On Earthquakes — their causes and effects. 215 



belongs or originates in that formation, commit an error of 

 judgment, and conclude the mass to be of the same nature 

 with that part which is sensible to the sight. The base of 

 the mountain and bosom of the volcano, may, and is most 

 likely to be, of primitive formation, underlying of course the 

 more recent strata. 



If we admit of the extended existence of subterranean 

 fires, it is far more reasonable to suppose they have their 

 origin in the most deeply seated rocks, frequently exhibiting 

 their destructive fires through a formation of their own 

 creation. 



Humboldt informs us that during an earthquake near 

 Guanaxuarto, in Mexico, the sound was confined to primitive 

 schist. The same able naturalist tells us that earthquakes 

 occur in the granite of Lima and Acapulco — gneiss of Ca- 

 raccas — mica-slate of Araya — primitive clay-slate of Tepe- 

 cuacuilco, Mexico — secondary limestone of the Appenines, 

 Spain, and New Andalusia — and the trappean porphyries of 

 Quito and Papayan. Thus we have earthquakes exhibiting 

 themselves in almost all formations, though more frequently 

 in the primitive, and better defined there ; but it does not 

 follow, although we may be sensible of the motion in a 

 secondary soil, that it has its origin in that formation. I 

 hope I have proved satisfactorily the great probability of the 

 existence of subterranean channels, communicating with 

 each other, and stretching out to great distances — some 

 communicating with the ocean, and others very likely un- 

 derlying it. Judging from the facts produced, I think the 

 deduction may be fairly made. 



In conclusion, 1 have to observe, this is a subject that ad- 

 mits of great latitude in speculation, and if I have been for- 

 tunate enough to add one new idea to the stock of knowl- 

 edge, I shall rest satisfied that my labour has been rewarded. 



