CHARLES B. WARRING. 259 
they would form vast depressions, and these now filled 
with water are the present oceans. He adds; The cool- 
ing of one part of the crust before the rest must have 
been a consequence of less vivid heat and violent action 
in the liquid rock of that part, and this, he says, may 
have been connected, with the exterior of the solid 
nucleus, being nearer in that part than elsewhere to the 
outer surface of the sphere, that is to there being less 
depth of liquid rock over it to cool, 
To this explanation two objections present themselves. 
It is, I think, extremely doubtful whether solid granite 
would sink in the molten mass, a doubt that becomes 
almost certainty, when we reflect that astronomy has 
proved that the interior of the earth is very much 
heavier than its crust. But admitting that, what Prof. 
Dana says, explains the existence of the vast ocean 
basin which now separates the continents, it does not 
even attempt to explain the parallelisms which are 
found, except to refer them to lines of cleavage existing 
in the earth’s crust. It is diffleult to see how lines of 
cleavage could affect coasts which had always been 
thousands of miles out of reach of each other. If 
cleavage planes exist, they could easily produce their 
full effect without the parallelism which we now see. 
South America, for example, might have had that pro- 
jection which now terminates at cape St. Roque, turned 
equally far to the west of the meridian which runs from 
cape Horn to the gulf of Maracaibo. The coast lines 
would remain parallel to the same great circles as now, 
only ina quite different order, and the parallelism be- 
tween South America and Africa—the thing for which 
we are seeking a cause—would disappear. 
Humboldt offers a very different explanation. It is 
that the Atlantic basin was eroded by an ocean current. 
The present Atlantic currents are too feeble and too 
superficial to produce any erosive effect on the bottom 
143 
