,507.J AND CONTRACTION OF THE EARTH. 273 



causes together produce the observed earthquakes. This explains 

 the occurrence of a large earthquake belt south of the Himalayas, 

 the existence of which heretofore has been so perplexing. Where 

 the crust is badly broken and fissured, the surface water more read- 

 ily aids the original tendency to relief, depending on the secular 

 leakage of the adjacent ocean. We have treated this question in 

 section 20 of the paper on the cause of earthquakes, but the Hima- 

 layas were not treated in great detail, and it seemed well to point 

 out specifically that the same tendency to relief, which we find 

 around the margins of the Pacific, holds equally true in the great 

 western extension known as the Indian Ocean. Here at present 

 relief is obtained chiefly along the northern and northeastern bord- 

 ers, though some is afforded in islands and various other places. 

 The Indian Ocean not only has high mountains on its margins, but 

 also a considerable number of islands and volcanoes. In this re- 

 spect it resembles the Pacific. 



A study of the principal mountain chains suggests that when- 

 ever a serious break has once occurred in the crust of the earth, 

 the strain arising in the underlying layer continues to find relief 

 by the escape of steam-saturated lava into the avenue thus opened, 

 which is the path of least resistance. The formation of the Alps, 

 Andes and other mountains, as well as the Himalayas, illustrate 

 this principle. 



Whether the motion of lava towards the avenue of escape is 

 by creeping flow, or by small earthquakes, we do not know ; pos- 

 sibly it may be by both methods. It is only in violent world-shak- 

 ing earthquakes that we can be sure that lava is moved beneath 

 the crust en mass over considerable distances. These streams of 

 lava are proved to exist by the uplift of coasts and by the sinking 

 of the sea bottom implied in seismic sea waves. 



Such subterranean movements correspond closely to Plato's 

 Pyriphlegethon, but when once a fault has moved for a long time 

 upward, the strain may become so great from the way in which the 

 blocks of the earth's crust are wedged together, that the upward 

 movement there becomes more difficult, and a neighboring region 

 affords easier relief. And in pushing up the neighboring area, the 

 previously elevated block may be let down again, by the relief 



