SECTION III.—RESUMED LABOURS AT MIRAN
The close vicinity to Charkhlik of the two sites just described and the consequent facility of employing comparatively large numbers of diggers had allowed me to complete their clearing by the evening of January 14th, and I was glad of this. The state of unwonted animation in which the little oasis was kept by requisitions for the troops marching on towards Keriya, by executions of captured ` revolutionaries ', &c., had only added to the difficulty of securing the transport and supplies needed for my future desert explorations. I felt the strain of delay all the more because I knew that out of the limited winter season which alone could be used for these explorations some time would have to be devoted to supplementary work at the site of Mirân before setting out for the main tasks northward. Only a small portion of the additional camels and food-stuffs required had been procured when this consideration of time obliged me to leave Charkhlik for Mirân on January 15th.
The last day of my stay had brought me the great satisfaction of seeing R. B. Lai Singh rejoin me in safety after fully four months of separation. Having left me in September beyond the Chichiklik pass, he had pushed on by forced marches through Yarkand and Khotan and had been able by the middle of October, in accordance with my instructions, to start triangulation of the main K`un-lun range from near Kapa (Map No. 23. B. 2), where our triangulation of 1906 had reached its eastern end. The work had to be carried on at great elevations and, owing to the lateness of the season and the total absence of local resources, was attended by very considerable hardships. But my indefatigable assistant faced them with the zeal and endurance of which he had so often given proof, and succeeded in extending his system of triangles, along with a careful plane-table survey, along the northernmost range, for over five degrees of longitude eastward, before excessive cold and snowfall obliged him to desist in the mountains to the north-east of Lop-nor. A full account of the survey work thus accomplished has been given in the Memoir on the maps embodying the surveys of all my three journeys.1
Not satisfied with having pushed this task as far as climatic conditions would permit, he continued his survey work with the plane-table along the route that leads towards Tun-huang through the inhospitable outer ranges of the Altin-tagh, snow-covered at the time. After reaching the small oasis of Nan-hu, explored on my second journey, he struck through the desert north and returned to join me by the track leading along the southern shore of the dried-up ancient sea of Lop. The difficulties of this track, fully described in Serindia 2 and the only one through the Lop desert that now, as in Marco Polo's time, is practicable for caravans, were illustrated by the fact that Lai Singh's party found no ice yet formed at the most brackish of the springs along it, and consequently suffered much from the want of drinkable water.3
The two marches that brought me to Mirân led along the desert track already twice followed by us in 1906-7 and offered no opportunity for fresh observations. Nor had any change, in the interval, come over the ruined site which extends to the east of the present course of the Mirân or Jahan-sai river, and which marks the position of the earliest capital of the ` Kingdom of Shanshan or Lou-lan ', corresponding to the present Lop region. Its remains, as explored in 1907, and the abundant finds of interest they had yielded, have been fully described in Serindia.4 There, too, will be found discussed all questions relating to the role that the site of Mirân has played in
1 See Memoir on Maps, pp. 28, 109 sqq.
2 Cf. Serindia, ii. pp. 549 sqq.
3 It was, no doubt, the same cause that accounted for the break-down of the small caravan whose misfortunes