THE SAND-BURIED RUINS OF CHIRA 181
able from the erosion of surrounding areas. Thus the maximum size of the mounds is determined by the extent to which the water supply has diminished. The actual size, of course, depends partly on the length of time since the water was withdrawn. Judging from the relation of mounds to ruins, I should say that mounds fifty or sixty feet high must be nearly two thousand years old, and those twenty-five feet high from five hundred to a thousand or more. Sometimes tamarisk mounds are formed without a change in the water supply, but such can be easily distinguished from those of which we are speaking. They are characterized by a rounded form, by a vigorous growth of bushes on all sides as well as on the top, and by the fact that they are composed entirely of æolian sand, and not partly of sand and partly of river deposits eroded by the wind. In general, tamarisk mounds are an unequivocal evidence of a diminution of the water supply. When, as in the region we are now traversing, they occur over broad areas where man's activity has had nothing to do with the water supply, they furnish one of the strongest possible proofs of climatic changes.
At the end of our first day from Malakalagan, when we had almost reached the border between the tamarisk mounds and the sandy desert, we came upon the pottery-strewn site of a very ancient town, lying fifteen miles north of Karakir, on the line which the Kara Su River would follow if prolonged. Karakir, to which I made a flying visit on the way to Imamla, is now a little hamlet of about thirty houses, although in the chronicle of the Mohammedan conquest, it is spoken of, under the name of Bowa Zengir, as a village of some importance. A peculiar event took place there
IIIFカラー高解像度
