ing that sand is entirely absent on the opposite northern and western shores. He does indeed tell us, that the dunes on the shore reach an altitude of 36o feet, and that those farther away to the south attain an altitude of at least 500 feet, above the surface of the lake. The altitude of 36o feet was obtained with the aneroid, and is doubtful, because Roborovskij states that the dunes beside the Tschertschendarja, when measured during the same journey, attained an altitude of 30o feet, which is indeed very much over the mark. The altitude of 50o feet is estimated from the surface of the lake, and is valueless, because we are ignorant both of the distance and of the ascent of the ground in that direction. But his description of the dunes that begin 15 versts north-east of Kimur-chani is more valuable: »Here gigantic dunes have advanced to the very shore of the lake and plunge straight down into its deep and transparent waters. Along the shore extends a narrow belt of thick kamisch, the 'individual stalks of which reach a height of 3 sashen, and a diameter of about one inch. These reeds led us to consider seriously the possibility of our farther advance. To penetrate into them was impossible; the stalks were so strong, the camels were quite unable to force their way through. Immediately on the right of them rose the steep slopes of the barkhans, and on the left was the deep lake . . . The prevailing winds in this region blow from the southwest, and consequently all the barkhans lay athwart the route we desired to follow, stretching from north-west to south-east, and their north-eastern slopes were very steep.»
The concluding sentence is not quite clear, for if the prevailing winds blow from the south-west, then it would seem to follow almost certainly that the dunes ought to turn their steep slopes for the most part towards the north-east. But the first part of the sentence would seem to indicate that they may also look towards the south-east. With regard to the winds in the neighbourhood of Kara-schahr, I was in 1896 given the following information: »Der vorherrschende Wind soll im Frühling vom Juldus-Thale kommen; im Spätsommer and Herbst ist Ostwind gewöhnlich. Auch jetzt, am 14. März, raste ein ungemein gewaltiger Sturm vom Juldus-Thale, erst um 5 Uhr nachmittags hörte er auf, nachdem er 6 Stunden gedauert and den Weg vom feinen, lockern Staub reingefegt hatte.»**
The wind that comes out of the Juldus valley would blow from the west-northwest. Across the basin of the Baghrasch-köl it would therefore appear that the prevailing winds blow in spring, and come from the west, whereas in the Desert of Lop they blow at the same season from the east and north-east.
Where then do the immense masses of sand in the Ak-bel-kum come from? Are they lacustrine dunes, or are they ordinary desert-dunes, formed in the same way as the dunes in the Takla-makan? With the slender amount of knowledge that we possess with regard to the basin of the Baghrasch-köl it is difficult to answer this question. What excellent material for a physico-geographical monograph a study of that basin would afford! The relations which the basin bears to the ring of mountain-chains which • girdle it round, as well as its relations with the Juldus . valley, the bathymetrical relations of the lake, its currents, its waves, its floral and
* Op. cit., p. 87.
** Peterm. Mitteil., Erginzhft. 131, p. 67.
IIIFカラー高解像度
