A. Vaughan — Corrugation of the Earth's Surface. 265 



So far we have reasoned as if the whole globe were made up of 

 homogeneous shells. But the crust is composed, and has been so 

 throughout all geological time of which we have any accurate 

 knowledge, of rocks differing greatly in their capabilities of ex- 

 pansion and contraction ; and further the greater part of the crust 

 has its temperature determined by earth-heat, or in other words 

 possesses heat which it must be gradually losing. It is consequently 

 important to obtain some idea of what would occur during the 

 cooling of a large heated mass, whei'e outer shells were composed 

 of materials with different coefficients of expansion. 



The first effect, caused by the cooling of the outermost shell, will 

 be to produce a pressure upon the interior; but the strain thus set 

 up in the outer shell will be gradually relieved by the splitting 

 permitted between materials of different powers of contraction. 



Thus the outermost shell will consist of a great number of 

 slightly disjointed portions. This process will be continued 

 throughout the cooling of the outer heterogeneous shells. When 

 the interior, supposed homogeneous, begins to contract, the dis- 

 jointed parts of the outer shell will be brought closer together to 

 accommodate themselves to a smaller area — a movement of which 

 they will always be capable without the formation of folds or rucks, 

 since the loss of temperature is necessarily greater the nearer the 

 surface. 



Applying this reasoning to the contraction of the crust of the 

 earth, it seems probable that jointing must ensue in the manner 

 described above in all such rocks as have heat to lose. We do not, 

 however, maintain that this is the only way in which such splitting 

 can be brought about, for, no doubt, the shrinking from loss of 

 moisture has always been a very effective cause. 



We will now attempt to obtain some estimate of the amount of 

 jointing necessary in order that the outer shells may follow a con- 

 tracting interior. Let us assume a shell at the depth of a few miles 

 to have its teuiperature diminished 10°, and further, let us suppose 

 that this contraction is unresisted by a more slowly contracting 

 interior. 



For the purposes of actual calculation let us suppose the material 

 of which the shell is composed to have a coefficient of linear con- 

 traction equal to -00001174 (this is the coefficient of sandstone, 

 which is intermediate between the coefficients for cast and wrought 

 iron, and is greater than that for marble). From these data the 

 contraction of one mile is found to be 7-2 inches. This, then, 

 is the amount by which one mile length of the next overlying 

 layer must be able to contract to avoid rucking up. 



Putting this statement into another form ; every yard of shell 

 just overlying the contracting interior must be capable of shrinking 

 ^io- of an inch to avoid rucking. The amount required at the 

 surface will not appreciably exceed the above amount ; for, if we 

 imagine contraction, to take place at a depth of five miles, the 

 required amount of shrinking in each yard of arc at the surface will 

 be approximately if ff X 2^io of an inch. 



