322 REID— THE DISTRIBUTION OF LAND 



remain the same. All that is necessary is that the temperatures 

 should change rapidly near the bases of the cones. 



Let us calculate the difference of mean temperature of the land 

 and sea cones under some simple distributions of temperature. 

 Suppose the apices of the cones to have the same temperature and 

 the bases to differ by i6° C. ; and suppose the temperature gradient 

 to be uniform in each cone from the apex to the base. An easy 

 calculation shows that the difference of the mean temperature of 

 the cones would be three quarters of i6°, or 12°. And with the 

 coefficient of expansion adopted by Trabert, this would account for 

 a difference of level of the bases of 770 meters, or about one seventh 

 of the actual amount. A constant temperature gradient in the 

 earth is of course impossible. With a gradient of 1° per 100 meters, 

 which is rertainly smaller than that observed at the surface, we 

 should have a central temperature of over 127,000°. 



As a second example let us suppose that the earth has cooled in 

 accordance with Lord Kelvin's theory. We shall take the original 

 temperature at 1170°, the present land surface temperature at 16°, 

 the sea bottom at 0°. For the sake of making the difference as 

 great as possible, we shall assume an age for the earth of 500 mil- 

 lion years. We find that below a depth of one tenth of the radius 

 the two cones have practically the same temperature and that the 

 mean difference in the two shells above this depth is somewhat less 

 than 4°, accounting for a diff'erence of level of the bases of the two 

 cones of about 25 meters. 



If we ascribe the earth's heat to radioactive substances, we are 

 confronted with our ignorance of the relative quantities under the 

 land and under the sea. They seem to be somewhat more abundant 

 in the more siliceous rocks of continental areas, though the red clay 

 of the deep sea seems to have a high content ; on the other hand the 

 less siliceous rocks of the oceanic areas have a lower conductivity 

 for heat. We may then as a rough approximation assume that the 

 temperature curve has the same form under the two regions, 

 differing, however, by 16° in temperature at the same depth below 

 the surface. This would only hold for moderate depths, say, for a 

 few hundred miles. Farther than this there would be a diminution 



