146 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 
lands of southeastern Pennsylvania, for there the streams have formed 
narrow flood plains and the valley sides are for the most part smoothly 
graded even in crystalline rocks ; yet the elevation of these uplands is 
not of remote date. If it is thought unsafe to make a comparison be- 
tween canyons in the arid interior basin of Utah and young valleys 
in our better-watered Atlantic slope, the Wahsatch canyons with their 
perennial streams may be compared with the dry side canyons of the Colo- 
rado canyon in Arizona. The expression of the two is much the same, 
allowance being made for the unlike attitude of the rocks. The chief 
difference between these two groups of canyons is this: those of 
the Arizona plateaus were cut down ina rising plateau mass by inter- 
mittent wet-weather streams working with respect to the sinking local 
baselevel, the intrenching Colorado ; those of the Wahsatch were cut 
down in a rising mountain mass by more persistent streams working 
with respect to a relatively fixed local baselevel. The erosion of the 
Arizona canyons, trunk and branch, cannot have been begun earlier than 
the latter part of Tertiary time; the erosion of the Wahsatch canyons 
may well have had an even later beginning. The date assigned to the 
Wahsatch fault by King, on incomplete geological evidence, is the close 
of the Eocene ; but this seems inadmissibly early in view of the sharpness 
of the Wahsatch peaks and spurs and of the enormous amount of erosion 
accomplished in the plateau province in post-Eocene time. 
It should be noted, however, that certain through-going streams in the 
Provo district have valleys that are more maturely opened than the 
canyons just considered. The so-called canyons of Provo river, Hobble 
creek, and Spanish fork are all relatively open, with moderately steep, 
frequently graded side, slopes. This seemed to me in part due to the 
occurrence of weaker rocks where the through-going streams have cut 
down their valleys, but Iam not sure that this explanation applies in 
all cases. It may be that in some examples the more open valleys are 
connected with differences in the date, amount, and rate of faulting. 
Some of the through-going valleys are nevertheless of true canyon-like 
form: such is Weber canyon, which is followed from the east by the 
Union Pacific railroad into the Great Basin at Ogden; and also Ogden 
canyon, a few miles further north, if the maps may be trusted. 
The part of the Wahsatch range next north of Spanish fork canyon, 
here called the Spanish Wahsatch, Plate 1, a, is beautifully carved with 
sharp ravines which preserve their narrow floor and steep walls directly 
to the mountain base. Two of these ravines were visited. The beds of 
their wet-weather streams pitch forward at an angle of from 22° to 34°, 
