588 Mr. H. Jeffreys on the Compression 



heat to pass outwards in time, yet the ocean will neverthe- 

 less act as a thermal blanket, reducing the cooling below it 

 and hence the surface compression. 



For both reasons, then, an oceanic area will tend to 

 require round its margin a smaller horizontal pressure 

 than the mean, if it is to remain spherical. Similarly, 

 a continental one requires a larger pressure round its 

 margin. These two statements are inconsistent, since by 

 the Third Law of Motion the pressures should be equal. 

 Evidently, then, the earth cannot remain spherical. The 

 pressure is inadequate to compress the surface layers of 

 the continental areas so as to keep both these and the 

 ocean areas part of the same sphere, so that the outer 

 layers on the continents must expand somewhat relative 

 to the inner ones, the margins remaining fixed by cohesion. 

 The only way in which this can happen is by a reduction 

 of the radius of curvature of the crust — in other words, the 

 continents will tend to rise. Similarly, the ocean bed will 

 sink. The effect capable of being thus produced is very great. 

 Thus, if the level of no strain were at a depth of 70 km., 

 and a surface layer, 3000 km. in length, were 3 km. too long 

 to continue to fit the inner sphere, the radius of curvature 

 would be increased by as much as 500 km., and the elevation 

 would be 15 km. in the centre. The estimate of the available 

 excess of length is probably conservative. This differential 

 compression is then probably a very important cause of 

 the elevation and maintenance of continents and oceans. 

 The adjustment of the figure of the earth to make the 

 compression constant all over is not likely to be complete, 

 however, as further stresses would be produced in the 

 crust, which would cause fracture. The adjustment is 

 not then likely to be much greater than would give iso- 

 static compensation of the changes produced by denudation. 

 The remaining part of the differential compression will be 

 shown in extra folding in the continents, and diminished 

 compression, or even, in extreme cases, a tension, in the 

 ocean bed. 



Mr. Holmes has called my attention to the importance of 

 another factor. The crumpling here considered is that which 

 has occurred at any depth since the level of no strain reached 

 that depth. Before it had gone so far down, there would be a 

 considerable tension in the interior of the continents, which 

 might be capable of causing rifts and accelerating intrusions. 

 Such rifts would occur very soon after the denudation. The 

 flexure of the crust here indicated would, on the other hand, 

 increase steadily with the time, and is essentially subsequent 

 to the denudation. 



