refuse, covered the ground to a distance of about io feet from the foot of the tower. By clearing this we recovered seven Chinese records on wood ; the fragment of a large tamarisk stick, T. xLiv. a. 03 (Pl. XLVI), bearing Chinese characters at one end and charred at the other ; also a number of small wooden objects, a horse-hair strainer and fragments of coarse fabrics described in the List below. Quite close to the south-west foot of the tower were found the fragment of a Wu-chu coin and a bronze arrow-head of Han type, such as is common along the Limes. The plentiful potsherds in the vicinity were also of the usual Han type, dark grey and mat or string-marked. There could be no doubt that the period of occupation was approximately contemporary with that of the Limes elsewhere. [With this conclusion M. Maspero's notes on the inscribed Chinese slips agree. T. XLIV. a. oi8 forms part of a private letter with touching references to the condition of the exiled writer. In another letter the officer commanding a Ping 42 of which the name appears to be incomplete complains of his inadequate pay, & c. I n 019 the receipt and dispatch of certain letters are recorded.]
Less than three miles to the south-east was found the tower T. )(LINT. b, seen in Fig. 222, rising in very fair preservation to a height of 2I feet. I is construction closely agreed with that of T. XLIV. a, but its base, correctly orientated, measured 27 feet square. Only at the western corner had a small breach been made, and practically no effect of wind-erosion was noticeable at the foot. A layer of refuse, quite 6 feet high, adjoined on the north-eastern side, and this had protected a hurdle revetment, made of tamarisk branches and secured by twisted reed strings, of which Pl. 16 shows an elevation. No doubt, a similar revetment had once protected all sides of the tower, which accounted for its good preservation.
The top of the tower was covered with a good deal of straw and refuse. Amidst this, two inscribed Chinese tablets were picked up practically on the surface and yet with their writing intact. Here, too, was found the pottery jar T. XLIV. b. 015 (Pl. XXV) with a string still tied round its neck and a hole at the bottom closed by two tiny wooden boards, stuck close together. From the refuse below were recovered nearly two dozen Chinese records in the form of wooden slips, as well as an unfinished and uninscribed wooden tablet, T. XLIV. b. 034. This shows at one end a raised seal socket with five string-grooves, an arrangement otherwise unusual but found also in Chinese wooden tablets from the ancient castrum L.E. to the north-east of the Lou-lan site.2a Among the numerous small objects in wood, described in the List, may be mentioned several square prisms, T. XLIV. b. oi, 04-6 (Pl. XLVII), which may have served as ` chessmen ', and T. XLIV. b. 037, which looks like a counter for some game. The small wooden spatulas, 02, OI I, 035-6, like those found at stations in sections of the Limes to the west, were perhaps used as eating-sticks' for rice, &c.
[Among the documents noted by M. Maspero T. XLIII.024,026 are of interest as showing dates which correspond to the years A. D. 62 and 112, respectively. The former mentions the commandant of the ` signal post ' (or company ?) Chên-chung m [p, referred to also in T. XLIII. a. 013. In T. XLIV. b. 2 the commandant of the same post is referred to by name and official orders for certain
officers mentioned as received from N an-ho f â . The name of the locality of Wan-nien &F,
mentioned also in T. XLIII. a. 018, is of interest as it occurs in Chavannes, Doc., No. 415 (T. xviii. i. 2 1), as that of a canton in the sub-prefecture of Lung-lo west of Tun-huang.3 The reference made to a locality so distant is curious, and so is also that to Yit-mên in T. XLIV. b. 3. It occurs in connexion with an order sent to the commandant of a fing for a levy of men. The draft of a letter, T. XLIV. b. 017, contains significant complaints of an exile serving on this inhospitable border.]
2 Regarding the term t`ing, frequent in the documents 2a See above, p. 262.
from the Tun-huang Limes, cf. Serindia, ii. pp. 747 sqq. 3 Cf. Serindia, ii. pp. 62o sq., 69o, &c.