Keriya River reached the Tarim as late as the sixteenth century." About 1000 A. D., according to the legend which I heard at both Malakalagan and Keriya, the ancient Buddhists, fleeing from the Mohammedan invaders, went down the river and so to the north, which would be quite impossible to-day. At a still earlier time, there seems to have been an important road along this route, as appears from the small ruins of Kara-dong discovered by Hedin forty miles above the present flood limit. Stein, who later explored them, assigns their abandonment to the third or fourth century of our era, the time when Rawak was abandoned. He thinks that Kara-dong was " a fortified frontier post or roadside Sarai'" guarding the road from Keriya to Kucha. Such a road could have existed only if the Keriya River once flowed farther than it does now. That it did so flow is proved conclusively by the dead river jungle which Hedin found for twenty or twenty-five miles beyond the present terminus of floods. The greater length of the river does not appear to have been due to less extensive use of the water for irrigation, for from the earliest times, Pimo, or Pein, seems to have been at least as large and prosperous as modern Keriya.
The obvious inference is that here again we have evidence of a change of climate.
On October 11, 1905, after four days in Keriya, I started eastward once more, without having stopped a day longer than I wished. My caravan, as then constituted, consisted of five splendid camels, eight horses, and five men, not including guides. The numbers sound large to one in the habit of traveling in less remote lands. As a matter of fact, they
IIIF Color HighRes
