268 Transactions of the Boijal Society of South Africa. 
land at some former time, but at present Kangnas is 65 miles from the 
nearest outlier to the north and about 75 from the main Dwyka area to 
the south ; the position of the Dwyka boundary to the east or south-east 
is not known exactly. 
At the present day the only ways in which material can leave the 
interior of Bushmanland are (1) wind action, (2) solution by water which 
percolates underground towards lower levels and carries away the dis- 
solved substances, and (3) by water action at the surface along the valleys. 
The third of these agents is certainly every insignificant, for one can- 
not see any channels in the valleys. No doubt every heavy rain moves 
some of the sand along the valleys, but owing to the porous nature of the 
ground and the dryness of the air in this region the low rainfall there 
cannot produce any appreciable effect by transport along the valleys, 
though it certainly tends to fill those valleys with sand. The removal of 
matter in solution is obviously a slow process, for the springs emerging 
where the bed-rock crops out at the edge of the sand- veld are few and 
small, though figures cannot be given. Wind action, like that of surface 
water, is much more efficacious in levelling up local valleys than in 
carrying material aw^ay from the district altogether. 
A consideration of the conditions now in force in Bushmanland leads 
one to believe that the valleys now existing may well represent extremely 
old, even meaozoic, valleys which have changed only through having been 
tilled up by locally derived material, provided that the climate has 
remained fairly constant throughout the intervening period. A condition 
of greater rainfall would have cleaned out the valleys, and a much 
diminished rainfall-^ would have been accompanied by the destruction 
of the vegetation, which is at present an efficient protection of the flat 
ground against the devastating effect of wind. 
The vegetation in Bushmanland may be divided into two classes — that 
which appears soon after the rain and dies off completely after seeding, 
and the short, stubby bush which has great power of resisting drought 
and never disappears under present conditions unless stamped out by 
stock on paths or round watering-places ; it is the second class that is 
of such great importance in protecting the ground from wind. The mere 
existence of this drought-resisting flora must be regarded as evidence of 
the great length of time during which the country has had a dry climate, 
though there are no means of correlating the adaptive changes in the 
plants with geological periods. 
It is well to remember that we have no conception of the time 
* There are no figures available for the interior of Bushmanland ; the rainfall there 
probably lies between 3-79 (Pella) and 5-75 (Lilyfontein) inches per year, for which 
stations figures are given in Buchan's "Discussion of the Rainfall," etc.. Cape 
Town, 1897. 
