Sec. x] 'I HE LOU-LAN SITE IN CHINESE HISTORICAL RECORDS 421
our surveys of 1914. So the correctness of the information given by Li Tao-yiian is here, too, borne out by topographical evidence.
What the passage following in Li Tao-yiian's text has to tell us of ` the town of Lou-lan ' is of particular interest for us. ` This is, no doubt, the place where the colony of soldiers sent to clear the fields [for cultivation] was established, and this is why the town inherited the name of the kingdom.' In order to understand the reference made here we must turn to an anecdote related by Li Tao-yuan in an earlier portion of his commentary where he deals with the course of the River of the South on its passage north of Shan-shan.25 The source from which the anecdote has been derived cannot be traced, and Ch`üan Tsu-wang, a Chinese editor of the Slaaii cluing, whom M. Chavannes quotes, has pointed out valid critical reasons for doubting the .authenticity of the anecdote.26 But whatever its origin and its value from an historical point of view may be, there can be no doubt about the fact that it must have originated, probably locally, at a time preceding the abandonment of the Lou-lan Site, and that it has preserved for us evidence as to the name of the Chinese colony situated there and at least a popular tradition as to the time of its foundation.
The following is an abstract of Li Tao-ytian's story as rendered by M. Chavannes :—`So Man
whose appellation was Yen-i g A and who was a native of Tun-huang, was a capable man.
At the request of the prefect Mao It. he was charged with the functions of an ` Êrh-shih general'.
At the head of a thousand soldiers of Chiu-ch`üan and Tun-huang he came to Lou-lan in order to establish an agricultural colony there. He raised a white house. He summoned soldiers of Shanshan, Yen-ch`i (Kara-shahr), and Ch`iu-tzü (Kuchâ), at the rate of a thousand for each of the three kingdoms, in order to construct a transverse barrage in the Chu-pin River. On the clay when the river was dammed up the water threw itself against the obstacle, bounding with violence, and the waves covered the dyke.' So Man is then said to have summoned with a grave voice the divinities of the river to submit to his authority as in old times those of the Huang Ho and the Hu-t`o River of China are believed to have submitted to certain dignitaries. ` [So] Man in person performed prayers and sacrifices ; but the water did not diminish. Thereupon he drew up [his soldiers] in battle array and put them under arms. They beat the drums, raised great shouts, at times making blows with their swords, at others shooting arrows, and thus for three days fought a great battle [against the river]. Then the water receded and fell ; it supplied irrigation and produced fertility. The Hu (barbarians) proclaimed this a miracle. [So Man] made great arable lands, and at the end of three years gathered a million measures of corn ; his fame impressed the foreign countries.'
It appears to. me not very difficult to separate certain obvious facts embodied in this story from the embellishments which popular imagination has woven around them. It is clear that tradition knew of the foundation of a Chinese military colony in a locality which unmistakably corresponds to that of our Lou-lan Site, and of its creation having depended on the success of an engineering scheme intended to secure adequate irrigation for the Lou-lan tract by a barrage thrown across a great river. Whether by this river, which the story as related by Li Tao-yuan calls Chopin, was meant the river once filling the dried-up Kuruk-daryâ bed or a main branch of the Tarim, water from which may have been needed to supplement the supply furnished to the Kuruk-daryâ by the Konche-daryâ or Kara-shahr River, is a question we need not attempt to examine here. Considering the great changes to which hydrographie conditions are necessarily subject in a terminal basin where differences of level are so slight as in the Lop depression, and considering also the scantiness of historical records and archaeological data, no convincing solution of such a question can, in my opinion, be hoped for.
2b See Chavannes, T`oung pao, 1905, pp. 567 sq.; also 26 Cf. T'oung pao, 1905, p. 568, note I.
above, pp. 325 sq., where the preceding text has been analysed.