strata above the canon, making at one time a thickness of over a
mile, were worn away in Pliocene times, before the cutting of the
Grand Canon began. Had they remained, and been cut through, we
should have had a chasm two miles deep instead of one mile.

The cutting power of a large, rapid volume of water, like the
Colorado, charged with sand and gravel, is very great. According to
Major Dutton, in the hydraulic mines of California, the escaping
water has been known to cut a chasm from twelve to twenty feet deep
in hard basaltic rock, in a single year. This is, of course,
exceptional, but there have, no doubt, been times when the Colorado
cut downward very rapidly. The enormous weathering of its side walls
is to me the more wonderful, probably because the forces that have
achieved this task are silent and invisible, and, so far as our
experience goes, so infinitely slow in their action. The river is a
tremendous machine for grinding and sawing and transporting, but the
rains and the frost and the air and the sunbeams smite the rocks as
with weapons of down, and one is naturally incredulous as to their
destructive effects.

Some of the smaller rivers in the plateau region flow in very deep
but very narrow canons. The rocks being harder and more homogeneous,
the weathering has been slight. The meteoric forces have not taken a
hand in the game. Thus the Parunuweap Canon is only twenty to
thirty feet wide, but from six hundred to fifteen hundred feet deep.

I suppose the slow, inappreciable erosion to which the old guide
alluded would have cut the canon since Middle Tertiary times. The
river, eating downward at the rate of one sixteenth of an inch a
year, would do it in about one million years. At half that rate it