160 HISTORICAL NOTICES OF KHOTAN [Chap. VII
predicted that after his death the lake would dry up and become a country called Li-yul, with the city of ` U-then ' (Hu.tlzen in Mr. Thomas' transcription) as its capital.
Mr. Rockhill has compared with this tradition the legends relating to the draining of the lakes which were believed to have once occupied the present valleys of Kashmir and Nepal, as well as Bod-yul or Tibet. Its resemblance to the ancient legend which represents Kashmir as having been originally the lake of Satisaras deserves special attention, in view of that close connexion of Khotan legendary lore with Kashmir which we shall have occasion to note hereafter. I have fully discussed the Kashmir legend, as told by the Nilamata Purana and also by Kalhana, in my ` Memoir on the ancient geography of Kashmir ' ". A reference to it will show how closely the rôle here ascribed to Buddha and his two companions corresponds to the part played in the Kashmir legend by Brahman and the gods Visnu and Balabhadra, who upon the former's command drained the lake by piercing the mountains. Buddha's stay on Mount Go§rnga seems to reflect the position taken up by Brahman and his divine host on the lofty peaks of the Naubandhana Tirtha in the mountains south-east of Kashmir. This close relation between the Khotan legend and the one told of Kashmir in the Nilamata Purana is all the more noteworthy because the Kashmir legend had assumed in Buddhistic lore a materially different form, as seen from the account of Hsüan-tsang and Mr. Rockhill's extracts from the Tibetan canon 11a. Thus the transfer of the Kashmir legend to Khotan cannot be attributed to any specifically Buddhist channel or agency.
The Annals of Li-yul ' begin the story of the foundation of Khotan with a reference to King Moka, who is said to have ruled over India 234 years after the death of Buddha 12. ` At that time the lake had dried up, but Li-yul was uninhabited.' In the thirtieth year A§oka's consort bore a son, whom the king, alarmed by the soothsayers' prediction that he would be king in his father's lifetime, caused to be abandoned. ` But when the child had been abandoned, there arose a breast on the earth from which he derived sustenance.' For this reason he was called Kustana or `breast of the earth' (sa.nu)13. This child was miraculously carried off by Vai§ravana to the king of China (Rgya), who had 999 sons, but wanted one more to complete a thousand, and brought up the boy. Kustana having found out his true origin
wanted a kingdom for himself ', and accordingly when twelve years old he ` got together a host of io,000 men, and with them went to seek a home in the west, and while thus employed he came to Me-skar of Li-yul '.
About that time Yagas 14, a minister of Moka, had been obliged to leave India as his relatives had become obnoxious to the king ; so he left the country with 7,000 men, and sought a home to the west and to the east, and thus he came into the country below the river of U-then' 15. Two followers of Kustana who had run away from Me-skar, came at Tola
" See Stein, Râjat., II. pp. 388 sq.; also i. 26-27 with note. "` See Mlmoires, i. pp. 168 sqq. ; Rockhill, loc. cit., pp. 167 sq.
12 If this date refers to the year of A§oka's accession, it approaches closely to the probable chronology of Buddha and Moka as ascertained by recent researches. Asoka's accession took place about 272 a. c. and circ. 487 B. c. appears to have been in Asoka's time the accepted date for Buddha's death ; see V. Smith, History of India, pp. 4o, 137. That the chronology of the Annals of Li-yul ', such as it is, was derived from sources different from those which Tibetan Buddhism generally relied on, becomes evident from the fact that, according to Mr. Rockhill, none of the Northern Buddhist texts
known to him place Moka the Pious later than a hundred years after Buddha ; comp. Life of the Buddha, p. 233, note 4.
'3 Life of the Buddha, p. 234. The spelling Kusthana, regularly followed in Mr. Rockhill's extracts is evidently an error of transcription. The text, according to Mr. Thomas, has always sa.nu. The meaning indicated shows that Skr. Kustana is intended.
14 Mr. Rockhill gives two forms of this name Yak and Yâ as; but it is evident that the Sanskrit name Yaks is meant, borne by a person who figures in the traditional story of Moka as current in Northern Buddhism.
Is Mr. Rockhill notes that here and elsewhere the rivers of Khotan are referred to by the expression shel-lchab or,