CH. VII RELICS OF TIBETAN OCCUPATION 113
number of observations made it appear a priori very probable that here a site of considerable antiquity had after abandonment been reoccupied, as I had noted before at the site of Endere, between Niya and Charchan.
It seemed hard to leave behind such a promising site, even for a time, without exhausting it. But a variety of practical considerations, mainly connected with climatic conditions, made the postponement imperative. And here I may point out, by the way, that only by such carefully planned adaptation to the widely different climatic conditions prevailing in the various regions did it become possible to extend our geographical and archaeological explorations over the vast areas covered by the surveys of my second and third expeditions. So it came about that the clearing of that desolate Tibetan stronghold could not be resumed by us until my return from the absolute desert by January 23, 1907. Camp was then pitched close under the walls of the fort. But the hope that they would afford some protection from the icy winds ever sweeping this desert glacis of the mountains proved vain; so frequently they veered round to catch us.
The ruined fort quite fulfilled the promise held out by the first experimental digging. The rooms and half-underground hovels which had sheltered its Tibetan garrison during the eighth to the ninth century A.D. were rough enough in design and construction, but proved to contain in some respects the most remarkable refuse accumulations it has ever fallen to my lot to clear. In the midst of inconceivable dirt, sweepings from the hearth, litter of straw, remnants of ragged clothing and broken implements, there were to be picked up in plenty Tibetan documents on wood
I
IIIF Color HighRes
