Sec. i] BY THE EASTERN COAST OF THE DRIED-UP SEA 319
twenty-four miles over the same difficult salt surface before they reached soft shay beyond the edge of the eastern belt of Yardangs (Map No. 32. c. 3).
The lines of terraces he here encountered appear to have been of the same type as those we had
to make our way through to the south-east of Camp ciii. But owing probably to a more rapid rise of the ground, salt-incrustation ceased sooner, and an open gravel Sai was reached within about nine miles from Camp 9o. Thence Lâl Singh's route, as the map shows, crossed the westernmost offshoots of a succession of low barren ranges all running parallel to that which overlooks the Bésh-toghrak valley from the north. This agreed with what we had been able to observe on our way along the eastern coast-line, and fully confirmed the view already expressed by Dr. Hedin as to the non-existence of a continuous range bordering that side of the sea-bed.' Lal Singh did not meet with any vegetation whatsoever until he descended into the Bésh-toghrak valley near the point where we found his Camp 93. But when passing some six miles to the south-east of Camp 91 between two outliers of those barren hill chains, they came twice upon the footprints of a string of camels and of a solitary horseman. They were half effaced in the gravel and, to the experienced eyes of Abdurrahim, seemed several years old. He took them to mark the passage of some Mongols making for Tun-huang from the western Kuruk-tagh.
Abdurrahim, when questioned by me about this point, gave me interesting information which
bears out this last inference. He remembered hearing from his father, who coming from Deghar had established the little colony at Singer in the Kuruk-tagh and died there as an octogenarian, that his grandfather, who like his father had been a hunter of wild camels and familiar with the wastes of the Kuruk-tâgh, knew vaguely of a route leading through them to the Tun-huang side. This grandfather was believed to have died in his hundredth year. The Kuruk-tâgh valleys to the west of Singer are known to have been much frequented by Mongols from the mountains about Kara-shahr in the times preceding the great Tungan rebellion,7a and it appears to me likely that the family tradition related by Abdurrahim was originally derived from a Mongol source. Migrations of Mongol families from the Central Tien-shan to the mountains south of Tun-huang and An-hsi take place occasionally even nowadays. As these hardy nomads generally like to keep off the great lines of traffic, some more enterprising individuals among them may well have been tempted recently, as in former times, to make their way by the most direct line through the desert region of the Kuruk-tagh and the westernmost Pei-shan.
The watering of our brave camels on that first day of our halt was a long business, and
threatened to exhaust for a time the scanty supply of rather brackish water oozing out at the bottom of our well. It was still in progress during the afternoon of March 7th when a dust-cloud was noticed approaching by the track from the south-west. It proved to be the party at the head of my convoy from Mirân, bringing the ponies laden with fodder supplies, also my hapless Chinese secretary, alive but as silent and inert as ever. It was followed before nightfall by the hired camels with our heavy baggage, under the care of ever faithful Ibrahim Beg. With his accustomed energy he had managed, in spite of Loplik indolence and the poor condition of the hired camels, to secure the timely start of the caravan from Mirân and to bring it safely through to meet us. Thus within less than three weeks of our separation at the Lou-lan Site the anxiously awaited con-
centration of my several columns was successfully achieved.
Even if our own camels had not been urgently in need of a short rest after what they had gone through, another two days' halt at Kum-kuduk was rendered imperative by the heavy bags of mails which had arrived with the caravan. The largest among them had come direct via Khotan under the care of Badruddin Khan's old Dâk carrier Turdi, the same who, on my second journey,
7 Cf. Iiedin, Central Asia, ii. p. '14. 73 See below, Chap. xx. sec. ii.