SEPT. 19, 1923 WASHINGTON: COMAGMATIC REGIONS 343 
uniformly characterized by being rather high in both soda and lime 
and rather low (for granite) in potash. Although much granite exists 
in the corresponding parts of western Europe it does not show the 
uniformity in composition that is the striking feature of the Appa- 
lachian region. 
In the first edition of his book Wegener assumed that France, the 
# Iberian Peninsula, and northwest Africa were conterminous with the 
southern Atlantic States and the West Indies, but in later editions 
they are separated by a rather narrow inlandsea. The interior of the 
Iberian Peninsula is largely granitic, but we know practically nothing 
of these rocks, especially as to their chemical characters. They may 
or may not correspond to the granites of the Appalachian region, but 
the different configurations of the two masses is against such a corre- 
spondence. Several areas of highly sodic rocks occur in Spain and 
Portugal, but these do not have any corresponding areas in the most 
southerly states, South Carolina, Georgia, and Florida, which should 
have adjoined Portugal, although similar highly sodic areas exist 
many miles inland and to the west in Arkansas and Texas. 
The Lesser Antilles form a region with somewhat peculiar and 
unusual petrologic features, the voleanic rocks showing as their chief 
and most notable character simultaneously high silica and lime, al- 
though the underlying plutonic igneous foundation appears to be of a 
more usual character. The Mexican region, to the west, is somewhat 
similar to the Antillean, especially in its high silica, but is rather more 
normal. Of the petrology of the west coast of the great African bulge 
or protuberance we know little, but that little, such as the abundance 
of basaltic and charnockitic rocks in Liberia, Senegal, and French 
Guinea, indicates that the eastern shore of the Atlantic in these 
latitudes is very different from the western. 
We come now to the crucial portion of the line of fission—the north- 
eastern corner of South America and the north and east shores of the 
Gulf of Guinea. The apparent exactness with which the one of these 
could be fitted into the other first suggested the idea of former juxta- 
position and. subsequent separation. Unfortunately we know com- 
paratively little in detail of the igneous rocks of much of these two parts 
of the globe, but we have sufficient knowledge of their broader features 
to permit us to consider the correspondence or non-correspondence of 
the rocks as very weighty evidence. 
The eastern part of the north littoral of South America, comprising 
the Guianas and the States of Pard, Maranhao, Ceard, and Rio 
