solution of clay suspended in the water does no harm, if the water be allowed to settle a little before use.
The depth of the Hoang-ho is nearly uniform throughout, varying according to the high or low level of the river. I measured the width opposite the town of Ding-hu with a compass, and found it to be 1,421 feet. Opposite Bautu I had no opportunity of measuring it, owing to the strict watch which the Chinese kept on us during the crossing. But, when the waters are at an average height, its width there is much the same, perhaps a little broader. The depth of the river is considerable, and it is unfordable in any part; it could be easily navigated by river steamers. At all events, large boats constantly pass backwards and forwards with supplies for the Chinese garrisons on its left bank. The voyage from Bautu to Ning-hia-fu is said to take forty days, whereas only seven days are required to float down stream between these points.
In that part of its course which we surveyed, the Hoang-ho is not subject to inundations, but flows between low level banks; the clayey soil and rapid current accounting for the constant abrasion and falling in of its banks.
From the meridian of the westernmost end of the Munni-ula mountains several channels, 170 to 290 feet wide, separate on either bank from the main river, soon uniting again with the parent stream; only one, the Baga-khatun, continues to flow in an independent stream for some distance to the east.
