242 
THE GEAl^D CAS^OjST DISTEICT. 
in volume at flood time, has less velocity than the torrents of the side 
chasms. The river has, however, sufficient power to sweep onward masses 
of considerable size, which are rapidly ground up as they are rolled along. 
It is apparent that in the work of corrasion an important part consists 
of the work of grinding up and destroying the masses Avhich are brought 
into the main chasm by the spasmodic, floods in the side gorges. Indeed, 
this constitutes by far the greatest part of the entire work. The coarse 
material — the large rocks, bowlders, and rubble which pile up at the mouth 
of the lateral chasm — are gradually spread out below the dam, and the 
tendency is to build up and increase the grade of the smoother reach 
below. But this tendency is quickly checked and brought to a stop by 
the increased power of the main current due to the increased slope. The 
body of fragments thus rolled in is of great amount in the aggregate. On 
the whole, the amount at the present epoch is not sufficient to prevent 
the river from cutting down its channel, though the process is of course 
greatly retarded. The river is still sinking its chasm in the strata. There 
are many stretches of comparatively still water where there is an equilib- 
rium between the tendency to cut still deeper, and the tendency to build 
up the bottom by the accumulation of ddbris. But a great part of the 
river bed is in the bare rock of the Paleozoic and Archjean strata, and 
wherever it is so the corrasion is proceeding at a rapid rate. Still other 
cases occur where the rate of corrasion is retarded but not completely 
counterbalanced by the accumulation, and these no doubt constitute the 
greater part of the extent of the chasm. 
Tlie great corrasive power of the Colorado is due to the large quantity 
of sand which it carries, and the high velocity given to its waters by its 
great declivity. As we have already shown, the quantity of fine material 
brought into the canons is very large and the supply is almost continuous 
throughout the year. As Mr. Gilbert has well shown, a river may be pow- 
erless to corrade when its waters carry no sediment, and also when the sedi- 
ment is so excessive that it cannot transport the entire supply. Neither of 
these extremes is found in the Colorado. The waters are heavily charged 
but not overloaded. It must be noted, however, that in the smoother por- 
tions where the declivity is locally very small, the sand is often dropped, 
