

CAUSES WHICH ORIGINATE EARTHQUAKES. 325 



tions of the earth's crust ; and a similar conclusion is indicated by the 

 well-known fact that the Swiss earthquakes mostly happen at night. 

 But the great earthquake vibrations, which have been measured to 

 extend to depths which vary from a mile or two down to more than 

 thirty miles, are clearly connected with the great internal earth-move- 

 ments which are consequences of the contraction of the earth's crust 

 from cooling. And while volcanoes are intelligibly accounted for as 

 distributed in lines of predominate anticlinal folding, we are unable to 

 account for many earthquake phenomena unless they are produced by 

 increased compression in regions of predominant synclinal folding. 

 For if the base of a synclinal fold is fractured owing to augmented 

 contraction of the rocks, and the fissure is formed at a depth of miles 

 beneath the surface, then the distinction of earthquakes from vol- 

 canoes becomes intelligible, and their frequent development in plains 

 and at sea, rather than in mountains, is such as might theoretically 

 have been anticipated from their frequent independence of volcanic 

 outbursts, and the dimensions of the areas affected. 



Changes of Level in Land. 



On the Elevation and Depression of Land. Whatever may 

 have been the earth's earliest cosmical relations, it appears first in 

 geological history as a spheroid of revolution, whose parts have taken 

 their relative place under the joint influence of gravitation to the centre 

 and rotation on an axis. The density increases toward the centre, the 

 surfaces of equal density are elliptical to the same axis as the external 

 oblately spheroidal surface. 



This spheroid cools by radiation ; contraction of the whole miss 

 follows, so that the crust is pressed into accommodation with the 

 interior. Thus inequalities of the surface would be occasioned ; and 

 from the beginning a continual system of reciprocal depressions and 

 elevations would be established. The consequence would be, that the 

 surface of the spheroid would be wrinkled by folds of elevation and de- 

 pression, growing more and more deep, and with the progress of time 

 more and more complicated. In remarkable harmony with this view 

 is the well-known fact of the frequency of anticlinal and synclinal and 

 more complicated flexures of the palaeozoic strata in all parts of the world, 

 flexures which were often completed before the close of that period. 



In later periods of the earth's contraction, local inequalities of 

 consolidation partly dependent on the earlier flexures, partly pro- 

 duced by the inequality of molecular aggregation, as by the separation 

 of different orders of silicates, calcium sulphate, or magnesian car- 

 bonates may have overcome the phenomenon of reciprocal depression 

 and elevation, and limited the areas in which elevation or depression 

 might take place and in which one might follow the other so that 

 the same tract might be alternately raised and depressed. 



In nearly all cases depression must be supposed to be real and 

 gradual, that is to say, part of the earth's surface affected by it must 

 be gradually carried nearer to the centre than it was before ; the 



