Commelina mascarenica (Commelinaceae): an overlooked Malagasy species in Africa Faden, Robert B. Adansonia 2008 3 30 1 47 55 XFN7 [677,1042,945,971] Liliopsida Commelinaceae Commelina Plantae Commelinales 1 54 Tracheophyta species mascarenica  The identification of  Commelinasp. Dand  Commelinasp.#5 as  C. mascarenicahappened as a result of a study that I began in 2000 of an apparently undescribed species of  Commelinafrom coastal and subcoastal Kenyaand Tanzania. That species (Faden in press) also has elongate, smooth seeds, so it became necessary to determine whether it might actually be  C. mascarenica. I had not studied that species before, so in March 2002, in a loan from the Muséum national d’Histoire naturelle, Paris (P) to the Royal Botanic Gardens, Kew (K), I examined 14 of the 15 collections – all except Perrier de la Bâthie 9018, which was not sent – that were cited as  C. mascarenicain Flore de Madagascar( Perrier de la Bâthie 1938). Thirteen specimens appeared to belong to a species that significantly differed from my new East African species. The fourteenth specimen, Bojer s.n., was a mixed collection, one part of which – designated by Perrier de la Bâthie (1938)as “ Bojer B” – was indeed my new East African  Commelinaspecies, and was not conspecific with the other 13 collections. The other part of the Bojer collection, “ Bojer A”, belonged to an unrelated species,  C. lyalliiC.B.Clarke. The 13 remaining collections from Madagascarand the Comoro Islandsthat were treated by Perrier de la Bâthie (1936, 1938) as  C. mascarenicawere recognized as conspecific with my  Commelinasp. Dand  Commelinasp.#5 because some of them, e.g., Decary 791(P) from the Comoro Islands, had mature seeds that were identical to those of the African plants. However, because I now had evidence that my putative new species also occurred in Madagascar, it became necessary to examine the typeof  C. mascarenica, in order to determine to which species the name should be applied. In August 2002, I was able to study the typeof  C. mascarenica, sent on loan to Kew from G. Although the typespecimen was rather scrappy,and the capsule and seeds in the packet were immature, I had no doubt that this was same species as the 13 collections examined from P that were cited as  C. mascarenicain Flore de Madagascarand also matched African  Commelinasp. Dand  Commelinasp.#5. Therefore the African plants were  C. mascarenicatoo. The first reference to  C. mascarenicain Africa was by Faden (2006), in an appendix to the Flora of Somalia. The species name was applied to a previously unnamed species (  Commelinasp.#5) and the species was noted to also occur in Kenya, Tanzania, Mozambique, Comoro Islandsand Madagascar.