HISTORICAL NOTICES OF LOP, SHAN-SHAN, AND LOU-LAN [Chap. IX
position of Lop-nor. It incidentally also shows that the older name of Lou-lan or Lao-lan must have long survived in popular use. ` This is why the local people ordinarily call the lake Lao-lan
- im '1* This is the term used by the Shih shih hsi yli chi when it says : " The River of the South, coming from Yü-tien (Khotan), proceeds towards the north-east for 3,000 li, and on its arrival at Shan-shan enters into the Lao-lan lake." ' This is not the place to enter upon the ` Lop-nor question ' in general, complicated, perhaps, quite as much as elucidated by the controversial literature which has accumulated over it since the days of the protagonists Prejevalsky and Richthofen. But for the historical student the Lao-lan lake in this passage can scarcely refer to anything but the Kara-koshun marshes, which occupied then much the same place which they do now.
We may now resume our task of tracing back historical notices of the Lop tract previous to Fa-hsien's journey. I can find no mention of it in the passages concerning the Western Regions which M. Chavannes has rendered accessible from the records of the Chin dynasties (A.D. 265-419), though the successful expedition of Lu Kuang in A. D. 383, which on the one side touched Karashahr and Kucha and on the other Chu-mo or Charchan, must have also passed through Shanshan.16 But it is of interest to remark that a Chinese notice quoted by Rémusat refers, under the date of A.D. 280, to Shan-shan as the key of the southern route leading from China to Khotan, and to its chiefs as having friendly relations with the Empire.16 In A.D. 283 the Chin Annals mention the dispatch of young men of princely descent from Shan-shan to take up service at the Imperial Court.17 We shall see in the next chapter how striking a confirmation these statements have received from the discoveries of Chinese records belonging to the first half of the Chin epoch which Dr. Hedin's and my own excavations brought to light from the ruins of the Chinese military station of ` Lou-lan ', in the north of the Lop desert and on the ancient ` middle route ' connecting the Tarim Basin with Tun-huang.18
For the immediately preceding ` Epoch of the Three Kingdoms ' (A.D. 220-265) the extant portion of the Wei lio, composed between A. D. 239 and 265,1° furnishes particularly important notices about the three routes which were then distinguished as leading from Tun-huang to the Western Regions, and to which it will be necessary to refer repeatedly hereafter. The description of the foreign territories along the Southern route opens with ` the kingdom of Chii-mo (Charchan), the kingdom of Hsiao yuan, the kingdom of Ching-chzieh, the kingdom of Lou-lan, which are all dependencies of Shan-shan', this list being followed by an enumeration of territories westwards, dependent upon Khotan.20 There can be no doubt that by Shan-shan is here meant the present Lop tract with its main oasis of Charkhlik. The identity of Chii-mo, Hsiao-yüan, and Ching-chüeh with the ancient oases stretching from Charchan to the end of the Niya River has been fully discussed already.21 With regard to the ` kingdom of Lou-lan', here mentioned as distinct from Shan-shan, it will be best to reserve our views until we have examined all archaeological data now available for the ` Lou-lan Site' north of Lop-nor, and until we have ascertained the light which they throw on the few historical notices relating to it.
The essential interest of the Wei Ho record lies in the fact that it mentions the dependence on
16 Cf. Chavannes in Ancient Khotan, i. p. 544 ; Voyage de Song Yun, p. 13, note 2. For the extracts from the Chin shu cf Ancient Khotan, i. pp. 542 sqq.
16 Cf. Rémusat, Remarques sur l'Extension de l'Empire Chinois, 1825, p. 109, quoted by Ritter, Asien, v. p. 324. The original publication is not accessible to me.
17 Cf. Chavannes, in Ancient Khotan, i. p. 537.
18 See below, pp. 407-8.
Shan-span during Chin times.
Niya or Chingchtieh dependent on S han shan.
19 Cf. Chavannes, Les Pays d'occident d'aprls le Wei lio, in T`ornrg-pao, i 905, pp. 519 sqq.
R0 Cf. Chavannes, T`oung pao, 1905, pp. 535 sq. The necessity of substituting Chit-mo 11* for the faulty reading of the Wei lio has been conclusively proved by M. Chavannes, ibid., p. 536, note 3.
R1 See above, pp. 219, 296.