________________
302
THE INDIAN ANTIQUARY. .
[DECEMBER, 1909.
8. The Domoko Oasis. From the Hanguya Tati I passed on to a group of small ruined sites exhibiting in a typical form the fate of destruction, to which ancient remains are exposed in the belt of sandy jungle often intervening between the still cultivated areas of the Domoko Oasis and the open desert of drift-sand In 1901, I had passed some completely eroded dwellings, forming the northernmost of those sites, in a maze of tamarisk-covered sand-cones not far from the village tract of Domoko, on the route from Khotan to Keriya. But information about the rest had become available only since, a few years later, an enterprising village, stimulated by my old desert guide, Ahmad "the Hunter," bad began to prospect there for "old papers" to sell in the antique market of Khotan. The site of Khadalik, from which one of my old treasure-seeking guides bad extracted some manuscript remains of interest, and to which the promise of good reward now induced him to take me, seemed disappointing at first sight; for its principal ruin, which soon proved to be that of a large Buddhist temple, presented itself merely as an extensive low debris heap covered with sand. But scarcely bad we begun systematic clearing of it, when pieces of paper manuscript began to crop out in numbers. Jt soon became evident that the destructive operations of those who in early days had quarried the ruined temple for timber, and the more recent burrowings by "treasure-seekers" like my gnide Mullah Khoja, had failed to disturb the votive offerings of the last worshippers, which, being mainly deposited on the floor, had long before passed under a safe cover of sand. So we were able to recover here, in spite of the almost complete disappearance of the superstructure, a large number of manuscript leaves in Sanskrit, Chinese, and the "unknown" language of Klotan, besides many inscribed wooden tablets in the same language, and some in Tibetan. Most of them probably contain Buddhist texts, like some excellently preserved large rolls, which on one side presents the Chinese version of a well-known Buddhist work, with what evidently is its translation into the “unknown" language on the other. The clue thus offered for the decipherment of the latter may yet prove of great value. Plentiful remains of stucco relievos and fresco pieces once adorning the temple walls, together with painted panels, had also found a safe refuge in the sand covering the floor. Their style pointed clearly to the same period as that ascertained for the Buddhist shrines I had excavated six years before at the site of Dandan-Oilik in the desert northward, i.e., to the latter half of the eighth century A. D. It was gratifying when the subsequent discovery in a second shrine close by of stringed rolls of Chinese copper pieces, no doubt deposited by some of the last worshippers, supplied definite numismatic confirmation of this dating.
We worked hard bere with a large number of diggers, and in spite of heat and smothering dust, practically without interruption from daybreak until nightfall. Yet it took us fully ten days to clear these temples together with some smaller adjoining shrines and dwellings. I was eager to move on to the east towards sites further away in the desert, and hence likely to have been abandoned far earlier. Yet I was doubly glad in the end to have spared time and labour for Khadalik at the outset, for when I returned to this tract nearly eighteen months later I found that the area containing the ruins had just been brought under irrigatiom from the stream which passes within three miles of it. I cannot do more than allude here to a problem of geographical interest presented by Khadalik and another small site, Mazar-toghrak, near the opposite (southern) edge of the Domoko oasis, where I subsequently excavated & considerable number of records on wood both in Chinese and the Brahmi script of old Khotan, indicating, as at Khadalik, abandonment about the end of the eighth century A. D. Now it is noteworthy that the large ruined settlement of Dandan-Oilik, which I explored in 1900, and which, as duly recognized also by my friend Mr. Huntington, who has carefully studied since the physiography of this whole region, must have receired its water from the same drainage system, was deserted about the same period. Dandan-Oilik is situated fully 65 miles further north in the desert, and if shrinkage of the watersupply needed for irrigation were to be considered as the only possible cause of abandonment of these sites, the chronological coincidence in the case of localities dependent on the same streams and yet 90 widely separated would certainly be curious.
(To be continued.)
