16 A JOURNEY IN BRAZIL. 
materials to powder ? We know that such an agency has 
been at work on the northern half of this hemisphere. We 
have now to look for its traces on the southern half, where 
no such investigations have ever been made within its warm 
latitudes ; though to Darwin science is already indebted for 
much valuable information concerning the glacial phenomena 
of the temperate and colder portions of the South American 
continent. We should examine the loose materials in every 
river we ascend, and see what relation they bear to the dry 
land above. The color of the water in connection with the 
nature of the banks will tell us something. The waters of 
the Rio Branco, for instance, are said to be milky white ; 
those of the Rio Negro, black. In the latter case the color 
is probably owing to the decomposition of vegetation. I 
would advise each one of our parties to pass a large amount 
of water from any river or stream along which they travel 
through a filter, and to examine the deposit microscopically. 
They will thus ascertain the character of the detritus, 
whether from sand, or lime, or granite, or mere river mud 
formed by the decomposition of organic matter. Even the 
smaller streams and rivulets will have their peculiar char- 
acter. The Brazilian table-land rises to a broad ridge 
running from west to east, and determining the direction 
of the rivers. It is usually represented as a mountain 
range, but is in fact nothing but a high flat ridge serving as 
a water-shed, and cut transversely by deep fissures in which 
the rivers flow. These fissures are broad in their lower 
parts, but little is known of their upper range ; and whoever 
will examine their banks carefully will do an important 
work for science. Indeed, very little is known accurately 
of the geology of Brazil. On the geological maps almost 
the whole country is represented as consisting of granite. 
