2 32 The American Geologist. 
April, 19115 
origin. The shelves partly cut by erosion and partly built 
up by waste may be compared with raised cut-ledges and 
raised beaches and terraces built up by waste, (v, p. 193). 
The combined sub-aerial and marine denudation have had 
the greatest effects during periods of oscillations of the 
shore lines. During elevation the streams would incise 
the coastal plains, and these dissected would furnish the 
most favorable conditions for marine erosion. By such 
repetition broad shelves would be formed, and most coasts 
have probably undergone many oscillations. "The effect 
of marine denudation can never become great on smooth 
undissected coasts ; dissected coasts will be cut back till 
they become smooth." The continental shelves of the 
world have been subjected to changes of level simultan- 
eously with and after, their formation ; they have only to 
a very small extent been formed after the time when the 
present shore line was attained (p. 194). • 
The shelves of Iceland and the Faeroe islands have 
been built up in Pliocene and Pleistocene times, and Nan- 
sen believes "that the continental shelves of other regions 
are to a great extent of similar age" (p. 186). He con- 
cludes with Hull, that those of the Scotch islands belong 
to a similar date, and he cites the pres,ent reviewer as show- 
ing that the age of the valleys dissecting the American 
continental shelf corresponds to that of Europe (p. 187). 
The canyons and channels must have been made subse- 
quently to the shelves. For example, he says that Fosse 
de Cap Breton" is perfectly incompatible with' the assump- 
tion that the shelf was formed after the last submergence 
of the drowned valleys, if we assume that this remarka1)le 
ravine is an ancient river canyon, other than which I (he) 
think there is no feasible explanation" (p. 190), thus strong- 
ly supporting Hull and others in this and other canyons 
off the coast of France. He sees no evidence against the 
existence of the submarine valleys and their being filled 
with drift and leveled over with wave washes during os- 
cillations. 
"We see that it is difficult to find conclusive evidence that 
the continental shelf * * * has been elevated into dry land 
after formation; but the drowned valleys and fjords at many 
