, divergentes, circiter 0.4 mm. supra basim bifidi, ramis 0.5- 

 0.6 mm. longis apice curvatis sed vix incrassatis. Fructus pedicellis 1-3 mm. 

 longis suffulti, tepalis persistentibus non vel vix accrescentibus, nonnunquam 

 purpurascentibus; capsula depresso-globosa, 3 mm. lata, 1.8 mm. alta, laevis. 

 Semina brunnea, 1.4 mm. longa, dorso lineis valde obscuris tenuibus circiter 

 10-12, sed has exacte enumerare difficile est. 



Mlanje District: Mlanje Mountain, 1891, A. Whyte s.n. (Herb. Kew.); ibid., 

 2440-2740 m., Mar. 1897, G. Adamson 354 (Herb. Kew.); ibid., Nov. 5, 1898, /. 

 Mahon s.n. (Herb. Kew.); ibid.; Luchenya Plateau, gregarious in brushy forest- 

 edges, shrub 30-120 cm. high, branches numerous, erect, red, flower yellowish- 

 green, 1820 m., June 25, 1946, 16426; ibid., abundant on sunny seepage-slopes 

 in grasslands, herb 10-30 cm. high, reddish, flowers green, 2000 m., June 27, 

 1946, 16472 (TYPUS in Herb. Kew.). Without precise locality, 1891, /. Buchanan 

 987 (Herb. Kew.). 



P. hutchinsonianus S. Moore was based on a specimen collected by Swynnerton 

 in the Chimanimani Mountains of Portuguese East Africa. When Hutchinson dealt 

 with this species for the Flora of tropical Africa he identified with it some speci- 

 mens from Nyasaland and one from Tanganyika Territory. With the material then 

 available this treatment was probably justified. Since then more specimens from 

 Nyasaland and Southern Rhodesia have come in, which show that the Nyasaland 

 plants differ constantly in several points from those from Southern Rhodesia and 

 Portuguese East Africa, and I am therefore describing the former as a new spe- 

 cies, P. confusus. I have not seen the Tanganyika specimen (Goetze 794 from 

 the Livingstone Mountains); from the locality it is more likely to be P. confusus 

 than P. hutchinsonianus. 



P. confusus differs from P. hutchinsonianus mainly in that the fruits are borne 

 on shorter pedicels 1-3 mm. long (3-5 mm. in P. hutchinsonianus), in the shorter 

 tepals on the fruit, 2.5-3 mm. long (4-4.5 mm. in P. hutchinsonianus), in the 

 smaller capsules 3 mm. across and 1.8 mm. high (4 mm. across and 3 mm. high 

 in P. hutchinsonianus), in the smaller seeds 1.4 mm. long- (2.3-2.5 mm. long in 

 P. hutchinsonianus). In addition the leaves of P. confusus are in general smaller 

 and thinner than those of P. hutchinsonianus, and they lack the incrassate and 

 somewhat recurved margins of the latter species. The two species may normally 

 be separa.ed on facies, without recourse to the quantitative fruit characters. 



P. confusus has a strong general resemblance to the South African P. 

 meyerianus Muell. Arg. (P. woodii Hutch.) but the disc-glands of that species 

 are strongly verruculose, not smooth as in P. confusus. 



Phyllanthus sp. 



Mlanje District: Mlanje Mountain; Luchenya Plateau, common on banks of a 

 forest-stream subject to flooding, shrub 40-100 cm. high, flowers green, 1820 m., 

 July 1, 1946, 16571. 



This much resembles P. confusus Brenan, but is apparently dioecious, show- 

 ing only $ flowers. I have not exactly matched it, but in the absence of flowers 

 it is not feasible to discover its exact taxonomic position. 



Phyllanthus sp. 



North Nyasa District: Nyika Plateau, grassy edges of montane forest, £ shrub 

 80-100 cm. high, stem and leaves brownish, flowers green, $ not seen, 2340 m., 

 Aug. 19, 1946, 17332. 



I have not been able to match this plant, but consider that more material, in- 

 cluding ^ flowers, is necessary to place it certainly. 



