National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Sec. v] THE DOMOKO-YAR AND THE REMAINS AT MAZAR-TOGHRAK 207

the one hand by the character of both Chinese and Brâhmi writing, and on the other by the structural remains found in the immediate vicinity. It is true, no exact chronological indication

could be derived from the small ruined mound some six feet high which I found amidst low dunes about 150 yards to the south-west of the site ; it undoubtedly represents the remains of a little

Stûpa, dug into long ago. But when eighteen months later I explored the ruin of a Buddhist shrine at Kara-yantak, scarcely more than a mile to the east of Mazar-toghrak, the art-remains there unearthed showed the closest possible agreement with those of the Khadalik temples.

The probability thus presents itself that the sites on opposite edges of the Domoko OasisKhâdalik with Balawaste and Farhâd Beg-yailaki below, and Mazâr-toghrak with Kara-yantak

above, the present area of cultivation—were abandoned about the same period, towards the close of

the eighth century A. D. The problem raised as to the cause of this simultaneous abandonment of sites, the extreme points of which are separated by not less than eighteen miles in a direct line

from north to south, is in itself of considerable interest for the archaeologist and the geographer alike. But it is thrust still more forcibly upon our attention when we remember that the same period must have seen the desertion of the large ruined settlement of Dandan-oilik.7

According to the observations which I made during my explorations of 19oo-o1, and which I have discussed at some length in my former Detailed Report,8 the Dandân-oilik Oasis received

its water from a canal fed by one or several of the streams now irrigating the oases of Chira,

Gulakhma, and Domoko. The careful examination which Professor Huntington has since made of this ground, and the physical changes undergone by it, has fully confirmed this view.° Now it is

of special importance to note that Dandan-oilik lies fifty-six miles farther north in the desert than Khâdalik, and not less than sixty-four beyond Mazâr-toghrak. Were shrinkage of the water-supply to be considered the only possible cause of abandonment, this chronological coincidence in the case of localities dependent on the identical drainage system, and yet so widely separated, would certainly be very curious.

That such shrinkage of the available water-supply has taken place in the Tarim Basin during historical times, and that it must be connected with a general desiccation period affecting the whole

of Central Asia and apparently most regions of the continent, if not of the whole earth, is a con-

clusion which a mass of steadily accumulating evidence is forcing upon the geographical student. It is Professor Huntington's special merit that he has brought out the central fact of that shrinkage and has emphasized the importance of the proofs which systematic archaeological investigation of ancient sites in the desert and near the present oases is able to furnish.10 At the same time he has

looked towards the results of this investigation to support a theory of his own which supposes

that the general process of desiccation has been diversified during the historical period known to us by a succession of minor though important climatic changes partaking of a pulsatory nature. By

a series of ingenious observations Professor Huntington has endeavoured to show that the climatic

pulsations thus assumed, i. e. periods of increased dryness extending over certain centuries followed in turn by periods of a reverse tendency toward more abundant rainfall, have exercised a determining

influence on history. He believes them to be reflected with particular clearness in the history of Central Asia, where a strictly mid-continental position would tend to increase the intensity of any

climatic variations.   -

Cf. Ancient Kholan, i. pp. 283 sq.   character of the climatic changes assumed by him and their

See ibid., i. pp. 285 sqq.   relation to the great cycles of the Glacial period and the

9 See Huntington, Pulse of Asia, pp. I7osgq., i88.   small ones known as Brückner's cycles', cf. ibid., pp. 365

   10 For a lucid summary of Prof. Huntington's views on   sqq. For his views as to the effect wrought by progressive

   the physical changes experienced by Central Asia in historical   desiccation in the Chira-Domoko region, see pp. 17o sqq.,

   times, cf. his Pulse of Asia, pp. I3 sqq. ; on the pulsatory   and elsewhere, passim.

Simultaneous abandonment of sites near Domoko.

Ancient localities dependent on same drainage.

Desiccation and climatic pulsations.