Page:Mongolia, the Tangut country, and the solitudes of northern Tibet vol 2 (1876).djvu/28

This page needs to be proofread.



the western side of the Kara-narin-ula. At length, after satisfying ourselves that this range does not throw out lateral spurs into the centre of the plateau, which it borders, we crossed to the other side by the defile of the river Ugyn-gol, and on November nth descended into the valley of the Yellow River. Here we passed suddenly from winter into mild autumnal weather, such as we had left behind us in Ala-shan. Not a particle of snow lay on the ground, and the thermometer, which on the uplands stood below zero at noon, now often rose above the freez- ing point. This change in the temperature occurred in an extent of only thirteen miles of country covered by the border range.

Winter, however, soon began to make itself felt also in the valley of the Hoang-ho. The water was covered with ice, and the morning frosts rapidly increased in intensity. The mercury fell to —26'0° Cent. (—14° Fahr.) at sunrise, but during the day it was warm, especially in calm weather ; the sky was almost always clear.

We saw no inhabitants on the western side of the Kara-narin-ula. All the Mongols had fled to the valley of the Hoang-ho, alarmed at the appear- ance of a small band of brigands who came from the environs of Lake Koko-nor. Such incursions were not unfrequent in those parts of Mongolia which lay on the borders of the districts disturbed by the Dungan rebellion. The bands of robbers which con- tinually made their appearance in these districts were composed of all kinds of vagabonds armed with