REVISION OF EXTRA- TROPICAL SOUTH AFRICAN ASCLEPIADACE^. 313 Asclepias L., Cynanchum L., DcBmia R. Br., Pentarrhinum E. Mey., and several Ceropegmia, as Orthanthera Wight, Decaceras Harv. and DichcBlia Harv. Not much is known as yet about the Kalahari region, very few plants having been brought from there. The coast- belt between the mouth of Orange River and Walfish Bay is almost unexplored, especially south of Angra Pequena, Prof. Schinz and Dr. Marloth in the western, and myself in the eastern Kalahari having collected no Asclepiads in this region within our boundary. Before proceeding to the south-eastern region, I may be allowed to make a few remarks concerning the eastern limits of the Kalahari region. When Mr. Bolus wrote his valuable Sketch of the Mora of South Africa, nothing was known about these parts, and hence the author preferred to adopt Grisebach's boundaries. Since then, however, extensive collections were made in the Transvaal and the Orange Free State, through both of which countries I went myself, collecting extensively in the former, but little only in the latter, yet sufficient to give me an idea of the distribution of the main types. From my own observations and from those of others, I think that the Kalahari should be limited in the south-east by the southern boundary of the Orange Free State, and comprise only the south-western half of this country ; from the neighbourhood of Potchefstroom the line should then go north to the head-waters of the Limpopo or Crocodile River, and from there north-east to the western extremity of the Zoutpansbergen. All the country lying to the east of this line should be included in the south-eastern region, with which it has much more affinity in its flora than with the Kalahari. The south-eastern region as defined above is, as far as Asclepi- adaceoi are concerned, probably the richest country in the world. Of the fifty-four South African genera of Asclepiadacece, we find forty in this region. Some of these extend north to the tropical Africa, others are typical to the region, such as Chlorocyathus Oliv., Ilhombonema Schltr., Fanninia Harv., Periglossum Dene., Krebsia Harv., Lasiostelma Bth., Macropetaluin Burch., Sisyranthus E. Mey., Riocreuxia Dene., and Anisotoma Fenzl. With the single exception of Ectadium virgatum E. Mey., all the species of the suborder Peri- plocoidecB have their home here. As in the other features of the flora, there is a decided affinity to the tropical African species, especially to those in Angola. Little or no affinity we find to Madagascar in the presence of Pentopatia natalensis Schltr., a species being, however, so well distinguished from the Madagascarian ones that it may be regarded as a type of a well-marked subgenus, Leptopcetia, a name already proposed by Harvey. The want of marked affinity to the high mountains of tropical Africa, especially the Kilimanjaro, the Kenia, the Ruwenzori, and the highlands of Abyssinia, is best explained by the circumstance that Asclepiadacem are not fond of the windy and cloudy mountain-heights. Even here in the south- eastern region, where they are so numerous, their number is but very small on the high mountains. By far the largest genera in the south-eastern region are Schizo- JouRNAL OF Botany. — Vol. 34. [July, 1896.] y
