256 The Land of Deep Corrosions 
nearer the Mekong; moreover the tremendous rains in the 
Salween valley south of the arid region have dissected the 
mountains on that side to a far greater extent, and as the 
torrents nearly always flow southwards parallel to the main 
river for some distance before their junction, a tiresome 
succession of deep valleys and high spurs needs to be 
crossed; this however is not the case with the Doker-la 
and the passes of the arid region to the north. 
From the Mekong to the Yang-tze is a journey which 
can easily be accomplished in four days over any of the 
passes I crossed, though personally I always set out from 
A-tun-tsi, which stands at a considerably higher elevation, 
thus requiring only three days for the journey. Following 
small roads by a more direct route, one could undoubtedly 
cross from river to river in three days without much diffi- 
culty. 
one too, I found the journey longer when made from 
the Yang-tze side than when made from the Mekong side, 
and for exactly the same reason, namely that south of the 
arid region the Yang-tze valley receives a greater rainfall 
than does the Mekong valley, though a much smaller rain- 
fall than the Salween valley. 
This is curious. Evidently the enclosing cliffs of the 
Mekong gorge—a mere rift in the rocks, its basin for three 
degrees of latitude not exceeding fifty miles in extreme 
breadth—are of insufficient height and extent to check the 
rain-bearing winds from the west, which, having drenched 
the mountains as far east as the Salween, pass right over 
the Mekong basin and throw down the remainder of their 
moisture on the higher peaks towards the Yang-tze. How 
thoroughly this rift-like character of the Mekong is main- 
tained is well illustrated by the passage of the watershed to 
the west in latitude 26°. Ascending from the river, we 
reached the crest of the watershed in five hours without 
any effort, the porters, who were carrying moderately heavy 
loads, taking things quite easily. But, from there, the 
descent to the Salween occupied a day and a half. Such 
is the great divide which separates the water flowing to the 
Indian Ocean from that going into the China Sea. 
This cramming of the western watershed against the 
Mekong is no doubt largely a result of the heavier rainfall 
