190 ANCIENT SITES OF THE KHOTAN OASIS [Chap. VIII
legend which, as related by the ` Annals of Li-yul ', made Buddha symbolize at this spot the future draining of the waters covering Khotan ? 18
In conclusion, the name now borne by the hill deserves brief notice. As pronounced by the people it sounds Komdy% but the written form adopted for it in the short Tadhkirah of the shrine and probably elsewhere is Kohmdyi. According to a statement of M. Grenard, Kohmdyi is interpreted to mean ` the snake of the mountain '—evidently a ` popular etymology ' which takes the Persian words kôh mountain' and mar snake' for its foundation, without troubling itself about the non-Persian form of the supposed compound. M. S. Lévi ingeniously recognized in this etymology, whether true or not, a reminiscence of the Grhapati (Chi-li-ho-po-ti) Naga whom the Suryagarbha-sutra names as inhabiting the site 10. I am unable to express any definite opinion as to the real derivation of the name, except that it manifestly belongs to that extensive portion of the local nomenclature of Khotan which is of pre-Turki origin. If it were shown that the Eastern Iranian tongue once current in Khotan possessed the phonetic equivalents of Persian kôh and mar, the notice of the Suryagarbha-sutra would permit us to hazard the conjecture that the name originally meant ` the hill of the snake (i.e. Naga)'.
SECTION II.-THE CULTURE-STRATA OF YÔTKAN
Supply of antiques from Yôtkan.
Apart from Mount Gogrriga, the position of which was indicated by unmistakable natural features, no attempt could be made to locate any of the other ancient sites of the Khotan oasis mentioned in our texts without previously determining the position of the ancient capital, for it is invariably with reference to the latter that Hsüan-tsang records those bearings and distances which alone can now guide us to the identifpation of minor sites. We have seen above that, according to the Hsi yü-chi, Mount Gogrnga lay 20 li to the south-west of the capital, and that various notices of the Chinese Annals placed the latter between the Yurung-kash and the Kara-kash rivers. These data would not by themselves have sufficed for an exact location of the site had not a fortunate chance in recent times brought to light ample archaeological evidence which supplements them and removes all doubt.
To M. Grenard belongs the merit of having first clearly recognized that the little village of Yôtkan in the Borazan canton, which had furnished a constant supply of antiques, such as ancient pottery, coins, gems, &c., to Khotan traders, as well as to previous European travellers, stands on ground once occupied by the ancient capital I. A steady flow of antiques from Yôtkan has since M. Grenard's visit in 1891 reached public collections in Europe and India, largely through the purchases of the British and Russian representatives at Kashgar, or else through acquisitions of European visitors to Khotan ; but the information available as to the exact conditions in which those relics were found in such remarkable quantities, and the general character of the site, remained of the scantiest 2.
18 Compare above, p. r59; Rockhill, Life of the Buddha,
19 Compare Noies chinoises sur l'Inde, iv. p. 4o note.
1 See Grenard, Mission D. de Rhins, iii. pp. 127 sq. 128 sq.
2 For brief notices of the locality compare Hedin, Through Asia, pp. 759 sq.; Hoernle, Report on C.A. antiquities, i. pp. xii sq. (based on communications from Mr. Macartney and the Swedish Missionaries of Kashgar).
In all these the site is inaccurately designated as ` Borazan', in reality the name of the canton to which the village of Yôtkan belongs. Dr. Hedin's chapter on ` Borasan and its archaeological remains' (loc. cit., pp. 759 sqq.) contains, besides good illustrations of antiques coming from Yôtkan, extracts from a paper published by M. Kiseritsky on the late M. Petrowsky's large collection of similar acquisitions. The original paper, quoted as being in the Journal of the Imp. Russian Archaeological Society, is not accessible to me.