horizontal; but since it has been exposed to wind and weather it has fallen an easy prey to their machinations. A slight irregularity at one point has given rise to a grooving, and down this a rainwater torrent has made its way in the rainy season, while during the dry season the gully has been excavated deeper and deeper by the wind. The reason that they have not been cut down to a greater depth than what they now possess — they are seldom more than two meters deep — is of course this, that the surface of the tabular elevations and platforms themselves are in their turn exposed to the filing and planing actions of the wind. One difference is noticeable between the gypsum formations in the lake depression and those which we studied on the shores of the Lakor-tso. In the case of the latter they are softer and more rounded, but here harder and mostly with vertical sides. A strange, an extraordinary landscape, — deathly white, flashing in the sunshine! It is however the same phenomenon as that which we encountered in the Desert of Lop, the principal difference being the diversity of the material.
Next we travelled for a space on hard saj, then crossed over an expanse of level sedimentary clay, and then once more entered a gypsum region, its elevations rising on both sides of us. At last however the old lake bottom came to an end, tapering away to a point in the west, close to the foot of the southern mountains. Beyond that the ground became hard, with a thin sprinkling of fine gravel, and some grass and scrub, though these latter but sparingly. Close to the western margin of
Fig. 150. THE MOUTH OF THE GORGE SE OF CAMP CXXXIII.
IIIF Color HighRes
