370 ON VOLCANOES AND EARTHQUAKES. 



of evidence which appears nncontrovertible. It is 

 equally probable that, like the American and Italian 

 volcanoes, those of the Canary isles are merely the 

 separate vents of one volcanic abyss, which must 

 there also be very deeply situated. I may here also 

 remark that the volcanic region of Sardinia extends 

 in a line through the whole length of that island. 



The causes of the volcanic fire have, naturally, been 

 subjects of much speculation. The theory which at- 

 tributes this to coal has already been examined. The 

 ignition of pyrites has also been a favourite theory 

 with those who saw nature only in the minutiae of 

 their laboratories and cabinets. There is no such 

 pyrites present, the products of volcanoes are not 

 those which would result from such a cause, and it is 

 neither able to explain their duration, their intermis- 

 sion, nor their deep and distant connexions. The 

 same answer nearly may be given to the theory which 

 supposes sulphur to be the cause ; and that which 

 attributes it to bitumen is equally unworthy of further 

 notice. 



As, in justification of these theories, it has been 

 asserted that the temperature of volcanoes was low, 

 and that the rocks were not melted by the ordinary 

 process of fusion, I must bestow a few words on a 

 speculation which derives no merit but from its ex- 

 traordinary absurdity. That the bitumen of Kirwan 

 or the sulphur of Ferrara, should produce the fluidity 

 of lavas without fusion, are propositions which do 

 not deserve a serious examination ; even could it be 

 proved that either of these substances were present in 

 lavas ; which they are not. The project of Dolomieu 

 for causing the parts to slide on each other, must be 

 passed over, in compassion to a well-meaning ob- 

 server. They who could adopt it from him, under 



