200 ANCIENT SITES OF THE KHOTAN OASIS [Chap. VIII
impossible to accept as evidence of this identification either the alleged local tradition of Yôtkan or that attaching to the mound of Naghâra-khâna, 22 miles to the south-east, which M. Grenard —on insufficient grounds, I believe—has assumed to have belonged to the ancient capital'. The careful inquiries made by me among the old villagers of Yôtkan conclusively prove that until the accidental discovery of the first finds in Niâz I.Iâkim Bég's time nothing whatever was known of the ` old town' beneath the ground, nor of any tradition existing about it. Even now the villagers' opinion about the origin of the remains is of the vaguest. In fact, the only statement made to me about an ancient city having once stood here was clearly traceable to the view expressed by a former European visitor. A tradition of a somewhat more genuine kind certainly locates an old town, the residence of the heathen ruler of Khotan, in the immediate vicinity of Naghâra-khâna. But it does not in any way connect the latter site with Yôtkan, and its own derivation is, as we shall see, subject to doubts 2.
In reality the proof for the location at Yôtkan of the old Khotan capital lies in the exact agreement of the site with the topographical indications furnished by the Chinese Annals, in the character and abundance of its antiquarian relics and—last but not least—in the ease with which we are able to identify from this starting point the positions assigned by Hsüan-tsang to the prominent Buddhist shrines he visited in the vicinity of the capital. The evidence derived from the latter sites must be left for discussion in the last section of this chapter.
Among the notices which the Annals contain relative to the position of the capital of Yü-t`ien, those derived from Kao Chü-hui's report on the Mission of 938 A. D. are certainly the clearest and most accurate. We have already seen that they place the White Jade river or the Yurung-kash at a distance of 3o li to the east of the city, the Green Jade river, corresponding to the present Yangi-Darya branch of the Kara-lash, 20 li to the west of it ; and the Black Jade river or Kara-kash another 7 li further to the west 3. A look at the map of the Khotan oasis will show how closely the distances here indicated agree with those actually measured between Yôtkan and the nearest points of the rivers. Taking Khalche as our starting point, we have a little over seven miles to the left bank of the Yurung-kash near the present Ilchi. The 3o li of the report would, according to the average equation of 5 li to the mile, take us within a mile of the actual river-bed ; and it is certainly noteworthy that this is exactly the width of the low-lying waste ground, once undoubtedly part of the Yurung-kash flood-bed, which separates the eastern edge of Khotan town from the river. The remains of an old embankment marked in Dutreuil de Rhins' plan of Khotan show clearly that the floods of the Yurung-kash must even in recent times have approached the town quite closely 4. Turning to the western river-beds the agreement is, if possible, still more striking. From Khalche to the nearest point on the Yangi-Darya measures four miles, which corresponds exactly to the 20 li of the report ; while the distance of i i miles separating the Yangi-Darya from the right bank of the Kara-lash is as close an approach to the 7 li of the notice as we could hope for.
The close accord between the actual distance separating the Yurung-kash and Kara-lash rivers and the aggregate of the measurements recorded in the report of the mission of 938 A. D. attests the accuracy of the latter, irrespective of the position assumed for the old capital. The firm evidence thus secured makes it of less consequence that earlier records of the Annals
I See Mission D. de Rhins, iii. pp. 127, 138.
2 Compare for Naghâra-khâna and the lake of Aiding-Kul,
s See above, p. 179.
' See Mission D. de Rhins, i. p. 96. The river is now
endeavouring to shift its bed again further to the west, and embankments have to be maintained in order to keep it out of the stretch of ground which evidently once formed part of its flood-bed ; see Hedin, Reisen in Z.-A. p. 26.