204 ANCIENT SITES OF THE KHOTAN OASIS [Chap. VIII
visits to Yôtkan an aggregate of close on 200 specimens. All are of copper ; I did not succeed in tracing any gold or silver coins, nor even information about such.
I much regret that neither during my journey nor subsequently after my collections had been deposited at the British Museum did I obtain sufficient leisure for an accurate numismatic analysis of these purchases ; nor could I claim full competence for it, seeing that the vast majority of the coins secured from Yôtkan belong to purely Chinese issues. A considerable number are too much worn or corroded for exact determination. Of the rest, however, i 50 have been classified with the kind help of Dr. Bushell and Prof. Rapson (see App. D) ; and the result of this preliminary analysis is sufficiently definite to be discussed here in its chronological bearing. Numismatic details in support of it will, I hope, be made accessible hereafter.
In the first place it is important to note that the collection includes over forty pieces of the Sino-Kharosthi type, bearing Chinese legends on the obverse and Indian Prakrit ones in Kharosthi characters on the reverse. A far larger number of these interesting coins has been examined by Dr. Hoernle, who, in Part i. of his ` Report on the British Collection of Central-Asian antiquities ', has set forth the results of his painstaking researches with great lucidityjand thoroughness'18. Dr. Hoernle recognizes in these coins a Chinese currency of Khotan issued by its native rulers at a period later than 73 A.D., when, as we have seen above, Khotan appears to have first passed under effective Chinese supremacy. ` It is distinctly a Chinese currency, because the standard of the coins is Chinese, inscribed in Chinese language and characters. . . . On the other hand, the reverse of the coins bears the symbols and names of the native kings in native (Indian) characters—a fact which indicates both that native kings still continue to reign, and that the language and characters used by the native administration were Indian.' The use of the types of the bare horse and more rarely of the Bactrian camel (both turned to the right) onjhese coins leads Dr. Hoernle to suggest some special connexion between the Khotan rulers and the ` Scythian ' kings of north-western India, Maues, Azes, and their successors, who are assumed on numismatic grounds to have ruled in the Punjab and adjoining tracts from about 50 B. C. to 8o A. D. ; on their coins the type of the bare horse occurs, with a great variety of others. In view of the obscurity which still )hides from our view the origin of those ' Scythian ' rulers of India and the true extent of their dominion, this resemblance in the coin types scarcely warrants at present far-reaching historical conjectures. But the justness of the main conclusions above indicated as to the character of the Sino-Kharosthi coins of Khotan and their historical significance cannot be subject to any reasonable doubt.
My collection includes specimens of both the large and the small Sino-Kharosthi coins described by Dr. Hoernle (intended to represent according to their Chinese legends a weight in copper of one Liang and four Chu and of six Chu, respectively). Several varieties of the latter class distinguished by the arrangement of the Kharosthi legends, can also be recognized. But the specimens, as will be seen from those reproduced in Plates XLIX, LXXXIX, 2-4, seem too much worn to give hope of their proving useful for the determination of the legends and of the royal names they contain 19. On the other hand, it is an observation possessing numismatic interest that whatever specimens of the Sino-Kharosthi currency I could secure or see were with some doubtful exceptions confined to the Yôtkan site. Considering the large number of ancient
is See Report on C.-A. antiquities, i. pp. r—r 6, with Plates I, III.
19 Dr. Hoernle distinguishes five varieties of the royal name : Gugramada, Gugradama, Gugramaya, Gugramoda, Gugratida. None of these names can be traced otherwise.
The recurrence of the identical Gugra- in the first part of these names finds, however, its parallel (as noted by Dr. Hoernle) in the Tibetan list of royal names all beginning with Vù'gya ; see Report on C.-A. antiquities, i. pp. 8, 16.