258 THE LOP-NOR PROBLEM.
sand increases in quantity from east to west, this does not in any way warrant us in drawing the further inference, that the vegetation increases in a similar way. All we know with certainty is, that this abundant vegetation, far more vigorous than that which is found at the present day beside the Kara-koschun, was necessarily associated with water. For instance, the banks of the Kuruk-darja were clothed with forests and steppes, and there was a luxuriant vegetation round the shores of Lop-nor. It is also to be assumed, that the islands and promontories in the lake were likewise forest-clad, at least I infer so from the scattered patches of forest which we saw in different parts of the desert, and which the close conjunction of vast quantities of mollusc-shells suggested had formerly belonged to the lake. When the Tarim flowed due east through the bed of the Kuruk-darja, and emptied itself into the Lop-nor, its banks were all the way adorned with forests, as at the present time, and they formed similar narrow belts. And as the distribution of water is the essential condition of the distribution of plant-life, the forests spread themselves also along the shores of the lake. But when the river burst out of that course, and turned to the south-east and south, the Kuruk-darja and Lop-nor dried up, and the forests died out. In time however new forests grew up along the new course, though as yet they have not advanced lower down than to Tschigelik-uj. We may be quite certain, that so long as the Kuruk-darja and Lop-nor were under water, they were bordered on the south by the same barren and desolate desert regions as at the present time. One has only to travel south from any point on the Tarim, say for example towards the Kerija-darja, to ascertain how narrow the belt of forest is. Precisely the same thing held good in the case of the Kurukdarja. The sand accumulated there in small quantities only, its passage being hindered by the river, so that dunes were only able to form at a considerable distance from it. Apart from this, the country was bare desert. I suspect therefore that, if from our western route we had travelled west across the desert, we should have found that the forest thinned, and finally ceased, before we came into contact with the Kontsche-darja and its lakes.
Apart from its forests, the Lop-nor bore a great resemblance to the Karakoschun. In both we have the mouth of the Tarim in the west and a string of lakes stretching out towards the east-north-east. In the Kara-koschun the fresh river-water ensures in the west a more plentiful organic life, both faunal and floral; in the east the water grows salter and salter, and consequently the vegetation thins out. The causes of the rapid decrease of vegetation from west to east in the northern lake-basin were clearly in every respect the same. The forest at all events continued east from Lôu-lan, but those parts of the lake which contained salt water were also destitute of plant-life. Our western route across the desert probably traversed the tracts of forest which grew on the western shore of Lop-nor.
In my earlier books I have dealt more or less exhaustively with the Lop-nor problem and its history;* but, now that the problem has been definitively solved, I
IIIFカラー高解像度
