which accompany the Indo-Chinese rivers and the upper course of the Jang-tse-kiang. The main range of the central Kwen-lun is everywhere double, indeed in places three-fold. Like most of the border mountains in Central Asia, this range also is remarkable for its rugged and mountainous character, a feature especially pronounced on its northern side, whereas the southern flank next Tibet is very much shorter and possesses much softer outlines. In the eastern half of the middle Kwen-lun, which I explored during my third journey, we find overlooking Tsajdam the outside ranges of Burchan-Budha, Go-schili, Tolaj, "horaj, Tsosone, and Dsukha, all of which have an east-west strike. Parallel with this outer barrier run the ranges of Schuga with its continuation to the snowy mountains of Amne-matschin, Gurbugundsuga, and Gurbu-naidschi. The third parallel range is composed of the Marco Polo chain, which begins at the Schuga river; this I explored as far as the snowy group of Charsa. Thence the great Tibetan border range extends west-north-west, still retaining its double or three-fold character. Thus we find next to Tsajdam the glaciated massive of Dschin-ri, from which the Garinga-ula stretches towards the east-south-east, and the Columbus chain towards the west-north-west. Fifty versts south of Dschin-ri lies the great range which the Imperial Geographical Society has called Prschevalskij Chain. North of the Columbus and Garinga Chains comes the Tsajdam Chain, separated from them by a narrow valley, through which flows the Chatin-san. Corridor-like long-axial or latitudinal valleys of this kind are characteristic of this part of the Kwen-lun. The westward continuation of the Columbus Mountains is formed of the Moscow Chain, which again is connected with the eastern part of the Tokus-daban. North of these the Altin-tagh breaks away to the east-northeast, as do also the two ranges of the Tschamen-tagh and the »Nameless» Range. The Tokus-daban itself runs towards the south-west, and joins the immense main range of the western Kwen-lun, which, covered with perpetual snow, forms the southern boundary of the Tarim basin. I have called its eastern portion the Russian Range. It is in this part of the system that the western end of the Prschevalskij Range is to be sought for. In the western part of the middle Kwen-lun the prevailing rocks are quartz phyllite, granite, and quartzite. The following are especially distinguishing features — the considerable altitude, the great number of peaks crowned with perpetual snow, the absence of cliffs, particularly in the snowy groups, the small number and insignificant size of the rivers, and the amazing sterility. Both the fauna and flora are poor, but gold occurs plentifully, a circumstance which in the future will be sure to have the strongest attraction for the covetous European.»
Finally, I will venture to reproduce one or two short extracts from Prschevalskij's description of the Tschimen valley, which he discovered during his fourth journey. Just as I had my headquarters at Temirlik, so he had his at Tschong-jar, and from there he made an excursion of close upon two months' duration into the far north of Tibet, covering in the course of it 784 versts. Here too his nomenclature is in a high degree confusing: sometimes he seems to use Mongol names, sometimes he employs invented Russian names, which give no hint whatever as to the nature of the object that they describe, and it is but rarely that we find the true, traditional, and illuminating name. »We started on the 19th November (0. S., 1884) and decided in the beginning to make westwards through the big, boundless
Hedi n, Tourney in Central Asia. IV. 6i
IIIF Color HighRes
