EXPLORATION IN CHINESE TURKESTAN. 757 



to their invuriahle kindness and attention with all the more gratitude, 

 as it was shoN\'n at a time when, as the}^ knew well, the conflict with 

 European powers Avas convulsing the Empire in the East. 



Such imperfect explanations and illustrations as, with an interpreter's 

 help, I could give of the historical connection of ancient Indian culture 

 and Buddhist religion with Central Asia, probably helped to dispel any 

 doubts and suspicions which might otherwise have been roused by the 

 intended excavations, etc. In this respect I found m}^ references to the 

 8i-yu-ki, the records of Iliuen Tsiang's travels, singularly helpful. 

 All educated Chinese officials seem to have read or heard legendary 

 accounts of the famous Chinese pilgrim's visit to the Buddhist king- 

 doms of the "western countries." In my intercourse with them I 

 never invoked in vain the memory of " the great monk of the Tang 

 dynasty (Tang-Sen)," whose footsteps I was now endeavoring to trace 

 in Turkestan, as I had done before in more than one part of India. 



Busily engaged as I was during my stay at Kashgar with practical 

 preparations, I managed also to survey a number of instructive ancient 

 remains, chietly ruins of Buddhist Stupas, in the vicinity, and to con- 

 tinue my studies of Turki. On September 111 finally set out on the 

 journey to Khotan. Choosing for the first portion of the march the 

 track which crosses the region of moving sands around the popular 

 shrine of Ordam-Padshah, I was able to fix the position of that curi- 

 ous pilgrimage place more accurately than is shown in existing maps. 

 From Yarkand onward I followed the ordinary caravan route, which 

 leads along the southern edge of the great desert, and mostly through 

 barren, uninhabited wastes of sand or gravel, toward Khotan. For 

 me it had a special historical interest; a variety of antiquarian and topo- 

 graphical observations which 1 was able to make proved beyond doubt 

 that we were moving along the identical great thoroughfare by which 

 in earlier times the trade from the Oxus and the Far West passed to 

 Khotan and on to China. 



It is impossible to refer here in detail to au}^ of this evidence. But 

 I may brief! 3^ mention at least the curious patches of ground frequently 

 passed on the route beyond Guma, where the eroded loess is thickly 

 strewn with fragments of coarse pottery, bricks, slag, and similar 

 refuse, marking the sites of villages and hamlets long ago abandoned. 

 Such debris areas, locally known as "tatis," are to be found in many 

 places beyond the present limits of cultivation in the Avhole Khotan 

 region; in some places they extend over whole scjuare miles. They 

 exhibit everywhere most striking evidence of the powerful erosive 

 action of the winds and sand storms which sweep over the desert and 

 its outskirts for long periods of the spring and summer. The above- 

 named fragments rest 011 nothing Init natui'al loess, either hard or more 

 or less disintegrated. Having alone survived l)y tiie hardness and 

 weight of their material, these fragments sink lower and lower as the 



