324 DISCOVERY OF DATED DOCUMENTS [CHAP. XX.
same cause, the difficulty of maintaining effective irrigation for these out-lying settlements.
I cannot attempt here to investigate the question to what extent this receding of the cultivated area may be attributed to neglect of irrigation works, caused probably by political troubles and consequent depopulation, or to a change in the physical conditions attending the supply of water from those streams. I may, however, with advantage call attention here to • my subsequent observations at certain villages of the Gulakhma and Domoko oases, the cultivated area of which has, owing to the difficulty of carrying the irrigation water sufficiently far, been shifted, within the memory of living men, as much as six to eight miles further to the south. The crumbling ruins of the deserted village homesteads which I saw there, stripped of all inaterials that could be of use, and the miles of once cultivated ground which the desert sand is now slowly over-running, but on which the lines of empty canals, irrigation embankments, &c., can still be • made out, were the best illustration of the process by which the lands of Dandan-Uiliq became finally merged in the desert.
In this connection I may note that our survey furnished no evidence in support of the assumption put forth by Dr. Sven Hedin, that the Keriya Darya in historical times flowed close to DandanUiliq, and that the abandonment of the site was connected with the subsequent shifting of the river to its present bed, some twenty-eight miles in a direct line further to the east. Probably, the distinguished explorer would have hesitated to make this suggestion had he known the indisputable antiquarian evidence which shows that the ruins to which, mainly on the basis of conjectural calculations as to the movement of the sand-dunes, he was prepared to assign an age of about two thousand years, were in reality abandoned only about the close of the eighth century of our era.
IIIFカラー高解像度
