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. iii]   THE RUINED WATCH-STATIONS T. xi AND T. xii. A   669

Skirting the western edge of this plateau tongue I came upon the remains of the wall Limes again running from the edge of the marsh towards T. xii. a, in the direction of ENE. The wall wall near

is

again was traceable to within about 3o yards of the actual water line, where its remains disappeared in thick reed beds, on a level about so feet higher than that of the marsh at the time. The wall, built in the usual fashion, ran in an unbroken line, and in places still not less than 6 feet in height, as far as the ruined tower T. xii. a (Fig. i 77). There it made a sharp turn to the south-east and descended over the steep slope of the plateau to the edge of the deep swamp which fills the depression eastwards and is visible on the left of Fig. 175. This stretch of wall beyond T. xii. a could be traced till, after about 8o yards, it disappeared among the reeds fringing the swamp. The total distance covered by its length from one marsh to the other was a little under half a mile.

The tower of T. xii. a was badly decayed, rising only to about 8 feet from the level of the Watch-

ground. Its base, as the subsequent clearing proved, was about 23 feet square (see plan T'wxeir,. a.

in Plate 39). It was built of sun-dried bricks, measuring about 17 by 8 inches, with a thickness of 5 inches ; after every three courses a layer of reeds was inserted. Heavy débris found on the east and south sides showed where portions of the superstructure had fallen, and

Toghrak branches and bundles of reeds, sticking out underneath on the east side, indicated that

remains of quarters lay buried here. Mere scraping with the heel of my boot, on the occasion of

my first reconnaissance visit, sufficed to unearth here from the foot of the débris the curious little

tablet, T. xii. a. i, Doc. No. 6o6 (Plate xVIII), 4 inches square, with a boldly inscribed direction

about ` two guards to be placed at each door '. Subsequent excavation on the east side of the tower Miscel-

brought to light a considerable quantity of fragments (T. xii. a. 001-41 ; i. ooi-6), mainly from finds in

wooden fittings, household implements, clothing, etc. ; besides eight Chinese records on wood. quarters.

Owing to the heavy mass of masonry which had fallen from the tower the thin walls of the quarters

had been badly crushed, but two small rooms, 12 feet wide, could still be distinguished.

Most of the miscellaneous relics found here were of types already mentioned. Among such, e. g.,

were the carved brackets for hanging clothes, etc., T. xii. a. 001-3, and wooden seal cases, 005-

I I, 0015-18. For specimens of these as well as some others, among which are the little bow,

T. xi'. a. 0024, and the curious carved wooden finial, 0025, see Plates LIII, LIV.

In consequence of the heavy earth-work the clearing of the débris adjoining the south foot of Clearing of

asses

the tower had to be left for a second day's work under the joint care of Naik Ram Singh and âdjoin ng

Chiang Ssti-yeh, while I myself was kept away, busy with prospecting the remains to the north and tower.

north-east of T. xiv. It was then that the most interesting finds at this station were made.

Immediately against the south face of the tower was a space about 4 feet wide, which seemed to

have been filled up on purpose with broken bricks and loose earth. Next to this came a still

narrower passage (marked ii in plan), only i' io" wide, enclosed between walls of single bricks and

divided by an equally thin partition into two little compartments, each about II feet in length.

A thick layer of straw and stable refuse covered this passage as well as a little room, measuring only

5 by 6 feet, which adjoined it and the south-west corner of the tower. The passage, as I

convinced myself by subsequent inspection, had its walls still standing to a height of over

4 feet.

Refuse of all kinds had completely filled the passage, and within it was found embedded the Early Sog-

remarkable collection of Early Sogdian documents on paper, T. xii. a. ii. 1-8 (Plates CLIII-CLVII), to dian, Chi-

nese, and

be discussed presently. According to the Naik's statement, which I have every reason to accept Kharosthi

as accurate, their position was about 3 feet above the floor. In the refuse below them there turned documents.

up three Chinese slips, among them two complete ones, Doc. Nos. 607, 609. From the little