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

Sec. iv]

SEARCH OF RUINED LIMES STATIONS T. XXVIII-xxx   • 6oI

(T. xXIX. k, 1. 12) and two of undoubted Chinese porcelain (T. xxix. i, j). As the latter is

definitely known to have first made its appearance among the keramic products of China at the

beginning of the Sung period (n. n. 963),' occupation of the site, continued or intermittent, down

to that period, and possibly even later, may now be considered as proved. The glazes of the

stoneware pieces vary greatly, including a number of fine colours, such as black, creamy, greenish

turquoise, brown, etc., some with minutely crackled surface, as well as several mottled tints.

The presence here of later keramic débris, the first I had come across on this ground, helps Ruin oI

towards the approximate dating of the sculptured remains which were brought to light by the Buddhist

clearing of a small ruined structure about fifty-two yards to the west of the south-west corner of the shrine.

enclosure. It proved to be a little shrine, measuring only nine feet by eleven inside and having

its completely broken entrance on the narrower side to the south. The walls, preserved elsewhere

to a height of four or five feet, were twenty inches in thickness and built with fairly hard bricks of two

sizes, one twelve by seven and a half inches with a thickness of five inches, the other twelve by six

inches and three inches thick. Débris of broken bricks, reed wattle, and partially charred timber

filled the interior. This was mainly occupied by a brick-built platform, four feet broad to the north

and three feet elsewhere, running all round it except at the entrance on the south. Numerous

fragments of stucco sculpture, found mainly in the débris covering the platform along the north wall,

left no doubt that the little ruin had been a Buddhist shrine.

The fragments, T. xxix. 002—o0I 2, though all badly broken, are of unusually hard clay which, Remains of

as the darkened colour of the surface and the partially charred wooden core in the larger pieces stucco

suggest, probably owes this quality to accidental firing in a conflagration. They comprise hands sculptures.

(003) and portions of arms (00' 2. a, b) probably from more than one relievo figure, life-size or

somewhat smaller, together with miscellaneous pieces of fingers, drapery, and ornaments. The

details of the modelling and appliqué decoration show close dependence on the models of Graeco-

Buddhist art. Considering the very conservative development of this Buddhist art on Chinese soil,

and also the present inadequacy of our chronological knowledge concerning it, I do not think it safe

to attempt any very accurate dating. But, with this reservation, I should be inclined to attribute

these remains to a period not later than Tang times. Instructive and interesting in this respect

is the fragment, T. xxix. oo2 (Plate Cxxxlx), which has two small heads, one above the other, each

only about three inches high, but very carefully modelled. The look of placid contemplation in the

upper head and the intense anger and passion in the lower one, with its frowning brows and eyes

and mouth wide open, are very cleverly expressed. As a third head is evidently missing below, it

is probable that a ` Trimurti representation of some Buddhist divinity was intended. We may

trace a certain resemblance in style to the small naturalistically treated relievo heads recovered in

numbers from the ruins of ` Ming-oi ' in the Karashahr district (cf. Plates CXXXII, CXXXIII).4a

It is clear that the remains of the small Buddhist shrine here uncovered must have had some Later origin

relation to the watch-station close by and the wall which passed it. That they belong to a period of shrine.

much later than the construction and maintenance of the wall is equally certain. The question was

how to account for their presence at a point of the Limes which, as proved by the dated records

found at T. xxvll, xxvlli, must have already been abandoned during the Later Han period, or at

least soon after. The same question obviously arose as regards the fragments of keramic products

of manifestly later origin found within and outside the enclosure adjoining T. xxix. From the first

there presented itself the conjectural explanation that it was probably the tenacity of local worship—

such as I had so often seen exemplified elsewhere, and last among the ruins of Shih-pan-tung-

which had here caused a small shrine to be maintained and restored centuries after the wall was

' Cf. Brinkley, China, its Ilislory, Arl, etc., ix. p. 12.   i' See below, chap. xxix. sec. iii.

1971

4 11