III. From the statements already made above it will be clear that III lies higher than I, and that II lies lower than I. Hence the flow of the current is as shown by the arrows. Between basin III and basin II there would appear to exist a threshold (indicated by diagonal shading in the fig.), for it is between these two lakes that the greatest difference of level is found; and a connection might easily be established between them along the gullies excavated in the clay soil by the wind, did there not exist a barrier higher than the water-level to prevent it. As basin III is clearly one of vast extent, an active evaporation must ensue from its surface, and a large quantity of water must be lost in the relatively recently moistened ground over which it spreads. Again, seeing that the lake discharges in its turn 8 cub.m. in the second, it is obvious that it must receive a heavy influx from the Tokus-tarim. This has even been estimated at 21 cub.m.; but if the circumstances are the same as those which obtain in the Tarim and Kara-koschun, and in the arms between I and II, then the amount ought to be 37 cub.m. But III is of course much less than I (Kara-koschun), and the evaporation is proportionally less. What relation existed between IV and the other three lakes I do not know ; but the probability is that in the spring of 190I the lake in question emptied itself into the Kara-koschun.
I crossed the zone of land between the lowermost Tarim and the Kara-koschun on the one side and the Astin-tagh on the other by three different routes, all starting from Abdal—in 1896 I went by the Kara-buran, now almost completely dried up, and by the district of Nadschi-bidschin to Tscharklik; in 1900 by the well of Jan-daschkak to Tscharklik; and in 190I east-south-east to the well of Dunglik. The first-named route I described in Peterm. Mitt. With regard to the other two I will add a few words here.
The middle route was traversed on the 6th, 7th, and 8th of April. From Abdal the road led first south-south-west, across the extreme westward extensions of the Abdal lakes, or rather shallow marshes (for they are nothing better), linked together by small shallow canals. They are supplied exclusively from the east, so that the water enters them from the opposite quarter to that from which the Tarim enters them. Their bottom is treacherous in the extreme; for here the natives of Abdal had just lost three horses literally »drowned» in the ooze, and it was only by dint of constructing a temporary bridge of sheaves of kamisch and tamarisks that we were able to get our camels over.
These quagmires are succeeded by a belt of perfectly barren and naked schor, 21/2 to 3 km. broad—dry, hard, saliferous sedimentary matter, which was laid down in the lake, for at that place the lake extends that far. The old shore-line appears to be marked by a thin belt of tamarisks, long dead and withered, but still stand-ding on their mounds. Here there is a broad shallow watercourse, called Mijanajaghi, down which the rain-torrents sometimes stream into the lake. Coming from Mijan, this watercourse continues on to Jaruk (pron. Jajuk) on the Tarim, though the water seldom gets down as far as that. Even then there was a tiny rill of muddy red water, excessively salt, trickling down it. South-west of Mijan-ajaghi the steppe begins again, and is dotted over with kamisch, jantak, and other desert plants, as well as with numerous tamarisks, dead and living, perched on the usual mounds. At the first line of tamarisk-mounds we distinctly felt the beginning of the ascent towards the foot of the mountains. Thus the former strip of shore beside the flat-bottomed lake is here not more than 3 km. broad. At Mijan-ajaghi there is nothing
IIIFカラー高解像度
