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

Sec. i]   THE SHRINES OF KHADALIK   159

parts of the same building may have been due to the fact that the pious worshipper depositing them had endeavoured to please with his offerings as many as possible of the divinities represented.9

But votive deposits of far greater archaeological value came to light in the shape of Chinese

copper coins, discovered deep down under the débris which filled the north-west corner of the enclosing passage (marked a in Plan). First there was found on the floor a scattered batch of thirteen ` cash ' pieces all belonging to the Tang period, and then, as if to satisfy my craving for exact chronological evidence, quite close to the foot of the wall two completely preserved rolls of coins, counting twenty and fifty-four pieces respectively, still held together by the original string which the last owner had passed through their square holes. Rapid examination showed me that these rolls were made up, apart from a few Wu-shu pieces, of Tang coins only, the latest being issues of the Ta-li period (A. D. 766-79). Almost all these coins were in very good preservation. The detached set comprised nine coins of Chien-yüan (A. D. 758-60), three of Ta-li, and one Wu-shu. In the smaller string Mr. J. Allan found besides one Wu-shu and one cash' of the K`ai-yüan issue, current from A. D. 6 18-2 7, sixteen Chien-yüan and two Ta-li pieces. The second roll was made up of two Wu-shu, two K`ai-yuan, forty-two Chien-yuan, and eight Ta-li coins.10

Votive deposits of this kind must obviously belong to the period immediately preceding

1   the abandonment of the shrine, and only current coins are likely to have been used for them.11
None of the Tang coins shows any marks of long circulation, so that it is safe to fix the date

'ï   of abandonment for the whole Khâdalik site, with its closely adjoining ruins and remains of
identical type, at the close of the eighth century A. D. In confirmation I may also note that of the

i   sixteen Chinese coins, found by me elsewhere at the site and most of them close to the west of

I   Kha. ii, all with the exception of one Wu-shu piece belong to the Tang periods already noted,

ï   and that of the four coins brought to me from the vicinity of the ruins none goes down later than the

ï   Chien-chung period (A.D. 780-3).

The structure, Kha. iii, immediately adjoining this shrine on the south proved to contain

a single hall measuring some forty-seven by forty-two feet. Its southern portion was occupied by a plastered platform, fifteen feet wide, rising ten inches above the floor which itself lay two feet higher than that of Kha. ii. The foot of the platform showed a bold moulding three inches wide. The finds in this large place were very scanty, and besides a small wooden tablet with traces of Brâhmi writing comprised merely a small disc of malachite, of uncertain use, and a flat wooden food-bowl. To the west of the large shrine and within eleven yards of it, a trial trench subsequently laid bare a small structure relatively well preserved and undoubtedly once used as a dwelling (Fig. 46). It consisted of a room to the south, Kha. iv, built with walls of sun-dried brick still

standing to a height of about five feet, and of a small room and veranda, Kha. v, adjoining it on the north. Both were constructed with timber and plaster walls. The room Kha. iv, about seventeen by twelve feet, still retained its substantially built clay fire-place, and in the corner beside it a plastered sitting platform. Fallen rafters and reed bundles from the roof filled the interior which closely recalled to my mind the small monastic dwelling D. Iii at Dandân-oilik.12

The veranda, Kha. v, and the small apartment at its back showed walls of timber and wattle, with horizontal layers of reeds to retain the surface plaster, exactly after the method I had found prevailing in the Dandân-oilik structures. The wall to the right of the door leading into room iv had a small recess, about a foot deep, which had evidently served as a cupboard. The finds made in Kha. IT were few ; but several presented distinct interest. Along with broken pieces of posts and

Find of stringed coins.

Excavation of Kha. iii and small dwellings.

Finds in
quarters,
Kha. v.

° This was the case in the shrine excavated at Endere in 1901 ; see Ancient Kholan, i. p. 425. 10 Cf. for details Appendix B.

" Cf. the remarks on similar coin deposits found at Rawak, Ancient Kholan, i. p. sor.

12 See Ancient Kholan, i. p. 256; Figs. 32, 33.