Secondly: The position of volcanos, _always near the sea_, together with the _agitations of the sea_, previous to, and during an eruption, as well as the _saline_ matter in the ejected substances, render it very clear, _that the sea, by subterranean communication, supplies water at the seat of volcanic action_. “The sea seems to sympathise with the agitations of the adjoining volcanos, rising and falling with rapid alternation--_caused by the sudden deflux of a great body of water into the vast volcanic caverns_.” _Ure’s New Sys. Geol._ p. 388. This fact is so well known in the history of volcanos, that it needs no further proof. It has, however, led to the remark, that volcanos are generally situated in islands, or near the sea coast. Indeed many of them are _submarine_, and have actually been seen in operation, throwing up vast columns of water to an immense height, until the edge of the crater appeared above the surface of the sea, and increased into islands, which have become permanent. At such times the water of the sea for a great distance round became _hot_, fishes died; and even the pitch melted from the hulks of the vessels in the neighborhood. 2. Another theory has been proposed which does not differ from the first, in regard to the _materials_ employed at the seat of volcanic action, nor in the _manner_ of the process; but in regard to the _condition_ of those materials when they _begin_ to operate in the production of the elastic agents. These materials may be in a state of _igneous fusion_ in the interior of the earth. This state is supposed to have resulted thus: