DAVIS: THE PLATEAU PROVINCE OF UTAH AND ARIZONA. 29 
cycle of erosion, and this makes the Hurricane fault as old as the Sevier- 
Toroweap fault. 
The point from which these observations were made is not surely 
identified on Dutton’s map, but it is probably the basalt-capped bluff at 
lat. 36°221/, long. 113°16}/, on Atlas sheet VIII. The lava-capped 
Permian slope lies just beyond a ravine next north of the bluff, from half 
a mile to a mile distant, and quite unmistakable as to structure; but 
on Dutton’s map the corresponding slope is colored to show a slanting 
sheet of basalt, and it is especially on this account that I am uncertain 
as to its identification. 
An important corollary of the above conclusion regarding the date of 
the fault is that when a general uplift of this broadly eroded region took 
place, introducing the canyon cycle, the renewed work of erosion was not 
limited to the carving of the canyon alone; it must have included the 
sweeping of weak strata such as the Permian clays from wide areas, as 
was indicated in my previous essay (b, p. 136). If the history of the 
region is thus correctly interpreted, the present inequality of level be- 
tween the Shivwits and the Uinkaret plateaus is only very indirectly 
the result of faulting. It is directly the result of erosion by which a 
heavy mass of weak Permian clays has been quickly worn off of the 
Shivwits plateau, while only a small amount of carving has been accom- 
plished in the resistant Aubrey beds of the adjoining Uinkaret plateau. 
The Hurricane ledge is therefore a cliff developed by erosion on a faulted 
mass, as in the foreground of Figure 8; it is not a “fault cliff” in the 
proper sense of that term, except in so far as modern faulting, late in 
the canyon cycle, may have increased the escarpment of erosion. We 
were unable to recognize any such element in the view from the bluff 
near Coal spring: if modern movement has taken place in this locality, 
it can have hardly occurred on the ancient surface of dislocation, for the 
lava bed that crosses the fault line did not seem to be broken. 
Renewed movement has, however, taken place elsewhere in association 
with the Hurricane fault. Dutton states that certain modern lava beds 
are somewhat dislocated on a branch of the Hurricane fault in Quanto- 
weap valley not far north of the canyon (b, p. 116, 117), just as they are 
at the mouth of south Toroweap valley. But much greater modern 
movement, long after the strong ancient movement, has taken place 
seventy miles north of the canyon where Virgin river crosses the fault 
line and beyond, as will be fully set forth in the report by my com- 
panions, Messrs. Huntington and Goldthwait. 
The great contrast in the resistance of different rocks to erosion may 
