of the highlands of Quito became acquainted with the existence of the little cyclopic fishes termed by them the Preñadilla. On the night between the 19th and 20th of June 1698, when the summit of Carguairazo, a mountain of 19,720 feet in height, fell in, leaving only two huge masses of rock remaining of the ledge of the crater, a space of nearly thirty-two square miles was overflowed and devastated by streams of liquid tuffa and argillaceous mud, containing large quantities of dead fish. In like manner, the putrid fever, which raged seven years previously in the mountain town of Ibarra, north of Quito, was ascribed to the ejection of fish from the volcano of Imbaburu.

Having delineated the structure and dynamic activity of volcanoes, it now remains for us to throw a glance at the differences existing in their material products. The subterranean forces sever old combinations of matter in order to produce new ones, and they also continue to act upon matter as long as it is in a state of liquefaction from heat, and capable of being displaced. The greater or less pressure under which merely softened or wholly liquid fluids are solidified, appears to constitute the main difference in the formation of plutonic1 and volcanic rocks. The mineral mass which flows in narrow, elongated streams from a volcanic opening (an earth-spring) is called lava. Where many such currents meet and are arrested in their course, they expand in width, filling large basins, in which they become solidified in superimposed strata. These few sentences describe the general character of the products of volcanic activity.

Rocks which are merely broken through by the volcanic action are often inclosed in the igneous products. Thus I have found angular fragments of feldspathic syenite imbedded in the black augitic lava of the volcano of Jorullo in Mexico; but the masses of dolomite and granular limestone, which contain magnificent clusters of crystalline fossils, (vesuvian and garnets, covered with mejonite, nepheline, and sodalite,) are not the ejected products of Vesuvius, these belonging rather to very generally distributed formations, viz., strata of tuffa, which are more ancient than the elevation of the Somma and of Vesuvius, and are probably the products of a deep-seated and concealed sub-marine volcanic

1 Rocks formed by fire are termed Plutonic; those ascribed to water are Neptunianfrom the names of the deities of fire and water in mythology.

action. We find five metals amongst the products of existing volcanoes-iron, copper, lead, arsenic, selenium. The vapours that rise from the fumarolles1 cause the sublimation of the chlorides of iron, copper, lead, and ammonium; iron glance and chloride of sodium (the latter often in large quantities) fill the cavities of recent lava streams and the fissures of the margin of the crater.

*

The different volcanoes over the Earth's surface, when they are considered independently of all climatic differences, are acutely and characteristically classified as central and linear volcanoes. Under the first name are comprised those which constitute the central point of many active mouths of eruption, distributed almost regularly in all directions; under the second, those lying at some little distance from one another, forming, as it were, chimneys or vents along an extended fissure. Linear volcanoes, again, admit of farther subdivision, namely, those which rise like separate conical islands from the bottom of the sea, being generally parallel with a chain of primitive mountains whose foot they appear to indicate, and those volcanic chains which are elevated on the highest ridges of these mountain chains, of which they form the summits. The Peak of Teneriffe, for instance, is a central volcano, being the central point of the volcanic group to which the eruption of Palma and Lancerote may be referred. The long rampart-like chain of the Andes, which is sometimes single, and sometimes divided into two or three parallel branches, connected by various transverse ridges, presents, from the south of Chili to the north-west coast of America, one of the grandest instances of a continental volcanic chain. The proximity of active volcanoes is always manifested in the chain of the Andes by the appearance of certain rocks, (as dolerite, melaphyre, trachyte, andesite, and dioritic porphyry,) which divide the so-called primitive rocks, the transition slates and sandstones, and the stratified formations.

In the volcanic chain of the new continent, the separate volcanoes are occasionally, when near together, in mutual dependence upon one another; and it is even seen that the volcanic activity for centuries together has moved on in one and the same direction-as, for instance, from north to south

1 From fumus, (Lat.) smoke.

in the province of Quito. The focus of the volcanic action lies below the whole of the highlands of this province; the only channels of communication with the atmosphere are, however, those mountains which we designate by special names, as the mountains of Pinchincha, Cotopaxi, and Tunguaragua, and which, from their grouping, elevation, and form, constitute the grandest and most picturesque spectacle to be found in any volcanic district of an equally limited extent. Experience shows us in many instances that the extremities of such groups of volcanic chains are connected together by subterranean communications; and this fact reminds us of the ancient and true expression made use of by Seneca, that the igneous mountain is only the issue of the more deeply seated volcanic forces. In the Mexican highlands a mutual dependence is also observed to exist among the volcanic mountains Orizaba, Popocatepitl, Jorullo, and Colima; and I have shown that they all lie in one direction between 18° 59′ and 19° 12′ north lat., and are situated in a transverse fissure running from sea to sea.

The geographical distribution of the volcanoes which have been in a state of activity during historical times, the great number of insular and littoral volcanic mountains, and the occasional, although ephemeral, eruptions in the bottom of the sea, early led to the belief that volcanic activity was connected with the neighbourhood of the sea, and was dependent upon it for its continuance. "For many hundred years," says Justin, or rather Trogus Pompeius, whom he follows, "Etna and the Eolian Islands have been burning; and how could this have continued so long, if the fire had not been fed by the neighbouring sea?" In order to explain the necessity of the vicinity of the sea, recourse has been had, even in modern times, to the hypothesis of the penetration of sea-water into the foci of volcanic agency —that is to say, into deep-seated terrestrial strata. When I collect together all the facts that may be derived from my own observation and the laborious researches of others, it appears to me that everything in this involved investigation depends upon the questions, whether the great quantity of aqueous vapours, which are unquestionably exhaled from

Justinus, a historian of the age of the Roman emperor Antoninus, epitomised the history of the Eastern and Roman Empires, by the preceding historian Trogus Pompeius, of the age of Julius and Augustus Cæsar.

volcanoes, even when in a state of rest, be derived from seawater impregnated with salt, or rather, perhaps, with fresh meteoric water; or whether the expansive force of the vapours (which at a depth of nearly 94,000 feet is equal to 2800 atmospheres) would be able at different depths to counterbalance the hydrostatic pressure of the sea, and thus afford them, under certain conditions, a free access to the focus; or whether the formation of metallic chlorides, the presence of chloride of sodium in the fissures of the crater, and the frequent mixture of hydro-chloric acid with the aqueous vapours, necessarily imply access of sea-water; or, finally, whether the repose of volcanoes (either when temporary or permanent and complete) depends upon the closure of the channels by which the sea or meteoric water was conveyed, or whether the absence of flames and of exhalations of hydrogen (and sulphuretted hydrogen gas seems more characteristic of solfataras than of active volcanoes) is not directly at variance with the hypothesis of the decomposition of great masses of water?1-HUMBOLDT.

THE OCEAN BASINS.

The continents determine the general outlines of the great ocean basins. The indentations of their coasts give the articulation of the shores of the oceans; the islands, by their disposition, by their less or greater frequency, give what else is wanting to complete their character; the one is the counterpart of the others; they are the same forms, but in an inverse order.

Two great oceans, the Pacific and the Atlantic, corresponding to the two worlds, surround, on almost all sides, the principal terrestrial masses. We may detach from the Pacific the Indian Ocean, which, though belonging to it, has some special characteristics; and separate from the Atlantic the Northern Frozen Ocean, the position of which gives it a particular physiognomy. As to the great Southern Sea, we may consider it less as an ocean by itself than as a common reservoir, from which issue, so to speak, all the seas of the globe, to make their way into the lands.

Vesu

The great volcanic districts are,-1, The line of shore round the Pacific and Indian Oceans, along the ridges of the Andes and Rocky Mountains on the American, and among the islands on the Asiatic side; 2, The Atlantic and the Mediterranean. vius is the only active volcano on the European Continent; France and Germany contain many extinct volcanic mountains.

The Pacific, the Indian Ocean, the Atlantic, correspond to the three double worlds which are composed of the great terrestrial masses, and separated one from the other; each of these seas being divided into a northern and a southern basin, except the Indian Ocean, which, on this account, is only a half ocean.

The general forms of the contours of these three oceans have, as a common feature, a wide opening towards the south, and are narrowed to a point on the north, just the reverse of the continents. Each of these has, meantime, a form which is peculiar to itself. The Pacific Ocean is an oval, wide open on the south, the sides of which go on approaching each other towards the north, so as to leave, between America and Asia, only the narrow passage of Behring's Straits, by which it communicates with the Frozen Ocean.

The Indian Ocean has the form of a triangle, the vertex of which is turned to the north; the Atlantic that of a valley with nearly parallel sides, which, narrowed for a moment, then broadens into the Frozen Ocean.

The oceans differ, moreover, in the mode of articulation of their shores. These indentations have very various forms, which we will classify under three species: the gulfs, like that of Bengal; the land-locked seas, isolated from the rest of the ocean by peninsulas and chains of islands, like the Sea of Japan, the Sea of Okhotsk; the inland seas, surrounded on all sides by the land, in a continuous manner, like the Mediterranean and the Baltic.

Considered with reference to the indentations, the three oceans have their own respective characters, and we find that in each one of these three forms predominate.

The Pacific Ocean is that of the land-locked or closed seas; for there are no less than five of considerable size along the coast of Asia: the Sea of Behring, closed in by the peninsula of Aljaska, and the chain of the Aleutian Islands; the Sea of Okhotsk, enclosed by the peninsula of Kamtschatka, and the series of the Kurile Islands; the Sea of Japan, shut in by the island chain of this name; the Northern Chinese Sea, locked by the islands of Lieu-Khieu and Formosa; the Southern Sea of China, locked by the Philippines, Borneo, and the peninsula of Indo-China. We may almost call the Vermilion Sea, or Gulf of California, an inland sea, it being the only indentation of this ocean somewhat marked, on the American coast.