774 C. B. BROWN ON THE ANCIENT 
a time when the Amazon commenced flowing over the lowest level 
portions of a great plain, which rose gradually above the sea to the 
west of the vicinity of the great Tertiary estuarine strata on the 
Javary river, and extended eastward until it reached considerably 
beyond its present limits; for it is probable that the Amazon (a 
shorter river then by 2000 miles) was one of the rivers which flowed 
from the Andes into the estuary in which the Tertiary beds were 
formed. The rocks of the plain were chiefly of comparatively soft 
materials, such as sandstane, shale, and limestone ; although in some 
parts of its borders they were of gr anite, gneiss, and quartz- Sporphyzy. 
On these substances the river ihesitin to exert its wearing action, and 
to move gradually sideways in a southerly direction in one part, and 
in a northerly in another, over the plain, cutting slowly deeper and 
deeper as it thus moved, and depositing the materials in its current- 
less portion (see diagram, fig. 8). 
By the time it had crossed its valley at any given part (see fig. 9) it 
was at a much lower level than at the side it was deflected from, and 
then, the direction of its curves becoming altered, it travelled back ~ 
to the vicinity of its starting-point at the extreme limit of its 
valley on that side (see fig. 10). There, from having cut deeper and 
deeper as it went, it has a cliff composed of materials of its own de- 
position as a limit ; while on the other side it left a cliff, D (see fig. 10), 
of the rocky strata of the plain ; and stretching from that latter cliff 
to the river’s edge is a great alluvial plain, partially submerged when 
the river is .at its highest stage. By supposing that the river once 
more recrossed its plain as in fig. 11, I think we have the conditions 
that are now in force on the Amazon in the vicinity of Monte Alegré 
and its barreiras. For if the upper portion of the higher tableland 
is composed of an old deposit of the river, it was formed under the 
conditions shown in fig. 8, and is represented in fig. 11, at B, as the 
first laid down alluvium ; while the old river-deposit is represented 
by G in the same diagram, and was formed under the conditions 
shown in fig. 10. The recent alluvium a present bordering the 
Amazon is given at I in fig. 11. 
In some places on the Amazon (at Obidos, for instance), where the 
river in its ordinary state comes only to about the level of the base 
of its old-deposit cliff, we must infer that in crossing and recrossing 
its valley it cut gradually downwards a distance equal to about twice 
its depth. There we find that the thickness of the old deposit cor- 
responds in measurement with the depth of the river, as it naturally 
should. Here, then, we have a proof that the thickness of materials 
deposited by a river is not greater than the depth of the river; and 
likewise that the strata beneath the bottom of its bed are not composed 
of alluvial matters, as we are led to believe by viewing some sections 
in geological works. By a study of river-action, as shown in dia- 
grams 8 to 10, we see that it is impossible for a river, in its non-tidal 
portion, to have deposited materials at a lower level than the deepest 
part of its bed. This can only take place near its mouth, where it 
is forming a delta, or in the case of a gradual rise of country near its 
mouth arresting its flow and converting it into a lake. In both 
