The Sudden Movements Of The Earth's Crust

( Originally Published 1915 )

SEISMIC PHENOMENA

WHEN the terrestrial crust was formed by the solidification of the superficial layers of the nucleus of fused matter, the spheroidal form of which constituted the Earth at the commencement of its history, a colossal reserve of energy was imprisoned inside it. This energy results from the heat of the central nucleus, which exists at inconceivably high temperatures.

Now the crust is far from homogeneous; it was not formed all at once, but in pieces of which the earlier ones constituted scoriae, isolated and floating on the surface of the spherical liquid mass. These became gradually united with each other and, being of various thicknesses, thus formed the first irregularities of the Earth's crust, which, as previously stated, has been likened by Lapparent to a marqueterie. Its lack of continuity and homogeneity involves an important consequence; if we compare it to a boiler, this boiler will not be equally strong everywhere but will have weak places in its sides, flaws as they are called in metallurgy, and it will burst at these places if the interior pressure increases beyond a certain limit.

The internal energy may manifest itself by the upward expansion of the material forming the superficial layers of the central nucleus, through a fissure in the crust. Such a manifestation constitutes a volcanic eruption.

As regards the cause which makes this material rise up through the fissures and fractures of the crust, it is probable that under the influence of the progressive cooling of the nucleus, a cooling which although very slow continues incessantly, gases are given off from the still liquid upper layers of the internal part, and accumulate under the crust, upon which they consequently exert a pressure. Possibly, also, the water of the seas infiltrates through the crust, which is of less thickness under the seas than under the continents, as we have already seen ; such infiltration would lead to the contact of the water with the igneous masses and the consequent dissociation of the water into its constituent gases. This would be another cause of an increase of internal pressure, tending to break open the crust and let the gases and the igneous material shoot out, or in any case to disturb violently the sides of the boiler, so to speak, which the terrestrial crust forms. The thermal energy accumulated at the centre can thus manifest itself exteriorly in two distinct ways: either by an expansion of the inner liquid and gaseous material and the forcing of this through the crust which the pressure has made to yield at some point; or by sudden movements and agitations imparted to the crust, by the internal pressure, which moves or bends the crust without breaking it, this sometimes resulting only in a vibratory phenomenon transmitted as true waves. The first is a volcanic eruption; the second a seismic phenomenon. Volcanoes and earthquakes are consequently two manifestations of the same cause, but they are in no wise directly connected together. As an example, in the case of Japan, which is the classical earthquake country, so to speak, the internal activity which is constantly manifested by very frequent earthquakes does not awake the old volcano of Fusiyama from its long quiescence.

The general characteristic of a volcano is that it occupies the summit of a mountain and gives off permanently, or at intervals, a greater or less abundance of vapours; from time to time out of an opening whose mouth is at the summit, and which is called the crater, it ejects a rain of stones and cinders accompanied by thick clouds of vapours and sometimes by burning gases. Such were the thick burning clouds observed at Martinique during the courageous and profitable study which Professor Lacroix made of the volcano, Mt. Pelée. These clouds are often the seat of violent electrical manifestations and are, consequently, furrowed with lightning flashes. While these emissions of gases and vapours take place into the atmosphere, a river of fused rock and similar materials, called lavas, emerges from the crater, streams down the sides of the mountain, and covers the surrounding country, sometimes to a considerable distance, retaining for a long time a very high temperature.

The volcanic mountains have been formed by such lavas and eruptive rocks. The first eruption takes place through a fracture in the crust; the rocks and lava accumulate around the orifice as an ever-increasing cone, since each new eruption expels new material onto the flanks of the small mountain so formed, increasing at the same time its extent and height. Little by little the cone becomes a mountain, which remains pierced with a passage traversing the crust, called a funnel, preserving communication between the exterior of the crust and the fluid incandescent magma which forms the upper part of the central nucleus. It is the extremity of this funnel which is called the crater.

When a volcanic mountain attains a certain height, for example over 4000 metres [13,000 ft.], as is the case with Mauna Loa in the Sandwich Isles, it needs a great pressure to support and eject the high column of lava which fills the funnel. For such an eruption to take place, the cause of which we have already spoken must operate, viz., the setting at liberty of enormous quantities of gases in the upper part of the interior magma. These gases result from the effervescence of the fused masses and are expelled violently through the orifice open to them, being forced out in con-sequence of their extreme pressure. In the case of Mauna Loa, the mountain stands on the floor of the Pacific and rises, as stated, to a height of more than 4000 metres [13,000 ft.] above sea-level; the gases which eject lava to its summit must therefore exert a pressure of several thousand atmospheres. These lavas constitute a veritable lake of fire in the great crater which exists at the summit of the mountain; they overflow and spread in great rivers of fire down the sides of the gigantic cone. We have here a typical case of a continuous outpouring of lava. Such a volcano is a true safety valve in this region of the Earth's crust.

But other kinds exist which are subject to frequent eruptions, notably the European volcanoes such as Vesuvius and Etna. When the lava arrives at the summit, already cooler and in less quantity, or when after an eruption, there is a slackening in the upward propulsion of the fused matter, that which fills the upper part of the crater solidifies, and the internal energy can then manifest itself again only by the emission of more or less abundant gases and vapours, which are prevented from escaping by the mass blocking the funnel of the volcano, and so accumulate under the crust. Since the pressure gradually increases, sooner or later it becomes greater than the obstructing mass can sustain. The eruption, therefore, has the character of a veritable explosion and often the entire mountain bursts like a shell does under its charge of melinite. The debris of the explosion is projected high up, ashes being carried to a height of several kilometres [or miles], and rocks are flung on to the surrounding country for a distance of hundreds of kilometres [or miles]. The eruption in this case always attains the character of a catastrophe; it will suffice to recall the eruption of Mt. Pelée in June, 1902, and that of Krakatoa, in the Sunda Isles in 1883. In cases where the volcano rises directly out of the sea, constituting a small island, the latter may entirely disappear in the course of the cataclysm; often, however, it only partially disappears, leaving only the crater above the sea-level. In this case a pierced or broken portion frequently admits the sea to the interior of the crater, thus forming an almost closed bay as may be seen in Saint Paul Island, in the Southern Seas. This crater-isle has been carefully studied by Professor Vélain. The Greek Archipelago has often been the seat of similar cataclysms, for example that of Santorin.

The mass of material ejected by volcanic eruptions may attain considerable proportions. This is obvious from a consideration of the volcano in the Sandwich Isles, before mentioned, when we recollect that the mountain itself, more than 4000 metres [13,000 ft.] high, is formed by material ejected during successive eruptions, and accumulated as a cone around the original orifice. The island itself, which is entirely constituted of lava, forms a mass of more than three hundred thousand cubic kilometres, since the cone is continued downwards under the water to the submarine floor of the Pacific. Even Vesuvius, one of the smallest of volcanoes, has given forth streams containing fifteen and twenty millions of cubic metres [or yards] of lava. The statement of these quantities, if we remember the considerable number of craters that are active at the present time and also the very large number of extinct volcanoes, shows that the exterior appearance of the Earth is incessantly modified by the addition of new material which alters its relief and strews its surface with mineral matter brought from the interior of the globe.

For a volcano to come into being, there must initially be a cleft, fissure, or crack in the terrestrial crust. Now there are regions of the Earth which are especially prone to such fractures, viz., the border of the oceans, chiefly those where an elevated coastal region dips suddenly to the sea. The seas, in fact, mark the lowered portions of the kind of marqueterie formed by the terrestrial crust while the continents represent the raised portions.

Maritime shores are therefore volcanic regions, places given over to volcanic activity; it is only necessary to glance at a map of the world (Fig. 22, p. 215) for direct confirmation of this fact. The Pacific is bordered everywhere, even on the shores of the Antarctic continent by a girdle of active volcanoes which constitute a veritable fiery circle ; so also a long chain of volcanoes lies along the shores of the Mediterranean, and extends by Asia Minor and the Persian Gulf to the Sunda Isles. Another line of craters lies in the midst of the Atlantic from Jean-Mayen and Etna in the north to the Antarctic volcanoes in the south passing by the shores of the Azores, Madeira, and the Canary Islands, where stands the imposing Peak of Teneriffe whose activity has recently begun to manifest itself anew.

The number of active volcanoes actually known is to be reckoned by hundreds, and this does not include the numerous submarine volcanoes which doubtless exist beneath the oceans, especially under the Pacific and whose existence and activity are only made known to us by abnormal waves on the surface of the oceans which cover them. Be-sides volcanoes proper, other phenomena at various points on the Earth's surface clearly point to internal activity; geysers, hot springs, and emanations of sulphur vapour which burst through fissures in the crust bear witness to the heat energy accumulated in the interior regions below. Volcanic eruptions and gaseous emanations do not always form a sufficient vent for the manifestations of this energy, which relieves itself by means of other phenomena that we will now study, viz., seismic phenomena.

The causes of the instability of the Earth's exterior envelope are numerous. We have noted, in the course of preceding chapters, those of them that are periodic, but there are others, the explanation of which must be sought in the situation of the crust itself in relation to the heated nucleus which it covers.

This nucleus cools at a constant rate, and in so cooling contracts. In the course of time, therefore, a space would be left between the upper layer of the nucleus and the lowest part of the crust which floats on its surface. The part of the crust which is immersed in the liquid mass below it has in all cases an upward thrust exerted upon it by this mass. Now this thrust will diminish in proportion as the internal mass contracts. After a sufficient interval of time the crust becomes insufficiently supported from below and so tends to sink down, and this sinking gives rise to a disturbance which affects a larger or smaller area around the principal centre of action. As the cause operates continually, this result will occur at all times of the Earth's history, though perhaps in a discontinuous way.

This is not all. Volcanic eruptions throw on to the surface of the Earth's crust a quantity of material which was formerly below it; such material is not replaced and its removal creates a space below the crust and at the same time adds to the weight of the latter. Gravity consequently tends to make the surface fall in and fill up the space when the upward thrust from below is insufficient, and this constitutes another reason why the external envelope of the Earth sinks.

It is, therefore, to be expected that each such sinking will be manifested by a shock, sometimes feeble and sometimes great, according to the degree of the fall which produces it. Furthermore, currents circulate in the upper regions of the central nucleus, the part that is still fluid, and so cause waves and undulatory movements which come into contact with the parts of the crust which project below its interior surface. The force of these agitations results in the shaking of these projecting parts, and the disturbance is transmitted by them to the rest of the solid crust. There are thus numerous reasons why the crust is never in repose.

In consequence of the incessant disturbances to which the crust is subjected, the importance of the study of these sudden movements that have such a disturbing effect upon it will be obvious.

Some of the shocks are devastating, others are feeble, sometimes even so feeble that only delicate instruments called seismographs, which are always based on the principle of inertia, can disclose and register them. There are, therefore, earthquakes and earth tremors, of varying intensity.

It is customary to divide the shocks which the Earth's crust undergoes into three categories: vertical shocks which, if intense enough, may project buildings upward into the air, as if an explosion had taken place; horizontal shocks which displace objects on the ground laterally and which are capable, among other results, of displacing an upper course of masonry with respect to a lower one; and finally undulatory shocks, the most numerous and the most terrible, which spread through the ground surface in the same way as the swell of the ocean spreads through the water surface. When such a seismic wave occurs the surface of the Earth's crust is agitated and disturbed just as the waves of the sea are. But these shocks produce a permanent alteration in the solid surface, whereas the waves of the sea give rise to only a passing perturbation; numerous deep crevices appear, buildings are destroyed, trees torn up, and whole towns may be annihilated. Recent examples are Valparaiso, San Francisco, and, still more lately, Messina, where the earthquake destroyed more than 200,000 human lives in a few seconds.

The centre of disturbance, the point from which the waves seem to radiate, is almost always below the surface of the ground, sometimes at very considerable depths, even up to 20 kilometres [12.5 miles]. This point is analogous to that where a stone thrown into water strikes the water; circular waves originate there and travel outwards. The orientation of the crevices and their inclination to the vertical enable the position of this point to be fairly accurately obtained. The projection of the centre of disturbance onto the ground surface, that is to say the place where the centre would be marked on a map, is commonly called the epicentre. The crevices crop out in the ground around the epicentre, forming concentric curves, roughly circular when there is only one centre of disturbance, but often elongated, in which case the existence of several such centres seems to be indicated. The movements are propagated at the surface of the ground with velocities varying between 150 and 800 metres [500—2500 ft.] per second ; we shall see later on that the rate of propagation for the total mass of the Earth is much more rapid.

These great disturbances are happily not very frequent; it is the earth tremors, detected and registered only by means of seismographs, which by their frequency prove the continual quivering of the solid crust of the Earth. It appears that more violent earthquakes occur when the barometric pressure is low, which is easily understandable, since the terrestrial crust supports a less quantity of atmosphere than usual and so the pressure inside it is not so much opposed as usual. A barometric fall of i centimetre [.39 in.] produces an increase of internal pressure of 130 kilograms per square metre [285 lbs. per sq. yd.], i. e., 130 millions of kilograms per square kilometre. Earth-quakes are also more frequent in winter than in summer and are especially numerous at the time of the equinoxes; the eruption of Mt. Pelée, in Martinique, in 1902, accompanied by a consider-able local earthquake and a tidal wave, occurred at a time when the Sun and the Moon were in a straight line with the Earth and so produced a combined attractive effect on the latter. Possibly internal tides arise forming a wave at the upper liquid surface of the central magma ; it is then readily understandable that, at the period of the equinoxes, when the luni-solar attraction is greatest, the internal tide, and consequently its wave-force, would be strongest. In this case, as M. Kovesligéthy believes, external factors would be the determining causes of the liberation of the internal energy which takes place in virtue of the weaknesses of the crust.

It is possibly in this direction that we must seek the solution of the important problem of the foretelling of earthquakes; such a result can only be attained by studying the laws which govern the movements of the superior fluid layer of the interior nucleus of the Earth, with the aid of modern physical methods, with their increasing precision. A remarkable coincidence between the years of maximum earthquakes, of maximum polar aurora, and of maximum magnetic storms has already been demonstrated. The periodicity of the three phenomena is the same, viz., eleven years, which is also the periodicity of the maximum activity of the solar spots. Our Sun, by the attraction of its mass, is the cause of the complex movements the Earth performs; by warming the Earth's crust it produces a daily deformation of the latter; it causes tides, not only at the surface of the seas, but also at the surface of the ocean of heated lava which exists beneath our feet. The question naturally follows whether the periodical variation in th number of the solar spots produces a variation of the kinds of radiation emitted by the Sun and what effect this would have upon the Earth. The Sun creates a field of force about it in space, and the intensity of this field is affected by the slightest variations in the solar activity. It may be, therefore, that we must make a fuller study of the Sun in order to determine the law of the vicissitudes of the Earth's thin and incessantly quivering crust. When we have described the magnetic and electrical phenomena of which the Earth is the seat, we shall still better understand the unquestionable influence that the Sun exerts on the terrestrial globe.

Seismic phenomena should not be treated as isolated occurrences, for the same reason that applies in the case of volcanic eruptions, viz., that they are different manifestations of the internal energy, having no "laws" or necessary interconnection in time or space, but they nevertheless arise from one sole cause, so some law should govern them when taken together.

The universal prevalence of seismic phenomena is established, just as in the case of volcanic eruptions. As regards the last, we know that nearly four hundred active craters exist on the Earth's surface, and that more than double this number of extinct or sleeping ones can be distinguished, and this takes no account of the unknown number, which is perhaps very large, that the oceans cover with their vast area of water. In recent years there have been signs of awakening of many of these centres of eruption, for example, in 1909, as has previously been mentioned, the old volcano of Tenerife, which had seemed definitely extinct has given proof of a renewal of activity.' A week never passes without the telegraph bringing news from some part of the world of earth movements, sometimes devastating, sometimes less important, but always clearly perceptible; especially in Turkestan, India, the Caucasus, the Philippine Isles, Japan, Sicily, and Provence, the Earth quivers and suffers incessant disturbance. Islands even disappear suddenly. We have thus isolated occurrences which are various forms of phenomena all resulting from one general cause. Lallemand has investigated whether it would not be possible to account for the seismic manifestations of the internal activity on the basis of the tetrahedral theory of the formation of the terrestrial crust, which theory was proposed by Lowthian Green in 1875. We have already said a few words about it near the beginning of this work, and we must now return to it in fuller detail. The English scientist had shown that when tubes of india-rubber were compressed from the outside, these tubes instead of being flattened took a form, the section of which was a triangle with concave sides. If the air contained in a glass globe, which is softened by heat, be exhausted, the globe, originally spherical, takes a form in which four hollow faces are clearly seen, these being the regions of flattening under the influence of the relative external increase of pressure. This form of triangular pyramid, or rather the tendency to take this form, is, more-over, a consequence of the principle of least action; starting with a fixed surface area, the terrestrial crust should, nevertheless, diminish as regards its enclosed volume, since the force of gravity makes it remain in contact with the internal nucleus and since this is continually contracting as it cools. In order to enclose a minimum volume, and at the same time obey the double condition of maintaining a fixed surface area and a symmetrical form, the crust must tend appreciably towards the figure of a regular tetrahedron, i. e., a triangular pyramid with equilateral faces, which is a regular solid occupying the minimum volume for a given area of surface.

It seems, however, at first sight, that the pyramidal form, with its edges, apices, and faces, is far removed from a spheroidal one, but we shall see that such dissimilarity is only apparent and that, on the contrary, the resemblance becomes marked when we study the matter more closely.

The exterior appearance of the Earth, i. e., the aspect it would present to an observer placed far away from it in space, is the result of the combination of the solid crust and its aqueous envelope, or in other words the lithosphere and the hydrosphere, the barysphere or central nucleus being in the interior of the first two.

If, since the time of its definite solidification, the crust tended to take the tetrahedral form, its foldings, and consequently the general orientation of the features of its relief, would have been made under the influence of that tendency (Fig. 4, p. 32) Hence the regions near the summits of the pyramid would be the only ones emerging above the hydro-sphere. Moreover it is natural to suppose that the terrestrial axis coincides with one of the four axes of symmetry of the tetrahedron; there ought, thus, to exist in one of the hemispheres three continental elevations, represented by three summits, the corresponding pole being occupied by an ocean, the bottom of which is represented by one of the flattened faces of the pyramidal figure. On the other hand, the opposite pole would be at the fourth summit of the pyramid and consequently a continental mass would emerge there above the spheroidal surface of the oceans.

Voyages made in the polar regions, both arctic and antarctic, during recent years fully confirm these aspects of the theory. Nansen, in the course of his circumnavigation around the North Pole has shown that that region was occupied by a sea whose depth reaches nearly 4000 metres [2.5 miles] ; I on the other hand Ross, de Gerlache, Charcot, Scott, Shackleton, and Amundsen have verified the existence around the South Pole of an immense continent whose centre is occupied by a highly elevated plateau and above which rise peaks whose height surpasses 4000 metres. The diametrical opposition of the continents and seas is consequently demonstrated with remarkable clearness, as regards the polar regions.

It is equally verified by terrestrial geography as a whole; the three continents Europe, Asia, and America, widened at the north and narrowed towards the south, are separated by three oceans, narrowed in the northern parts and broadening in the southern hemisphere. It may be said that Europe and Asia are connected together in their northern parts, but this is rather a superficial objection, since beyond the Caspian Sea and the Sea of Aral obvious signs of an actual depression between these two continents exist. Also precise measurements have shown that the western half of Siberia has only a very slight elevation above sea-level; a very slight lowering of the level would transform that part of the continent into a sea. Possibly, at a not very far distant period this depression, lying along the foot of the Ural Mountains, was covered by an actual sea. The pointed terminations of the continents towards the south, Cape Horn, the Cape of Good Hope, the point of Tasmania prolonging Australia, which is itself a continuation of the Asiatic continent, indicate that the base of the terrestrial tetrahedron is towards the north. The northern widened parts of America and Asia are very nearly connected together by the elongations between which passes the Strait of Behring.

But we can carry still further the conclusions that may be deduced from this theory of the terrestrial tetrahedron. So far, we have only considered the tendency to the tetrahedral form in the case of an immovable Earth. We know, however, that the Earth is not immovable, but, on the contrary, performs a number of combined movements of which one of the most important is its movement of rotation.

What would be the effect of the Earth's rotation on the tetrahedral figure at the time when this was being formed? We shall see that it would deform the lines and produce on the solid crust a geographical modification of which indisputable evidence is found and which would be difficult to explain in any other way. A familiar comparison will lead us to understand the nature and origin of this deformation.

Let us take an old umbrella, the covering material of which has been removed, and at the end of each rib attach a little leaden ball. Then let the umbrella be opened and an effort be made to make it turn between the fingers, holding it vertically with one hand and rotating the curved part of the handle with the other; this curved part furnishes a lever arm to the motive force given by the hand. A resistance will be felt which tends to retard the rotation of the umbrella and such resistance is due to the moment of inertia of the apparatus, this tending to resist the turning force that we apply. If we apply a too violent force to the handle in this way, we shall cause a torsion of the ribs and their supporting pieces and these will twist if their attachment be sufficiently strong to stand it.

A similar thing occurred at the time of the solidification of the crust. The emergent continental apices, though acted upon by the force of rotation, remained in place by reason of their inertia. But the influence of the rotation continued to make itself felt and the edges connecting the northern continental apices with the South Polar apex (Fig. 4, p. 32) were twisted in the middle so that while the northern parts of the continents remained retarded, towards the west, their southern parts, narrowing down to points, became deflected towards the east because of a common deviation. A glance at a map of the world (Fig. 21) clearly shows the fact of this deviation.

Now, in twisting, the edges of the tetrahedron became weakened in their middle points and the crust gave way there; this line of rupture actually exists, and has received from geographers the name of the intercontinental depression. This depression is a kind of furrow, a marine girdle which completely surrounds the terrestrial globe nearly at its middle, that is to say in the neighbourhood of the equator, of which it is north in some parts and south in others. Europe, in fact, is separated from Africa by the Mediterranean Sea; Asia is separated from Australia by a series of seas almost blocked with chains of islands which are mountains whose summits only appear above the waters. Finally North America is connected to South America only by the frail junction known as the Isthmus of Panama (Fig. 22).

The remarkable conception of a terrestrial tetrahedron has yet another consequence; it allows of a very simple explanation of the distribution of volcanoes and of centres of seismic disturbances on the Earth's surface.

When the interior crust, under the influence of the contraction of the central nucleus, became folded and wrinkled in order to remain in contact with the nucleus, which gravity forces it to do, the foldings showed the tendency to conform to the tetrahedral character. These foldings were formed in a still plastic crust, but later, when it became rigid, the action of the same forces tended to produce, not foldings, but fractures. Hence the continuous shocks which disturb the crust are due to its deformation.

But the regions where the foldings were produced are the regions of least resistance; if a boiler plate be bent and then made use of, the pressure of the steam will cause a fracture exactly in the place where it was bent. Now the edges of the tetrahedron and the neighbouring regions are the foldings of the crust, and so its resistance ought to be more feeble there. The same applies to the whole length of the intercontinental depression where the crust has already suffered a twisting tending to enfeeble its resistance to rupture. The great continental ridges, such as the American Cordilleras and the chains of islands bordering the Pacific, will thus be the special regions of earthquakes and of volcanoes, the latter formed about the fissures of the folded and weakened crust. A fortiori, the greatest number of volcanoes ought to be situated at the points of inter-section of the continental ridges with the great intercontinental depression. That this is so may be seen at once from a glance at a map giving the distribution of volcanoes over the Earth's surface (Fig. 22). The Pacific in particular is surrounded by a veritable fiery circle. On the other hand, craters are not met with on the gentle slopes whose uniform inclination shows that there have been no foldings and sudden deformations.

The mining engineer, M. Lallemand, to whom we owe the extremely precise methods of modern determinations of altitude, has shown that the tetrahedral theory also allows of a very natural explanation of the anomalies that have been proved to exist in the values of gravity. As we know, gravity is weakest in the middle of a continent and strongest on the oceanic islands, whereas if we were guided only by the consideration of the density of the immediately neighbouring regions, the reverse should be the case.

Actually, the exterior of the terrestrial globe comprises two different things, the lithosphere which is the foundation, supporting the hydro-sphere which covers the greater part of it. The latter, because of its fluidity, obeys the combined actions of gravitation and centrifugal force. If its surface be prolonged in imagination beneath the earth, as a result of the operations of determination of altitude, we arrive at a surface called the geoid, as previously stated, which is the fundamental surface of elevation in the consideration of gravity. But in the neighbourhood of the summits of the tetrahedron, that is to say, in the central regions of the great continental masses, this surface must project above the normal ellipsoid of the geodesists, for the tendency to the tetrahedral form manifested by the lithosphere since its original solidification ought also to be found in a smaller degree, in the fundamental surface of elevation. Consequently there should be corresponding irregularities in the values of gravity reduced to sea-level, that is to say, after allowance has been made for the attraction of the subjacent crust. In the neighbourhood of the edges of the tetrahedron, rising above the theoretical ellipsoid, the attractive force ought therefore to be weaker and the centrifugal force stronger than in the middle of the oceans, where the stronger attraction and the weaker centrifugal force produce, on the contrary, an excess of gravity. This excess in the oceanic islands and the deficit in the interior of the continents are shown by actual experimental results.

This new explanation due to M. Lallemand is in complete accord with the theory of Professor Lippmann. It is also another confirmation of Lowthian Green's conception of the terrestrial tetrahedron. One last source of confirmation has not yet been realised, but will certainly be so sooner or later, thanks to the efforts of the International Geodetic Association, viz., the measurement of three long meridian arcs in the southern hemisphere, as near as possible to the South Pole and in particular in the Argentine and the Antarctic continent. If the Earth has really tended to a tetrahedral form in becoming externally solidified, and considering, besides, the large proportion of water and the small extent of land in this hemisphere, it may legitimately be supposed that the flattening of the Earth will be a little less in the southern hemisphere than in the northern one. The greater number of the long meridian arcs on which geodesists have centred their efforts during the last century and a half are situated in the northern hemisphere. The Peru arc, measured in the eighteenth century by Bouguer and La Condamine, and remeasured recently with great trouble and remarkable precision by the equatorial mission under the direction of Colonel Bourgeois, and the arc of the Cape of Good Hope, the work of English astronomers, are the only ones furnishing us with data for the southern hemisphere. It is to be hoped that a continuous arc, from Cape Horn to the equator, will soon be measured in South America. Then we shall perhaps have the ultimate data that we lack, in default of a method and instrument permitting navigators to measure the intensity of gravity with precision on board ship in the open sea, in the regions of the geoid where, up to the present, all accurate results have been found impossible.

The scientific observation of earthquakes, which registering seismographs enable nowadays to be made continuously, gives us new information as to the rigidity of the terrestrial globe, which as previously shown, we have deduced from the tides of the crust.

When a strong earth tremor is produced at any point whatsoever of the Earth, the most distant seismographical observatories, those which for example are situated 6000 or 8000 kilometres [3600 to 4800 miles] from the initial centre of disturbance, are affected by it after several minutes, when the seismographs become agitated. If the time of the first registering of the phenomenon, propagated through the entire mass of the Earth and not simply over the surface of the crust, be compared with the actual time of its occurrence at the place of origin, it may be proved that the movement is propagated with a velocity of about 10 kilometres [6.2 miles] per second. This is a speed three hundred times greater than that of the most rapid of our express trains.

After a further few minutes, the apparatus will be again disturbed, more strongly and for a greater length of time than previously. If, as in the preceding case, we compare the times of origination and registering of the original shock we find that these other seismic waves are propagated with a velocity of 5 kilometres [3 miles] per second, viz., about half of the preceding. If we compare these results with those which the mathematical theory of elasticity gives, we find an exact agreement. In fact this theory, which is based on experimental evidence, teaches us that if an instantaneous disturbance be produced at one point in a perfectly elastic solid, two series of waves arise in the solid, the first of which propagate themselves with a velocity double that of the second. This is precisely what the study of seismographical observations shows us, and so we have a remarkable accordance between theory and observation.

By making use of the seismographical results in the elasticity calculations it is found that the elasticity of the Earth, considered in its entirety, is of the same order of magnitude as that of steel, actually a little greater. This agrees wonderfully with the results deduced from the study of the terrestrial tides and of nutation.

We can now understand, from this knowledge of the Earth's elasticity, why observations of the propagation of seismic disturbances to the immediately neighbouring regions has never shown a velocity of more than 800 metres [1/2 a mile] per second. Waves are transmitted to the neighbouring points by the crust itself, while transmission to distant places takes place through the elastic medium constituted by the Earth as a whole. Thus the density of the central nucleus of our globe is confirmed; although at inconceivable temperatures this acquires, by reason of the pressure to which it is subjected, a physical state practically equivalent to the solid state, and consequently possesses a rigidity of the same order as that of the best kinds of steel.

Earthquakes result not only in sudden shocks but also in permanent deformations of the terrestrial crust. We must seek for the origin of these, not in explosive eruptions, but in settlements, i. e., in movements which affect the juxtaposed portions of the marqueterie which the terrestrial crust resembles, when such portions exhibit a certain amount of play with regard to each other. This conclusion is verified by the permanent cracks which accompany great earth-quakes and which sometimes attain up to 5o or even 10o kilometres [30 to 6o miles] in length. Usually, also, one of the edges of the crevice so formed is raised with respect to the other side, which is lower. Often there is a displacement of level in the horizontal sense, and if, for example, the region affected by the earthquake is traversed by a road, this may be cut in such a way that the two pieces have no longer either the same direction or the same level. Earthquakes thus give rise to permanent deformations of the crust, deformations the origin of which is a sudden movement of the latter. The exact surveys carried out by the government officials of the different countries have demonstrated permanent differences of level of more than 2 metres [or yards] in the regions affected by the more important earthquakes, such for example as that established by the geodetic operations carried out in Croatia after the Agram earthquake in 1885.

But in addition to these permanent deformations of sudden origin, there are also slow deformations that our Earth's crust undergoes continuously. We can only perceive these movements by an advance or recession of the seashores, which appear to encroach on the land or move seawards as the case may be. Examples of such occurrences are abundant; in the Red Sea may be seen lines of coral reef of relatively recent date, emerging above the actual sea-level, and which could only have been constructed by their microscopic builders when under a protective layer of water, which must thus have covered them not long since. In Scandinavia a kind of sea-saw movement is taking place; the bottom of the Gulf of Bothnia is rising while the southern part of the peninsula seems to be gradually sinking into the sea.

By comparing the coast reference marks traced, by Celsius, on the rocks on the shores of Sweden, in 1730, we are able to prove ground movements reaching nearly 2 metres [or yards] per century, and similar facts have been verified in Norway, Finland, and Siberia. Everyone knows the classical alternation of risings and fallings exhibited by the columns of the Temple of Serapis at Pozzuoli, a movement which averages I millimetre [.039 inch] per year. In the Indies, subterranean forests have been discovered; in Prussia, lakes of relatively recent formation exist in depressions of the ground. Finally, in mountainous countries, a spectator placed upon a summit may see distant mountains just above a near hill in front of him; now in several cases such visibility of the distant peaks has ceased, on account of a slight upraising of the interposed hill or of a sinking of the distant peak in question. These phenomena have taken place in French Jura, in Spain, Bohemia, Switzerland, and Thuringia. The sinkings of the constituent portions of the Earth's crust are thus a general and permanent phenomenon ; when they do not occur suddenly, they operate slowly, but they take place unceasingly, giving a perpetual mobility to the ground that seems to us so firm. From this, it will be recognised how great is the importance, from the point of view of the Earth's history, of those precise determinations of altitude, which are the only means of detecting and measuring relative changes of altitude of different points on the solid. surface of the terrestrial globe.

There is one further manifestation of the interior activity of the globe which may lead to terrible catastrophes, viz., the violent movements of the sea, which are wrongly called tidal waves, for they have no connection with the periodic phenomena of flux and reflux of the waters of the sea.

A tidal wave originates in a seismic disturbance occurring at a place on the bottom of the sea. Such a phenomenon may happen as a result either of a sudden upraising or a sudden lowering of the submarine surface. Let us suppose, for example, that the cause is the former; a liquid protuberance immediately forms at the upper surface of the sea over the part of the submarine surface raised. This protuberance is bordered by two hollows, the depth of which is in proportion to the height of the uplifted mass of water, and so gives rise to a wave which extends outwards over the surface of the ocean, reproducing on a vast scale the phenomenon of rings which is caused by drop-ping a stone into water. A wave which is called the seismic wave of translation is thus propagated over the surface of the sea. Its arrival at the coast is preceded by a lowering of the water to the same extent, which at the moment of reaching the shore produces there a retreat of the sea. Ships are suddenly grounded on the bottom of the ports where they are at anchor. But an instant later the high wave follows this retreat, and the vessels are sometimes carried a considerable distance on to dry land by the crest of the great wave which has refloated them. Water breaks over low coasts, submerging habitations and drowning men and animals in its passage; this is what occurred at Lisbon, in 1755, when a terrible tidal wave followed a great earthquake. Thirty thou-sand persons were killed by these two causes combined.

Tidal waves are formidable even on coasts like those of Portugal, but are far more so in the case of low shores such as exist in the archipelagoes of Polynesia. This is especially so as they are situated in the Pacific, a region where earthquakes are numerous and therefore are more exposed to these terrible phenomena. These islands have but a very slight elevation above sea-level and when a tidal wave descends upon one it destroys everything upon it.

The velocity of propagation of these waves of seismic origin attains considerable values; they may move over the surface of the sea at the rate of 350 to 400 nautical miles an hour. The marine mile equals 1852 metres and is the length of one minute of arc measured on a terrestrial meridian. Therefore these seismic waves of translation are propagated with the velocity of 750 to 800 kilo-metres [465 to 495 miles] per hour. The eruption of Krakatoa was accompanied by a violent local seismic phenomenon and this produced a gigantic wave of translation, which made itself felt two days afterwards by the tide recorder at Rochefort. In the regions of Japan or Peru earthquakes frequently accompany volcanic eruptions, and on every occasion it has been shown that the wave traverses the entire width of the Pacific in twelve hours.

Oceanographers have made a study of this propagation and found a simple law governing it. If the mean depth of the ocean at whose surface the wave is moving be multiplied by the mean intensity of gravity along the track of the wave, the square root of the product is equal to the velocity of propagation. As a consequence of this simple law we may in nearly every case deduce the mean depth of the ocean over whose surface the wave passes. For we can usually deter-mine the velocity of propagation of an important seismic wave of translation, since the earthquake giving rise to it occurs at a known hour, and the moment the wave reaches the opposite shore of the sea or ocean also can be precisely found by means of tide recorders. The mean intensity of gravity along the path of the wave is also known. A remarkable agreement is found between the results of direct soundings and those furnished in this manner, constituting a beautiful confirmation of the theory of the propagation of the waves.