210 F. Kingdon Ward—The Land of Deep Corrosions. 
Many travellers crossing Western China between Burma and 
Ssuchuan have regarded the endless parallel ranges which have to be 
erossed as spurs thrown out from the main Himalayan range or from 
the Tibetan plateau, at least as far south as Likiang. Thus Captain 
Gill writes: ‘‘The great plateau that extends over the whole of 
Central Asia throws down a huge arm between the Chin-sha-chiang 
(Yangtze) and the Lan-tsang-chiang (Mekong), gradually diminishing 
in altitude as it extends south. The northern portion of this arm 
partakes more or less of the character of the main tableland . . . this 
arm is not more than 35 miles wide in the latitude of Batang and . . . 
it is little more-than a ridge of mountains running due north and 
south between the two streams” (Zhe River of Golden Sand, vol. ii). 
West of Batang the summit of this divide is, as Captain Gill indicates, 
an undulating grass-land plateau cut up by streams flowing in shallow 
valleys, with lakes occupying hollows here and there, from 12,000 to 
14,000 feet above sea-level; in appearance it closely resembles the 
grass-land country of North-East Tibet, on the Kansu border. But 
trom Atuntsi southwards to Likiang the divide narrows down to a mere 
rock wall. 
However, it is inconceivable that the Mekong—Yangtze divide was 
formed independently of the Mekong—Salween and Salween—Irrawaddy 
divides, which so closely resemble it in structure, as well as in their 
flora; if the Mekong—Yangtze divide is an arm of the great plateau, 
or a spur of the Himalayas, so too are its immediate neighbours, the 
Mekong—Salween and Salween-Irrawaddy divides. 
The same characteristics are repeated throughout the country to the 
east of Batang, though in a less sensational manner. The whole of 
the immense mountainous region lying between Batang and Tatsienlu 
and between the parallels of 32° and 26° partakes of the same nature, 
and though I do not know it at first hand the accounts of the few 
travellers who have been there emphasize the fact that it is a region 
of high parallel mountain ranges, often snow-clad, trending north and 
south, with rivers flowing south in deep narrow valleys between. 
However, in all this country there is no river to compare with those 
of the Yunnan—Tibet border, the Yalung being the only one of any 
size, though even beyond Tatsienlu the T'a-tu, and beyond that again 
the Min, also flow south. 
Captain Gill says of the country between Tatsienlu and Batang the 
day after leaving the former city: ‘‘ A few yards more, and reaching 
the summit of the Cheh-toh-shan we at length looked upon the great 
Himalayan plateau . . . and from this point, with the exception of 
a dip into the Yalung-chiang, the road is always at an altitude 
of 12,000 feet above the sea until the descent into the valley of the 
Chin-sha-chiang . . . ” (Zhe River of Golden Sand, vol. i1). 
The object of the present paper is to insist on the relationship and 
common origin of the physical features of the entire region from the 
Brahmapootra in Assam, across the sources of the Irrawaddy in 
the Burmese hinterland and the Yunnan—Tibet border, into Western 
China. There is no reason whatever to believe that this region may 
be regarded as a dissected plateau—that it was uplifted in its entirety 
at the same time as the Tibetan plateau and subsequently dissected 
