(1) _Atmospheric Waves._--These phenomena have been ably handled by General Strachey,[8] from a large number of observations extending all over the globe. From these it has been clearly established that an atmospheric wave, originating at Krakatoa as a centre, expanded outwards in a circular form and travelled onwards till it became a great circle at a distance of 180 degrees from its point of origin, after which it still advanced, but now gradually contracting to a node at the antipodes of Krakatoa; that is to say, at a point over the surface of North America, situated in lat. 6 N. and long. 72 W. (or thereabout). Having attained this position, the wave was reflected or reproduced, expanding outwards for 180 degrees and travelling backwards again to Krakatoa, from which it again started, and returning to its original form again overspread the globe. This wonderful repet.i.tion, due to the spherical form of the earth, was observed no fewer than seven times, though with such diminished force as ultimately to be outside the range of observation by the most sensitive instruments. It is one of the triumphs of modern scientific appliances that the course of such a wave, generated in a fluid surrounding a globe, which might be demonstrated on mathematical principles, has been actually determined by experiments carried on over so great an area.

(2) _Sound Waves._--If the sound-waves produced at the time of maximum eruption were not quite as far-reaching as those of the air, they were certainly sufficiently surprising to be almost incredible, were it not that they rest, both as regards time and character, upon incontestible authority. The sound of the eruption, resembling that of the discharge of artillery, was heard not only over nearly all parts of Sumatra, Java, and the coast of Borneo opposite the Straits of Sunda, but at places over two thousand miles distant from the scene of the explosions.

Detailed accounts, collected with great care, are given in the Report of the Royal Society, from which the following are selected as examples:--

1. At the port of Acheen, at the northern extremity of Sumatra, distant 1,073 miles, it was supposed that the port was being attacked, and the troops were put under arms.

2. At Singapore, distant 522 miles, two steamers were dispatched to look out for the vessel which was supposed to be firing guns as signals of distress.

3. At Bankok, in Siam, distant 1,413 miles, the report was heard on the 27th; as also at Labuan, in Borneo, distant 1,037 miles.

4. At places in the Philippine Islands, distant about 1,450 miles, detonations were heard on the 27th, at the time of the eruption.

The above places lie northwards of Krakatoa. In the opposite direction, we have the following examples:--

5. At Perth, in Western Australia, distant 1,092 miles, sounds as of guns firing at sea were heard; and at the Victorian Plains, distant about 1,700 miles, similar sounds were heard.

6. In South Australia, at Alice's Springs, Undoolga, and other places at distances of over 2,000 miles, the sounds of the eruption were also heard.

7. In a westerly direction at Dutch Bay, Ceylon, distant 2,058 miles, the sounds were heard between 7 a.m. and 10 a.m. on the morning of the 27th of August.

8. Lastly, at the Chagos Islands, distant 2,267 miles, the detonations were audible between 10 and 11 a.m. of the same day.

Some of the above distances are so great that we may fail to realise them; but they will be more easily appreciated, perhaps, if we change the localities to our own side of the globe, and take two or three cases with similar distances. Then, if the eruption had taken place amongst the volcanoes of the Canaries, the detonations would have been heard at Gibraltar, at Lisbon, at Portsmouth, Southampton, Cork, and probably at Dublin and Liverpool; or, again, supposing the eruption had taken place on the coast of Iceland, the report would have been heard all over the western and northern coasts of the British Isles, as well as at Amsterdam and the Hague. The enormous distance to which the sound travelled in the case of Krakatoa was greatly due to the fact that the explosions took place at the surface of the sea, and the sound was carried along that surface uninterruptedly to the localities recorded; a range of mountains intervening would have cut off the sound-wave at a comparatively short distance from its source.

(3) _Oceanic Waves._--As may be supposed, the eruption gave rise to great agitation of the ocean waters with various degrees of vertical oscillation; but according to the conclusions of Captain Wharton, founded on numerous data, the greatest wave seems to have originated at Krakatoa about 10 a.m. on the 27th of August, rising on the coasts of the Straits of Sunda to a height of fifty feet above the ordinary sea-level. This wave appears to have been observed over at least half the globe. It travelled westwards to the coast of Hindostan and Southern Arabia, ultimately reaching the coasts of France and England. Eastwards it struck the coast of Australia, New Zealand, the Sandwich Islands, Alaska, and the western coast of North America; so that it was only the continent of North and South America which formed a barrier (and that not absolute) to the circulation of this oceanic wave all over the globe. The destruction to life and property caused by this wave along the coasts of Sunda was very great. Combined with the earthquake shocks (which, however, were not very severe), the tremendous storm of wind, the fall of ashes and cinders, and the changes in the sea-bed, it produced in the Straits of Sunda for some time after the eruption a disastrous transformation. Lighthouses had been swept away; all the old familiar landmarks on the sh.o.r.e were obscured by a vast deposit of volcanic dust; the sea itself was enc.u.mbered with enormous quant.i.ties of floating pumice, in many places of such thickness that no vessel could force its way through them; and for months after the eruption one of the princ.i.p.al channels was greatly obstructed by two islands which had arisen in its midst. The Sebesi channel was completely blocked by banks composed of volcanic materials, and two portions of these banks rose above the sea as islands, which received the name of "Steers Island" and "Calmeyer Island"; but these, by the action of the waves, have since been completely swept away, and the materials strewn over the bed of the sea.[9]

(_g._) _Atmospheric Effects._--But the face of nature, even in her most terrific and repulsive aspect, is seldom altogether unrelieved by some traces of beauty. In contrast to the fearful and disastrous phenomena just described, is to be placed the splendour of the heavens, witnessed all over the central regions of the globe throughout a period of several months after the eruption of 1883, which has been ably treated by the Hon. Rollo Russell and Mr. C. D. Archibald, in the Royal Society's Report.

When the particles of lava and ashes mingled with vapour were projected into the air with a velocity greater than that of a ball discharged from the largest Armstrong gun, these materials were carried by the prevalent trade-winds in a westerly direction, and some of them fell on the deck of ships sailing in the Indian Ocean as far as long. 80 E., as in the case of the _British Empire_--on which the particles fell on the 29th of August, at a distance of 1,600 miles from Krakatoa. But far beyond this limit, the finer particles of dust (or rather minute crystals of felspar and other minerals), mingled with vapour of water, were carried by the higher currents of the air as far as the Seych.e.l.les and Africa,--not only the East coast, but also the West, as Cape Coast Castle on the Gold Coast; to Paramaribo, Trinidad, Panama, the Sandwich Isles, Ceylon and British India, at all of which places during the month of September the sun a.s.sumed tints of blue or green, as did also the moon just before and after the appearance of the stars;[10] and from the latter end of September and for several months, the sky was remarkable for its magnificent coloration; pa.s.sing from crimson through purple to yellow, and melting away in azure tints which were visible in Europe and the British Isles; while a large corona was observed round both the sun and moon. These beautiful sky effects were objects of general observation throughout the latter part of the year 1883 and commencement of the following year.

The explanation of these phenomena may be briefly stated. The fine particles, consisting for the most part of translucent crystals, or fragments of crystals, formed a canopy high up in the atmosphere, being gradually spread over both sides of the equator till it formed a broad belt, through which the rays of the sun and moon were refracted. Towards dawn and sunset they were refracted and reflected from the facets of the crystal, and thus underwent decomposition into the prismatic colours; as do the rays of the sun when refracted and reflected from the particles of moisture in a rain-cloud. The subject is one which cannot be fully dealt with here, and is rather outside the scope of this work.

(_h._) _Origin of the Eruption._--The ultimate cause of volcanic eruptions is treated in a subsequent chapter, nor is that of Krakatoa essentially different from others. It was remarkable, however, both for the magnitude of the forces evoked and the stupendous scale of the resulting phenomena. It is evident that water played an important part in these phenomena, though not as the prime mover;--any more than water in the boiler of a locomotive is the prime mover in the generation of the steam. Without the fuel in the furnace the steam would not be produced; and the amount of steam generated will be proportional to the quant.i.ty and heat of the fuel in the furnace and the quant.i.ty of water in the boiler. In the case of Krakatoa, both these elements were enormous and inexhaustible. The volcanic chimney (or system of chimneys), being situated on an island, was readily accessible to the waters of the ocean when fissures gave them access to the interior molten matter. That such fissures were opened we may well believe. The earthquakes which occurred at the beginning of May, and later on, on the 27th of that month, may indicate movements of the crust by which water gained access. It appears that in May the only crater in a state of activity was that of Perboewatan; in June another crater came into action, connected with Danan in the centre of the island, and in August a third burst forth. Thus there was progressive activity up to the commencement of the grand eruption of the 26th of that month.[11] During this last paroxysmal stage, the centre of the island gave way and sunk down, when the waters of the ocean gained free access, and meeting with the columns of molten matter rising from below originated the prodigious ma.s.ses of steam which rose into the air.

(_i._) _Cause of the Detonations._--The detonations which accompanied the last great eruption are repeatedly referred to in all the accounts.

These may have been due, not only to the sudden explosions of steam directly produced by the ocean water coming in contact with the molten lava, but by dissociation of the vapour of water at the critical point of temperature into its elements of oxygen and hydrogen; the reunion of these elements at the required temperature would also result in explosions.

The phenomena attending this great volcanic eruption, so carefully tabulated and critically examined, will henceforth be referred to as const.i.tuting an epoch in the history of volcanic action over the globe, and be of immense value for reference and comparison.

[1] The eruption of Krakatoa has been the subject of an elaborate Report published by the Royal Society, and is also described in a work by Chevalier R. D. M. Verbeek, Ingenieur en Chef des Mines, and published by order of the Governor-General of the Netherland Indies (1886). See also an Article by Sir R. S. Ball in the _Contemporary Review_ for November, 1888.

[2] Verbeek, _loc. cit._, p. 4.

[3] The account of this eruption is a free translation from Verbeek.

[4] Verbeek, _loc. cit._, p. 160.

[5] Judd, _Rep. R. S._

[6] A fuller account by Prof. Judd will be found in the _Report of the Royal Society_, p. 14. Several vessels at anchor were driven ash.o.r.e on the straits owing to the strong wind which arose.

[7] Judd, _Report_, p. 21.

[8] _Report_, Part ii.

[9] In this eruption, 36,380 human beings perished, of whom 37 were Europeans; 163 villages (_kampoengs_) were entirely, and 132 partially, destroyed.--Verbeek, _loc. cit._, p. 79.

[10] Verbeek, _loc. cit._, p. 144-5. The dust put a girdle round the earth in thirteen days.

[11] Verbeek, _loc. cit._, p. 30.

CHAPTER II.

EARTHQUAKES.

_Connection of Earthquakes with Volcanic Action._--The connection between earthquake shocks and volcanic eruptions is now so generally recognised that it is unnecessary to insist upon it here. All volcanic districts over the globe are specially liable to vibrations of the crust; but at the same time it is to be recollected that these movements visit countries occasionally from which volcanoes, either recent or extinct, are absent; in which cases we may consider earthquake shocks to be abortive attempts to originate volcanic action.

(_a._) _Origin._--From the numerous observations which have been made regarding the nature of these phenomena by Hopkins, Lyell, and others, it seems clearly established that earthquakes have their origin in some sudden impact of gas, steam, or molten matter impelled by gas or steam under high pressure, beneath the solid crust.[1] How such impact originates we need not stop to inquire, as the cause is closely connected with that which produces volcanic eruptions. The effect, however, of such impact is to originate a wave of translation through the crust, travelling outwards from the point, or focus, on the surface immediately over the point of impact.[2] These waves of translation can in some cases be laid down on a map, and are called "isoseismal curves,"

each curve representing approximately an equal degree of seismal intensity; as shown on the chart of a part of North America affected by the great Charleston earthquake. (Fig. 37.) Mr. Hopkins has shown that the earthquake-wave, when it pa.s.ses through rocks differing in density and elasticity, changes in some degree not only its velocity, but its direction; being both refracted and reflected in a manner a.n.a.logous to that of light when it pa.s.ses from one medium to another of different density.[3] When a shock traverses the crust through a thickness of several miles it will meet with various kinds of rock as well as with fissures and plications of the strata, owing to which its course will be more or less modified.

(_b._) _Formation of Fissures._--During earthquake movements, fissures may be formed in the crust, and filled with gaseous or melted matter which may not in all cases reach the surface; and, on the principle that volcanoes are safety-valves for regions beyond their immediate influence, we may infer that the earthquake shock, which generally precedes the outburst of a volcano long dormant, finds relief by the eruption which follows; so that whatever may be the extent of the disastrous results of such an eruption, they would be still more disastrous if there had been no such safety-valve as that afforded by a volcanic vent. Thus, probably, owing to the extinction of volcanic activity in Syria, the earthquakes in that region have been peculiarly destructive. For example, on January 1, 1837, the town of Safed west of the Jordan valley was completely destroyed by an earthquake in which most of the inhabitants perished. The great earthquakes of Aleppo in the present century, and of Syria in the middle of the eighteenth, were of exceptional severity. In that of Syria, which took place in 1759, and which was protracted during a period of three months, an area of 10,000 square leagues was affected. Accon, Saphat, Baalbeck, Damascus, Sidon, Tripoli, and other places were almost entirely levelled to the ground; many thousands of human beings lost their lives.[4] Other examples might be cited.

(_c._) _Earthquake Waves._--We have now to return to the phenomena connected with the transmission of earthquake-waves. The velocity of transmission through the earth is very great, and several attempts have been made to measure this velocity with accuracy. The most valuable of such attempts are those connected with the Charleston and Riviera shocks. Fortunately, owing to the perfection of modern appliances, and the number of observers all over the globe, these results are ent.i.tled to great confidence. The phenomena connected with the Charleston earthquake, which took place on the 31st of August, 1886, are described in great detail by Captain Clarence E. Dutton, of the U.S. Ordnance Corps.[5] The conclusions arrived at are;--that as regards the depth of the focal point, this is estimated at twelve miles, with a probable error of less than two miles; while, as regards the rate of travel of the earthquake-wave, the estimate is (in one case) about 3.236 miles per second; and in another about 3.226 miles per second.

On the other hand, in the case of the earthquake of the Riviera, which took place on the 23rd of February, 1887, at 5.30 a.m. (local time), the vibrations of which appear to have extended across the Atlantic, and to have sensibly affected the seismograph in the Government Signal Office at Washington, the rate of travel was calculated at about 500 miles per hour, less than one-half that determined in the case of Charleston; but Captain Dutton claims, and probably with justice, that the results obtained in the latter case are far more reliable than any hitherto arrived at.

(_d._) _Oceanic Waves._--When the originating impact takes place under the bed of the ocean--either by a sudden up-thrust of the crust to the extent, let us suppose, of two or three feet, or by an explosion from a submarine volcano--a double wave is formed, one travelling through the crust, the other through the ocean; and as the rate of velocity of the former is greatly in excess of that of the latter, the results on their reaching the land are often disastrous in the extreme. It is the ocean-wave, however, which is the more important, and calls for special consideration. If the impact takes place in very deep water, the whole ma.s.s of the water is raised in the form of a low dome, sloping equally away in all directions; and it commences to travel outwards as a wave with an advancing crest until it approaches the coast and enters shallow water. The wave then increases in height, and the water in front is drawn in and relatively lowered; so that on reaching a coast with a shelving sh.o.r.e the form of the surface consists of a trough in front followed by an advancing crest. These effects may be observed on a small scale in the case of a steamship advancing up a river, or into a harbour with a narrow channel, but are inappreciable in deep water, or along a precipitous open coast.

(_e._) _The Earthquake of Lisbon, 1755._--The disastrous results of a submarine earthquake upon the coast have never been more terribly ill.u.s.trated than in the case of the earthquake of Lisbon which took place on November 1, 1755. The inhabitants had no warning of the coming danger, when a sound like that of thunder was heard underground, and immediately afterwards a violent shock threw down the greater part of their city; this was the land-wave. In the course of about six minutes, sixty thousand persons perished. The sea first retired and left the harbour dry, so forming the trough in front of the crest; immediately after the water rolled in with a lofty crest, some 50 feet above the ordinary level, flooding the harbour and portions of the city bordering the sh.o.r.e. The mountains of Arrabida, Estrella, Julio, Marvan, and Cintra, were impetuously shaken, as it were, from their very foundations; and according to the computation of Humboldt, a portion of the earth's surface four times the extent of Europe felt the effects of this great seismic shock, which extended to the Alps, the sh.o.r.es of the Baltic, the lakes of Scotland, the great lakes of North America, and the West Indian Islands. The velocity of the sea-wave was estimated at about 20 miles per minute.

(_f._) _Earthquake of Lima and Callao, 28th October, 1746._--Of somewhat similar character was the terrible catastrophe with which the cities of Lima and Callao were visited in the middle of the last century,[6] in which the former city, then one of great magnificence, was overthrown; and Callao was inundated by a sea-wave, in which out of 23 ships of all sizes in the harbour the greater number foundered; several, including a man-of-war, were lifted bodily and stranded, and all the inhabitants with the exception of about two hundred were drowned. A volcano in Lucanas burst forth the same night, and such quant.i.ties of water descended from the cone that the whole country was overflowed; and in the mountain near Pataz, called Conversiones de Caxamarquilla, three other volcanoes burst forth, and torrents of water swept down their sides. In the case of these cities, the land-wave, or shock, preceded the sea-wave, which of course only reached the port of Callao.

[Ill.u.s.tration: Fig. 37.--The lines represent isoseismal curves, or curves of equal intensity, the force decreasing outwards from the focus at Charleston, No. 10.]

(_g._) _Earthquake of Charleston, 31st August, 1886._--I shall close this account of some remarkable earthquakes with a few facts regarding that of Charleston, on the Atlantic seaboard of Carolina.[7] At 9.51 a.m. of this day, the inhabitants engaged in their ordinary occupations were startled by the sound of a distant roar, which speedily deepened in volume so as to resemble the noise of cannon rattling along the road, "spreading into an awful noise, that seemed to pervade at once the troubled earth below and the still air above." At the same time the floors began to heave underfoot, the walls visibly swayed to and fro, and the crash of falling masonry was heard on all sides, while universal terror took possession of the populace, who rushed into the streets, the black portion of the community being the most demonstrative of their terror. Such was the commencement of the earthquake, by which nearly all the houses of Charleston were damaged or destroyed, many of the public buildings seriously injured or partially demolished. The effects were felt all over the States as far as the great lakes of Canada and the borders of the Rocky Mountains. Two epicentral _foci_ appear to have been established; one lying about 15 miles to the N.W. of Charleston, called the _Woodstock focus_; the other about 14 miles due west of Charleston, called the _Rantowles focus_; around each of these _foci_ the isoseismic curves concentrated,[8] but in the map (Fig. 37) are combined into the area of one curve. The position of these _foci_ clearly shows that the origin of the Charleston earthquake was not submarine, though occurring within a short distance of the Atlantic border; the curves of equal intensity (isoseismals) are drawn all over the area influenced by the shock.

As a general result of these detailed observations, Captain Dutton states that there is a remarkable coincidence in the phenomena with those indicated by the theory of wave-motion as the proper one for an elastic, nearly h.o.m.ogeneous, solid medium, composed of such materials as we know to const.i.tute the rocks of the outer portions of the earth; but on the other hand he states that nothing has been disclosed which seems to bring us any nearer to the precise nature of the forces which generated the disturbance.[9]

[1] The views of Mr. R. Mallet, briefly stated, are somewhat as follows:--Owing to the secular cooling of the earth, and the consequent lateral crushing of the surface, this crushing from time to time overcomes the resistance; in which case shocks are experienced along the lines of fracture and faulting by which the crust is intersected. These shocks give rise to earthquake waves, and as the crushing of the walls of the fissure developes heat, we have here the _vera causa_ both of volcanic eruptions and earthquake shocks--the former intensified into explosions by access of water through the fissures.--"On the Dynamics of Earthquakes," _Trans. Roy. Irish Acad._, vol. xxi.

[2] Ill.u.s.tration of the mode of propagation of earthquake shocks will be found in Lyell's _Principles of Geology_, vol. ii. p. 136, or in the author's _Physiography_, p. 76, after Hopkins.

[3] "Rep. on Theories of Elevation and Earthquakes," _Brit. a.s.s. Rep._ 1847, p. 33. In the map prepared by Prof. J. Milne and Mr. W. K. Burton to show the range of the great earthquake of j.a.pan (1891), similar isoseismal lines are laid down.

[4] Lyell, _loc. cit._, p. 163. Two Catalogues of Earthquakes have been drawn up by Prof. O'Reilly, and are published in the _Trans. Roy. Irish Academy_, vol. xxviii. (1884 and 1886).