TO HAMI AND TURFAN
SECTION I.—FROM AN-HSI TO HAMI ; HSÜAN-TSANG'S DESERT CROSSING
AT An-hsi I was kept busy for twelve days by manifold exigent tasks which have been recorded in my Personal Narrative.' Here it will suffice to mention that, besides the preparation of a full report to Government on my previous operations, of proposals, accounts, etc., they included arrangements for the return via Khotan of Surveyor Rai Ram Singh, now invalided to India, and the secret acquisition, successfully managed through Chiang Ssû-yeh, of extensive further ` selections' of manuscripts from Chien-fo-tung. As soon as my collection of antiques had received this precious addition of four camel-loads of texts from the hidden temple library, I set out with R. B. Lai Singh for the journey which was intended to take us along the foot of the Tien-shan to Hami, Turfan, and Kara-shahr, and then for another winter campaign into the desert of the Taklamakân.
Regard for economy of time, and also for the safe transport of our heavy loads of antiques, obliged me to follow to Hâmi the present Chinese high road where it crosses the stony desert of the Pei-shan in eleven weary marches aggregating a total distance of about 218 miles. It has been followed too often by European travellers to need any detailed description here. , For the traveller who is not a geologist there is little to observe in this great waste of gravel and crumbling rock. The much-decayed ranges of hills, through which the route passes in succession and which alone break its monotony, rise nowhere much above 7,000 feet, and the wide detritus-filled valleys or plateaus between them lie rarely more than a few hundred feet below the saddles crossed. But there are certain aspects of this true ` Gobi ' which have their interest for the student of ancient geography, and which in view of the historical importance of the route deserve to be noted here.
As we moved along from one wretched little roadside station to another, each established with its mud hovels, tiny post of soldiers, and big heaps of refuse at points where some shallow depression offers a scanty supply of water in well or spring, and occasionally patches of equally scanty grazing on scrub or reeds, I was able to observe conditions of traffic which certainly could have changed but little since ancient times. Ever since the Chinese in A. D. 73 first acquired a firm foothold at Hami,2 this northern route ', with the few alternative tracks practicable on the west from Tunhuang, on the east from the side of Su-chou and the Su lo Ho bend, had formed an important, if not the principal, line of communication to the territories on both sides of the Tien-shan and to the Tarim Basin. Whenever Chinese power could assert itself in those regions and protect them against barbarian inroads from the north, this route must have seen heavy traffic carried on much in the same way as it now is. All information available from Chinese records and later Western sources points to the conclusion that, ever since the more direct route from Tun-huang to Lou-lan and the northern oases of the Tarim Basin was finally abandoned in the fourth century, this ' northern road '
' Cf. Desert Cathay, ii. pp. 338 sqq. Chou, T'oung pao, 1907, p. 156.
2 Cf. Chavannes, Les pays d'occident après le Heou Han
Second stay at An-hsi.
High road from An-hsi
Importance of 'northern route' after A.D. 73.