312 THE SEARCH FOR THE ANCIENT CHINESE HIGH ROAD (Chap. VIII
habitable territory of Lou-lan, I specially directed the Surveyor to revisit the ground where we first struck the salt-coated Yârdangs on the western coast of the dry sea-bed. Its vicinity was to be carefully searched by him for any features which might throw light on the direction here followed by the ancient route.
As Afrâz-gul's diary shows, he reached the Yardang belt to the east of the Mesa where we had found relics of ancient traffic in the shape of Han coins, a dagger, &c., on February 22, 1915, after two marches from Altmish-bulak. Having made his way south-eastwards between the salt-coated Yârdangs, he found a patch of open clayey ground half a mile from the well-marked western shore of the sea-bed and pitched there his Camp ccxxxviii. a (Map No. 32. B. 3). On the same day he made sure of the exact relative position of his camp by searching the ground to the north-west until he reached and identified the find-place of the previous year at the Mesa marked on the map and situated about three miles from his camp. On the following morning, leaving his camp where it stood, he proceeded to the north-north-east, with a single companion, and came upon the footprints of my camels where, about three miles east of that Mesa, we had changed our direction to the north-east 25 Thence he turned eastwards to reach the shore of the open sea-bed. He had moved only one mile in this direction when he found on the shôr-covered ground a number of small fragments of oxydized iron, evidently the last remains of a completely decayed implement, C. ccxxxviii. a. 02-6. Discovered in the direct continuation of the line which had led us from Camp ci to the find-place of the Han coins, dagger, &c., these insignificant fragments can confidently be recognized as relics of the ancient traffic which had passed there.
Two miles farther to the east Afrâz-gul arrived at the last line of Yàrdangs overlooking eastwards the open expanse of hard salt crust. The slope descending to it from the plateau-like ground which bears the shôr-covered Yârdangs was very marked, and Afrâz-gul subsequently estimated the difference of level at this spot between the flat floor and the shore-line of the dried-up sea-bed at about seventy feet. In contrast to the salt-coated terraces he had passed through, which rose from twenty to thirty feet and more, he noticed from this point that the Yàrdangs fringing the sea-bed far away in a north-easterly direction were all small and apparently clear of shôr. It did not occur to him at the time that the ancient route might have followed this easy stretch of the coast-line for some distance to the north-east before striking across the salt sea-bed. Accordingly he did not himself reconnoitre in that direction, but merely directed Abdulmalik, his companion, to proceed there, while he himself was busy with the plane-table and subsequently with prospecting to the south-west, in which direction their farther progress was to lie. Abdulmalik rejoined him after a time without having found any more traces of the ancient route. How far his search had actually extended remains doubtful. Those few fragments of iron remain, therefore, for the present the last indication of the line which the route from Lou-lan probably followed before its passage across the dried-up sea 2° And here we may take our leave of the ` White Dragon Mounds ' and resume our search for the old Chinese route over the easier ground to the east.
25 See above, p. 298. shore of the dried-up sea with a cairn, depositing a record
26 I may note here that Afraz-gul, in accordance with my below it. This may help a future explorer to verify details.
directions, marked the point from which he had surveyed the