National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Sec. ii]   THE SOUTH-WESTERN FLANK OF THE LIMES   641

Passing thence westwards through luxuriant Toghrak jungle, I ascended a long, tongue-like Watch-

plateau which bears, near its south-western edge, the ruined watch-tower T. v (see Plate 36). tower T. V.

Neither on its gravel-covered surface nor elsewhere along the flanking line of towers could any

indication be discovered that the wall of the Limes was ever extended in this direction beyond the

stretch seen just south of T. iv. a. The elevated position it occupies, together with its own height,

macle the tower T. v a very conspicuous object even from afar. Though a portion of its west face

had fallen, it still rose to over thirty feet in height. It measured nineteen feet square at the base,

and on its top preserved heavy beams of Toghrak wood, together with the remnant of a low brick

parapet. The material used was stamped clay in layers about three inches thick. On the north of

the tower were the remains of some small quarters, having walls of greatly varying thickness built

with bricks approximately of the size usual along the ancient Limes, about sixteen by seven and a half

inches and five inches thick.

In the little apartment marked i on the plan there were found half a dozen Chinese tablets, Finds of

mostly well preserved, including one, Doc., No. 428, Plate XII, dated 39 B. c., and another, Doc., documents

No. 429, Plate XIII, which contains a calendar for the same year.1 Among the contents, which refer at T. v.

mainly to military matters, I may note the mention made in No. 436 of ` the captain of the company

of Ta-chien-tu ', which bears out the previous suggestion about the connexion of this local name

with the south-western flank of the Limes.2 The miscellaneous finds in the quarters were few, but Wooden

they included four wooden brackets, painted in several colours, T. v. 001-4 (Plate LIV), which are brackets for

of interest as representing a class of barrack fittings regularly used in these watch-stations. They hanging

cluttl&c.

differ considerably in size, and to some extent also in style of decoration, as is seen from the

illustrations in Plate LIV and the general description given in the List below, under T. viii. oo4.

In each case, however, there is a solid tenon, square in section, for insertion in the wall, and a hooked

bracket resembling a claw turned upwards. The fact that the paint has ordinarily been rubbed off

the upper surface of the indented portion next to the claw-like end clearly shows that these brackets

were used for hanging up clothes, articles of equipment, etc., and arms, too, if they were big enough.

There were plentiful refuse-heaps on the slopes below the watch-station, proving that it had been

occupied for a long time ; but their contents were only dung of horses and camels, reed straw, and

the like.

The position occupied by T. v was typical of the watch-stations which were found to stretch Position of

far away to the south-west along the edge of the great marshy basin. The distances at which they posts chosen

were placed from each other, in the case of T. v, T. vi. a, c, d, always about five miles, clearly signalling.

showed that they were meant mainly as signalling posts along a line, and not for warding off inroads.

The objects which the wall was intended to serve along the front of the Limes were here, on its

south-western flank, secured by practically impassable marshes. Yet even thus the links of the

chain of posts could not have been kept so far apart, with the consequent saving in trouble and cost,

had not the configuration of the ground offered ideal positions all along for signalling stations visible

far away.

As I have already had occasion briefly to notice, and as the map in Plate 33 clearly brings out, Ridges

the bare gravel-covered plateau which marks the foot of the alluvial glacis of the range south of watched for

Nan-hu here stretches out a succession of finger-like ridges of clay. They project into the wide stations.

marsh-filled terminal basin of the Su-lo Ho like headlands of a coast-line indented by deep fjords,

rising with very steep faces to heights from about 120 to over 200 feet above the intervening de-

pressions and commanding extensive views, as can be seen from the vistas presented in Figs. 169, 170.

   ' Through a clerical oversight, Nos. 428, 429, 436, 439   T. iv. d.

   (Documents, pp. 95 sqq.) were given the wrong site-mark   2 See above, p. 636.

1374

4N