CHAPTER XVII.

THE UPPER YANG-TSZE.

THIS chapter, being specially devoted to what may prove dull and uninteresting to the general reader, I am sorry that it is not in my power to make it more scientific in character; but it must be recollected that the "Upper Yang-tsze Expedition" was not composed of savans, but simply of persons travelling together for the sake of amusement and adventure, at the same time that they kept in view the opportunity afforded them of adding to the geographical knowledge of their countrymen. Neither the botany nor zoology of the country is treated of, because the only specimens brought back in those departments were,-the ferns collected by Colonel Sarel, of which a list appears in the Appendix; a few seeds; and some birds collected during idle moments: these last have but just arrived in England. It is accordingly hoped that those who peruse this chapter will bear with its imperfections.

PHYSICAL FEATURES OF THE RIVER.

Glancing at a map of Eastern Asia, it will be seen that, rising by several sources in the elevated region of Tibet, and among the mountains separating that country from China proper, the Yang-tsze Kiang first takes a southerly direction into the province of Yu-nan, where, recurving northwards, it enters Sz'chuan, through which province its course is a general one of east-north-east, and in which it receives several important tributaries. Passing into Hoo-peh, it dips to the south-east to the edge of Hoo-nan, where its volume

is swelled by the waters of the Tung-ting Lake, which serves as a reservoir for the drainage of that province, as well as of the mountainous confines of Kwei-chow and Quang-si. Immediately altering its course, it proceeds north-east till it is joined by the Han, a considerable river coming from Shen-si and Ho-nan; after which it again dips to the southward to another of the great lakes, the Poyang, in the province of Kiang-si, where it is augmented by the drainage of the whole of that province. Thence its course through An-hoei is northeast till it reaches Nanking, 200 miles from the sea, where the influence of the tide begins to be felt, and beyond which it gradually widens into the great estuary by which it is connected with the ocean.

Lower portions of this magnificent stream have been described geographically, if not geologically, by former travellers; and as the few specimens of rocks obtained (see Appendix) do not allow of any geological description by myself, the following remarks are confined to its course above Hankow, at present the uppermost foreign mercantile port, 600 miles above its mouth.

The Hoo-peh Plain.-The valley of the Yang-tsze, in the province of Hoo-peh, is one immense plain for about 200 miles, being only bordered by mountains to the southward at that part where it touches on Hoo-nan and the Tung-ting Lake; with this exception, from Hankow to near I-chang, 366 geographical miles by its windings, its banks are alluvial, and the great triangular space included by its course between those two places is so low, that an extensive system of embanking is requisite to keep its waters within bounds. This country is sufficiently described in the narrative,* so that no

*See Chapters V., VI., VII., and XIX., describing the constant action of the water on the banks, continually exposing fresh sections of the alluvial soil, and the effects of the annual overflows in adding year by year fresh deposit, and thereby steadily raising the level of the land between the river and the embankments.

special notice is here required, and I shall therefore at once pass on to consider the two remaining districts, viz. the confused mountain mass between I-chang and Wan, on the confines of the provinces of Hoo-peh and Sz'chuan; and its course through the latter province.

The Hoo-peh and Sz'chuan Mountain District.-The direct distance from I-chang to Wan is 140 geographical miles, throughout which the general direction of the Yang-tsze is east and west. In this section of the river the great gorges and strongest rapids occur; and, as will be seen by reference to the chart, it is the narrowest portion during nearly eighteen hundred miles of its course.

In our tedious and monotonous junk voyage through Hoopeh, the first place at which we observed any change in the one prevailing feature of alluvial plain was at Kiang-kow, a small town which will be found marked on the map attached to this volume, about twenty miles above Kin-chow (foo). Here we found a shingly beach, and the land on the left bank commenced to rise. Above this point the bed of the river was in places stony; the soil of the adjacent country was of a clayey or gravelly nature; near the village of Yang-chi lime was being quarried; and opposite Chi-kiang, just above, a pinkish limestone was exposed on the east side of the river, in low cliffs. (See Specimen No. 4, Appendix V.)

From this point to I-chang the Yang-tsze skirts the eastern edge of a mountainous country, and before reaching that place you have to pass between vertical cliffs of conglomerate of a very coarse nature. The adjoining country to the westward is mostly of the same formation, split in every direction by gorges and chasms, and of a very broken and rugged nature. Specimen No. 5 of this conglomerate contains pebbles of quartz, jasper, and limestone. On the side of the river opposite to the city of I-chang, sandstone is associated with this conglomerate; it is of a soft porous nature. (See Specimen No. 6.)

For the eighty miles between I-chang and Quai-chow, the Yang-tsze cuts through a very mountainous country by a succession of deep gorges. The cliffs often rise several hundred feet in sheer vertical walls, affording excellent opportunities for the examination of its geological structure; but, as I have no pretension to the character of a geologist, I fear my notes on the subject will be rather confused and undefined.

In the first portion of "I-chang Gorge" we still found conglomerate and sandstone, but at six miles above the city were limestone and chert. (See Specimen No. 7.)

The rapids commence fifteen miles above I-chang, and the bed of the river is full of granite boulders, many of very large size, which run out into points or form islands, much to the hindrance of the navigation. No. 8 is a specimen from one of these boulders, and gneiss (No. 9) was obtained before entering the second or "Lu-kan" gorge. In this and the "Mi-tan" or third gorge the cliffs rise nearly vertically from the river to eight and nine hundred feet, and the strata dip considerably to the north-west. The hills through which the river breaks and forms these gorges seem to have a general lay of north and south, though they vary considerably. At Kwei (see map) the strata run N.N.W. and S.S.E., dipping about 75° E.N.E.; and in this neighbourhood we met with the first signs of coal. Specimen No. 10 of sandstone is from two miles below, while 11, 12, and 13 are the coal and sandstone associated with it, only a few miles above. No. 12 can hardly be called coal, being no more than carbonaceous shale with traces of plants; but it was the material the native Chinese were working out of the hill-sides by galleries driven horizontally, the seams being from three to four feet thick. When got out it was pounded up and mixed with water and loam, being sent to market in the form of bricks, as already described. Above this we observed coal worked in numerous places, and these workings have been inserted in the larger charts published by Mr. Arrowsmith from our survey.

The village and rapid of Kwan-du-kow are at the mouth of the fourth or Wu-shan gorge, so named from its terminating near that town. This gorge monopolises twenty miles of the river without interruption. At its lower end is a fine-grained mouse-coloured limestone (Specimen No. 14); and at the boundary between the provinces of Hoo-peh and Sz'chuan, which occurs in the middle of this gorge, the general rock is a blackish limestone, of which No. 15 is a specimen.

is

In this portion of the river the disturbance of the strata is very great; and, the stream being hemmed in by high cliffs, it is impossible to get much idea of the general form of the mountains when boat-travelling. Above Wu-shan the hills recede from the river slightly, and are less steep for a few miles, until a fifth, the Fung-siang or "Wind-Box" gorge, entered, which is in no way inferior to the most magnificent parts of the others; and after a course of three miles, one emerges once more through a pass not over a hundred and fifty yards wide, into the open light of day at Quai-chow, where specimen No. 16, a dark-grey limestone, was obtained, and where there is found a red clay suitable for making tiles. So much for the gorges. The general features can be best gained by a glance at the large chart published by Mr. Arrowsmith, where the hills, originally sketched in on a scale of half an inch to the mile, have been reduced one-half.

From Quai-chow to near Wan, a direct distance of fortyfive miles, the country is much of one character, the general rock being sandstone of a rough calcareous nature (Specimen No. 20), usually horizontal. Coal is worked for in places, and the specimen (No. 17) obtained by us here is considered to be anthracite. We also found, a few miles below Wan, a very fine white lime in powder (No. 21), which is used as cement, but whether it is procured in the immediate locality or not I cannot say. With regard to the sandstone which is so very general throughout the upper Yang-tsze coal-fields,

U