PHYSIOGRAPHY OF SOUTHERN SHEN-SI. 333 
1,400 feet, 420 meters, or about 20 feet, 6 meters, to the mile. There are 
long stretches with scarcely perceptible current (Fig. B, Plate XLVIII), 
rapids over shallow gravel bars, and occasional falls of a few feet over the 
edge of a limestone stratum into a basin in soft shale below. The stream 
is, however, navigable for boats which resemble Venetian gondolas, and 
which, being both light and well-modeled, are safely guided through the 
rapids by skilful boatmen or tracked up by means of tow lines. The com- 
merce on the river, especially in coal, which is mined at T’an-mu-shu-p’ing, 
and in rice and goods, which are used in the mountains, is considerable. 
Our route was unfavorable for physiographic study, the limestone 
precipices shutting us in so closely that even distant views of the adjacent 
heights were rare (Plate LI). We saw, perhaps, a seemingly individual 
mountain form several thousand feet above us in the distance; approached 
it, passed beneath it, and turning from some further point found that we 
could no longer identify it among many similar crags. Hence it was diffh- 
cult for Sargent to even sketch the most prominent of the cliff tops next 
the canyon, and greater eminences a little back from its margin escaped 
us altogether. 
At certain points, however, we obtained wider views, as between 
San-shi-li-p’u and Kiu-shi-li-p’u, where the valley widens in the red beds. 
We were then able to recognize the types of topography which we had seen 
in crossing the Ki-sin-ling, and to observe that the mountain ranges are 
but little lower near the Yang-tzi than they are on the main divide. Thus 
the canyons become deeper as we proceed downstream, and two of the 
narrowest and most beautiful occur just above and below Pa-ho. ‘The 
fall of the river increases as we approach the Yang-tzi-kiang, and several 
of the heaviest rapids occur in the canyon two miles north of Wu-shan-hién. 
The course of the Ta-ning-ho is nearly at right angles to the structure 
of the region, except in the red valley between San-shi-li-p’u and Kiu-shi- 
li-p’u. Nearly all of the smaller streams have developed in belts of shale, 
whereas the great mountain ridges are outcrops of limestone. The lime- 
stone has often been stripped, exposing long dip slopes, and the mountain is 
a more or less perfect anticlinal arch. It is evident that the course of the 
Ta-ning-ho, across very pronounced folds and in spite of very marked 
differences among the rocks, is not one which could have been acquired 
under existing conditions of relief. The stream could only have originated 
upon a surface not far below the summits of the present mountains, and 
on a slope which sufficed to give every advantage to brooks that flowed 
southward. It probably developed by the growth of such brooks, which, 
being accelerated, were enabled to divert to their transverse course any 
streams flowing on the strike of the rocks. 
