国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

Sec. iv]   PAST THE MAZAR-TAGH OF KHOTAN   91

D. 4). At a point called Kizil-ziârat (Map No. 8. c. i) we passed a small area of cultivation recently opened by a few Dolan families from the Abaci tract of Ak-su, and flocks of sheep belonging to Ak-su ` Bais ' were found grazing-even higher.-up the river.

Beyond Acha-dong living Toghraks became rare, and much dead forest and dead tamarisk-cones (Fig. 84) showed that this area had for a long time past ceased to receive water from the river, though a maze of dry beds evidently connected with it was passed here.' Towards the end of a thirty-one miles' march over desolate waterless ground, living vegetation became again more frequent, until at last at Palâs-yepti we reached a marshy bed holding a sluggish stream. I t was said to receive its water from the Kara-köl branch of the dying Kashgar river, and the survey effected a year and a half later along the Ak-su-Tumshuk high road (Map No. 7. c. 4) has shown reason to believe that this bed represents, indeed, the final termination of the Kâshgar-darya.2 Making thence our way to the N E., past patches of abandoned cultivation and marshes, we reached on the morning of November 9th the extensive but poorly cultivated village tract known as Ghôra-chöl. It forms the southernmost portion of the Abaci canton belonging to the district of Ak-su (Map No. 12. A. 4) and receives its irrigation from the Ak-su river. Settled by indolent Dolâns who have only in modern times taken to agricultural life,3 the tract offered but scanty resources. My arrival not having been foreseen and arranged for at district head-quarters, it was only with considerable difficulty that I secured the half-dozen ponies that were needed to enable me to push up the Khotan river by forced marches.

Fortunately a great change in the Khotan river's terminal course, that had taken place since my passage in 1908, helped us considerably in shortening this journey. From a point known as Yalghuz-kum the river had turned off to the north-west into what before was probably an old dry bed of its delta (Map No. 13. A, B. I). This opportune change allowed us on November IIth to strike straight across to the SSE. of Ghöra-chöl and to reach the Khotan river end by a single march, instead of having to make the great detour towards the old route at the junction of the Ak-su and Yarkand rivers (Map No. 12. A. 4). Crossing the latter at Kochatlik, where we found a ferry installed,' we moved for about ten miles across a tamarisk-covered steppe. There numerous dry beds lined with living Toghraks clearly showed by their direction that they had once carried water derived from the Yarkand river. Some six miles beyond we passed deep ` Yârs ' recently cut by the Khotan river floods, and then arrived at the new bed, nearly a mile wide but quite dry. Water, however, was easily found, as elsewhere along the lower Khotan river, by digging a well under the steep bank of the bed. Next day a march of some twenty-eight miles along the new river-bed, itself repeatedly split up into several branches in true delta fashion, brought us to the point known as Yalghuz-kum, where it diverged from the old bed as surveyed in 1908 (Map No. 13. B. I). Most of the ground traversed by the new river-bed showed but scanty scrub, and beds of reeds had begun to grow up only near its head. This confirmed the experience gained in 1908

Journey to Ghöra-chöl.

New terminal course of Khotan river.

1 The indication of living trees and tamarisk-cones along the route followed to the north-east of Acha-kuduk (Map No. 7. D. 4), as taken from Surveyor Muhammad Yagiib's plane-table, is misleading and ought to have been corrected.

2 It is possible that the Ghöra-akin (or Göre-akin), a now completely dry terminal branch of the Kashgar-darya that

I noted in 1908 north of Tumshuk (cf. Serindia, iii. p. 1309), and again in 1915 south of Yaidc (Map No. 7. c. 4), may also at one time have reached the Yarkand river belt in this vicinity.

Perhaps the name Ghôra-chöl borne by the recently

colonized ground in the extreme south of the Abad tract of Ak-su may have some connexion with it.

3 Cf. Serindia, iii. p. 1296. Ghöra-chöl was said to have been settled under Chinese orders in 1877, on the reconquest of the ` New Dominion '. The abandoned cultivation I had noticed at Sailat-terek on my way from the Kara-köl termination was ascribed to an attempt at colonization made about a century ago.

4 We found here water flowing in a channel, about 75 feet wide and 51 feet deep in the middle, with a velocity of 1.7 foot per second.

N2