AMAZONIAN VEGETATION 361 
apparently repeated on all of them. For example, 
many of the trees of the inundated margins of the 
Tapajoz (some of them undescribed when I first 
gathered them) I found afterwards on the Rio 
Negro up to its very sources — although none of 
them inhabit the shores of the Amazon, either 
between the mouths of those two affluents or else- 
where. A few recur on the Teffe and other black- 
water streams entering still farther to the west, 
and even on similar affluents of the Orinoco. 
Here, at least, would seem to be a case of the 
vegetation depending on the distribution of the 
running waters ; but in reality both the kind of 
water and the vegetation nourished by it depend 
entirely on the nature of the soil, those rivers which 
run chiefly through soft alluvial bottoms being 
•turbid, while those that have a hard rocky bed run 
clear ; and the two classes of rivers are repeated over 
and over throughout the length and breadth of the 
Amazon region. Into the black Rio Negro runs 
that whitest of rivers, the Rio Branco, and imparts 
to the vegetation of the former, for a little way 
below their confluence, quite an Amazonian char- 
acter.^ The two largest tributaries of the Casi- 
quiari, namely, the Pacimoni and the Siapa, run 
nearly parallel through a longish course, and at 
rarely more than 15 miles apart ; yet the former has 
clear dark water and the latter is excessively muddy. 
Moreover, when I explored the Pacimoni to its 
very sources, I found it divide at last into two 
nearly equal rivulets, whereof the one had white 
and the other black water. The true riparial vege- 
^ Here, for instance, is the only locality throughout the Rio Negro for 
Botnbax Ahtnguba, a fine Silk-Cotton tree abounding on the Amazon. 
