W. O. Crosby—Origin of Continents. 243 
not supported by a vestige of evidence, viz. that the earth was 
originally, and is now, of unlike composition along different radii, or 
on different sides; the continental portions of the crust, according 
to Prof. Dana, being composed of denser materials than the oceanic, 
while according to Prof. Le Conte, the relation is just the reverse. 
If the liquid globe had possessed this constitution, the ellipsoidal 
form of the equatorial section, which under the most refined 
measurements of geodesy almost disappears, would, in obedience to 
the laws of hydrostatic equilibrium, be strongly marked. 
It seems strange that an assumption so vital to these theories 
should have been made without any attempt to demonstrate its 
validity. What are the facts that support it? Where are the 
analyses showing an essential difference in composition between the 
continents and ocean-bottoms? We may safely say that the known 
facts and the probabilities are all against the supposition that such 
a difference exists. But without this unproved difference in com- 
position there could be no difference in conductivity and radial 
contraction, and the theories entirely fail. 
However, granting the possibility of this diversity of composition, 
it still fails as a foundation for the theory that the continents are 
fixed, since it could only have been a very transient characteristic 
of the crust. For, as Prof. Dana admits, the elevation or subsidence 
of large areas of the crust must involve a horizontal displacement of 
liquid material beneath, material under the Pacific, for example, 
being squeezed under the bordering continents. This process, of 
course, mixes up the matter which by cooling forms continents and 
ocean-basins through unequal contraction; and the areas of high 
and low conductivity are no longer kept distinct. 
But we may grant farther the possibility of a permanent difference 
in composition, and still doubt the necessity of Professor Dana’s 
inferences. As a rule, dense bodies are not only good conductors of 
heat, but also have low fusing points. This is eminently true of 
the main constituents of the earth’s crust. The most favourable 
supposition that could be made for Prof. Dana’s theory would be 
that the continents have, or had originally, the composition of 
basalt; and the sea-floors the composition of granite; and in any 
case the difference in composition must be regarded as similar to, 
but less rather than greater than, that between basalt and granite. 
If, however, areas of liquid basalt and granite have the same 
initial temperature, and cool under identical conditions, it does not 
necessarily follow that the basalt areas will solidify first. The 
basalt, on account of its greater conductivity, loses heat more rapidly 
than the granite; and yet, in view of the higher fusing point of the 
granite, the probabilities are that it would first assume the solid 
state, the two rocks being wider apart in their fusing points than 
in their power of conducting heat. This conclusion is abundantly 
sustained by the observations made on the relative liquidity of basic 
and acidic lavas. 
In this connexion, Prof. Dana offers the following explanation of 
the fact that the land is mainly in the northern hemisphere :—The 
