5. A vicariant pair in sect. Fissimura

– Our novelty B. pauciflora is similar and closely related to Barleria neurophylla, a species described by Clarke (1900) from D.R.Congo. There is no isotype nor any identified material of that species in BR where it has hitherto been overlooked as being a species from Central Africa. The holotype (Carson s.n., K) comes from Kavala Island, in the Congolese part of Tanganyika Lake. There is a second Congolese collection from the western shore of the lake, Kässner 3048 (K). B. neurophylla is endemic from that part of the lake and from the next Mahali mountains, a small range fringing the eastern shore of the lake in Tanzania (fig. 11).

B. neurophylla is very closely related to B. pauciflora and has the same ecology, also growing in gallery forests, but several minor differences in characters and the disjunct distribution of the two taxa justify their treatment as distinct species. They may be considered as vicariant taxa. Several specimens of B. neurophyll a were found in K and a duplicate from K in the unidentified material of BR.

As B. neurophylla is placed by Balkwill & Balkwill (1997) in sect. Fissimura, B. pauciflora should be placed there as well. It is an unarmed plant, with bracteoles never spinous, and a corolla with 4 + 1 configuration.