212 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 1932 



Greenland-Iceland elevation, runs through both Atlantics until it 

 finally ends in the seventieth parallel of latitude south. 



The basins are laiown to be subdivided by transverse small rises, 

 so that we have the whole floor divided into basins and domes. So 

 far, however, the folded mountain chains which end, as it seems 

 abruptly, against the Atlantic on the Eurafrican side have not been 

 traced with any definiteness out into the Atlantic, although repre- 

 sentatives of these folds, of like age and similar direction, are found 

 on the western side, as pointed out long ago by Marcel Bertrand and 

 Eduard Suess. The mid-Atlantic Rise has been regarded by dif- 

 ferent writers as due to different causes. Haug suggested that the 

 whole Atlantic may be regarded as an enormous geosyncline, and 

 the Atlantic Rise he looked upon as the first evidence of the folding 

 into mountain chains of the accumulated deposits. By others the 

 depressions on each side are regarded as rift valleys let down on 

 account of the movement of America and Eurafrica and so widening 

 the Atlantic Ocean. There is, of course, nothing impossible nor 

 inconceivable in such prodigious faulting as is here implied. The 

 boundary faults of the Central Valley of Scotland are of the same 

 order of magnitude vertically, and in length the Atlantic faults 

 may be compared with the Palestine-Red Sea-East Africa rifts. 

 Such a structure would also be in agreement with that surrounding 

 the Pacific Ocean as to magnitude, but there we have the Pacific 

 type of coast line with its mountain chains parallel to the coast and 

 all overturned toward the Pacific Ocean, while except for the 

 AUeghanies in North America, the Atlantic coast line is marked by 

 mountain chains which run transversely to it. It is to be noticed 

 that the parallel mountain chains of eastern North America are over- 

 turned away from the Atlantic Ocean, and not toward it. 



As I have already pointed out, we can not expect from the nature 

 of the evidence to learn much of the past history of the " dry-land " 

 areas from the biological side. On the other hand, investigations 

 on the physical side can at once yield results of far greater value. 

 This investigation may be applied to test the case of the North 

 Atlantic, and the particular physical test which may be worth while 

 undertaking is that of the character and distribution of the sedi- 

 ments which through the successive ages have been derived from the 

 area now occupied by the North Atlantic and the adjacent lands. 



III. THE ARCHEAN OR PRE-CAMBRIAN PERIOD 



The Torridonian rocks are composed in large part of red felds- 

 pathic sandstones or arkoscs and represent continental deposits, as 

 is evident in their petrographical characters and their relation to 



