230 The American Geologist. October, im;7 
trast to much of the land adjoining, but wliat is more surpris- 
ing, the valleys beneath the water do not in some cases seem 
to bear any relation to the topography of the land lying op- 
posite. The most of the depressions do however appear to be 
related to corresponding depressions in the land toward which 
they head. 
Now the question arises as to Ihe manner in which this 
bench has been formed. Is it due to truncatit»n or sedimenta- 
tion or to structure? In its broad features I believe that We 
must hold it to be of structural origin, but nevertheless mod- 
ified by both the other agencies. In general the width of the 
plateau bears very little relation to the important drainage 
features of the land, and even whereat first sight there would 
appear to be a correspondence supporting the view that the 
plateau has been built up by sedimentation a closer examina- 
tion shows that such is probably not the case. The plateau 
widens opposite the Golden Gate where tlie floods from the 
Great Valley sweep out into the ocean, but even here the out- 
er edge of the bench is bordered by the Farraloues islands 
which are formed of the ancient granites of the Coast Ranges. 
If the even surface is due partly to truncation and partly to 
sedimentation Avithout any greater elevation than the present 
ever having been experienced it is most diflicult to account 
for the transverse valleys. Faulting cannot possibl}'^ explain 
them. It seems to me that as first suggeste^l hj professor Le 
Conte, and in harmony with the other lines of evidence, the 
position and character of these valleys point to a former ele- 
vation of from 2000 to 3000 feet. It is true that the subma- 
rine valleys may have been originally cut out at the post-Mi- 
ocene elevation, or even earlier, but the fact that tliey Iiave 
not been filled and extend up so close to the shore in j)Iacep 
points most conclusively to a coniparativelj'- recent elevation 
of the coast. How faulting without subaerial erosion could 
have produced the contours of the valleys as given by David- 
son it is impossible to conceive, and when to this is added the 
fact that the}^ do not conform as a general thing to the im- 
portant structural features of the Coast Ranges We are com- 
pelled to seek a different explanation. In cases where a sub- 
marine valley comes up to the shore directly opposite the 
mouth of a land valle}^ where the Pleistocene is at least hun- 
