356 ON VOLCANOES AND EARTHQUAKES. 



The geological connexions of volcanoes have not 

 often been clearly ascertained, nor perhaps are they 

 of much moment. If the seat of the fire is as deep 

 as the phenomena would indicate., their appearance 

 and place can have no necessary connexion with any 

 of the rocks at the surface; since they must break 

 out through whatever obstacles happen to oppose 

 them. It is generally difficult to make the obser- 

 vations, because the surrounding rocks are, in most 

 cases, covered by the produce of their eruptions, so 

 as to be rendered invisible. Hence volcanic rocks 

 alone are to be found in the neighbourhood. Those 

 of Spanish America are often indeed situated among 

 traps and porphyries ; but there can be no doubt that 

 these are volcanic products, ejected under a degree 

 of pressure which has prevented the disengagement 

 of the aeriform matters. If they could be proved to 

 be trap rocks, formed therefore before the emergence 

 of the present strata, or at that time, they would esta- 

 blish the identity between trap and volcanic pro- 

 ducts, and also prove that the seat, as well as the 

 origin of both, was the same ; thus explaining the 

 tendency of the trap rocks to recur more than once 

 in the same place, and further, the limited and local 

 nature of these productions. As these porphyries are 

 connected with gneiss and micaceous schist, their vol- 

 canic origin is the more probable ; for, with occasional 

 exceptions sufficiently intelligible, the trap rocks are 

 in contact with secondary strata. 



In Auvergne, the volcanic matter has issued 

 through granite ; whence the lavas of that country 

 repose immediately on this rock. In the African 

 islands, the volcanoes appear to have elevated 

 themselves from beneath the secondary strata ; 

 carrying up with them the calcareous beds now 



