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

610   THE OASIS OF NAN-HU AND THE YANG BARRIER [Chap.XVI

I do not regret the prudent restraint which both my own feeling and Chiang Ssti-yeh's advice imposed upon me as regards these cemetery sites. I would, however, recommend them to the attention of some future archaeologist visitor—whenever ` modern progress ' may have swept away the traditional respect for the habitations of the dead even among the pious folk of that conservative back-water of Kan-su. Only let him do the work with systematic thoroughness of research and not leave too much to be destroyed by the mere greed of local exploitation that seems to be at work further east !

We camped that night near where the Tang Ho debouches from a deep canon-like depression flanked by steep conglomerate cliffs, and about three miles from where the main western canal of the

Tun-huang oasis takes off on the left bank. A march of some thirty miles, prolonged until after

nightfall, then brought me on April 6 to Nan-hu. For the first half of this distance the route kept close to the southern edge of a gravel-covered plateau which falls off with precipitous cliffs, generally

from about 8o to zoo feet high, to the deep-cut bed of the Tang Ho (see Map No. 78. c. 4). The

latter seemed like a huge fosse, with a glacis stretching away northward from the brink of its counter-scarp. Above the scarp on the south there rose the absolutely barren foot-hills of the Nan-shah,

covered here, as also to the south of Tun-huang, with those formidable dunes of drift-sand which

account for the latter's alternative name of Sua-chou, the ` City of the Sands '. Considering that these forbidding sand-covered slopes are really impassable, and that there is, judging from Captain

Roborovsky's surveys, no practicable route up the deep-cut gorge of the Tang Ho from where it

turns sharply south-east into the mountains, it was easy to realize how well protected Tun-huang was from any nomadic attacks which might otherwise proceed from the high plateaus in the south.

Both near the debouchtïre of the river and where it makes its big bend (Map No. 79. c. i) there were half-ruined watch-towers, built on the cliffs above the left bank and commanding a view of the deep-cut bed. But they bore no ancient look, nor did the two small brick Stûpas, well plastered and manifestly still receiving worship from wayfarers, which I passed at the roadside shelter of Shih-wu-fou.

It was after about two miles from the point where the route, continuing to the south-west, leaves the vicinity of the river near the above-mentioned bend, and as we were skirting the last off-

shoot of a bare ridge rising gently towards the south, that I first noticed what seemed like a low dyke,

or agger, of gravel and stones. It was the embankment which Mr. St. George Littledale, in the account of his pioneer journey of i 893, mentions as having been noticed by him on the last march

before reaching the Sha-chou oasis, and which the Royal Geographical Society's Map of Tibet has duly marked.2 His brief description of it proved quite correct, and the shrewd guess he had hazarded as regards its character seemed a priori deserving of consideration. The ` embankment' rose only four or five feet above the bare ` Sai ', and could easily have been mistaken for a natural swelling, had it not stretched away steadily to S.255°W. in a line perfectly straight and keeping close by the route for upwards of five miles. From where the route first approached it I could see it also continuing, though less distinct, in the opposite direction, until, as the plane-table showed, it must have struck the canon of the Tang Ho just at the river's sharp bend. The agger was broad, measuring about twenty-four feet at its base, and, as the surface on its top was hard, it seemed to have been used as a cart-track. Of watch-towers or any other structural remains along it I

March along left bank of Tang Ho.

Dune-covered foot-hills.

Line of ancient embankment.

' Cf. his paper A Journey across Central Asia, in Geogr. Journal, 1894, iii. p. 458: `The day before reaching Saitu our path for seven or eight miles followed an embankment which was four or five feet high and about ten yards across ; where our path branched off, the embankment continued as

tar as we could see. I never heard that the Great Wall of China extended beyond Suchau, but this certainly much resembled parts of the wall that we afterwards saw, and if it was not the wall I am quite at a loss to say what its use could have been.'