The classification to be used in forthcoming treatments (this was posted in 1999) of Rhipsalis and allied genera (Cactaceae subtribe Rhipsalidinae Britton & Rose) is briefly explained, and 26 new names proposed.

Introduction
After extensive discussion, a Working Party of the International Organization for Succulent Plant Study (IOS), of which the author was a member, recently recommended the acceptance of five genera in the Rhipsalis group (Hunt, D. R. & Taylor, N. P., eds., in Bradleya 4: 65-78(1986), Group II), with the reservation that two of the genera, Lepismium and Pfeiffera, might be combined but for the confusion likely to arise from the amplification of Lepismium (the older name) in a sense very different from that proposed by Backeberg (Die Cactaceae 2: 682-697.1959). Enquiries were subsequently made concerning the possibility of conserving the name Pfeiffera, but it is understood that a proposal to this end would be unlikely to succeed and that the priority rule would have to take its course. Even so, the author feels that the taxonomic grounds for combining the two genera must override nomenclatural considerations and the Pfeiffera-Lepismium merger should go ahead.

The nomenclatural changes necessitated by the IOS Working Party's report and by the author's decision to amplify Lepismium are made below so that the new names will be available for use in a treatment of the Cactaceae for Kubitzki et al., The Families and Genera of Vascular Plants, and in detailed treatments of Rhipsalis and allied genera which are in preparation.

The proposed framework of genera and subgenera in the Rhipsalidinae is based on morphological (vegetative and floral) and micromorphological investigations (predominantly SEM [Scanning Electron Microsopy] studies of seedcoats and epidermises) carried out by the author and others between 1971 and 1986. In addition, papers on the systematics of the family as a whole, notably the survey of pollen-morphology by B. E. Leuenberger (Diss. Bot. vol.31, Vaduz 1976), the various publications on individual species and genera by M. Kimnach (in Cact. Succ. J. (US), vols. 28-57.1956-1985) and the recently published survey of the Rhipsalidinae by S. A. Volgin (in Feddes Repert. 97: 553-564.1986) have been fully considered.

The four genera recognized here form a natural group and are probably not closely related to the other epiphytic genera of Cactaceae; there is evidence that the great resemblance with some Hylocereinae (notably the genus Pseudorhipsalis) is a result of convergent adaptation to the same epiphytic habitat.

No hybrids between the genera ( or even the subgenera) accepted here are known. On the basis of seed-morphology, and various other characters, the Rhipsalidinae seem to have had a common origin with the South American tribe Notocacteae F. Buxb. (resembling most closely the genus Corryocactus).

The character-basis and possible phylogenetic relationships of the component genera and subgenera of the Rhipsalidinae are summarized in the following key. The scheme will be justified in more detail in future papers devoted to the individual genera.