Interior of the Globe. 41 



of an almost continued line of craters in the intermediate 

 spaces, strengthens the probability of such a connection. The 

 three volcanoes were, Osorno in lat. 40° S., Concagua in 32° S., 

 and Ooseguina in lat. 13° N., and the day of their simultane- 

 ous eruption was 20th Jan. 1835. — (Mr Darwin, Trans. Geol. 

 Soc, March 1838.) If the volcanoes at the two extremities 

 of the line sympathize in their action, we would expect all the 

 intermediate parts to be disturbed less or more ; and the ex- 

 treme frequency of earthquakes in the Andes (often several in 

 a day) may thus be the natural consequence of the long series 

 of craters, the activity of any one of which will agitate the 

 whole chain in a less or greater degree. 



The hypothesis helps us to explain local elevation and local 

 subsidence. Fused matter must be subject to greater changes 

 of temperature than solid matter. The one is influenced by 

 conduction and circulation, the other by conduction only. A 

 small contraction or expansion in a fluid mass 5 miles or 50 

 miles deep, is sufficient alone to account for a portion of the 

 external crust which forms its roof being raised or depressed, 

 without calling in the agency of water at all. Then as to the 

 causes of contraction and expansion, the fluid mass, indepen- 

 dently of mere variations of temperature, may augment its 

 bulk by dissolving a part of its walls, or diminish its bulk by 

 partial or entire solidification. Again, let us suppose a large 

 volume of steam to be permanently in contact with the fused 

 mass, and so placed that it cannot escape, it will act like the 

 air in the air-chamber of a fire-engine, and a small addition to 

 its volume will force upwards, and eject, a part of the fused 

 matter. I think, then, that the hypothesis of partial deposits 

 of fluid matter has material advantages over that of a fluid 

 nucleus. Firstly, it allows us to assign a thickness to the crust 

 of the globe which is more in accordance with its known sta- 

 bility. Secondly, it offers greater probabilities of such changes 

 taking place in the condition of the matter itself, as the phe- 

 nomena require us to suppose. Thirdly, by placing each fluid 

 mass within solid walls, it better explains how these changes 

 may manifest themselves at the surface ; because it allows us 

 to suppose that the fused matter, confined and compressed 

 within its rocky envelope, may ascend, on the same principle 



