Sec. iii] DISCOVERIES IN AN ANCIENT REFUSE-HEAP, L.A. vi. ii 385
bones, however, were plentiful in the refuse, just as bones of sheep, donkeys, and camels. The absence of straw from wheat and other cereals in the midst of the masses of reed straw was curious, pointing to the fact that the cultivated area which supplied food for men and fodder for animals must have been at some distance. Grains of millet (the present färigh, Panicum miliaceum),
L.A. vi. 003, were, however, found, and a subsequent close examination of the wall plaster in L.A. II. vi revealed the presence of wheat straw in it.
Before concluding the account of the finds made in L.A. vi. ii, I may briefly refer to the numismatic evidence which it furnishes. It has its special interest on account of the certainty with which the numerous finds of dated Chinese documents, as mentioned above,16 allow us to attribute the accumulation of the refuse deposits between the structures of L.A. III, v, and vi in the main to the second half of the third and the early decades of the fourth century A. D. Among the nineteen Chinese copper coins found at L.A. v and L.A. vi there are seven Wu-chu pieces or fragments of such, one Ho-ch`iian coin, and eleven much-clipped pieces of a type illustrated by Nos. 29-34 in Plate CXL.164 The proportion between the full Wu-chu and these small pieces, clipped down often to quite diminutive sizes, is not materially altered if we include the coins found at L.A. III and Iv, the totals being then raised to twelve and nineteen respectively. We have here clear archaeological evidence proving that such clipped pieces must already have been extensively in circulation during the latter half of the third century A.D., and that the view which would attribute the introduction of these thin coins, known to Chinese numismatics under the graphic term of ` goose's eyes', to the short-lived reign of Fei-ti, A. D. 465, and his successors of the Sung dynasty, is misleading.17 There seems good reason to believe that Chinese numismatists, too, are cognizant of these much-clipped pieces going back at least to the reign of Hsien-ti, the last emperor of the Later Han dynasty, A. D. 189-220.18 In reality the process of constant debasement to which this quasi-subsidiary currency owes its origin is likely to have set in far earlier.
SECTION IV.—REMAINS OF A WALLED ENCLOSURE
The clearing of the other structural remains which had survived within the ancient station was effected with ease on December 22. They proved unfortunately scanty, owing to the terrible havoc worked by wind-erosion, which in several places, e.g. to the north and north-west of L.A. vi,
had left nothing but big pieces of timber débris scattered over the slopes of bare Yârdangs to mark the position once occupied by substantial dwellings. At a distance of about a hundred yards to the
south of the ` Ya-mên', a group of small quarters, L.A. vii (see Plates 23, 24), still showed its walls, built with vertical tamarisk rushes, to a height of two or three feet. Within these walls and the adjoining fences, which the same pliable material has saved from erosion, light drift-sand had accumulated over the original ground. But besides some small metal fragments, including a fairly well preserved ear-ring, L.A. VII. 002. e (Plate xxix), and a hollow Toghrak trunk probably once used for storing grain, there was found here only an oblong Kharosthi tablet, L.A. vu. i. I. An almost completely destroyed structure, on a terrace rising about twenty-six feet above the adjoining eroded depression, yielded only three Wu-chu coins, a wooden comb, and a few miscellaneous fragments in lacquered wood and metal.
The ruin L.A. Ix, seen in Figs. 92, 95, lay within about forty yards of the Stiipa, and furnished a striking illustration of the physical factors at work here. Of the substantial dwelling which, as the heavy foundation beams seen in Fig. 92 on the north slope of the isolated terrace show, had