356 TO TUN-HUANG AND AN-HSI [Chap. X
district. The packets of Chien-fo-tung rolls that I was subsequently able to rescue by purchase at Su-chou and Kan-chou clearly showed that such pilfering had continued en route as the carelessly guarded convoy slowly made its way to distant Peking. It was easy to form some idea of the extent of this leakage by the numerous specimens of such rolls which had been carried off into Hsin-chiang and were subsequently shown to me at different Ya-mêns along the route or were in some cases to be secured from Chinese petty employees.
There was accordingly special reason to feel satisfaction when Wang Tao-shih's cordial invitation to the ` Thousand Buddhas ' was, on the occasion of a further visit, supplemented by a discreetly conveyed hint that his store of old manuscripts was, notwithstanding all that had happened, not yet altogether exhausted. I could feel sure that he would be there in person to show me what he had managed to save from well-meant but hopelessly inefficient official interference, and not merely the big new shrine, hospice, &c., which he proudly claimed to have built with the silver ` horse-
shoes ' received from myself.
Lakelet In the midst of the efforts required to secure the timely completion of the preparations for my
and sound- further travels; I managed to pay a renewed visit to the far-famed sanctuary of the ` Crescent ing sands of
Yüeh ya- Spring Yüeh ya-ch zian A 1. g.8 That fine spring-fed lakelet, strangely hidden away amidst
ch'uan. the high dunes beyond the southern edge of the oasis, is considered a wonder of nature and, together
with the resounding sand-hill above it, forms a sort of ` Svayarimbhü Tirtha ', or pilgrimage place for the people of Tun-huang. The site is duly mentioned by the Chinese envoy to Khotan who passed through Tun-huang in A. D. 938,9 as also by Marco Polo some three hundred and fifty years later. For other notices, ancient and modern, and for parallels to the natural phenomenon of the ` sounding sands ' which has made the site renowned among the Chinese, I must be content to refer to Sir Henry Yule's and Professor Cordier's notes on Marco Polo's chapter dealing with the ` Province of Tangut ' 10 and to a very interesting chapter in Lord Curzon's recent work.loa To the Chinese notices must be added the account of the Tun-huang Lu, a text recovered by me from the hoard of the ` Thousand Buddhas ' and translated by Dr. L. Giles, dating from the close of the Tang period.1'
I cannot attempt to discuss here in detail the interesting physical features presented by these huge accumulations of drift-sand which cover the foot-spur of the outermost Nan-shan range where it projects towards the oasis between the mouths of the Tang-ho and the Chien-fo-tung valleys. They rise to heights of several hundreds of feet, as seen in the photographs reproduced in Figs. 207, 208, 209. But I may briefly draw attention to the evident connexion between
` Thousand Buddhas' invited.
Sand accumulations on foothills.
8 For a description of my former visit to the site, see Desert Cathay, ii. pp. 16o sqq. A very careful account of the physical features of the remarkable lakelet and the surrounding sand ridges is given by Professor L. de L6czy, Grdf Széchenyi Keletdzsiai utja, i. pp. 481 sq.
9 Cf. Rémusat, Ville de Khotan, p. 77.
10 Yule, Marco Polo, i. pp. 202 sq., 207.
1°' See Curzon, Tales of Travel, pp. 263 sqq.
u See J.R.A.S., 1914, pp. 710 sq. ; 1915, pp. 43 sq. The description of the ` Hill of Sounding Sand ' given by this account of the mirabilia of Tun-huang is as accurate as could be expected from a writer familiar with the locality.
The mention of the mysterious hole which the sand has not been able to cover up evidently refers to the strange hollow between the high sand ridges which the crescent-shaped lakelet occupies. The indications given as to the total area covered by the high ridges of drift-sand south of
the oasis are borne out by the recorded results of our surveys.
As Map No. 38. s. 4 shows, these sand ridges, where they abut on the gravel glacis separating them from the cultivated area, extend for a distance of about 17 miles in a direct line. This corresponds very closely to the stretch of ` 8o li east and west ' which the Tun-huang Lu mentions. Similarly the stretch of ` 40 li north and south ' and the ` height of 500 feet in places ' are anything but exaggerated.
I regret not to have ascertained, while at Tun-huang, the exact date when the annual fête at Yüeh-ya-ch`üan takes place. But my recollection is that it falls in the first week or so of June, which would correspond with the tuan-wu day (the Dragon festival on the fifth of the fifth moon) mentioned by the Tun-huang Lu as the date when it is customary ` for men and women from the city to clamber up to some of the highest points and rush down again in a body, which causes the sand to give forth a loud rumbling sound like thunder '.