his opportune appearance was all the greater for the disquieting news I had just received. As soon as he realized the seriousness of the transport problem before me, he offered to let a dozen of the camels he had hired from Tun-huang for Khotan be exchanged for the Charkhlik animals whose owners were clamouring to return. In addition he readily agreed to come back himself from Charkhlik, where his caravan was to halt, in order to take charge on animals of his own of the loads of antiques I was anxious to dispatch safely to Kâshgar before setting out for our desert explorations. Thus was solved the question of transporting our heavy baggage and stores to Tunhuang, as well as the fodder supply which our fourteen ponies would need on that long journey.
But great difficulties still remained. I had assigned to Lâl Singh, who in spite of the hardships already undergone was eager to be as soon as possible at work again, the task of carrying out an exact survey of the ancient river-bed, the ` Kuruk-daryâ ' and its branches, by which the waters of the Konche-daryâ once reached the area, now an absolute desert, south of the Kuruk-tâgh foot-hills. Through this had passed the earliest Chinese route into the Tarim basin, as marked by the ruins of the Lou-lan Site. The latter, first discovered by Dr. Hedin in 1900 and explored by me
in 1906, was to be our rendezvous. It was impossible to spare any of our own camels for Lal Singh's trying journey, nor any of the hired animals I had managed to collect, even if the owners of these could have been trusted to face such an expedition. I was therefore obliged to let him start on January 23rd with hired ponies, and proceed northward by the Tarim to Tikenlik ; I hoped that he would there be able to pick up the camels I had bespoken four months before from Kâshgar, having asked Abdur-rahim, the hardy hunter from Tikenlik and Lâ,l Singh's old guide in the Kuruk-tâgh, to hold them ready. No assurance had ever reached me that Abdur-rahim had received my instructions, nor was I even sure as to his whereabouts, to say nothing of the worrying doubt whether official obstruction from Urumchi might not defeat the arrangement.
As regards Surveyor Muhammad Yâqûb Khan, the question of his useful employment during
our explorations in the Lop Desert was complicated not merely by difficulties of transport but also by other considerations. I had originally planned that he should carry a survey round the easternmost limits of the present marshes of the Kara-koshun, where the salt wastes of the ancient but now dried-up Lop Sea adjoin the ` Lop-nor ' of the maps, and thence explore the north-east shores of the former to about the latitude of the Lou-lan Site, where he would have rejoined me. This programme, which presented considerable geographical interest, had, to my great regret, to be abandoned. On the one hand it was found impossible to provide Muhammad Yaqûb's little party with enough camels to carry sufficient ice for at least three weeks' work in an unexplored and waterless desert. On the other hand experience during our journey had convinced me that the Surveyor, however willing and brave by nature, could not be employed on an independent task of this kind without serious risk to his own and his party's safety. So I decided instead to send him with five camels by the desert track leading along the southern shore of the great salt-encrusted sea basin to a point near Kum-kuduk•(Map No. 32. D. 4) where on my previous journey we had approximately located the easternmost extension of the ancient Lop Sea. Thence he was to carry a line of exact levelling towards the termination of the Su-lo-ho drainage with a view to determining the geographical relation of the latter to the terminal basin of the Tarim more definitely than had previously been possible.
Among the tasks I had planned as my own the chief were the excavation of any ancient remains
that the intended exploration of the dried-up delta of the ` Kuruk-daryâ ' might reveal, and the search for, and exact determination of, the ancient Chinese route once leading from the Lou-lan Site eastwards to the terminal point of the old border wall west of Tun-huang. In order to assure adequate time for the latter rather hazardous task and for the survey of the unexplored northern