388 SURVEY OF ANCIENT STATION CH. XXXIII
between the north and south walls I had thus traced was just a little under a quarter of a mile.
It now remained to ascertain the position of the east and west walls which must once have completed the defences of this small fortified station. This proved no easy task. On the east side of the ruins, beyond the Stupa mound, I searched in vain for any sign of a continuous wall among the close-set Yardangs which furrowed the ground with the regularity of a ploughed field. Ultimately I was forced to the conclusion that the constant scouring of that terrible north-east wind and the sand it drives before it, must first have breached this eastern wall face at every point now marked by a Yardang trench, and in the end have broken down and carried off any fragments of the clay rampart that had at first survived on the tops of the erosion terraces. As these were quite narrow here and streaked by sharp-cut smaller ridges, the process could be fully explained ; yet its result looked very puzzling at first, until the badly breached eastern walls of ruined Chinese towns about An-hsi, surveyed half a year later, revealed to me the intermediate stages. That powerful wind, which had first breached and then completely effaced the east wall, had, of course, been at work on the west wall, too, which equally lay in the line of its progress. After the first obstacle had been literally blown away altogether, the force of its scouring attack on the second must have increased still further.
So the total disappearance of the western face of the enclosing wall proper was a fact which did not
surprise me. But luckily I could still determine the
line it had once occupied by two decayed mounds of stamped clay which rose facing each other exactly half-way between the extant northern and southern wall faces and to the west of the main group of ruined dwellings. A close examination showed that they were the remains of two massive bastions or towers which had once flanked the main gate of the little fortified station. They must have been over twenty-four feet in thickness, and still rose to sixteen feet or so above the intervening ground. This was strewn with pieces of heavy timber, evidently the
IIIFカラー高解像度
