Chapter XXVI (2)

We may compare this partial elevation of solid matter to the rise of a hardened crust of scoriae, such as is usually formed on the surface of lava-currents, even while they are in motion, and which, although stony and capable of supporting heavy weights, may be upraised without bursting by the intumescence of the melted matter below. That the upheaval was merely local is proved by the fact that the neighboring Kaimenis did not participate in the movement, still less the three more distant or outer islands before mentioned. The history, therefore, of the Kaimenis shows that they have been the result of intermittent action, and it lends no support to the hypothesis of the sudden distension of horizontal beds blown up like a bladder by a single paroxysmal effort of expansive gases.

It will be seen by the accompanying map and sections, that the Kaimenis are arranged in a linear direction, running N. E. and S. W., in a manner different from that represented in the older charts. In their longest diameter they form at their base a ridge nearly bisecting the gulf or crater (see sections, figs. 64, 65).

On considering these facts we are naturally led to compare the smaller and newer islands in the centre of the gulf to the modern cone of Vesuvius, surrounded by the older semicircular escarpment of Somma, or to liken them to the Peak of Teneriffe before described, as surrounded by its "fosse and bastion." This idea will appear to be still more fully confirmed when we study the soundings taken during the late hydrographical survey. Thera, which constitutes alone more than two-thirds of the outer circuit, presents everywhere towards the gulf, high and steep precipices composed of rocks of volcanic origin. In all places near the base of its cliffs, a depth of from 800 to 1000 feet of water was found, and Lieut. Leycester informs us[610] that if the gulf, which is six miles in diameter, could be drained, a bowl-shaped cavity would appear with walls 2449 feet high in some places, and even on the southwest side, where it is lowest, nowhere less than 1200 feet high; while the Kaimenis would be seen to form in the centre a huge mountain five and a half miles in circumference at its base, with three principal summits (the Old, the New, and the Little Burnt Islands) rising severally to the heights of 1251, 1629, and 1158 feet above the bottom of the abyss. The rim of the great caldron thus exposed would be observed to be in all parts perfect and unbroken, except at one point where there is a deep and long chasm or channel, known by mariners as "the northern entrance" (B, fig. 63) between Thera and Therasia, and called by Lieut. Leycester "the door into the crater." It is no less than 1170 feet deep, and constitutes, as will appear by the soundings (see map), a remarkable feature in the bed of the sea. There is no corresponding channel passing out from the gulf into the Mediterranean at any other point in the circuit between the outer islands, the greatest depth there ranging from 7 to 66 feet.

We may conceive, therefore, if at some former time the whole mass of Santorin stood at a higher level by 1200 feet, that this single ravine or narrow valley now forming "the northern entrance," was the passage by which the sea entered a circular bay and swept out in the form of mud and pebbles, the materials derived by denudation from wasting cliffs. In this manner the original crater may have been slowly widened and deepened, after which the whole archipelago may have been partially submerged to its present depth.

That such oscillations of level may in the course of ages have taken place, will be the more readily admitted when we state that part of Thera has actually sunk down in modern times, as, for example, during the great earthquake before alluded to, which happened in 1650. The subsidence alluded to is proved not only by tradition, but by the fact that a road which formerly led between two places on the east coast of Thera is now twelve fathoms under water.

MM. Boblaye and Virlet mention,[611] that the waves are constantly undermining and encroaching on the cliffs of Therasia and Aspronisi, and shoals or submarine ledges were found, during the late survey, to occur round a great part of these islands, attesting the recent progress of denudation. M. Virlet also remarks, in regard to the separation of the three islands forming the walls of the crater, that the channels between them are all to the W. and N. W., the quarter most exposed to the waves and currents.

Mr. Darwin, in his work on volcanic islands, has shown that in the Mauritius and in Santiago, there is an external circle of basaltic rocks of vast diameter, in the interior of which more modern eruptions have taken place, the older rocks dipping away from the central space in every direction, as in the outer islands of Santorin. He refers the numerous breaches, some of them very wide in the external ramparts of those islands, to the denuding action of the sea. Every geologist, therefore, will be prepared to call in the aid of the same powerful cause, to account for the removal of a large part of the rocks which must once have occupied the interior space, in the same manner as they attribute the abstraction of matter from elliptical "valleys of elevation," such as those of Woolhope and the Wealden in England, to the waves and currents of the sea.

Thera, Therasia, and Aspronisi are all composed of volcanic matter, except the southern part of Thera, where Mount St. Elias rises to three times the height of the loftiest of the igneous rocks, reaching an elevation of 1887 feet above the sea.[612] This mountain is formed of granular limestone and argillaceous schist, and must have been originally a submarine eminence in the bed of the Mediterranean, before the volcanic cone, one side of the base of which now abuts against it, was formed. The inclination, strike, and fractures of the calcareous and argillaceous strata of St. Elias have no relation to the great cone, but, according to M. Bory St. Vincent, have the same direction as those of the other isles of the Grecian Archipelago, namely, from N. N. W. to S. S. E. Each of the three islands, Thera, Therasia, and Aspronisi, is capped by an enormous mass of white tufaceous conglomerate, from forty to fifty feet thick, beneath which are beds of trachytic lava and tuff, having a gentle inclination of only 3 degrees or 4 degrees. Each bed is usually very narrow and discontinuous, the successive layers being moulded or dove-tailed, as M. Virlet expresses it, into the inequalities of the previously existing surface, on which showers of cinders or streams of melted matter have been poured. Nothing, therefore, seems more evident than that we have in Santorin the basal remains of a great ruined cone, or flattened dome; and the absence of dikes in the cliffs surrounding the gulf would indicate that the eruptions took place originally, as they have done in the last two thousand years, not near the margin but in the centre of the space now occupied by the gulf. The central portions of the dome have since been removed by engulfment, or denudation, or by both these causes.

An important fact is adduced by M. Virlet, to show that the gentle dip of the lava-streams in the three outer islands towards all points of the compass, away from the centre of the gulf, has not been due to the upheaval of horizontal beds, as conjectured by Von Buch, who had not visited Santorin.[613] The French geologist found that the vesicles or pores of the trachytic masses were lengthened out in the several directions in which they would have flowed if they had descended from the axis of a cone once occupying the centre of the crater. For it is well known that the bubbles of confined gas in a fluid in motion assume an oval form, and the direction of their longer axis coincides always with that of the stream.

On a review, therefore, of all the facts now brought to light respecting Santorin, I attribute the moderate slope of the beds in Thera and the other external islands to their having originally descended the inclined flanks of a large volcanic cone, the principal orifice or vents of eruption having been always situated where they are now, in or near the centre of the space occupied by the gulf or crater--in other words, where the outburst of the Kaimenis has been witnessed in historical times. The single long and deep opening into the crater is a feature common to all those remnants of ancient volcanoes, the central portions of which have been removed, and is probably connected with aqueous denudation. This denuding process has been the work of ages when the sea was admitted into an original crater, and has taken place during the gradual emergence of the island from the sea, or during various oscillations in its level.

The volcanic island of St. Paul in the midst of the Indian Ocean, lat. 38 degrees 44 minutes S., long. 77 degrees 37 minutes E., surveyed by Capt. Blackwood in 1842, seems to exemplify the first stage in the formation of such an archipelago as that of Santorin. We have there a crater one mile in diameter, surrounded by steep and lofty cliffs on every side save one, where the sea enters by a single passage nearly dry at low water. In the interior of the small circular bay or crater there is a depth of 30 fathoms, or 180 feet. The surface of the island slopes away on all sides from the crest of the rocks encircling the crater.[614]

Cone and crater of Barren Island, in the Bay of Bengal. Height of the central cone (according to Capt Miller, in 1834), 500 feet.]

_Barren Island._--There is great analogy between the structure of Barren Island in the Bay of Bengal, lat. 12 degrees 15 minutes, and that of Santorin last described. When seen from the ocean, this island presents, on almost all sides, a surface of bare rocks, rising, with a moderate acclivity, towards the interior; but at one point there is a cleft by which we can penetrate into the centre, and there discover that it is occupied by a great circular basin, filled by the waters of the sea, and bordered all around by steep rocks, in the midst of which rises a volcanic cone, very frequently in eruption. The summit of this cone is about 500 feet in height, corresponding to that of the circular border which incloses the basin; so that it can be seen from the sea only through the ravine. It is most probable that the exterior inclosure of Barren Island (_c_, _d_, fig. 67) is nothing more than the remains of a truncated cone _c_, _a_, _b_, _d_, a great portion of which has been removed by engulfment, explosion, or denudation, which may have preceded the formation of the new interior cone, _f_, _e_, _g_.[615]

Supposed section of Barren Island, in the Bay of Bengal.]

MUD VOLCANOES.

_Iceland._--Mr. R. Bunsen, in his account of the pseudo-volcanic phenomena of Iceland, describes many valleys where sulphurous and aqueous vapors burst forth with a hissing sound, from the hot soil formed of volcanic tuff. In such spots a pool of boiling water is seen, in which a bluish-black argillaceous paste rises in huge bubbles. These bubbles on bursting throw the boiling mud to a height of fifteen feet and upwards, accumulating it in ledges round the crater or basin of the spring.

_Baku on the Caspian._--The formation of a new mud volcano was witnessed on the 27th of November, 1827, at Tokmali, on the peninsula of Abscheron, east of Baku. Flames blazed up to an extraordinary height for a space of three hours, and continued for twenty hours to rise about three feet above a crater, from which mud was ejected. At another point in the same district where flames issued, fragments of rock of large size were hurled up into the air, and scattered around.[616]

Mud cones and craters of Hinglaj near Beila, district of Lus, 120 miles northwest of mouth of Indus. From original drawing by Capt. Robertson. (See Map, p. 460.)]

_Sicily._--At a place called Macaluba, near Girgenti in Sicily, are several conical mounds from ten to thirty feet in height, with small craters at their summits, from which cold water, mixed with mud and bitumen, is cast out. Bubbles of carbonic acid and carburetted hydrogen gas are also disengaged from these springs, and at certain periods with such violence, as to throw the mud to the height of 200 feet. These "air volcanoes," as they are sometimes termed, are known to have been in the same state of activity for the last fifteen centuries; and Dr. Daubeny imagines that the gases which escape may be generated by the slow combustion of beds of sulphur, which is actually in progress in the blue clay, out of which the springs rise.[617] But as the gases are similar to those disengaged in volcanic eruptions, and as they have continued to stream out for so long a period, they may perhaps be derived from a more deep-seated source.

_Beila in India._--In the district of Luss or Lus, south of Beila, about 120 miles N. W. of Cutch and the mouths of the Indus (see Map, fig. 71, p. 460), numerous mud volcanoes are scattered over an area of probably not less than 1000 square miles. Some of these have been well described by Captain Hart, and subsequently by Captain Robertson, who has paid a visit to that region, and made sketches of them, which he has kindly placed at my disposal. From one of these the annexed view has been selected. These conical hills occur to the westward of the Hara mountains and the river Hubb. (See Map, p. 460.) One of the cones is 400 feet high, composed of light-colored earth, and having at its summit a crater thirty yards in diameter. The liquid mud which fills the crater is continually disturbed by air-bubbles, and here and there is cast up in small jets.[618]

_Mineral composition of volcanic products._--The mineral called felspar forms in general more than half of the mass of modern lavas. When it is in great excess, lavas are called trachytic: they consist generally of a base of compact felspar, in which crystals of glassy felspar are disseminated.[619] When augite (or pyroxene) predominates, lavas are termed basaltic. They contain about 50 per cent. of silica, or much less than the trachytes, in which there is usually about 75 per cent. of that mineral. They also contain about 11 per cent. of protoxide of iron, and as much of lime, both of which are wanting, or only in insignificant quantities in the trachytic rocks.[620] But lavas occur of an intermediate composition between the trachytic and basaltic, which from their color have been called graystones. The abundance of quartz, forming distinct crystals or concretions, characterizes the granitic and other ancient rocks, now generally considered by geologists as of igneous origin; whereas that mineral is rarely exhibited in a separate form in recent lavas, although silica enters so largely into their composition. Hornblende, so common in hypogene rocks, or those commonly called "primary," is rare in modern lava; nor does it enter largely into rocks of any age in which augite abounds. It should, however, be stated, that the experiments of Mr. Gustav Rose have made it very questionable, whether the minerals called hornblende and augite can be separated as distinct species, as their different varieties seem to pass into each other, whether we consider the characters derived from their angles of crystallization, their chemical composition, or their specific gravity. The difference in form of the two substances may be explained by the different circumstances under which they have been produced, the form of hornblende being the result of slower cooling. Crystals of augite have been met with in the scoriae of furnaces, but never those of hornblende; and crystals of augite have been obtained by melting hornblende in a platina crucible; but hornblende itself has not been formed artificially.[621] Mica occurs plentifully in some recent trachytes, but is rarely present where augite is in excess.

_Frequency of eruptions, and nature of subterranean igneous rocks._--When we speak of the igneous rocks of our own times, we mean that small portion which, in violent eruptions, is forced up by elastic fluids to the surface of the earth,--the sand, scoriae, and lava, which cool in the open air. But we cannot obtain access to that which is congealed far beneath the surface under great pressure, equal to that of many hundred, or many thousand atmospheres.

During the last century, about fifty eruptions are recorded of the five European volcanic districts, of Vesuvius, Etna, Volcano, Santorin, and Iceland; but many beneath the sea in the Grecian archipelago and near Iceland may doubtless have passed unnoticed. If some of them produced no lava, others, on the contrary, like that of Skaptar Jokul, in 1783, poured out melted matter for five or six years consecutively; which cases, being reckoned as single eruptions, will compensate for those of inferior strength. Now, if we consider the active volcanoes of Europe to constitute about a fortieth part of those already known on the globe, and calculate that, one with another, they are about equal in activity to the burning mountains in other districts, we may then compute that there happen on the earth about 2000 eruptions in the course of a century, or about twenty every year.

However inconsiderable, therefore, may be the superficial rocks which the operations of fire produce on the surface, we must suppose the subterranean changes now constantly in progress to be on the grandest scale. The loftiest volcanic cones must be as insignificant, when contrasted to the products of fire in the nether regions, as are the deposits formed in shallow estuaries when compared to submarine formations accumulating in the abysses of the ocean. In regard to the characters of these volcanic rocks, formed in our own times in the bowels of the earth, whether in rents and caverns, or by the cooling of lakes of melted lava, we may safely infer that the rocks are heavier and less porous than ordinary lavas, and more crystalline, although composed of the same mineral ingredients. As the hardest crystals produced artificially in the laboratory require the longest time for their formation, so we must suppose that where the cooling down of melted matter takes place by insensible degrees, in the course of ages, a variety of minerals will be produced far harder than any formed by natural processes within the short period of human observation.

These subterranean volcanic rocks, moreover, cannot be stratified in the same manner as sedimentary deposits from water, although it is evident that when great masses consolidate from a state of fusion, they may separate into natural divisions; for this is seen to be the case in many lava-currents. We may also expect that the rocks in question will often be rent by earthquakes, since these are common in volcanic regions; and the fissures will be often injected with similar matter, so that dikes of crystalline rock will traverse masses of similar composition. It is also clear, that no organic remains can be included in such masses, as also that these deep-seated igneous formations considered in mass must underlie all the strata containing organic remains, because the heat proceeds from below upwards, and the intensity required to reduce the mineral ingredients to a fluid state must destroy all organic bodies in rocks included in the midst of them.

If by a continued series of elevatory movements, such masses shall hereafter be brought up to the surface, in the same manner as sedimentary marine strata have, in the course of ages, been upheaved to the summit of the loftiest mountains, it is not difficult to foresee what perplexing problems may be presented to the geologist. He may then, perhaps, study in some mountain-chain the very rocks produced at the depth of several miles beneath the Andes, Iceland, or Java, in the time of Leibnitz, and draw from them the same conclusion which that philosopher derived from certain igneous products of high antiquity; for he conceived our globe to have been, for an indefinite period, in the state of a comet, without an ocean, and uninhabitable alike by aquatic or terrestrial animals.