812 REVISION OF EXTRA- TROPICAL SOUTH AFRICAN ASCLEPIADACE^. Flora of Tropical Africa should be confined to those growing north of this line : no attempt should be made to draw a boundary on the east coast as far south as Delagoa Bay, or even farther, as has been done in the three published volumes of the Flora of Tropical Africa. Perhaps it may not be out of place to make a few remarks on the geographical distribution of the South African Asclepiadacem. It is certainly interesting and well worth mentioning how very different the distribution of AsclepiadacecB in this country is from that of the Orchids, although they have many points in common, and are adapted to the same mode of fertilization by insects. While the Orchids are fairly equally divided over a belt around the south and east coasts, and are vastly decreasing in number and proportion to the other orders towards the drier interior, the AsclepiadacecB undoubtedly have their head-quarters in the east, and are most distinctly decreasing in number towards the south-west. In the south-western region we find only three genera which may be regarded as truly typical to this region — MicroloviaB. Br., Fustcgia R. Br., and Glossostephanus E. Mey. Omitting these three genera, all the AsclejnadacecB in the south-western region seem to be outposts of more or less eastern or northern genera, which have migrated in, for most of them have a decided distribution to the east or north, at any rate they can hardly be regarded as belonging to the same type of flora which by botanists is generally called the true Cape flora, of which the extremely local distributions of many species is a marked feature, and seems to tell for the age of this flora. The number of species of AsclepiadacecB in the south-western region is, considering the richness of the whole flora, so small that it would be useless to attempt to ascertain the correct proportion. It may suffice to say that there are to be found hardly fifty different species ; almost half this number being common with either the northern or south-eastern regions. The largest genus here repre- sented is Schizoglossum, with thirteen species, of which only eight are endemic in our region, whereas other five have been recorded from the south-eastern region as well. The Stapeliece mostly occur only where the flora of the Karroo has found its way through the river valleys or deep mountain passes into the south-western regions, or where no sharp boundary exists between these two regions, as is the case in the Clanwilliam district in the north-west. Other genera represented in the south-western region are Astephanus R. Br., Asclepias L., Cynanchum L., Sarcostemma R. Br., Brachystelmaria Schltr., and Secamone R.Br. These comprise seventeen species, of which only six are endemic. The '* Karroo " and the -' Upper Region " are both very similar with regard to their Asclepiadaceous flora. Here the Stapeliece, rank by far the highest in number, for here are evidently the head- quarters of the large genus Stapelia and several of the smaller allied genera, such as Huernia, Caralluma, and Duvalia. A rather distinct flora we get towards the Namaqualand coast-belt, out of which may be mentioned several characteristic genera, viz. Ectadium E. Mey., Ryssolobium E. Mey., Decabelone Dene., Hoodia Dene., and Tricho- caulon N. E. Br. Other genera represented are Microloma R. Br.,
