quake which we shall examine later on near Nushki. In fact, both the
crack at Nushki and the collapse of the west side of this hill range, as
well as a great portion of that deep crack in the earth's crest in which
the Shela flows, have very likely been formed by the same cause. They are
within the same zone of volcanic formation. In the particular case of
this hill range in Afghanistan the collapse did not appear to me to be
due to the action of water, but to a sudden crumbling which had caused a
very sharp vertical cut.

[Illustration: Sand Hills.]

To the north of the salt wastes was another long belt of yellow sand
extending for some 40 miles, upon which there was absolutely no
vegetation, while intervening between the salt and this sand flat were
numerous sand barchans, like horseshoes, with a gradual slope on the
windward side (north) and a crescent hollow with a steep but not quite
vertical bank on the lee side.

I noticed all over Persia, and in Beluchistan as well as here, that these
sand barchans, or barchanes, will only form on level ground--generally on
extensive plains. All single sand hills, however, whether barchans,
conical, semi-spherical, or of more irregular shapes, are invariably
caused by a primary obstacle, however small, arresting the sand.
Various are the theories with regard to the formation of these barchans,
and especially with regard to the formation of the hollow on the lee
side.

[Illustration]

The explanation from my own observation has--if no other--at least the
merit of simplicity. The wind, on meeting the semi-circular back of the
barchan, is diverted on the two sides of it; these two currents come into
violent collision again on the lee-side, where, the air being more or
less still, a considerable portion of the wind is forcibly driven back
towards the barchan, corroding its side in a double rotatory way, each
such circle having for a diameter the radius of the barchan crescent
containing them. In fact in many barchans the sand ripples on the
windward slopes cross the direction of the wind at right angles. A line
of sand formed in the centre of the barchan crescent in the opposite
direction to the wind is often to be seen during wind storms or soon
after. I have also seen barchans, the inner crescent of which showed
beyond doubt that when there is a prevalent wind from one side only, the
above explanation, although less scientifically obscure and elaborate
than most, applies, an