| CIi. XXXVIII EXCAVATIONS RESUMED 439 seemed quite close, and in the early morning I found myself with surprise in an atmosphere full of mist and hoar-frost. As we marched out to the ruins the mist lifted, and when the clouds in the south also cleared off for a short time, we could see that snow was steadily falling on the distant mountains. It was strange to be reminded in this terribly barren plain that there was still such a thing as atmospheric moisture. Arrived at the ruined fort I had the camp pitched close under its walls (Fig. 135), in the hope that they might afford us some protection soon, when the icy winds should be loose again to sweep the desert glacis. The camels were sent back to the narrow jungle belt by the Miran stream to find there such grazing as dead leaves of wild poplars and dry roots could offer. The ease with which an ample supply of ice could be assured here meant a great advantage for us all, and spared me the usual anxieties about water transport. Then we set promptly to work to continue the systematic clearing of the interior where my trial excavation had stopped a month and a half earlier. It did not take long to get proof that the ruined fort was' likely to fulfil the promise held out by the first experimental digging. When I examined the ground between the row of apartments previously excavated and the eastern fort wall behind them, another line of small rooms was laid bare, built casemate-fashion partly into the rampart, and also crammed to the roof with refuse of all sorts (Fig. 134). In the midst of inconceivable dirt, sweepings from the hearth, litter of straw, remnants of old clothing and implements, and leavings of a yet more unsavoury kind, there were to be picked up in plenty Tibetan documents on wood and paper, fragments in many cases, but often quite complete (Fig. 136). From a single small apartment, measuring only some eleven by seven feet, and still retaining in parts its smoke-begrimed wall-plaster, we recovered over a hundred such pieces. The amount of decayed animal and vegetable matter which had found a resting-place in these walled-in dustbins had often caused the remains of written records to be encrusted so thickly that it required much attention and care to spot and extract them. An all- | | | |
IIIFカラー高解像度
