国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

Tiles with Bodhisattva head.

1198   KARA-SHAHR AND ITS RUINED SITES   [Chap. XXIX

the Far East which made itself felt with increasing strength in Central-Asian Buddhist art during the last centuries preceding its disappearance.

In a small recess spared from the wall which divides xv from another cella to the south, and opening towards the platform of the hall xvi, there were found, thrown together in a heap, seventeen

fragments of hard-burned tiles showing a Bodhisattva head in high relief in the centre and closely

resembling in type and size the tile Mi. xxiii. 1 already described. Plate CxXIX gives specimens of heads from them (Mi. xvi. 004, oo8). The fact that the walls of the recess, about 3 feet high,

looked bright red as if of an oven suggested at first sight that it had been used as a kiln for firing

such tiles, and that the fragments were left behind from malformed and rejected pieces ; but, since there were marks of what might be subsequent accidental burning, other explanations also may be

possible. To the west of xv there stretches a line of small shrines, partly built into the slope and hence of unusual shape. The partial clearing of their vaulted passages showed that the soil filling them was very damp and the fresco decoration of the walls almost completely perished.

To the north-east of the central group of temples there rises a conspicuous shrine, xvii, consisting of a rectangular cella and built on a high walled terrace, which Fig. 287 shows on the right

as seen from the north-west. Its interior was found filled to a great height with hard-burned débris.

The remains of stucco relievos brought to light by partial excavation at the entrance include the well-modelled heads, almost life-size, Mi. xvii. 003-5 (Plates CXxX, Cxxxi). Like the rest of the

relievos found here they show close correspondence in style to the work of Mi. x—xii. Of the bodies belonging to these heads no remains could be traced. Their plaster had evidently not been hardened sufficiently by fire before the walls fell in and completely smashed them. A Kaiyüan coin was found close to the entrance.

A group of closely adjoining small cellas and Stûpas to the north of xvii yielded finds only in two places. From the little cella xx there were recovered fragments of ornamented bronze bands,

Mi. xx. o01, and the well-carved wooden capital, Mi. xx. 002 (Plate CXxvIIi), of Indo-Corinthian style decorated with acanthus leaves. On the base of the almost completely destroyed small Stûpa xix there turned up a fragmentary sheet of paper bearing cursive writing which looked like Turkish ` Runic ' script, but has not yet been determined.

To the west of the group just mentioned there lie scattered half a dozen detached Stûpas and cellas, partly seen in Fig. 287 on the left. They form the north-west end of the site. None of

them yielded any finds of interest except the shrine xviii, the last which remains to be described.

It presented several unusual features. In the first place, it is worth noting that the ruin showed no sign of having suffered from fire, though marks of wilful destruction were only too obvious other-

wise. From an outer court or verandah facing north-east there was entered a kind of antechapel

about 17 feet square (Fig. 289) which does not appear to have been vaulted ; for the débris which filled it lay only to a height of about 3 feet. Judging from the remains of wall-paintings found in

its west corner, this antechapel must have been provided with a roof. So I suppose that it was built with timber, which in this case was not burned, but carried off for use elsewhere after the destruction of the site. Towards this antechapel opened a shallow cella, 91 feet by 6, surrounded on three sides by a vaulted passage a little over 4 feet wide which was in fair preservation but unfortunately showed only whitewashed walls.

In the corners of the antechapel there survived only a statue base on each side of the entrance,

and in front of them the miscellaneous rags of linen and silk fabrics, probably votive offerings, described under Mi. xviii. 0012. More interesting remains came lo light on clearing the débris

within the cella. The large image platform extending along most of its west wall retained only the wooden stumps of the stucco statues which once occupied it. But there was found also the profusely

Shrine xvii partially cleared.

Group of cellas xix,

XX.

Structural features of shrine xviii.

Finds in antechapel and cella.