ON ITS EXTERIOR. VOLCANOES. 231 



did not permit scientific institutions to give to this great 

 phenomenon the attention which, at a latter epoch (2d July, 

 1831) was devoted to the ephemeral apparition of the 

 igneous island of Eerdinandea, between the limestone coast 

 of Sciacca and the purely volcanic island of Pantellaria. ( 233 ) 

 The geographical distribution of the volcanoes which 

 have been in a state of activity within historic times, the 

 great number situated on islands or on coasts, and the re- 

 curring phenomena of eruptions from the bed of the sea, 

 early led to a belief that volcanic activity is connected with 

 the vicinity of the sea, and dependent on it for its continu- 

 ance. " Etna and the J^olian islands have been burning for 

 centuries/'' says Justin, ( 234 ) (or rather Trogus Pompeius, 

 whom Justin follows ;) " and how could they have lasted so 

 long if the neighbourhood of the sea did not feed the fire ?" 

 Even in recent times it has been attempted to explain the 

 supposed necessity of the vicinity of the sea, by the hypo- 

 thesis of sea water penetrating to the foci of volcanic acti- 

 vity, or to very deep-seated strata. After comprehending 

 in one view all that my own observation has furnished, and 

 all that I can gather from facts diligently collected else- 

 where, it appears to me that the conclusion in this intricate 

 investigation must depend upon the solution of certain 

 questions ; we must, -for instance, first determine whether 

 the great mass of aqueous vapour unquestionably exhaled 

 by volcanoes, even when in a state of repose, be derived 

 from sea water impregnated with salt, or from fresh water 

 obtained from meteoric sources. In the next place we 

 must decide whether the expansive force of aqueous 

 vapour (which, at a depth of 88000 French feet, is equiva- 

 lent to 2800 atmospheres,) would be sufficient, at the dif- 

 ferent depths of the foci of volcanic action, to counterba- 



