might be expected to die away gradually in a declining series of folds. This close packing, even overfolding, indicates the presence of an immovable obstacle against which the beds now composing the mountains have been squeezed up.

In the case of the Alps, as well as of the other mountain chains already mentioned, such a barrier exists-a 'horst' on a gigantic scale. It appears as a great and comparatively undisturbed mass in the 'Russian platform,' which is almost bare of rocks later than Cretaceous times and has apparently resisted any local plication. This platform may be traced westward, through Poland, Saxony, Hungary, parts of Austria, and most of Germany, into France. In this direction, however, its structure becomes complicated, for it has been affected by a process of mountain-making, later in date than the British coal measures, which set its mark on a region extending from Westphalia to southern Ireland, as well as by subsequent collapses. These have left great blocks of crystalline rocks standing up like islands, as they probably did above the sea in which Jurassic, Cretaceous and early Tertiary strata were deposited. Those blocks may still be identified in the highlands of Bohemia and Bavaria, of the Schwarzwald and the Vosges, and in that great upland of crystalline rock which is now crowned by the volcanoes of Auvergne.

The Alpine region-for we must restrict ourselves to this is the product of two sets of thrusts, acting apparently from the same direction, and radiating, if we include the Apennines and the Dalmatian mountains, from the depression now occupied by the Adriatic sea and the North Italian plain, from west of Turin to Venice. The rock-barrier to the north, which probably underlies, at no great depth, the lowland on that side of the Alps, was apparently strong enough to resist the making of any important fold except the Jura. Even this is really an outpost of the Alps, into which chain it passes near Chambéry. But the foundation, beneath the present Alpine chain, seems to have been weaker, perhaps because it had already twice undergone considerable disturbances, one before and the other immediately after Carboniferous times. So, between the thrust from the one side and the resistance on the other, it' buckled up.'

In his fourth volume Prof. Suess returns to this subject, for it had received much notice during the twenty years' interval; but it may be doubted whether some of this later literature has really added to our knowledge. Certain of his authors not only affirm the existence of folds on a most abnormal scale, but also carry this process so far as to make it transfer strata from one side to the other of the present chain. Such a process, in a mass like the crust of the earth, though the movements may be comparatively superficial, is difficult to understand if any regard be paid to mechanical principles; and the evidence adduced in support of some of the most surprising folds depends on identifications of rocks which are more than questionable. We think, therefore, that Prof. Suess would have been wise to commit himself less unreservedly to views which belong rather to the poetry than the prose of science. The main fact, however, that the Alps are a region of an intense folding, sometimes of an exceptional character, associated occasionally with very remarkable overthrusting, is quite separate from these exaggerations or mistakes, and will remain unchanged when they have joined universal deluges, precipitated basalts, and aqueous granites, in the limbo appointed for exploded hypotheses.

Earthquakes and volcanic eruptions stand in close relation to movements of collapse. A zone of ancient, mainly crystalline rocks, the remnant of an old mountain region, sweeps along the 'foot' of Italy into north-eastern Sicily. Its western margin is a great seismic area, which passes under the Strait of Messina, and is indicative of a line of fault, probably extending to Etna. The Lipari Islands form two lines, the one parallel with the Italian, the other with the northern Sicilian coast. On the former stand Stromboli and Vulcano, nearly in a line with Etna, the only volcanoes still active; the latter passes from Ustica (to the north of Palermo) through two smaller islands, to the cluster by Panaria (probably the remnants of a shattered cone) marking the junction of the two fissures. Here is a region of crustal disturbances, of which Suess remarks:

'Within a space bounded by the peripheral line of 1783 (one of the destructive Calabrian earthquakes) the crust of the earth has sunk down in the form of a dish, and the radial

fractures have been produced which converge to the Lipari Islands. Any disturbance in the equilibrium of the several fragments of the crust gives rise to increased volcanic activity on the islands and to earthquakes on the mainland of Sicily.'

The principle thus indicated is capable of wide application. Some earthquakes, such as those at Ischia in 1881 and 1883, are the direct consequence of volcanic disturbance; but these, though sometimes destructive, generally affect only limited districts. On the other hand, earthquakes which shake whole provinces, like those of Calabria, of Charleston and of Lisbon, are independent of eruptive centres. Both these types of disturbance affect Central America. Here the grouping of the volcanoes, which appear to be shifting their centres of discharge towards the Pacific, suggests a sinking of the coast in that direction. This ocean, indeed, may be a region of general subsidence. It is bordered by lines of lofty mountains, often associated with volcanoes. In the northern continent of America lies the triple series of the Rockies, the Sierra Nevada and the Coast Range, with occasional volcanic summits which become more numerous and active in approaching the Arctic region. Along the western side of the southern continent extends the great Andean chain, with its linear groups (in Ecuador, Peru and Chile) of volcanic mountains, few of which are now active, but some are the most lofty on the globe. The Pacific on its western side is bordered at intervals by lines of volcanoes, which here are more conspicuous than folded mountains. Their orifices may be traced on a curve from New Zealand through the Melanesian Islands to New Guinea, where the rents seem to separate, one branch sweeping round through Sumatra and the Andaman Islands, the middle and most important passing through the Philippines and Japan to Kamchatka, and the eastern, the least strongly marked, running through the Ladrones till it rejoins the second one. Volcanic islands also, often with a linear grouping, interrupt the surface of the Pacific, suggesting that the whole of that vast ocean is an area of comparatively recent collapse, associated with marginal foldings and extrusion of liquid material as a result of marginal

pressure.

Similar concurrences, though that of earthquakes

cannot, of course, be proved, may be traced far back in regions of mountain-making. The Alps, for instance, exhibit old lines of faultings transverse to the trends of the present ranges, associated with extrusions of igneous matter. The highland region, which, during the Carboniferous period, occupied no small portion of their site, was disturbed in the Permian. A progressive collapse began, especially in the eastern half, which lasted through more than the Secondary era, causing those outbursts of red porphyries, followed by basic lavas and tuffs, and, by intrusive masses of igneous rocks, of which the Fassathal, to the east of Botzen, was the more immediate centre, though it affected a much wider area. But tangential thrusts, operating first at the beginning and then at the end of Miocene times, produced the chain now known as the Alps. This chain has been affected by lines of fracture and more gentle folding, which have a general NNE.-SSW. trend, and may possibly be associated with the peculiar southward curvation which becomes conspicuous in Savoy and may be followed to the Mediterranean.

That sea, in many places, tells a similar story-regions of progressive collapse, associated with volcanic disturbances, with subterranean intrusions of molten rock, and with an extra-marginal zone of folding and thrusting. Large as is the Mediterranean Sea, it is but a remnant of a far wider expanse of water, which in Mid-Eocene ages extended, probably without interruption, from southwestern England to China. Incidentally also this sea indicates that, in the long lapse of ages, the earth's crust has been subject to movements diverse in direction. During that part of the Cretaceous period, when chalk was being formed over parts of our islands and France, the sea, as proved by its deposits, was invading the lands forming its northern margin. But the Mid-Eocene sea trespassed also in the opposite direction; for in northern Africa, while a sandstone (not always present) is the only representative of the Cretaceous period, extensive beds of limestone, which often rest directly on ancient crystalline rocks, were left by the Mid-Eocene sea. This limestone, and even the more sandy beds into which it locally passes, can be readily recognised by the presence of an unusually large foraminifer, which, from its disc-like form,

has received the name of nummulites. Abundant in the sandy clays on the shore of Bracklesham Bay, in Sussex, it has been raised in the Alps to a height of some 10,000 feet above sea-level, and to double that height (according to Suess) in the Karakoram Himalayas. The most conspicuous physical features in Europe-perhaps also in Asia-had no existence when these lowly organisms lived and died.

But this great expanse of salt water was comparatively short-lived, as geologists reckon time. It soon began to shrink, and after some oscillations distinctly contracted. Great earth movements were setting in, connected with the rise of the Alps and other nearly related mountain chains. It is at this point that the Professor enters on the history of the sea which was more directly the forerunner of the existing Mediterranean. He divides this into four stages. The first, corresponding generally with the Middle Miocene, has for its more northern representatives certain deposits in the French Rhone valley and the marine sandstone (Molasse) of Switzerland; and these, in some places, pass up into gypsiferous marls (Schlier) which afford the spectacle of a great expiring sea.' But the second stage, corresponding with the Upper Miocene, proves a movement of expansion. During it were deposited the Fahluns of Touraine, the Leithakalk near Vienna, and sundry strata in Baden, Hungary, Transylvania and Wallachia, ending in the Danube valley with a remarkable group of beds which the Professor designates the 'Sarmatian' stage. This deposit is interesting, for, though fossil mollusca are remarkably numerous, the number of species is comparatively small, probably not more than fifty. They prove that great changes in the conditions of life occurred during this second Mediterranean stage. In its earlier phase calcareous algæ, corals, bryozoa, echinoderms and foraminifers were abundant; and many of the molluscs represent genera now characteristic of subtropical seas. Practically none of these occur in the fauna of the Sarmatian phase, which indicates a temperate climate corresponding rather with that of the Black Sea than that of the present Mediterranean. The highest strata even became lacustrine, and may be said to foreshadow the Euxine and Aralo-Caspian basins.