terway.' He was,
however, correct enough, as usual, for the Chinese consider it to be the
main upper stream of the Yangtzu."--H.C.] Though our Geographies give the
specific names of Wen and Min to the great branch which flows by Ch'eng-tu
fu, and treat the Tibetan branch which flows through northern Yunnan under
the name of Kin Sha or "Golden Sand," as the main river, the Chinese seem
always to have regarded the former as the true Kiang; as may be seen in
Ritter (IV. 650) and Martini. The latter describes the city as quite
insulated by the ramifications of the river, from which channels and
canals pass all about it, adorned with many quays and bridges of stone.

The numerous channels in reuniting form two rivers, one the Min, and the
other the To-Kiang, which also joins the Yangtzu at Lu-chau.

[In his _Introductory Essay to Captain Gill's River of Golden Sand_,
Colonel Yule (p. 37) writes: "Captain Gill has pointed out that, of the
many branches of the river which ramify through the plain of Ch'eng-tu, no
one now passes through the city at all corresponding in magnitude to that
which Marco Polo describes, about 1283, as running through the midst of
Sin-da-fu, 'a good half-mile wide, and very deep withal.' The largest
branch adjoining the city now runs on the south side, but does not exceed
a hundred yards in width; and though it is crossed by a covered bridge
with huxters' booths, more or less in the style described by Polo, it
necessarily falls far short of his great bridge of half a mile in length.
Captain Gill suggests that a change may have taken place in the last five
(this should be _six_) centuries, owing to the deepening of the
river-bed at its exit from the plain, and consequent draining of the
latter. But I should think it more probable that the ramification of
channels round Ch'eng-tu, which is so conspicuous even on a small general
map of China, like that which accompanies this work, is in great part due
to art; that the mass of the river has been drawn off to irrigate the
plain; and that thus the wide river, which in the 13th century may have
passed through the city, no unworthy representative of the mighty Kiang,
has long since ceased, on that scale, to flow. And I have pointed out
briefly that the fact, which Baron Richthofen attests, of an actual
bifurcation of waters on a large scale taking place in the plain of
Ch'eng-tu--one arm 'branching east to form the To' (as in the terse
indication of the Yue-K