National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Belt of ancient cultivation.

274   REMAINS OF ANCIENT LOU-LAN   [Chap. VII

tagh and known to him by the Mongol name of Ulan-tementu.12 I myself, with Lai Singh, Afrâzgul and the few men who could be spared from the camel convoy, continued our former direction towards the reported cemetery site, with the intention of exploring it and thence rejoining our camp in the evening. The camels would thus be spared a considerable detour, and a day would be saved on our march to Altmish-bulak.

For about six miles from our starting-point the Yârdangs, which rose to about ten feet in height and preserved their usual direction, right across our route, proved very troublesome. This confirmed the wisdom of the arrangement above indicated, as the camels were enabled thereby to proceed straight to the intended halting-place for the night, thus keeping a bearing almost parallel to that of the wind-eroded ridges and trenches. As the latter portion of the preceding day's march had shown us no traces whatever of ancient human occupation, I was all the more interested to find, after we had covered about four miles, that fragments of well-made pottery, resembling that found around L.A., became frequent. Several bronze arrow-heads and miscellaneous metal fragments, as described in the List below, were also picked up on this ground, besides fragments of two inscribed Wu-chu coins. Among the former the well-made barbed arrow-head, C. xcvi. o16 (Pl. XXIII), with three sharp blades, may be specially mentioned. This type, to which one of Lai Singh's finds, Lal S. 015 (Pl. XXIII), and one from Kum-kuduk, Kum. of (PI. XXIII), also belong, differs markedly from that of the triangular bronze arrow-heads which formed part of the regular ammunition in use with Chinese crossbows of Han times, and which were recovered in such numbers both within the Lou-lan area and on the Tun-huang Limes.l3

At a point about seven miles distant from our last camp my attention was attracted by timber debris to the much-decayed ruin of an ancient dwelling (Fig. 167), occupying the top of a Yârdang about eight feet high. The ground covered by roughly hewn beams and posts of Toghrak wood measured about twenty-five by fifteen feet. Wind-erosion had bared the surface everywhere, except on the south, where a layer of reed-straw, probably fallen from what had been the roof,. had protected some slight remains of a clay-built wall. The scanty refuse surviving here and there contained only oat-straw and horse-dung. The remains of posts, made of slender tree-trunks with gabled ends which had once probably carried the rafters of the roof, suggested the coarsely built dwelling of a cultivator or simple roadside quarters. But the pottery debris found in the vicinity was all of good quality and manifestly of approximately the same period as that common at the Lou-lan sites. Several beads of glass and stone and small bronze fragments were also among the ` Tati ' remains picked up near the spot (C. xcvi. oI-5, 07-12, Pl. XXIII).

These finds and those previously recorded some miles nearer to our camp C. 84 (or C. xcvi) 14 clearly indicate a belt of ancient occupation, approximately contemporary with that of the other Lou-lan sites. Judging from its position as shown on the map (No. 29. D. 3), it may be conjectured that this belt lay along a direct route leading from the casirum L.E., the true bridge-head as it were of the desert route coming from Tun-huang, to a point at the foot of the Kuruk-tâgh glacis, from which the well-defined main course of the Kuruk-dary5. skirts it all the way to Ying-p`an and the present bed of the Konche-daryâ. Of Ying-p`an I shall have occasion to prove farther on that its ruins mark an important station on the ancient Chinese route connecting the Lou-lan area with Korla and with the string of oases which stretch westwards along the foot of the Tien-shan and constitute the natural northern highway of the Tarim basin.i5 The belt just referred to lies exactly

Finds of potsherds, metal frs.

Debris of ruined dwelling.

12 See C. 85, Map No. 29. D. 3.

13 Cf. below, pp. 290 sq. ; Serindia, i. pp. 358, 428 sqq. ; ii. pp. 599, 604, 704, &c. See also below, pp. 346, 416, T. xxll. f. 02-3.

14 The ` camp number ' C. 84 is that taken from Lai Singh's plane-table, C. xcvi that from Afràz-gul's and my own. I. See below, Chap. xxi. sec. i.