356 Quaternary Deposits^ Etc. — Mills. 
It follows from the foregoing considerations thai not only 
does the region generally owe its present elevation to Qua- 
ternary or recent uplifting, but that mountains rising above 
the general level of the region have been uplifted in Quater- 
nary or recent time. 
Canga is by no means limited to the areas above mentioned ; 
but is commonly associated with the other materials of the 
cascalho along the road through the the region of massive 
hematite deposits from Oaro Preto to Diamantina. I did not 
see it in process of formation anywhere, and with present 
slopes soft, precipitated oxides of iron could hardly find rest- 
ing place in considerable quantities. It is possible that when 
the faulting and uplifting were taking place hot springs added 
to the deposits of iron oxides. 
The cascalho consists of the harder and more resisting 
portions of the rocks (from which the finer disintegrated por- 
tions have been removed by wind or water or both) and the 
chemically deposited oxides of iron. A part of the quartzose 
gravels of the cascalho rests on the surface of the rock near 
where its materials had existed in place, while another por- 
tion lies along the streams, having been moved forward by 
them a greater or less distance. These stream-moved gravels 
are always in comparatively thin sheets. I nowhere saw 
them accumulated in thick masses as in California. It fol- 
lows that the uplifting which has taken place in Quaternary 
or recent time has not obstructed the drainage by lessening 
the slopes or raising mountain masses across the pathway of 
the streams faster than the streams could cut down through 
them. The general increase of elevation has increased the 
slope of the streams and increased their power of erosion, and 
caused them to cut their channels down deeper below the gen- 
eral surface, making the region one of steep slopes. The local 
uplifting has also increased the slopes of the streams in places, 
and as it nowhere, within the field of my observation, obstruct- 
ed the drainage so as to cause massing of the gravels, I infer 
that it took place quite generally at least along the same axis 
as the older uplifting which caused the divides and determined 
the position and direction of the streams. 
Near Itambe do Matto-dentro the streams flow over itacol- 
umite and I noticed that they have not eroded channels of 
considerable depth, but flow near the general elevation of the 
