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

574   THROUGH THE LOP DESERT TO TUN-HUANG   [Chap. XIV

But the stretches of deceptive desert ground intervening, and the great détours which marshy depressions were likely to cause, made it impossible for me to attempt this. Luckily the route allowed us to fix the positions of the towers with approximate precision on the plane-table. Thus it was seen that the distance from tower to tower varied considerably, with a general average of about two miles. This, too, confirmed the impression that the line which they were intended to guard had been adapted with care to the natural features of the ground. From the neighbourhood of T. x onwards I had noted frequent cart-ruts, some apparently recent, along or branching off from our route, and had drawn from them the conclusion that, desolate as the whole region seemed, it was yet at times being visited by Chinese from the Tun-huang oasis in search of fuel or marsh grazing. So I was not surprised when, on reaching after about ten miles the edge of another long-stretched depression full of luxuriant reed-beds and containing a series of spring-fed marshes (Map No. 74. D. 3), I came upon the remains of a hut and small Chinese shrine manifestly modern in their appearance.

Ruined fort   On a neck of higher ground within the depression there rose a ruined fort, T. xrv, small but of

T. xtv.   remarkably massive appearance. Fig. 183 shows it as seen from the north-east, and Fig. i 84 from

the south-west, with the gate in the west face. Its walls, built of very hard and well-laid strata of stamped clay, each about three inches thick, rose in very fair preservation to a height of close on thirty feet. Fully fifteen feet thick at their base, they formed a solid square, approximately orientated and measuring about eighty-five feet on each outer face (see plan, Plate 40). There was no trace of earlier quarters inside, and only scanty refuse from recent occupation by wayfarers. But the very massiveness of construction and the damage which the east and north walls had in spite of it suffered through erosion, as seen in Fig. 183, were enough to convey the impression of considerable antiquity.

Wide view   From the top of the little stronghold a wide and impressive view opened. To the south, the

commanded marshy depression was seen to merge soon in a belt of Toghrak and tamarisk jungle. Beyond it an

from fort.   Y P   g   g   J g   Y

absolutely bare gravel glacis rose towards the equally barren foot-hills of a great range far away, of which the snowy crest line then remained hidden. To the north-east, at least four towers, lit up by the sun behind us, could be sighted quite clearly in the distance. In faint streaks of brown. which my glasses seemed to show here and there over the flat expanse of grey in the same direction, I thought that I could still recognize remains of the line of wall of which those towers were the silent guardians. A fine position it seemed, this height of the fort wall, for a commandant surveying the whole line of watch-stations, and for those who were to look out for the signals sent along it. At a considerable distance beyond the line of towers, the sombre, barren hills of the Kuruk-tagh, rising in a succession of serrated chains and void of all life for ages past, formed a reddish-brown background. I knew that somewhere between the foot of the outer hills and the line which the towers marked the drainage of the Su-lo Ho was bound to have once cut its way westwards. But even from that commanding position it was in vain that I tried to locate it. On a later reconnaissance, too, made to the north of T. xrv, the deeply-cut bed of the river, sunk like a hidden fosse in the deceptive gravel ` Sai', escaped me, though I closely approached it.

Large ruined   But as the march continued across a sterile gravel plateau till the evening, I noticed that the

structure.   route was bringing us nearer and nearer to a wide marshy basin stretching approximately east

to west, as seen in Map No. 78. A. 3, and manifestly part of the true Su-lo Ho trough. We had been skirting its steep southern bank for about a mile, and were approaching a roughly-built and much-decayed tower, T. xv111, that stood near its edge, when the twilight showed me a huge structure rising from the low ground which fringed the basin (Fig. t 86). The first hurried inspection, made before it became quite dark, just sufficed to reveal the imposing dimensions of the building and its massive construction. But even when next morning I was able to revisit it from our camp,