The Identity of Hypseloderma jubense (Sapindaceae)

Abstract

Recent papers (Hall 1980, Friis 1984) have discussed the affinity and status of Hypseloderma Radlk. (1932), a monotypic genus described from the riverine forest of the Juba River in southern Somalia. It was concluded that the genus would appear to be an isolated endemic of the Zanzibar-Inhambane region (as defined by White 1983). No related taxa could be found in the tribe Schleichereae Radlk., to which Hypseloderma clearly belonged on account of its three-loculed ovary in which each locule contained only one ovule; its indehiscent, unlobed, + berry-like fruit; its seeds completely enclosed in a fleshy arillode, and its paripinnate leaves. These characters, and the ones mentioned in the following text, were observed by Friis on the extant material of Hypseloderma at FT, or are described in Radlkofer's account of the floral structure, based on his dissection of the only extant flowers of the same material (Radlkofer 1932: 886). It was later realized (see note added in proof to Friis 1984) by the present authors that the material of Hypseloderma looked very much like the material of an apparently very rare species of Camptolepis Radlk., C. ramiflora (Taub.) Radlk., known only from a few collections made near Dar-es-Salaam on the Tanzanian coast.