Page:The naturalist on the River Amazons 1863 v1.djvu/21

 although sailing sometimes at a distance of eight or nine miles from the eastern bank, the opposite shore was at no time visible. Indeed, the Pará river is 36 miles in breadth at its month; and at the city of Pará, nearly 70 miles from the sea, it is 20 miles wide; but at that point a series of islands commences which contracts the river view in front of the city.

It will be well to explain here that the Pará river is not, strictly speaking, one of the mouths of the Amazons. It is made to appear so on many of the maps in common use, because the channels which connect it with the main river are there given much broader than they are in reality, conveying the impression that a large body of water finds an outlet from the main river into the Pará. It is doubtful, however, if there be any considerable stream of water flowing constantly downward through these channels. The whole of the district traversed by them consists of a complex group of low islands formed of river deposit, between which is an intricate net-work of deep and narrow channels. The land probably lies somewhat lower here than it does on the sea coast, and the tides meet about the middle of the channels; but the ebb and flow are so complicated that it is difficult to ascertain whether there is a constant line of current in one direction. A flow down one of the channels is in some cases diverted into an ebb through other ramifications. In travelling from the Pará to the main Amazons, I have always followed the most easterly channel, and there the flow of the tide always causes a strong upward current; it is said that this is not so perceptible in other channels, and