as further Wellby throughout the whole of his journey across Tibet from west to east, and Bower again in the course of his journey, encountered similar lakes in the big latitudinal valleys; and as, finally, Littledale and I both discovered a quantity of elongated lakes in the valleys which we traversed on our way to the Panggong-tso, there exists every reason to believe that lakes lie scattered all over the Tibetan highlands as thick as the spots on a panther's hide. In the preceding chapters I have attempted to prove that on the fundamental plateau-base there rise long parallel mountain-ranges, with endlessly long latitudinal valleys between them. In the light of what we do know, we may therefore take it as fully established, that each of these valleys embraces a long series of lakes, exactly in the same way as my latitudinal valley and Wellby's do. Each and every one of these latitudinal valleys is divided into a great number of self-contained drainage-basins; and each such basin is bounded on north and south by the neighbouring mountain-ranges and on east and west by flat cross-thresholds in the latitudinal valley itself. The number of these self-contained drainage-basins on the plateau of Tibet is, to put it briefly, incalculable, and in point of size they vary from the tiniest pool to sheets of vast area, like the Kum-köls, Selling-tsos, and Panggong-tsos.
The regular, more or less east-west prolongation of these lakes grows generally less noticeable towards the south and towards the west. The Selling-tso is only an apparent exception, for two latitudinal valleys at least fall within the area of that lake. On the other hand the Dangra-jum-tso and one or two others of Nain Singh's lakes deviate markedly in point of shape from the general norm; though it ought at the same time to be observed that Nain Singh had only a fugitive acquaintance with these lakes, and their true shapes are pretty certainly very different from those which he has assigned to them. Great divergences are also
IIIF Color HighRes
