Pillans (1942) recognised 3 spp. in Madagascar: one of these, P. mauritiana (also in the Mascarenes, and now referred to as P. nitida) is known from Madagascar by only a single old Chapelier collection. This is a label error and that sp. is not actually present in Madagascar. Perrier (1950) only recognised a single species (P. emirnensis), which he regarded as endemic to Madagascar and made no mention of the other taxa, while authors working on the African flora later equated this with the African P. tropica. Provisionally we accept these conclusions. The Malagasy species is not longer endemic but is shared with East Tropical Africa. S. Haemmerli (University of Zurich) is currently revising the genus.