THE SEARCH FOR THE ANCIENT CHINESE HIGH ROAD
SECTION I.—TO THE EASTERNMOST OUTPOST OF LOU-LAN
THE halt of four days, from the 21st to the 24th of February, at Altmish-bulak, after the Halt at
fatigues and privations already undergone by men and beasts in the desert, was a much-needed Altmish-
and very helpful preparation for the difficult explorations immediately before us. The abundant bulak.
grazing on the reed-beds of the little oasis, uninviting as it might have appeared in its wintry
dryness, was a great boon for the camels, which had felt the prolonged want of food even more
than the lack of water. On their powers of endurance, -I knew, we should have mainly to depend, if
we were to extend our surveys in the waterless desert sufficiently far to complete the tasks I had
in view, with a reasonable margin of safety for us all. The camels were happily grazing and our
men free to attend to much-needed repairs, adjustment of saddles and loads, &c. ; while Lai
Singh was occupied with the record of his previous mapping preliminary for fresh surveys, &c.,
and I myself was busy with arrears of writing work and the careful preparation of our future plans.
There was nothing to distract my attention from these tasks in the surroundings offered by Position of
Altmish-bulak, delightful as this small patch of vegetation seemed to us men, too, after the dead Altmish-
world we had toiled in. The little oasis, only about a mile and a half across where widest, is situated near the point where an utterly barren valley debouches from the southernmost Kuruk-tâgh range, and lies on the most direct route between the ancient sites of Lou-lan and theTurfân basin. Though it may therefore have seen occasional traffic when those sites were occupied, yet no vestige of antiquity could be traced. Nor did the physical features, either of the small area where a fine sandy soil supports desert vegetation, or of the adjoining waste of decomposed rocky ridges, gravel, or stony ` Sai ', offer special interest. They had moreover been fully described already by Dr. Hedin.1
The atmosphere, with its dust haze raised by light north-easterly winds, betokened the approach Range
of spring, though a minimum temperature of 21 degrees Fahr. below freezing-point was still north of
registered during our stay. This, of course, precluded any chance of our obtaining a sight of the bulak. K`un-lun range to the south of the desert, and so securing a connecting link for the triangulation of the Kuruk-tâgh that I planned to carry out during the following winter. But on the last day of our stay, February 24th, the air cleared sufficiently to allow a view of a considerable portion of the utterly barren range to the north and north-east. Taking advantage of this chance for plane-table work, La1 Singh climbed a hill due north, which subsequently served as a triangulation station and is marked with the height 3,96o (actually 4,247) feet on the map.2 The view thence northward extended to the next much-broken hill range from which the drainage bed of Altmishbulak descends, and gave him a first glimpse of that region of much-decayed barren chains, with wide waterless valleys between them, in which so much of his work during the next winter was to lie.
On the same day our practical preparations were completed : our ice supply was replenished