inhabited.b Everything pointed to the conclusion that the structural remains of the ruined fort and the deep deposits of rubbish—rich in archaeological plums and remarkable, too, for their dirt—accumulated within them, all belonged to a protracted period of Tibetan occupation which in the light of available historical evidence could safely be assigned to the eighth or ninth century A.D.
It seemed hard to leave unexhausted, even for a time, so promising a mine as the profuse antiquarian haul of that first day had revealed. But when, on the following morning, I left the excavations in the fort to be continued under Naik Râm Singh's and Chiang Ssû-yeh's supervision, and proceeded to a rapid preliminary examination of the ruin, about a mile and a half to the north-east, of which Tokhta Akhûn had spoken as showing remains of sculptures, I soon convinced myself that, by settling down to a complete clearing of Miran, I should risk too long a delay of my expedition to the ruins in the north of the Lop-nor desert. And against such a course there were the gravest practical reasons. The ruin proved to be that of a Buddhist shrine, with its central portion rising as a solid mass of masonry about forty-six feet long and thirty wide. Above the débris encumbering the north-east side of the base there still showed remains of stucco relievos occupying niches divided by architectural decoration of good design. I cleared a small portion of the base on that side, and soon came upon fragments of stucco sculptures of large size, including a well-modelled colossal Buddha head, that closely resembled in style the relievos of the Rawak Stûpa.
It was impossible to be mistaken in attributing the temple to a period far more ancient than that of the Tibetan fort. Various observations made it probable that a site of considerable antiquity had been reoccupied here as in the case of Endere. It was clear that in order to secure adequate time for the careful excavation of this temple and of other ruins of earlier origin, I should have to revisit the site after my return from the ruins northward. Fortunately the vicinity of Abdal, where my base was to be established, and which would have to serve also as the starting-point for the desert journey to Tun-huang, made it easy to shape my plans accordingly. So after a further rapid reconnaissance of the whole site I returned to the fort, where the excavations carried on in my absence had yielded interesting finds. By nightfall I had had everything filled in again as a precaution against ` exploration ' by ` treasure-seekers '.
SECTION II.—PAST THE TERMINAL LAGOONS OF THE TARIM
On the morning of December Io I set out from Miran for the journey to the ruins of Start from
` Lou-lan '. On the preceding evening Surveyor R. S. Ram Singh had rejoined me from Mirân.
Charkhlik. He was suffering from rheumatic fever, and consequently needed a camel to ride from
the point where the want of water would oblige us to leave our ponies behind. This necessitated
a still more careful calculation of the weight of the indispensable food-supplies, baggage, and ice
which we could carry into the desert on the available twenty camels. I found myself obliged to
reduce the total number of labourers to be taken with us to thirty-five. In addition, there were
fifteen in our own party, including camel-men and Lop hunters, for whom the necessities of life had
to be assured during protracted operations in the waterless desert. The excavations at Miran had
provided a useful test, and it was the least efficient of our diggers who were paid off and sent back
to Charkhlik before starting.
The march of some nineteen miles which brought us northward to the Tarim led along the March to
These jars were carefully re-buried before I left the at 'treasure-seeking' which the men of a caravan encamped let at Abdal.
Mirân ruins on December z for the march across the Lop at Miran were said to have indulged in. For two frs.
desert. But on my return to the site in January I found recovered, see M.1. oo63 in List, Chap. xiu. sec. ix.
them unearthed again and broken, the result of some attempt