As for the contour of the new lake, its various creeks, bays, sounds, and other divisions were in almost every case drawn out from north-east to south-west, as one would naturally expect from the conformation of the surface prevailing throughout the Desert of Lop, where the north-east wind plays such a predominant role. When the water begins to invade a region of that character, it is self-evident that its first endeavour will be to fill all the depressions and hollows which preexist there.
The strip of desert which intervened between the newly discovered lake and the northern shore of Kara-koschun was 21 km. broad; this we crossed southwards on the and April. Only a few minutes after turning our backs upon the lake, we were once more travelling amongst dunes of the ordinary character. The only sheet of water we saw was in the east, but it soon disappeared. At first the sand increased gradually in height, rising from 7 or 8 m. to i r m., and in some places it was continuous; though at intervals there were depressions free from sand, corresponding to the bajirs of the Desert of Tschertschen, except that they extended from northwest to south-east and from north-north-west to south-south-east. Clay terraces and jardangs still continued to crop up amongst the dunes, as in the interior of the Desert of Lop, and were nearly always crowned on the top with a clump of old and dead kamisch-stubble, which appeared to be very much younger than that we saw beside the Kuruk-darja and on the northern shore of Lop-nor. Occasionally we came across a piece of withered tamarisk wood. Mollusc shells, which were absent north of the lake, became once more plentiful. Here I picked up a piece of green granite about 2 dm. in diameter, which no doubt came from the mountains that rose nearest to us on the south. In some of the larger bajirs I measured dunes with an altitude of I t m. and of an exceptionally regular scutiform shape. Small cylinders of sand and lime formed originally round old kamisch-stalks lay scattered about the level floors of these depressions, and intermingled with them were numerous sharp-edged crystals of gypsum. South of these depressions the aspect of the country changed again rather suddenly: the dunes became lower, the tamarisks grew on mounds, which stood closer together, and a little distance off were the trunks of dead poplars still standing upright, making a belt from east-north-east to west-south-west, and concentrating into thicker clumps here and there. These toghraks had been deprived of water whilst still comparatively young; and there were only a few specimens of a more mature and vigorous age. In their existing condition they were a good deal like the poplars of the Kuruk-darja, although in point of fact they were no doubt younger than these. I crossed a similar belt of dead forest immediately south of the bed of the Tokus-tarim at Jangi-jer. It can hardly admit of doubt that these two strips of forest were connected, and formed parts of one and the same long narrow belt of dead forest, running parallel to the lowermost Tarim and the western part of the Kara-koschun; and in the following year I ascertained that it did not extend any farther to the north-east than just this particular spot. It would almost look as if this strip of forest once ran along the northern shore of the Kara-koschun itself; in that case it would bespeak a rather great age for that lake. Such a conclusion would however be a hasty one; for this strip of forest has nothing whatever to do with the Kara-koschun and the Tarim. If
IIIFカラー高解像度
