83 
degree to the cutting action of wind-driven sand. Where we see 
drifting sand dunes, as on the south coast, it does not seem to 
matter much whether the ground is high or low; the dunes march 
forward before the ])revailing winds over high and low ground 
without discrimination. There is no visible tendency to build up 
on one spot more than another, excej)! right along the beaches, 
where the loose sand thrown n]) from the sea is first stopped by 
the defence of vegetation. Sand-hills build up most characteristi- 
cally along this frontier, l)etween the fresh supplies of sand ever 
throAvn on the beach by the waves and the barrier opposed to its 
inward drift by the fixing action of vegetation. Among the inland 
lakes excellent examples of sandhills round the margins of the 
lakes are quite commonly seen, for example on the south side of 
the Siberia lake on the road from Davyhnrst to Siberia, 
and on the shores of Lake Koorkoordine on the 
road from Southern Cross to Koolyaiiobbin. And just 
as we often find marginal sand dunes round the existing salt 
lakes, so the main belts of sandy country corre.spond with the posi- 
tion in which we might exjiect sand dunes to have lieen formed 
round the older and larger lake-beds, which now form the brown 
soil plains. They are still in ])osition where they were formed 
when the lakes were filled with water, and are therefore on the 
high country separating the lake basins. 
While on this suliject it is instructive to notice the very dif- 
ferent type of surface prevailing in the sand dune country from 
that on the hrown-soil plains. Where the surface has been moved 
obviously by wind it is uneven, and in small hills and hollows re- 
sembling waves, which in cases rise to the magnitude of dunes where 
the bodies of sand are larger. Wc at once get the typical wind- 
shaped surface, exemplifying the well-known fact that material 
accumulated by wind lies in waves and drifts and not in level 
dei)osits. But how then can we ascril^e the levelness of the plains 
to wind action? Wind is not a cutter of level surfaces, but of 
uneven ones. 
It is readily admitted that windstorms of much violence are 
not uncommon on our fields, and on any fine day in summer it is 
quite usual to he able to see several whirlwind clouds of dust 
dancing over the landscape at one time. But so far as the writer's 
observation has gone, it is very uncommon, if not unknown, to see 
any signs of the ground being worn away perceptibly by dust 
storms. If they were a strongly-operating cause of removal of 
the surface soil we should expect to see quite commonly, trees 
with the top portions of their roots laid bare by the wind, and 
standing well out of the soil, but the general appearance of the 
trees is usually quite distinctly the contrary of this, the soil often 
appearing rather to have accumulated round the stems than to have 
uncovered the roots. A good deal of attention has been given to 
observing this point, and the writer is quite satisfied that the vege- 
