side. A mile or two south of Yii-lin P'u we passed a massive stone bridge 

 spanning the river at a point where it cuts through an outcrop of rock, and 

 some miles further south came upon the remains of a swinging chain bridge. 

 So far as could be gathered, two chains, their ends fastened to rocks on either 

 side, had carried across the river a roadway of timber. In any case the 

 contrivance had long fallen into disuse, the chains were rusted nearly through, 

 and but fragments of the planking still visible. At other points the passage 

 of the river had to be effected by fords, or ferries. Seven small villages were 

 passed, and towards the end of the day, after covering twenty-one miles, we 

 reached Yii-ho-p'u, a small walled town which at one time held a garrison of 

 soldiers, though the present population cannot exceed two hundred all told. 

 Our entry was effected over the top of the north wall, where sand had banked 

 up to such an extent as to render this possible. Huge mounds of sand from 

 twenty to thirty feet high were also noticed inside the wall. 



December 6th saw us once more on the march. The road still continued 

 down the valley of the Yii-lin Ho, and was excellent throughout. Shortly 

 after leaving Yii-ho-p'u, the last of the sand was left behind. The valley 

 widened out considerably, being bounded on either side by low loess hills 

 overlying a thick, faultless and slightly undulating stratum of shale. The bed 

 of the river was cobbly, but no signs of limestone, or anything but slate, 

 sandstone, and shale were visible. The river flowed down a bed some ten to 

 twenty feet below the level of the rest of the valley. Round the villages we 

 noticed numerous plantations of jujubes valued for their sweet, date-like fruit. 

 It was ascertained that wheat, millet, and sorghum form the chief crops 

 grown in the valley, and on the surrounding hills. The latter are not so 

 carefully terraced as in Shansi, where the natives dislike cultivating any but 

 level surfaces. At a village named Yen-wa (literally " salt scrape " or 

 " scratch ") extensive salt works were found. To collect the salt, flat surfaces 

 are prepared, and allowed to remain for a few days, at the end of which time 

 an efflorescence appears. This is scraped off, and put into large perforated 

 earthenware jars, through which water is allowed to percolate, thus dissolving 

 out the salt. The impregnated liquid is next boiled down, and the salt 

 extracted. This product, containing as it did a considerable portion of alkali 

 which could not be separated, was found to be of very poor quality. The 

 whole population was engaged in this industry, and the village looks a very 

 dismal place. We noticed the remains of numerous forts, some perched on 

 the hills, others extending across the valley; whilst high mounds at half-mile 

 intervals marked the sites of watch towers, which had run in a regular chain 



25 



