MARVINE ON SUPERIMPOSED VALLEYS. 165 
structure and characters of the upper rocks, and when they gradually cut 
down through these and commenced sinking their canons into the under 
lying complicated rocks, these canons bore no relation whatever to their 
complications. It is but recently that the upper rocks have been completely 
removed from the summits of the mountain-spurs, the ancient level of sub 
aqueous erosion being still indicated by the often uniform level of the spurs 
and hill-tops over considerable areas, and large plateau-like regions which 
became very marked from certain points of view. Two or three such levels 
are indicated at a few places, showing not only that the sedimentaries have 
once extended up over what are now the mountain rocks, but that the uplift 
ing has been mainly confined along certain partly well-defined lines, the 
intermediate belts, though uplifted bodily, remaining comparatively level, a 
type of folding, probably, not uncommon farther west, and which will be 
referred to again in the following chapter. 
******* 
"It is true that the structure of the lower rocks has begun to affect the 
courses of the streams, and in places to a considerable extent. Meeting a 
softer bed a canon will often have its course directed by it, and follow it 
for some distance, leaving the adjacent harder beds plainly indicated by the 
ridges, and sometimes the sinuosities of structure are very curiously fol 
lowed by a stream in all its windings, but it soon breaks away and runs 
independently of the bedding. Many of the smaller ravines have had their 
positions determined by the structure; but in a broad sense the drainage is 
from the main mountain crest eastward, independent of structure. Thus, 
while in places geological features may find expression in surface form, yet, 
as often, there may be no conceivable relation between topography and 
geology. The subaqueous erosion, in smoothing all to a common level, 
destroys all former surface expression of geological character, and the 
present erosion has not yet been in progress sufficiently long to recreate the 
lost features." 
I fully concur with Mr. Marvine in the above explanation of the valleys 
in the main Rocky Mountains of Colorado, as my own observations in that 
country had led me to the same conclusion. There can be no doubt that 
the present courses of the streams were determined by conditions not found 
