2l8 
Bews . — The South-east African Flora : 
situations are very moist, others very dry ; some are shady, others fully 
exposed to the intense light of high altitudes ; some are free from frosts 
owing to the rapid cold air drainage, others near at hand are not, and 
all these varied types of habitat are usually thoroughly mixed up in any 
small area. 
Judging from the number of endemic species characteristic of mountain 
ranges, such variable and unstable conditions are favourable for the produc- 
tion of new species. These may be very rare, but in other cases mountain 
species extend for immense distances along the ranges without descending to 
lower altitudes. Mountain ranges are, therefore, looked upon as great 
highways of migration for their own characteristic flora, and further, like the 
river valleys, they act as barriers to migration across them. 
There is considerable difference of opinion regarding the origin of the 
temperate African flora. It is richest in numbers in the south-west of the 
Cape Colony, a region climatically most suited to it, and there it occurs 
down to sea-level. Eastward, as soon as the region of summer rainfall is 
entered, it becomes entirely a mountain flora. It was first investigated at the 
Cape and is best known there. It is natural to speak of the occurrences of 
Ericas, Proteas, &c., on the Drakensberg as ‘ outliers ’ of the south-west 
flora, a term which, to a certain extent, assumes an origin for it in the 
south-west. It has connexions with the flora of Australia and South 
America, and it is therefore looked upon as the remnants of the flora of a 
former temperate Antarctic continent. It is not, on the whole, phylogeneti- 
cally an old flora in spite of assumptions to the contrary, and it is extremely 
doubtful whether any Antarctic continent has existed since the rise of 
the Angiosperms. 
Other authorities believe that the mountain and South-western African 
flora has come from the north. The original immigrants travelled south 
along the mountain ranges crossing the Equator, and when they reached the 
more temperate south-western areas developed enormously and produced 
the great numbers of new species which now occur there. 
It is not, of course, necessary to assume single centres of origin for any 
of the widespread component elements of this flora (such as the Ericaceae). 
So long as we deal with the larger groups such as the families and large 
genera there are many reasons for believing 4 multiple origins ’ or polygenesis 
as the most likely, and the recent developments of genetics show that the 
polygenesis of species is also extremely probable. Without, therefore, 
arguing further concerning the exact geographical origin of the temperate 
African flora, we may turn to the question of real interest, viz. its relationship 
to the tropical-subtropical flora, with which geographically it is so closely 
associated. 
Sinnott and Bailey (6) have brought forward much evidence from 
palaeobotany, anatomy, and phylogeny to show that the tropical woody 
