as a Cause of Earthquakes. 89 



shocks most commonly followed the direction of chains of 

 mountains. The same observation has been made with re- 

 spect to the West Indies by Jonnes in his Hist. Phys. des An- 

 tilles, p. 104. This circumstance shews that earthquakes, 

 except in the vicinity of volcanos, are not owing to the ex- 

 pansive power of gases or vapours, which most probably do not 

 extend to any considerable distances, as they would soon be 

 condensed or absorbed by water, or else so far expanded as to 

 lose their power of acting with sufficient force. 



It is frequently observed during violent earthquakes, for in- 

 stance those of the Caraccas in March 1812, and of the plains 

 of the Mississippi in 1811, that the surface of the ground has 

 been in a continual undulating movement and heaving up like 

 a boiling liquid. This agitation may be explained with far 

 greater probability by the supposition that the action upwards 

 arises from an under body of water, than from the expansion 

 of gases or vapours. The reiteration of the shock of an earth- 

 quake after certain intervals, may be owing to the subsidence 

 of the elevated strata having closed the principal openings, 

 which caused another oscillation after a sufficient influx of 

 water had taken place. 



Earthquakes very seldom, if ever, occur, except where there 

 are volcanos, in very high latitudes that have the surface of 

 the ground covered almost perpetually with ice or snow. In 

 these countries the soil probably remains frozen to a very 

 great depth, consequently there can be no action of water ca- 

 pable of upheaving it. 



If hydrostatical pressure be the cause of earthquakes, they 

 would not occur, as is the case, on very elevated surfaces or 

 steppes, except in a very slight degree, from the motion com- 

 municated by distant convulsions of the earth. 



