FROM THE ANAMBARUIN-GOL TO THE DESERT-RANGE. 481
which two camels can pass one another. Almost everywhere the sharp-cut edges of the clay erosion-terrace project through the sand, in places quite distinct, in others only adumbrated as it were; but throughout they are as sharply determined as though they had been cut with a knife. They run I'/2 to 2 m. above the bottom of the watercourse. The material is clay. Very often they form projecting cornices, strewn with coarse sand. This formation would appear to originate in this way: layers of clay and sand are disposed one upon the other, and then the superimposed layer of clay, offering greater resistance than the underlying sand, is left hanging without support. Every now and again, however, these overhanging cornices break off; for in various places we found their crumbling remains littering the bed of the watercourse.
Fig. 209.
But nowhere did we observe any indications that water had recently flowed down that way, and the bottom was of course as dry as tinder. Nor do the sharp-cut edges of the original erosion channel remain, for the terraced banks, of which I have just spoken above, are higher than the original edges, and have been produced in part by the crumbling away of the original edges, in part by the erosive action of the wind. Nevertheless this remarkably clear-drawn hollow through the drift-sand is alone sufficient to prove that, at intervals of perhaps five or ten years, water does flow down from the northern versant of the little desert-range, and it must be after short, but sharp, showers. Did the winds blow here with the same regularity that they exhibit in the Desert of Tschertschen or the Desert of Lop, such a narrow channel as that would not be able to keep itself open so long, but it would be filled up by sand coming from the direction of the prevailing wind. So far as I was able to ascertain whilst crossing that part of the desert, the winds there are indeed locally less regular and more variable. And yet the region must be visited by violent storms, only they blow rather from the west and west-north-west; at any rate it was from that direction that they had come last, for most of the dunes turned their leeward slopes very decidedly towards the east and east-south-east. Probably the wind blows predominantly from that direction in summer, autumn or winter,
IIIFカラー高解像度
