[2] Strabo, lib. vi.

[3] Judd, _Volcanoes_, p. 8.

[4] Stromboli has also been described by Spallanzani, Hoffmann, Daubeny, and others. The account of Judd is the most recent. Of this island Strabo says, "Strongyle a rotundate figurae sic dicta, ignita ipsa quoque, violentia flammarum minor, fulgore excellens; ibi habita.s.se aecolum ajunt."--Lib. vi.

[5] _Poggend. Annal._, vol. xxvi., quoted by Daubeny.

[6] Communicated by Captain Petrie to the Victoria Inst.i.tute, 1st February 1892. See also a detailed and ill.u.s.trated account of the eruption communicated by A. Ricco to the _Annali dell' Ufficio centrale Meteorologico e Geodonamico_, Ser. ii., Parte 3, vol. xi. Summarised by Mr. Butler in _Nature_, April 21, 1892.

CHAPTER IV.

THE SANTORIN GROUP.

[Ill.u.s.tration: Fig. 13.--Ideal Section through the Gulf of Santorin, to show the structure of the submerged volcano.--_a._ Island of Asp.r.o.nisi; _b._ Island of Thera; 1. Old Kaimeni Island; 2. New Kaimeni Island; 3.

Little Kaimeni Island.]

(_a._) Before leaving the subject of European active volcanoes, it is necessary to give some account of the remarkable volcanic island of Santorin, in the Grecian archipelago. This island for 2000 years has been the scene of active volcanic operations, and in its outline and configuration, both below and above the surface of the Mediterranean, presents the aspect of a partially submerged volcanic mountain. (See Section, Fig. 13.) If, for example, we can imagine the waters of the sea to rise around the flanks of Vesuvius until they have entered and overflowed to some depth the interior caldron of Somma, thus converting the old crater into a crescent-shaped island, and the cone of Vesuvius into an island--or group of islands--within the caldron, then we shall form some idea of the appearance and structure of the Santorin group.

_Form of the Group._--The princ.i.p.al island, Thera, has somewhat the shape of a crescent, breaking off in a precipitous cliff on the inner side, but on the outer side sloping at an angle of about fifteen degrees into deep water. Continuing the curvature of the crescent, but separated by a channel, is the island of Therasia; and between this and the southern promontory of Thera is another island called Asp.r.o.nisi. All these islands, if united, would form the rim of a crater, in which the volcanic matter slopes outward into deep water, descending at a short distance to a depth of 200 fathoms and upwards. In the centre of the gulf thus formed rise three islands, called the Old, New, and Little Kaimenis. These may be regarded as cones of eruption, which history records as having been thrown up at successive intervals. According to Pliny, the year 186 B.C. gave birth to Old Kaimeni, also called Hiera, or the Sacred Isle; and in the first year of our era Thera (the Divine) made its appearance above the water, and was soon joined to the older island by subsequent eruptions. Old Kaimeni also increased in size by the eruptions of 726 and 1427. A century and a half later, in 1573, another eruption produced the cone and crater called Micra-Kaimeni. Thus were formed, or rather were rendered visible above the water, the central craters of eruption; and between these and the inner cliff of Thera and Therasia is a ring of deep water, descending to a depth of over 200 fathoms. So that, were these islands raised out of the sea, we should have presented to our view a magnificent circular crater about six miles in diameter, bounded by nearly vertical walls of rock from 1000 to 1500 feet in height, and ruptured at one point, from the centre of which would rise two volcanic cones--namely, the Kaimenis--one with a double crater, still foci of eruption, and from time to time bursting forth in paroxysms of volcanic energy, of which those of 1650, 1707, and 1866 were the most violent and destructive.[1] Of this last I give a bird's-eye view (Fig. 14).

The only rock of non-volcanic origin in these islands consists of granular limestone and clay slate forming the ridge of Mount St. Elias, which rises to a height of 1887 feet at the south-eastern side of the island of Thera, crossing the island from its outer margin nearly to the interior cliff, so that the volcanic materials have been piled up along its sides. The rocks of St. Elias are much more ancient than any of the volcanic materials around; and, as Bory St. Vincent has shown, have been subjected to the same flexures, dip and strike, as those sedimentary rocks which go to form the non-volcanic islands of the Grecian archipelago.

[Ill.u.s.tration: Fig. 14.--Bird's-eye View of the Gulf of Santorin during the volcanic eruption of February 1866.--(After Lyell.)]

[Ill.u.s.tration: _Ground Plan of Rocca Monfina_ Fig. 15.--Rocca Monfina, in Southern Italy, showing a crater-ring of trachytic tuffs, from the midst of which, according to Judd, an andesite lava-cone has been built up. Compare with the Santorin Group.]

(_b._) _Origin of the Santorin Group._--In reference to the origin of the Santorin group, Lyell regards it as a remnant of a great volcanic mountain which possessed a focus of eruption rising in the position of the present foci, but afterwards partially destroyed and the whole submerged to a depth of over 1000 feet. But another explanation is open to us, and one not inconsistent with what we now know of the physical changes to which the Mediterranean has been subjected since early Tertiary times. To my mind it is difficult to conceive how such a volcanic mountain as that of Santorin could have been formed under water; while, on the other hand, its physical structure and contour bear so striking a resemblance (as already observed) to those of Vesuvius and Rocca Monfina that we are much tempted to infer that it had a somewhat similar origin. Now we know that Vesuvius was built up by means of successive eruptions taking place under the air; and the question arises whether it could be possible that Santorin had a similar origin owing to the waters of the Mediterranean having been temporally lowered at a later Tertiary epoch. It has been stated by M. Fouque that the age of the more ancient volcanic beds of Santorin belong, as shown by the included fossils, to the newer Pliocene epoch. These are of course the unsubmerged, and therefore more recent strata, and may have been recently upheaved during one or more of the outbursts of volcanic energy. But it seems an impossibility that the Gulf of Santorin, with its precipitous walls and deep circular interior channel, as shown by the Ideal Section (Fig. 13), could have been formed otherwise than under the air. We are led, therefore, to inquire whether there was a time in the history of the Mediterranean, since the Eocene period, when the waters were lower than at present. That this was the case we have clear evidence. The remains of elephants, hippopotami, and other animals, which have been discovered in great numbers in the Maltese caves, show that this island was united to Sicily, and this again to Europe, during the later Pliocene epoch, so as to have become the abode of an Europasian fauna. According to Dr. Wallace, a causeway of dry land existed, stretching from Italy to Tunis in North Africa through the Maltese Islands--an inference involving the lowering of the waters of the Mediterranean by several hundred feet.[2] There is every reason for supposing that the old volcano of Santorin was in active eruption at this period; and its history may be considered to be similar to that of Vesuvius until, at the rising of the waters during the Pluvial (or Post-Pliocene) epoch, during which they rose higher than at present, Santorin was converted into a group of islands, slightly differing in form from those of the present day. This view seems to meet the difficulties regarding the origin of this group, difficulties which Lyell had long since clearly recognised.

(_c._) _Limit of the Mediterranean Volcanic Region._--With the Santorin group we conclude our account of the active European volcanoes. It may be observed, however, that from some cause not ascertained the volcanic districts of the Mediterranean and its sh.o.r.es are confined to the north side of that great inland sea; so that as regards vulcanicity the African coast presents a striking contrast to that of the opposite side.

If we draw a line from the sh.o.r.es of the Levant to the Straits of Gibraltar, by Candia, Malta, and to the south of Pantelleria and Sardinia, we shall find that the volcanic islands and districts of the mainland lie to the north of it.[3] This has doubtless some connection with the internal geological structure. The immunity of the Libyan desert from volcanic irruptions is in keeping with the remarkably undisturbed condition of the Secondary strata, which seldom depart much from the horizontal position; while the igneous rocks of the Atlas mountains are probably of great geological antiquity. On the other hand, the Secondary and Tertiary formations of the northern sh.o.r.es and islands of the Mediterranean are generally characterised by the highly-inclined, flexured, and folded position of the strata. Hence we may suppose that the crust over the region lying to the north of the volcanic line, owing to its broken and ruptured condition, was less able to resist the pressure of the internal forces of eruption than that lying to the south of it; and that, in consequence, vents and fissures of eruption were established over the former of these regions, while they are absent in the latter.

[1] Fuller details will be found in Daubeny's _Volcanoes_, chap. xviii., and Lyell's _Principles of Geology_, vol. ii. p. 65 (edition 1872). The bird's-eye view is taken from this latter work by kind permission of the publisher, Mr. J. Murray, as also the accompanying Ideal Section, Fig.

13.

[2] Wallace, _Geographical Distribution of Animals_ (1876). The author's _Sketch of Geological History_, p. 130 (Deacon & Co., 1887).

[3] The _volcanic area_ lying to the north of this line will include Sardinia, Sicily, Pantelleria, the Grecian Archipelago, Asia Minor, and Syria; the _non-volcanic area_ lying to the south of this line will include the African coast, Malta, Isles of Crete and Cyprus. The Isle of Pantelleria is apparently just on the line, which, continued eastward, probably follows the north coast of Cyprus, parallel to the strike of the strata and of the central axis of that island.--See "Carte Geologique de l'ile de Chypre, par MM. Albert Gaudry et Amedee Damour"

(1860).

CHAPTER V.

EUROPEAN EXTINCT OR DORMANT VOLCANOES.

We are naturally led on from a consideration of the active volcanoes of Europe to that of volcanoes which are either dormant or extinct in the same region. Such are to be found in Italy, Central France, both banks of the Rhine and Moselle, the Westerwald, Vogelsgebirge, and other districts of Germany; in Hungary, Styria, and the borders of the Grecian archipelago. But the subject is too large to be treated here in detail; and I propose to confine my observations to some selected cases which are to be found in Southern Italy, Central France, and the Rhenish districts, where the volcanic features are of so recent an age as to preserve their outward form and structure almost intact.

(_a._) _Southern Italy._--Extinct volcanoes and volcanic rocks occupy considerable tracts between the western flanks of the Apennines and the Mediterranean coast in the Neapolitan and Roman States, forming the remarkable group of the Phlegraean fields (Campi Phlegraei), with the adjoining islands of Ischia, Procida, Nisida, Vandolena, Ponza, and Palmarola; at Melfi and Avellino. All the region around Rome extending along the western slopes of the Apennines from Velletri to Orvieto, together with Mount Annato in Tuscany, is formed of volcanic material, and the same may be said of a large part of the island of Sardinia. From these districts I shall select some points which seem to be of special interest.

_Monte Nuovo and the Phlegraean Fields._--The tract of which this celebrated district forms a part lies as it were in a bay of the Apennine limestone of Jura.s.sic age. The floor of this bay is composed of puzzolana, a name given to beds of volcanic tuff of great thickness, and rising into considerable hills in the vicinity of the city of Naples, such as that of St. Elmo. Its composition is peculiar, as it is chiefly formed of small pieces of pumice, obsidian, and trachyte, in beds alternating with loam, ferriferous sand, and fragments of limestone. It is evidently of marine formation, as Sir William Hamilton, Professor Pilla, and others have detected sea-sh.e.l.ls therein, of the genera _Ostraea_, _Cardium_, _Pecten_ and _Pectunculus_, _Buccinum_, etc. It is generally of a greyish colour, and sometimes sufficiently firm to be used as a building stone. The Roman Campagna is largely formed of similar materials, which were deposited at a time when the districts in question were submerged, and matter was being erupted from volcanic vents at various points around, and spread over the sea-bed.

Such is the character of the general floor on which the more recent crater-cones of this district have been built. These are numerous, and all extinct with the exception of the Solfatara, near Puzzuoli, from which gases mixed with aqueous vapour are continually being exhaled. The gases consist of sulphuretted hydrogen mixed with a minute quant.i.ty of muriatic acid.[1] This district is also remarkable for containing several lakes occupying the interiors of extinct craters; amongst others, Lake Avernus, which, owing to its surface having been darkened by forests, and in consequence of the effluvia arising from its stagnant waters, has had imparted to it a character of gloom and terror, so that Homer in the _Odyssey_ makes it the entrance to h.e.l.l, and describes the visit of Ulysses to it. Virgil follows in his steps. Another lake of similar origin is Lake Agnano. Here also is the Grotto del Cane, a cavern from which are constantly issuing volumes of carbonic acid gas combined with much aqueous vapour, which is condensed by the coldness of the external air, thus proving the high temperature of the ground from which the gaseous vapour issues. This whole volcanic region, so replete with objects of interest,[2] may be considered, as regards its volcanic character, in a moribund condition; but that it is still capable of spasmodic movement is evinced by the origin of Monte Nuovo, the most recent of the crater-cones of the district. This mountain, rising from the sh.o.r.e of the Bay of Baiae, was suddenly formed in September 29th, 1538, and rises to a height of 440 feet above the sea-level. It is a crater-cone, and the depth of the crater has been determined by the Italian mineralogist Pini to be 421 English feet; its bottom is thus only 19 feet above the sea-level. A portion of the base of the cone is considered partly to occupy the site of the Lucrine Lake, which was itself nothing more than the crater of a pre-existent volcano, and was almost entirely filled up during the explosion of 1538. Monte Nuovo is composed of ashes, lapilli, and pumice-stones; and its sudden formation, heralded by earthquakes, and accompanied by the ejection of volcanic matter mixed with fire and water, is recorded by Falconi, who vividly depicts the terror and consternation of the inhabitants of the surrounding country produced by this sudden and terrible outburst of volcanic forces.[3]

(_b._) _Central Italy and the Roman States._--The tract bordering the western slopes of the Apennines northward from Naples into Tuscany, and including the Roman States, is characterised by volcanic rocks and physical features of remarkable interest and variety. These occur in the form of extinct craters, sometimes filled with water, and thus converted into circular lakes; or of extensive sheets and conical hills of tuff; or, finally, of old necks and ma.s.ses of trachyte and basalt, sometimes exhibiting the columnar structure. The Eternal City itself is built on hills of volcanic material which some observers have supposed to be the crater of a great volcano; but Ponzi, Brocchi, and Daubeny all concur in the opinion that this is not the case, as will clearly appear from the following account.

The geological structure of the valley of the Tiber at Rome is very clearly described by Professor Ponzi in a memoir published in 1850, from which the accompanying section is taken.[4] (Fig. 16.) From this it will be seen that "the Seven-hilled City" is built upon promontories of stratified volcanic tuff, of which the Campagna is formed, breaking off along the banks of the Tiber, the hills being the result of the erosion, or denudation, of the strata along the side of the river valley. As the strata dip from west to east across the course of the river, it follows that those on the western banks are below those on the opposite side; and thus the marine sands and marls which underlie the volcanic tuff, and are concealed by it along the eastern side of the valley, emerge on the west, and form the range of hills on that side. Such being the structure of the formations under Rome, it is evident that it is not "built on a volcano."

[Ill.u.s.tration: Fig. 16.--Geological Section across the Valley of the Tiber at Rome. 1. Alluvium of the Tiber; 2. Diluvium; 3. Volcanic tuff (recent deposits); 4. Sands, etc.; 5. Blue marl (sub-Apennine deposits).]

The tuff contains fragments of lava and pebbles of Apennine limestone, and was deposited under the waters of an extensive lake at a time when volcanic action was rife amidst the Alban Hills. This lacustrine formation rests in turn on deposits of marine origin, containing oysters, patellae, and other sea-sh.e.l.ls, of which the chain of hills on the right bank of the Tiber is chiefly formed.

The district around Albano lying to the south of Rome is of peculiar interest from the a.s.semblage of old crater-lakes which it contains; as, for instance, those of Albano, Vallariccia, Nemi, Juturna, and the lake of Gabii. The lake of Albano, one of the most beautiful sheets of water in the world, is about six miles in circ.u.mference, and surrounded by beds of peperino, a variety of tuff presenting a bright, undecomposed aspect when newly broken. The level of this lake was lowered by the Romans during the siege of Veii by means of a tunnel, so that the waters are 200 feet lower than the level at which they originally stood. In the same district is the lake of Nemi, very regular in its circular outline; that of Juturna lying near the foot of the Alban Hills, and that of Ariccia lying in a deep hollow eight miles in circ.u.mference;--all may be supposed to have been the craters of extinct volcanoes, both by reason of their shape and of the materials of which they are formed. All these old craters are, however, according to Daubeny, "only the dependencies and offshoots, as it were, of the great extinct volcano, the traces of which still remain upon the summit of the Alban Hills, and which is comparable in its form to that of Vesuvius, as it is surrounded by an outer circle of volcanic rock comparable to that of Somma."[5]

To the north of the city of Rome are several crateriform lakes, some of which are of great size, such as that of Bolsena, over twenty miles in circ.u.mference, and the Lago di Bracciano, almost as large, and lying about twelve miles from the city. These extensive sheets of water are surrounded by banks of tuff and volcanic sand, in which fragments of augite, leucite, and crystals of t.i.tanite are distributed. The town of Viterbo is built up at the foot of a steep hill called Monte Cimini, the lower part of which is composed of trachyte; this is surmounted by tuff, which appears to have been ejected from an extinct crater occupying the summit of the mountain, and now converted into a lake called the Lake of Vico. This crater is perfectly circular, and from its centre rises a little conical hill covered by trees.

(_c._) _Physical History._--s.p.a.ce does not permit of a fuller description of the remarkable volcanic features of the tract lying along the western slope of the Apennines; but from what has been stated it will be clear that volcanic forces have been in operation at one time on a grand scale in the Roman States and the South of Tuscany, over a tract extending from Mount Annato to Velletri and Segni.

This tract was separated from that of the Neapolitan volcanic region by a range of limestone hills of Jura.s.sic age between Segni and Gaeta, a protrusion of the Alban Hills westward; but the general structure and physical history of both regions are probably very similar, with the exception that the igneous forces still retain their vitality in the more southerly region. In the case of the Roman volcanic district, a bay seems to have been formed about the close of the Miocene period, bounded on all sides but the west by hills of limestone, over whose bed strata of marl, sandstone, and conglomerate were deposited. This tract was converted by subsequent movements into a fresh-water lake, and contemporaneously volcanic operations commenced over the whole region, and beds of tuff, often containing blocks of rock ejected from neighbouring craters, were deposited over those of marine origin.

Meanwhile numerous crater-cones were thrown up; and, as the land gradually rose, the waters of the lake were drained off, leaving dry the Campagna and plain of the Tiber. Ultimately the volcanic fires smouldered down and died out, whether within the historic epoch or not is uncertain; lakes were formed within the now dormant craters, and the face of nature gradually a.s.sumed a more placid and less forbidding aspect over this memorable region, destined to be the site of Rome, the Mistress of the World.

[1] As determined by Daubeny in 1825.

[2] Including the ruins of the Temple of Serapis, whose pillars are perforated by marine boring sh.e.l.ls up to a height of about 16 feet from their base; indicating that the land had sunk down beneath the sea, and afterwards been elevated to its present level.

[3] The account of Falconi, and another by Pietro Giacomo di Toledo, are given by Sir W. Hamilton, _op. cit._, p. 198, and also reproduced by Sir C. Lyell, _Principles_, vol. i. p. 608.

[4] Guiseppe Ponzi, "Sulla storia fisica del Bacino di Roma," _Annali di Scienze Fisiche_ (Roma, 1850).

[5] Daubeny, _Volcanoes_, p. 171.

CHAPTER VI.

EXTINCT VOLCANOES OF CENTRAL FRANCE.

(_a._) _General Structure of the Auvergne District._--From a granitic and gneissose platform situated near the centre of France, and separated from the western spurs of the Alps by the wide valley of the Rhone, there rises a group of volcanic mountains surpa.s.sing in variety of form and structure any similar mountain group in Europe, and belonging to an epoch ranging from the Middle Tertiary down almost to the present day.

This volcanic group of mountains gives rise to several important rivers, such as the Loire, the Allier, the Soule (a branch of the Loire), the Creuse, the Dordogne, and the Lot; and in the Plomb du Cantal attains an elevation of 6130 feet above the sea. Its southern section, that of Mont Dore, the Cantal, and the Haute Loire, is characterised by magnificent valleys, traversing plateaux of volcanic lava, and exhibiting the results of river erosion on a grand scale; while its northern section, that of the Puy de Dome, presents to us a varied succession of volcanic crater-cones and domes, with their extruded lava-streams, almost as fresh and unchanged in form as if they had only yesterday become extinct. A somewhat similar, but less important, chain of extinct volcanoes also occurs in the Velay and Vivarais, between the upper waters of the Loire and the Allier, in the vicinity of the town of Le Puy.[1] The princ.i.p.al city in this region is Clermont-Ferrand, lying near the base of the Puy de Dome, and ever memorable as the birthplace of Blaise Pascal.[2]

[Ill.u.s.tration: Fig. 17.--Generalised Section through the Puy de Dome and Vale of Clermont, distance about ten miles. The general floor formed of granite and gneiss (G); D. Domite-lava of the Puy de Dome; Sc. Cones of ashes and scoriae; L. Lava-sheets; A. Alluvium of the Vale of Clermont and Lake deposits.]