244 THE DESERT OF LOP.
it. A more eloquent and more convincing illustration of the way in which the origination of a sandy desert is initiated, and of the way it continues to grow, could not well be wished. To the north-east and east-north-east of us lay the regions of the Kuruk-tagh, from which the sand came; south-west and west was the eastern border of the sandy Desert of Lop; but in the neighbourhood where we then were there were no dunes at all. All the drift-sand which has gone to build up the dunes of the Desert of Lop has however been blown across the schor desert; but by reason of its almost absolute horizontality and the absence of even the smallest impediment, the sand has not been arrested, but has continued on unchecked until it reached a part of the desert in which the jardangs, tamarisk-mounds, and kötäk existing there provided the requisite hindrances for the dunes to accumulate against. Hence it would be rash to say, that the absence of dunes from this part of the desert is due to the relatively recent period at which the ancient lake dried up. If a dune, exceedingly small it is true, but nevertheless perfectly regular in formation, can grow up in a single night, then in the course of hundreds of years dunes of gigantic size could arise, were it not that the winds are so constant, so regular, and so powerful, and that the surface is so level. It is for these reasons too that dunes will be unable to establish themselves here in the future either. This little one-night-old circular dune illustrates also what I have already emphasised with regard to the filling up of the Kara-koschun. When we call to mind, that that shallow expanse of water is 170 to zoo years old, and couple with it the fact that a distinct dune can grow up in a single night, it is clear how immense must in 200 years be the masses of sand which have been transported by the same north-east and east-north-east winds into the Kara-koschun basin, and deposited there. It is amazing, that this depression was not long ago filled up by these stupendous transpositions of material blown into it.
On the 14th March we covered 11,25o m. in a direction S. r .79 W., and between the two encampments there was a rise of 0.644 m. Thus the desert still continued to be incredibly flat. The results of the survey thus far were that, whereas in the first two days we descended, during the next two we ascended; in other words, we crossed over a depression in the northern part of the desert. Hence the object of the survey was already achieved, and the problem already solved. In comparison with the proof of the existence of this depression, it was of less importance to ascertain the relative altitude of the Kara-koschun with regard to our point of departure. It was at any rate already clear, that we were now on the low threshold or swelling that separates from one another the two depressions of the Desert of Lop.
On the whole the landscape did not really change during this day's stage. We marched all day across the schor desert, and the only difference was that the upbulging parts of the schor surface were somewhat higher than hitherto, and now formed actual ridges or dykes i/2 to 3/4 m. high, seldom as much as i m. These dykes or ramparts possess no fixed or predominant direction, and in this respect differ from the jardangs. Sometimes they run in straight lines, sometimes they wind about, and point to every quarter of the compâss. Again, they converge and fuse, and form rings with a hollow in the middle, like miniature volcanic craters. Often
IIIFカラー高解像度
