SECTION II.—THE RUINS NEAR KHAN-UI
It was from the Chinese City Prefect or Hsien-kuan of Kashgar that I first learned of the existence of ancient remains on desert ground beyond the hamlet of Khan-ui, some twenty miles to the north-east of the ' Old City ' of Kashgar. A tradition communicated to me by that amiable official connected the site with a Chinese settlement supposed to have existed there during the time of the Han Dynasty. But whether this tradition rested on some historical information or was merely, as I suspected, derived by a kind of ' popular etymology ' from the name of the neighbouring hamlet Khan-ui (literally meaning ' the Khan's residence '), I was not able to ascertain. The march which brought me, on Sept. 4, 1900, to Khan-ui, through the fertile village tract of Bash-karim, has been described in my Personal Narrative. There, too, I have given an account of the picturesque shrine of Bû (Bibi) Mairyam Khanum, situated some three miles to the west of Khan-ui, which marks the resting-place of the saintly daughter of Satok Boghra Khan, the royal champion of Islam, famous in the local traditions of Kashgar l.
At a distance of about one mile to the north-east of Khan-ui hamlet all cultivation ceases ; and after another mile in the same direction, over absolutely barren ' Dasht', I found an extensive area of low denuded loess banks covered in plentiful patches with small decayed fragments of ancient pottery, glass, and slag. These remains, though far too small to show any characteristic decorative features, supplied unmistakable evidence of ancient habitation. But of the buildings from which they originated no other indication could be traced. The walls of sun-dried bricks or mud, of which these buildings must have mainly been constructed, had long ago disappeared—undoubtedly, as subsequent experience at so many other sites showed me, through the erosive action of wind and sand. The bare surface of relatively hard loess on which these fragments rested bore plain evidence, too, of the destructive forces at work here. It had, no doubt, been lowered considerably below the level of the ground at the time when the site was occupied ; but the total absence of constructive remains makes it impossible to estimate the extent of reduction in level. In the shallow depressions separating these loess banks, there were to be found here and there small accumulations of sand forming rudimentary dunes. I now regret not having secured at the time specimens of this sand ; for their microscopical analysis would probably have furnished clear proof as to whether this ' sand ' is disintegrated loess produced by erosion on the spot, or fine detritus washed down from the hills northward.
This area, strewn with diminutive débris, extends probably over more than a square mile, and is known by the name of Hasa-Tam 2. Popular tradition, as related to me by Sop Niâz Bowa, the old Aksakal of Bésh-karim, supposes an ' old town ' (kane-shahs), the capital of a ' Chinese Khâkân', to have stood here until Satok Boghra Khan destroyed it. Traces of an old canal, by which the town is believed to have received its water, are said to exist south of Hasa-Tam ; but the traditional connexion of this canal with the Yaman-yar river, far away to the south, is incompatible with geographical facts.
Proceeding eastward of Hasa-Tam for a distance of three miles, over ground where stretches of wind-swept bare loess alternate with low dunes of moving sand, I reached the ruins known as Topa-Tim (' the Sandy Mound '). They proved to consist of a conspicuous mound, roughly circular in shape, built of sun-dried bricks, and the much decayed remains of a great quadrangle closely adjoining it on the west. The mound, which rises to a height of 28 feet, and on the