Sec. iv] THE TERRITORY OF WEI-LI AND THE MODERN KARA-KUM 781
Korla with its Chinese garrison, and is easily accessible from the high road. It was scarcely surprising that with the lulling of official apprehensions almost all these unwilling settlers had soon found their way thence to the Kara-shahr side, where they could join an older colony of their kin and obtain better openings for their hereditary pursuits.
The conclusions to which the above information had led me as regards the latest shift of the head-quarters of modern Wei-li were fully confirmed by what I observed on the occasion of our move on March 29th from the ` new town ' of Kara-kum to the northward. For about two miles clearings had been made and fields laid out amidst patches still showing luxuriant tamarisk jungle. Then after crossing an unreclaimed area, where a dry bed was pointed out to me as the head of an old river-course connected with the lagoon of Gherilghan-köl near our Camp ccli, we arrived at the southern edge of old Kara-kum ' cultivation. Reeds and scrub had overrun most of the fields, but the people who still occupied two farms had no complaints to make of the unproductiveness of their soil: A mile farther on we reached the half-deserted Bâzâr of the old town ' with its Sarai and Ya-mên buildings. Here I was able to examine a number of settlers who had stuck to their land, among them some who had come when the colony was first established about 1900. They expressed themselves quite satisfied with the quality of the soil they were cultivating, and attributed the departure of others to some extent to the damage done in certain parts of the cultivated area by the accumulation of stagnant water below rice fields, which causes shôr to develop around. It was over-abundance of canal water and neglect of drainage that accounted for this. But the reasons for the abandonment of a far greater proportion of the land were to be found, firstly in the vagrant disposition of settlers of the type described above who were ever for trying their luck in new places ; and secondly in official pressure which had forced others to give up their holdings and to take up land in the newly ` boomed ' settlement on the river. In an orchard close to the Ya-mên various fruit trees and vines were still thriving, eloquent proof that both soil and water suited their cultivation. But with signs of neglect all round, their death through want of regular irrigation could not be long deferred. Yet the canal which brought water to the distributary channels of both old and new Kara-kum had been enlarged only a few years before with the help of some two thousand men collected for this beneficent corvée from Korla, Bugur, and other oases. Where I measured its discharge higher up close to Shinega, it amounted to over 200 cubic feet of water per second, and there was nothing to prevent this being greatly increased ; for the canal took off above Korla, where the volume of the Konche-dary5. was at all seasons greatly in excess of present irrigation requirements.
Leaving the last fields of ` old Kara-kum ', the road passed across a wide level plain with abundance of scrub and reeds, where water was said to be everywhere obtainable at no great depth and to be free from salinity. Having thus covered some eight miles through a district which, given an adequate population, could readily be transformed into a fertile tract, we turned off northeastwards, when within sight of Shinega village, to the spring of Suget-bulak at the foot of the gravel glacis. A ruined tower rises near it on a low projecting terrace of the Sai and is passed by the track coming from Sanje and Yâr-karaul. It has decayed into a shapeless mass of brickwork, about 37 feet in diameter and about 10 feet high, and has evidently suffered much through the burrowings of treasure-seekers. Its position leaves no doubt that it belonged to the line of ancient watch-stations guarding the route from Ying-p`an. I could not learn of any ruin intermediate between Y. ix and Yâr-karaul, about ten miles distant ; but from the configuration of the ground it seems probable that when this tower was intact, the post of Yâr-karaul on its high Mesa could be sighted from it.
This was the last of the watch-towers of which remains could be traced along the ancient
II 5 x