国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

the valley widens and has a 3 m high terrace at its right side; below Tsölung the
left side terrace is also developed. The gravel becomes more scarce and the detritus
material finer. The river-bed is very winding. The Chang-tang nature becomes
more and more pronounced: the mountains are low, flat and rounded and their re-
lative height above the bottom of the valley inconsiderable; no steep, rocky slopes
are seen, living rock is rare, everything is rounded, levelled, denudated as always
in the plateau-basins without outlet. The valleys are broad, flat and open. The fall
of the valley is very gradual and the Shak-chu very winding. The morphology is thus
quite another than at the eastern side of the Chang-la-Pod-la, where the deep-cut,
accentuated sculpture of the peripheric regions prevails. In this respect the Chang-la-
Pod-la plays the same important part as a morphological boundary, as all other passes
on the continental water-parting.

The Shak-chu widens out to a plain, where, from the Lapsen-tari, the Targo-
gangri first becomes visible to the N. 55° W., offering a brilliant spectacle as rising
above the whole rest of the plateau-land, and all over covered with eternal snow.
To the south and S.W. we also see a considerable range, the continuation of the
Pabla; it is covered with some snow and has several flat summits: a pyramidal top
straight S.W. is called Ditse-muri. The country is very open, and the view reaches
far away and is not hindered or closed in, as in the deep valleys east of the pass.
At Tsale-sekung a round stone-wall serves as protection to salt-caravans. The ground
consists of coarse sand. The only vegetation is moss and grass. To the north a
little ridge appears, cut through by many small valleys and ravines. At a little hill,
Kyangdam, the Shak-chu joins the Sha-chu from the S.W., and thus forms the upper
course of the Targo-tsangpo, which, during the following days, is left to the west of
our route. Sha-chu comes from Sha-la in the Pabla range, where, therefore, the source
of the Targo-tsangpo should be situated. Other informants regarded the Tarok-la
as the real source of the river. The Tarok-la is shown as situated S. 50° W.
Probably there are several equivalent branches from different passes, forming the
Sha-chu. To the east is a mountain, Gonak-kekar, with some snow; to the N.E.
is Golep, and N. 25° E. Jom, from which the valley Ponglok descends.

The section of the road from Camp 147, at 5 050 m, to Camp 148, Bumnak,
at 4 945 m, keeps very much at one and the same level, although slightly undulat-
ing, first falling slowly along the upper course of the Targo-tsangpo, then ascending
to a very flat threshold, Chumar-la, 5 108 m, and finally descending through the
Bumnak valley.

Several right tributaries from the N.E. enter the Targo-tsangpo: Chita-chur-
kang is the first; it is broad, comes from the E.N.E., and has a stream of open
water, whereas the Targo valley is full of ice. The Ponglok-jom valley has a small
brook; this is the case with Gapu-nitä, which is fed by springs in the neighbourhood.
The Uktsang brook was frozen, but had running water under the ice. Then follow
Chimkar-mäntang, Longlung and Lungchen, in the background of which the Jom