FROM CORRIENTES TO HUMAITA. 295
feet and ankles, and thus only the roaring, bawling gamblers, who sat lengthening out the night, were cleared away.
We are now about to leave the main branch of the riverine system, and to eviter the Paraguay proper, con- cerning which you may wish to have a few geographical details. The same authority that announces the topo- graphical homology of the Parana and the Ohio compares the Paraguay with the Missouri, and its great western influents, the Salado, the Bermejo, and the Pilcomayo, with the Red River, the Arkansas, and the Platte. It rises — according to a late explorer, the Cavagliere Bossi, who kindly sent me a copy of his work — two leagues from the Arinos of the Amazonian valley, and the source may be reached without passing through the wild and plundering Gnarani tribes. It floods properly in March to June, w^hen the supersaturated lands along the upper course dis- charge their surplus. The inundation of December- January, 1868-9, which precipitated the operations of the Brazilians against La Villeta, was caused by the Parana, which, forcing back and heaping up the waters of the Paraguay, poured them over both banks, whereas that to the east generally suff'ers. As a rule, the discharge of the upper bed is clear and that of the lower is muddy. At the junction, ships prefer to fill with the Parana, and higher up the crews drink of the springs and fountains. The free navigation of the Paraguay is a political necessity for the Brazilian Empire, which has had a line of steamers upon it since 1857. In six weeks they make Matto Grosso, some 2000 to 2200 miles from Buenos Aires; and in case of necessity they can easily efi'ect the passage in twelve days and nights, at the rate of eight miles an hour. For sailing craft at least six months must be allowed, and some have occupied seven in reaching Humaita; whereas the round trip from Buenos Aires to
