| scarcely altered on the Indus part; and the western limit now assigned to the whole range in the Karakash pass corresponds exactly with the coincidence of the range and the Tibetan plateau. It is thus concluded that the pass at the head of Mr. Hayward's Kara-kash valley divides the Karakorum from the Gang-dis-ri mountains. — The waterparting of the Indus reaches the probable source of the main stream in the sacred peak of Kailas Parbat, called also by the pundit 'Gang-ri' ; and continues along the Gang-dis-ri mountains, which now skirt the Sutlej as far as the sacred lakes of Rakas Tal and Manasarowar, where the Sutlej rises. Here the basins of the Indus and Brahmaputra barely meet, and the waterparting descending from the Gang-dis-ri mountains, crosses the continuous trough in which the great rivers have their rise, to flow away in opposite directions.» Thus Saunders regards the range on which the Aling-gangri is situated as being in immediate connection with the range he calls Gang-dis-ri, a problem that could not be solved in 187o. Even in 1908 I was not able to prove whether the Aling-gangri were the northwestern part of the range which I have called the Lunkar range. Saunders' memoir is adorned with two excellent maps. Several important geographical features in them are taken from Chinese originals, so, for instance, Lob Nor and the two great and many small lakes south of it. So is also the case with all the lakes on the Tibetan highland. The Tarok-tso and its river, the Tarcou-tsanpoutchou, the Chaktak-tsangpo breaking through the Gang-dis-ri, all is Chinese. In this respect everything is the same as on Hermann Berghaus' map in Stieler's Hand-Atlas for 1861. The same may be said of the mountains north of the Tsangpo, although they are, in some respects, for instance east of Tengri-nor, more correct on the German map. In both cases there is a principal range with several ramifications and there is a detached range between the Raga-tsangpo and the Tsangpo. The scarcity of reliable material becomes the more visible if we regard Saunders' watershed of the Bay of Bengal. The upper Mekong is represented, of course, as belonging to the Bay of Bengal, which makes the drainage area much exaggerated to the north-east. Further, the river which is identical with the Chaktak-tsangpo, and the river which is the same as d'Anville's Naouc-tsanpou, have their sources north of the Gang-dis-ri and therefore the watershed makes two apophyses, proceeding northwards towards the interior of the country. As the range is the traditional Dzang of Ritter, and the two rivers come down from d'Anville, one should not expect too much exactness from this watershed, where nobody had ever been, except the Lama surveyors. Of great interest are the two vertical sections at the top of the maps. The first represents Himalaya and Tibet between 79° and 8o° and shows a bulky range between the Indus and the Panggong-tso, corresponding to Burrard's Ladak range. On the second section we get a very clear idea of what Saunders means with his Gangdis-ri Mts, for he represents it as one very massive range with a small secondary range 24-141741 III. | |
IIIFカラー高解像度
