230 THE DESERT EDGE OF KHOTAN CH. XIX
the way were so high and sterile as to make me feel as if I were back again in the depth of the desert. The area of moving sands is here steadily being pushed back by resumed cultivation. The lands of Ak- kul had been brought under irrigation only some fifteen years before, and my return march next morning showed how cultivation was now being gradually extended over ground abandoned for centuries to the desert. In the fertile loess soil to which new cuts carried ample water, the poplars, willows, and Jigda trees usually planted along the edges of fields were shooting up rapidly. Hence it was easy to note at a glance the new conquests made each year from the desert sands. Already the belt of luxuriant reed beds and tamarisk scrub in which the overflow from the Ak-kul canal finds its end, was approaching from one side the ground where ancient occupation has left behind its pottery débris. And I wondered whether, in spite of slowly progressing desiccation, the time was not near when, under the pressure of increased population and a growing need of land, the oasis might victoriously recover most of this desolate waste of the Hanguya Tati.
I felt grateful for this advance of human activity, since the vicinity of cultivated ground made it easy for me early next morning to attack the ancient structure with an adequate posse of labourers and without any worries about their commissariat and water-supply. Within a few hours we laid bare what proved to be the north wall of a temple cella, sixty-four feet long, adjoined by a passage which must, as in the case of the Dandan-oilik shrines, have extended all round the four sides of the cella. All along the walls of this passage the finds of small terra-cotta relievos were plentiful, and the deeper they lay buried the more frequent were the remains of the original gilding. Of the larger sculptures and of the frescoed plaster surfaces of the walls, which in all probability adorned the enclosing passage, only the scantiest indications remained in the débris covering the original floor to a height of about two and a half feet. The friable clay of which they were made had evidently crumbled away just as in the strata of Yotkan.
But the careful examination of that débris revealed
IIIF Color HighRes
