The name Eugenia bukobensis has been consistently used for a well-known and widespread tree particularly frequent and often a co-dominant around the northern end of Lake Victoria in Uganda and Tanzania and extending from northeastern Zambia to western Zaire, southwestern Ethiopia and western Kenya. It is cited in all the relevant local literature, for example Engler in Die Vegetation der Erde 9, Pflanzenwelt Africas 3 (2): 733. 1921; Eggeling, Indigenous Trees of Uganda: 153. 1940; Hora, Check-Lists Forest Trees and Shrubs British Empire 5. Tanganyika Territory Part 1: 84. 1940; Brenan, Check-Lists Forest Trees and Shrubs British Empire 5, Tanganyika Territory Part 2: 377. 1949; Eggeling & Dale, Indigenous Trees of Uganda ed. 2: 272. 1952; White, Forest Flora of Northern Rhodesia: 302. 1962; Boutique, Flora Congo Belge, Myrtaceae: 25. 1968; Hamilton, A Field Guide to Uganda Forest Trees: 192. 1981; Friis in Fl. Ethiopia 2 (2): 75. 1996; Verdcourt in Kew Bull. 54: 47. 1999; Verdcourt in Flora of Tropical East Africa, Myrtaceae: 59 (2001). However, when Engler described Eugenia bukobensis he cited in synonymy E. cotinifolia Jacq. var. elliptica (Lam.) Bak. ex Engl. et Nied. (in Pflanzenwelt OstAfricas C: 287. 1895), which is based on Eugenia elliptica Lam. (Encycl. 3: 206. 1789), a species described from Mauritian material and not at all related to the plant from Bukoba in Tanzania. Technically the name E. bukobensis is, therefore, illegitimate and must be typified by the type of E. elliptica from Mauritius. There seems no doubt that Engler was referring back to a misidentification; his description is of a new species which had been misidentified in 1895 and he naturally drew attention to this. Unfortunately his method of doing so renders his own name illegitimate under the present Code. Conservation with a new type is therefore proposed here. In the protologue of E. bukobensis, Engler cited five Stuhlmann specimens which were all destroyed during World War II; I have therefore selected a specimen collected more recently just north of Bukoba as the conserved type. The tree is of ornamental value, mainly because of its beautiful white flowers and is sometimes planted as a live fence (Katende A.B. et al., Catalogue of economically important plants in Uganda. Botany Dept. Makerere University, Kampala, Uganda. 1998). Perhaps there is a great potential for developing this aspect as researchers in Uganda are seeking to encourage use of indigenous ornamentals. The fruits are widely eaten raw (Katende et al., Useful trees and shrubs for Uganda. Regional Soil Conservation, RSCU/SIDA. Technical Handbook Series 10, Nairobi. 1995; Bukenya Ziraba, R. The non-cultivated edible plants of Uganda. NAPRECA Monograph Series No 9. Addis Ababa, Ethiopia. 1996; Bukenya-Ziraba, R. Some non-cultivated edible plants of Uganda. In Timberlake, J. & Kativu, S. (eds.). African Plants. Biodiversity, Taxonomy and Uses. Proc. of the 1957 A.E.T.F.A.T. Congress, Harare, Zimbabwe, 1999). F. White tentatively sank E. bukobensis into E. capensis (Eckl. & Zeyh.) Sond. subsp. nyassensis (Engl.) F. White (Fl. Zamb.: 4: 190. 1978), but we do not think this is correct. Recent workers on the genus in southern Africa (cf. E. Retief in T. H. Arnold & B. C. de Wet, Plants of southern Africa: names and distributions. Mem. Bot. Surv. S. Afr. 62: 523. 1993) do not accept his work on the genus. There are no synonyms and the alternative to conservation would be to produce a new name which would be confusing and irritating to local workers seeing that the current name has been in use for over a hundred
Read full abstract