284 THE LOP-NOR PROBLEM.
to say between Lop-nor and the small lakes south of it. This absurdity is enough to show that the district south of Lop-nor is drawn without any knowledge, and that both the topography and the nomenclature are employed without any definite rules. And it becomes even yet more absurd when we proceed farther from Nukitu-schan-k'ou. We do not read »Farther west you come to Ike-ghaschon», which would mean that we were to proceed from Nukitu-schan-k'ou; but what we are told is, »Farther west from the southern shore of Lop-nor you come to Ikeghaschon». How we are to get from Nukitu-schan-k'ou to the southern shore of Lop-nor the Chinese text does not tell us; consequently there is here a lacuna in the itinerary. This makes it perfectly evident, that the cartographical material is derived from two different sources, and that the cartographer has made a desperate attempt to get them to fit, but without success. If we are to proceed west from Nukitusekin in order to reach Nukitu-schan-k'ou, then the former must lie east of the latter; and if from the southern shore of Lop-nor we are to travel west to reach Ikeghaschon, then Lop-nor must lie east of Ike-ghaschon. On the Wu-tschang map however Ike-ghaschon is shown as lying to the east of Lop-nor. In this case the map is wrong and the text right, for as a matter of fact Lop-nor is placed on the Wutschang map i'/Z too far to the west, but as near as possible at the right latitude. On the other hand, on the map that accompanies the Si yii-schuei-tao-ki. Lop-nor lies a good deal to the east, that is if we compare it with the position of Bostangnur (Baghrasch-köl). If on the Wu-tschang map we move Lop-nor to the position it actually occupies, according to my explorations, the Chinese text also is right, for then we should have to travel west from the southern shore of the lake in order to reach Ike-ghaschon, and continue thence in the same direction in order to get to Bagha-ghaschon.
I have already remarked, that, as I suspect, the author of the Si yii-schueilao-ki, in describing the southern route between Sa-tscheo (or Chara-nor) and Lop-nor has interwoven the descriptions of two different itineraries; that is to say, the first part, and by far the larger part, of his itinerary belongs to a more southerly route, which traverses mountainous country throughout, while the latter part belongs to the northern route, quite a different one from that by which Kosloff travelled from Abdal to Sa-tscheo. Probably some of the cairns of stones which I saw running south-west in the Kuruk-tagh north of Toghrak-kuduk show the direction in which it ran. In order to prove the correctness of my conception, it will be necessary to analyse the three roads which the Wu-tschang map shows as running east from the Lop-nor, and to compare them with the roads which exist now in the same quarter. The one farthest to the south has but little interest for us in this connection. It is shown with extraordinary distinctness on the Chinese map, and there is not the slightest difficulty in identifying it with the road which is still in use along the foot of the southern border-range of Tsajdam. I have myself travelled east along the same road from Ike-tsohan-gol and crossed the Bulungir-gol at the same spot as the road does on the Chinese map. The Dabusun-nur of the map corresponds to my Hollusun-nor, but it has been transposed as far towards the north-
IIIFカラー高解像度
