Notes

A record from Timor is erroneous. The various types of embryos are figured in F.C.B. Ochnaceae: 3, fig. 1 (1967).I have followed Robson’s clear division of the genus into three sections:• Sect. Renicarpus. Carpels and drupelets reniform attached by middle (sp. 1)• Sect. Ochna (Diporidium (Bartl. & Wendl.) Engl.). Anthers with pores (spp. 2–21)• Sect. Schizanthera Engl. Anthers with slits (spp. 22–36)Spp. 37–47 are those I have been unable to match and think may be new taxa but anthers are not available.It cannot be pretended that naming Ochna species is easy. Without young flowers with anthers it is not possible to use the key effectively although by using both leads results can be obtained. Characters are not always constant. In couplet 23 for example some of the species may have shoots and/or pedicels glabrous rather than papillose-puberulous. To make allowances for all exceptions would make the key too unwieldy. A list is given of the species occurring in each Flora area, which may help in identification. List of species occurring in the Flora areasU 1: 14, 16, 19, 28, 33, 34U 2: 1, 13, 18, 40U 3: 13, 22, 33, 34U 4: 1, 8, 13, 18, 34K 1: 4, 6, 8, 14, 16, 22K 2: 14, 16, 27K 3: 1, 8, 16, 22K 4: 8, 14, 16, 22K 5: 8, 16, 22K 6: 8, 16, 22K 7: 2, 3, 4, 5, 6, 7, 14, 16, 17, 22T 1 : 1, 7, 16, 26, 27, 28, 33, 34, 44T 2: 8, 15, 16, 22T 3: 2, 3, 4, 5, 6, 7, 16, 22T 4: 7, 16, 22, 26, 27, 28, 30, 31, 32, 33, 34T 5: 7, 16, 22, 26, 29, 33, 36T 6: 5, 6, 7, 9, 14, 16, 21, 22, 23, 25, 28, 33, 37, 39, 41, 42, 45, 47T 7: 7, 14, 15, 16, 22, 23, 24, 25, 26, 28, 29, 30, 31, 32, 33, 34, 35, 43, 46T 8: 2, 3, 5, 6, 7, 10, 11, 12, 16, 20, 21, 22, 25, 28, 33, 38Z : 4, 6P : 4, 5{/l}. ERRONEOUS RECORD: On 18 Apr. 1909 Baroa de Soutellinho sent to Kew a small specimen of an Ochna he had cultivated in Oporto from seeds said to be from “Ruwenzori, Central Africa”. Robson has identified this as the South African species O. atropurpurea DC. and the seeds could not have come from East Africa.