76 RECONNAISSANCE OF PART OF WESTERN ARIZONA. 
reaches. It is evident, therefore, that even in the newest channels, 
such as that at Bulls Head, bed rock is not to be expected as near 
the surface as it would be if the river were in an eroding stage. In 
other words, it is probable that the rock channel is deeper than the 
maximum depth of scour. 
The flood-plain deposits, so conspicuous in the lower reaches of the 
river, become less conspicuous upstream, until in Black Canyon rapids 
of notable proportions occur. In the absence of soundings in the 
channel for bed rock farther north than Bulls Head, it is impossible 
to say that the channel is not filled in the northern portion of the 
region as it is farther south where soundings have been made. The 
swift current washes the fine material away and leaves large bowlders 
clogging the channel even where the grade of the river is steep enough 
to form dangerous rapids. It is the writer's impression that the 
river is flowing essentially on rock bottom through Black, Boulder, 
Virgin, and Iceberg canyons, with only such obstructions of bowlders 
as are incidental to variations in the river's power of transportation; 
in other words, that there is no permanent filling in the bottom of 
those canyons. 
The deflection of the river previously described has resulted in the 
cutting of young channels in many places for short distances, as at 
Eagle Rock, Bulls Head, Great Bend, etc. It is in these young chan- 
nels that bed rock is to be expected at the least depth. The relation, 
however, to the old debris-filled channel at the side may be such as 
to influence this depth. For example, at Bulls Head the depth to 
bed rock in the channel now occupied by the river might be greater 
near the edge of the old channel, owing to the fact that the new 
channel might coincide with some lateral wash of the old, while north 
of Bulls Head Rock, midway of the young channel, bed rock might 
be much nearer the surface. The depth to which a large river works 
is known to be great, although there is little information at present 
to indicate the maximum depth to which Colorado River works. The 
records of Missouri River, previously cited (p. 66), indicate the 
possibility that a depth of 73 to 100 feet, or even more in narrow 
channels, such as that at Bulls Head, is to be expected as a result of 
river abrasion alone. 
Considering further the depth of scour, it is evident that the river 
will cut to a less depth in wide passages than in narrow passages 
where the current is swift. The narrow places are usually selected as 
dam sites on account of the presence of favorable abutments com- 
paratively close together. It is possible that in certain cases the 
depth to bed rock in the wide part of a gorge, where the abrasive force 
is weak owing to a diminution of velocity, might be enough less than 
in the narrow parts, where the current is swift, to more than compen- 
sate for the greater length of a dam. 
