hard white salt — we had great difficulty in chipping a piece off with a paddle — only I dm. below the soft sedimentary matter it is astonishing that the water of the lake is as fresh as it actually is. This salt layer was deposited in a desert lake that was drying up and disappearing; and there can exist no doubt that it marks an earlier stage in the migration of the Lop-nor — perhaps this was the position of the lake before it travelled north. Not having implements with me, I was unable to measure the thickness of the salt deposit; but from what my canoe-men told me, and they -fetch salt from that place in summer, partly for their own use, partly for sale in Tscharklik, the maximum thickness approaches 85 cm., though in one place, namely where we landed, the thickness decreases to io cm.
In the part of the lake called Avugh-köli, there juts out from the southern shore a flat cape, on which stands a nischan consisting of a faggot of branches; it is there that the Dung-chan-jol, or road to Tung-chuan, divides. While the right branch goes up into the mountains vi â Dunglik and Tatlik-bulak, the left branch continues on through the desert of Gobi to the town already mentioned, and is said to traverse schor and lschöl, i. e. a salt surface and desert. This branch seems to swing away from the Kara-koschun; I say »seems», because it is impossible to obtain, even from those who have travelled that way, really trustworthy information as to how far the marsh and its kamisch-beds extend to the east. A little east of the cape there was a black round island projecting . about one foot above the surface. Towards this we steered, with the intention of landing on it; but it was by no means so easy, for the canoes soon stuck fast in the mud and we had to plough our way on foot through the ooze and slush. At length we reached the road, which runs through a country absolutely destitute of organic life. The soil was genuine schor, that is a blend of dust and clay, perfectly dry and hard, and broken into »warts» and »blisters», which looked as if they had been formed by gases whilst the ground was still moist. Close to the water-line there was a belt of a different consistency, namely perfectly level silt, yellow, almost white in colour, and only recently exposed in consequence of the shrinkage of the lake.
Still farther east this long lake grows so shallow, that we had perforce to paddle amongst the jäkän in order to make any progress at all. One of the small lakes we , crossed over is called Jäkänlik-köl, while the most easterly part of the big lake is called Gölme-kätgen-köl, or the Lake of the Lost Fishing-net, a name which proves that fishing was, at any rate formerly, carried on there. At the present time the Ajagh-köl marks the eastern limit of regular fishing along this southern waterway, although the people do occasionally put down their nets in the jangal north of the Sate-köl. East of a schor island, which occupies the greater part of the breadth of the Gölme-kätgen, the water was so shallow that the canoes would no longer float, but had to be dragged. At length even that was no longer possible, and then we stopped. The water there was a light yellow, showing the reflection of the bottom, which is everywhere covered with a layer of fine yellow silt, the product of atmospheric dust. The greatest depth near the point where we turned back was 19 cm.; but large portions of the lake-bottom were either wholly exposed or just flush with the surface of the water. And yet there was a slight
IIIF Color HighRes
