natives. The Luma-ring-tso itself is a very small basin, only a few kilometers long; but if we add to it, as Nain Singh has done, all the salt-basins lying to the west of it, we get a very elongated depression, which perhaps does exceed 33 miles in length.
In our next two stages we shall have a better opportunity of becoming acquainted with this peculiar depression. We could have travelled equally well by its northern or its southern shore; but we chose the latter, because, according to the Tibetans, the grazing was better on that side. Here for a space we travelled between Nain Singh's route, to the north of us, and Littledale's, to the south. The routes of Bower, Wellby and Malcolm, and Deasy all ran to the north.
November 3rd. North-north-west of the camp we left a miniature lake, barely 100 - m. in diameter, and thus of the same dimensions as the Harschu. In fact, these two lakes will have been connected together not so very long ago, indeed they may perhaps be sometimes united even now. Their water is fresh, so that they may be regarded as spring-fed pools, which get rid of their surplus water by some trickling rivulet. Immediately west of Harschu rises a terrace, 20 m. high at the most, with a steep slope next the lake. Immediately after that comes a second terraced step, about i o m. high. From the top of the latter a steep slope of 3o m. leads down to level ground again on the west. Thus this is only a »pier-arm», quite narrow, but fairly long, stretching towards the south-east from the lowest slopes of the mountains on the north. It is built up of coarse sand and gravel, compacted into a hard solid mass. This pier is slightly curved in outline and concave towards the east; hence it forms the western boundary of the depression in which lie the present Harschu lakes. It narrows towards the south-east, and at the same time grows lower. Between its lowest extremity and the lowest terrace at the foot of the range on the south of the latitudinal valley there appears to be a break or gap. And at the foot of the northern mountains, but at a considerable height above the bottom of the valley, there are yet a couple of other terraces.
On the west side of the »pier» stretches a perfectly level expanse of hard, dry, white schor, a depression surrounded on all sides by sharply defined terraces. That water does sometimes gather there was evident from the deeply indented hoof-marks of the kulans, making black gaping holes. But at that time the ground was so dry and hard that our horses' hoofs scarce left any impression upon it. Next on the south rose a gloomy mountain-mass, with beach-lines fully wo m. above the bottom of the valley, and thus dating from a time when this depression was still connected with the Perutse-tso. The western promontory of this mountain-mass stands beside the outlet of an important transverse glen, at the head of which was visible the nearest range on the south, dotted with separate patches of snow and known as Dandur. The corresponding promontory on the opposite site of the glen is known as Sogele.
On the other side of the schor expanse the hard gravel begins again, and is sprinkled with grass and scrub. After that comes another pier-like threshold, about 15 m. high ; this is concave towards the west, narrowest in the middle of the valley, and widens out at the sides, where it gradually merges into the terraces which exist there. To the west of this threshold also there is an independent basin,
IIIF Color HighRes
