FORM OF THE CONTINENTS— WATTS 195 



ranges, the Armorican range, the continuation of the Pyrenees and 

 Atlas; and, on the American side, the uplands of Labrador, New- 

 foundland, and the eastern States, and the hill ranges of Guiana and 

 Brazil. The Atlantic is in disconformity with the grain of the land, 

 while the Pacific conforms with it. The Pacific has the rock-folds 

 of its ranges breaking like ocean waves toward it as though the land 

 were being driven by pressure to advance upon it, while the Atlantic 

 recalls the effects of fracture under tension. 



The middle and southern edges of the Atlantic, however, agree to 

 some extent with the Pacific type. The Caribbean Sea, with the 

 Antilles and the rest of its border girdle, recalls the similar structure 

 of the Mediterranean, as it stretches eastward, with breaks, to the 

 East Indian Archipelago ; while the Andes are continued to Antarctica 

 in a sweeping curve of islands. The rest of the Indian Ocean is of 

 Atlantic type, as seen in the shores of eastern Africa and western 

 Australia. 



Another feature of the Atlantic is the parallelism of much of its 

 eastern and western coasts, the meaning of which has often attracted 

 the speculations of geologists and geographers. With a little stretch 

 of the imagination, and some ingenuity and elasticity of adjustment, 

 plans or maps of the opposite sides may be fitted fairly closely, par- 

 ticularly if we plot and assemble the real edges of the continents, the 

 steep slopes which divide the "shelves" on which they stand from the 

 ocean depths. This has suggested the possibility that the two sides 

 may once have been united, and have since broken and drifted apart 

 till they are now separated by the ocean. 



This view, outlmed by others, has been emphasized by Wegener 

 and dealt with by him in full detail in his work on The Origin of 

 Continents and Oceans, and it now plays a leading part in what is 

 known as the Wegener theory of continental drift. The hypothesis 

 is supported by the close resemblances in the rocks and fossils of 

 many ages m western Europe and Britain to those of eastern North 

 America; by community of the structures by which these rocks are 

 affected; and by the strong likeness exhibited by the living animals 

 and plants on the two sides, so that they can only be referred to a 

 single biological and distributional unit, the Palearctic region. 



The hypothesis, however, did not stop at this; and in the South 

 Atlantic and certain other areas Wegener and his followers have 

 also given good reasons for believing that continental masses, once 

 continuous, have drifted apart. 



Broad areas in southern Africa are built of rocks known as the 

 Karroo formation, of which the lower part, of late Carboniferous 

 age, is characterized especially by species of the strange fernlike fossil 



112059—37 14 



