| || || || |
820 KUCHA AND SOME OF ITS ANCIENT SITES [Chap. XXIII
of the Inchike-darya, the main continuation of the Muz-art river, and also along those of the lower Tarim. This practice has distinct antiquarian interest, as I have already pointed out elsewhere ;1 for the Former Han Annals record early Chinese efforts to establish a military agricultural colony in the territory of Ch`ii-li •fp. This territory comprised the riverine tracts along the lower course of the Muz-art river and the Tarim to the east of Shahyar.
On the morning of May 3rd we resumed our march to the ENE., and passed more lately inundated ground, still holding fresh water in pools. After proceeding about four miles we reached another rough earthen circumvallation, resembling the ` Chong-shahr ' of the preceding day. It was of irregular oblong shape, measuring about 200 yards from north to south ; its ramparts rose about 15 feet above the shôr-permeated soil of the enclosed area. The abundànt grazing afforded by reeds and scrub in the neighbourhood suggested that this and similar circumvallations might have been constructed as places of temporary refuge for herdsmen and their flocks. But once again no indication of date could be looked for.
Leaving this salt-encrusted ground and entering a belt of tamarisk-cones, we came upon the well-marked line of an ancient canal, running to the north-east and 22 feet wide at the bottom. After following it for about three miles we reached the site of Khitai-be zâr, where Mir Sharif stated that he had carried out diggings and had found remains of paper manuscripts. We here found ruins of small structures, mostly of timber and wattle, scattered over a plateau which had apparently been formed by levelling old tamarisk-cones (see plan, Pl. 44). The ruins had been almost completely destroyed, down to the wall foundations, by repeated burrowings, and had previously suffered from wind-erosion. Near the northern end of the area, however, I was able to make out the walls, built of rough slabs of clay, of what had apparently been a small cella, measuring 11 - feet by 13 inside, and surrounded by an enclosing passage. On clearing away some refuse to the east of this, we came upon minute fragments of paper, inscribed with Brahmi characters ; they closely resembled in type a fragmentary manuscript leaf with Kuchean writing which Mir Sharif had previously shown me, stating that he had found it at this site. From this archaeological evidence it seems justifiable to conclude that these badly damaged ruins are those of a small Buddhist shrine and the monastic and other quarters once adjoining. They have perhaps been abandoned since Tang times.
Practical considerations demanded an early return to Kuchà town ; I was therefore unable to visit the ruins of Khitai-shahri, situated about three miles to the north-east. They had already been examined by Afraz-gul, on the devious route which he had followed, under my instructions, from Bugur to Kucha. According to his careful report he had found there an oblong circumvallation, measuring about 27o yards by 156 as shown by the sketch-plan reproduced in Pl. 41. His description indicated a close resemblance between it and the walled enclosure of Tonguz-bash both as regards constructive features and state of preservation. The walls, about 26 feet thick at the base, still rose in places to a height of about 18 feet. Afraz-gul observed thin layers of tamarisk brushwood interspersed at intervals of 2 feet 3 inches between the layers of clay used, in the form of pisé and rough slabs or kisek, in the construction of the walls. Within the enclosure the ground was permeated with shôr, and such remains of structures as could be traced, in the shape of clay wall foundations or debris of Toghrak timber, had been completely overturned in the course of burrowings. The same thing had occurred at two low mounds situated, as shown in the sketch-plan (Pl. 41), to the north-west of the circumvallation, which had apparently once carried structures of some kind. The shape of the nearer mound suggested to the Surveyor that it possibly marked a ruined Stûpa. Beyond these mounds in the same direction he noticed
1 Cf. Serindia, iii. p. 1236.