Page:Through the Brazilian Wilderness by Theodore Roosevelt (1925).djvu/156

 the first time, the Juruena and the Gy-Paraná, two important affluents of the Tapajos and the Madeira respectively. The Tapajos and the Madeira, like the Orinoco and Rio Negro, have been highways of travel for a couple of centuries. The Madeira (as later the Tapajos) was the chief means of ingress, a century and a half ago, to the little Portuguese settlements of this far interior region of Brazil; one of these little towns, named Matto Grosso, being the original capital of the province. It has long been abandoned by the government, and practically so by its inhabitants, the ruins of palace, fortress, and church now rising amid the rank tropical luxuriance of the wild forest. The mouths of the main affluents of these highway rivers were as a rule well known. But in many cases nothing but the mouth was known. The river itself was not known, and it was placed on the map by guesswork. Colonel Rondon found, for example, that the course of the Gy-Paraná was put down on the map two degrees out of its proper place. He, with his party, was the first to find out its sources, the first to traverse its upper course, the first to map its length. He and his assistants performed a similar service for the Juruena, discovering the sources, discovering and descending some of the branches, and for the first time making a trustworthy map of the main river itself, until its junction with the Tapajos. Near the watershed between the Juruena and the Gy-Paraná he established his farthest station to the westward, named José Bonofacio, after one of the chief republican patriots of Brazil. A couple of days’ march northwestward from this station, he in 1909 came across a part of the stream of a river