THE MESSINA EARTHQUAKE

MANY months, possibly several years, must elapse before the history of the Messina earthquake can be fully written. The official report on the Calabrian earthquake of 1905 is still unfinished. So also is that on the earthquake of 1907. And in the recent earthquake the area affected is so much larger than in the others, the phenomena were so much more varied and the destruction of many places is so complete, that before the final report can appear interest in the greatest disaster of modern times may well have begun to wane. In the present paper little can be attempted. No scientific investigation of the earthquake has yet been made. But from the maze of details furnished by the newspapers it is possible to prepare a connected account of the more important phenomena, and thence to trace the relations of the recent earthquake with its forerunners in the same district.

lines, which pass through places at which the intensity of the shock was the same, should not be regarded as accurately laid down. The careful surveys that will be made by the competent Italian seismologists will probably reveal slight discrepancies between those now and afterwards drawn; but, approximately at any rate, the lines indicate the manner in which the intensity of the shock varied throughout the central area. The three small curves, which are roughly circular in form, surround all those places which were entirely or almost entirely destroyed. The largest curve, which is twenty miles in diameter, contains Messina, Reggio di Calabria, Pellaro, &c. It will be noticed that its centre is submarine, and that it lies nearer to Reggio than to Messina. The other two curves are respectively eleven and ten miles in diameter, the former including Palmi, Bagnara, &c., and the latter Monteleone and the neighbouring villages. The Messina and Palmi curves are shown detached, but they should perhaps coalesce, forming a single dumbbell-shaped curve. In either case they appear to indicate the existence of two separate foci or two maxima of disturbance. The total area contained by the three curves is about 500 square miles. Outside them is one of oval form, which includes the places in which injury to buildings was considerable, though not complete. This curve is 100 miles long and fifty-six miles wide, and contains about 4500 square miles. The last curve, of which only small portions traverse the land, surrounds the places at which slight damage occurred and very few lives were lost.

The intensity of the shock, however, did not vary uniformly between these curves, and it is not at present clear to what cause or causes certain places owe their immunity. Taormina, for instance, lies but a few miles to the south of the Messina curve, yet only one important building within it seems to have been destroyed.

How far the earthquake was felt is not definitely known, but that the disturbed area was comparatively small is borne out by several accounts. At Palermo, for instance, which is only 125 miles from the Messina centre, a sharp shock was felt, but not much attention was paid to it— indeed, many of the inhabitants were not awakened. At Naples, 200 miles from the same point, the shock is described as slight, and it was also felt at Bari, distant 220 miles; but at greater distances it can hardly have attracted much notice. If we take 220 miles as the average radius of the disturbed region, the total area shaken, including that covered by the sea, would be about 150,000 square miles.

The significance of such figures-4500 square miles for the violently shaken area and 150,000 square miles for the disturbed area-will be evident when we compare them with the corresponding figures for other earthquakes. In the recent San Francisco earthquake the strongly shaken area contained about 40,000 and the disturbed area about one and a half million square miles. In the Assam earthquake of 1897 serious damage to brick buildings occurred within a district

including about 150,000 square miles, while the area shaken can hardly have fallen short of one and three-quarter million square miles, or nearly half the size of Europe. The Hereford earthquake of 1896, the greatest experienced in this country for many years, was in no place capable of throwing down more than a few chimneys, but was distinctly felt over an area of about 100,000 square miles. On the other hand, in the Neapolitan earthquake of 1857 the destruction of property was almost complete over an area of 950 square miles, while the disturbed area was estimated to contain less than 40,000 square miles. The Calabrian earthquake of 1894 was of destructive violence within a district measuring about 250 square miles, and disturbed an area of about 50,000 square miles. In the same category may also be placed the Jamaica earthquake of 1907, so disastrous to Kingston and the immediate neighbourhood, so little felt towards the limits of the island.

Thus we have two classes of earthquakes, or, rather, two limiting cases with many intermediate varieties. In both the ruin within the central area is almost complete; but at one end of the scale the intensity of the shock fades away slowly, and at the other end very rapidly, from the centre. The extent of the area of destructive violence in the San Francisco and Assam earthquakes is due partly to the immense magnitude of the seismic focus; but the difference in the rate at which the intensity declines outwards from the centre depends chiefly on the depth of the principal part of the focus-the slow decrease being connected with a deep-seated focus, and the rapid decrease with a shallow focus. It is, no doubt, to their comparatively superficial origin that the violence of the Calabrian earthquakes is to be traced. On the other hand, the form and small size of the ruined areas in the recent Messina earthquake indicate that the seismic foci were not of great linear dimensions; though, if the three foci corresponding to the Messina, Palmi, and Monteleone curves were all connected the complete focus may have been about fifty miles in length.

The destructiveness of the earthquake was due to several causes. The close neighbourhood of the origin to a populous town like Messina was one of the most important, this being also the case in the recent earthquakes of San Francisco, Valparaiso, and Kingston. The occurrence of the shock at about 5.20 a.m., when most people were indoors and asleep, was a factor of no less consequence. But even those who were awake were practically unable to escape owing to the sudden onset of the shock. Though a tremor seems to have been recorded ten minutes earlier in the underground observatory of Messina, there was not even the indefinite warning occasionally given by perceptible fore-shocks. To those who were awake the first symptom was a deep rumbling noise like a peal of thunder or the explosion of many bombs. After a few seconds at the most, this was followed by a rough jolting movement, the shock increasing so rapidly in strength that few

were able to quit their dwellings before the heavy stone floors and staircases, parting from the lofty outer walls, crashed down, followed in nearly every case by the tottering walls themselves, which encumbered the streets with débris. As to the exact nature of the shock within the ruined area, we can expect to know little from personal evidence; but, from the reports of several witnesses (including one on a steamer near Messina), there seem to have been two maxima of intensity, separated by a brief interval of lesser movement. The total duration of the shock is estimated by several observers at from thirty to thirty-two seconds-not a great duration for a destructive earthquake, but sufficient to lay in ruins every town and village within the central areas.

The shock was felt severely on several vessels that were close to Messina at the time. Ships quivered suddenly as though they had lost their screws, or run aground, or struck a piece of wreckage. Thousands of fishes must have died. After the sea-wave, the tongue of land protecting the harbour of Messina from the sea was found to be covered by the bodies of dead fishes, killed no doubt by the shock to which their whole bodies were simultaneously exposed.

As in the Lisbon earthquake, and in so many of the earthquakes on the coast of Japan and South America, the damage wrought by the shock itself was aggravated by the rush of the great sea-wave soon after the shock was over. Along the east coast of Sicily, at Giardini (near Taormina), Riposto, and Catania, the sea suddenly receded for fifty yards or more. It then returned, a great muddy wave, from ten to twelve feet in height, driving ships and boats from their moorings, sweeping over the low-lying coasts, and leaving them strewn with wreckage. The effects of the wave at Messina were at first exaggerated, but it flooded the lower parts of the town, washed the ruined lighthouses and neighbouring cottages into the sea, and damaged the fronts of the harbour-works.

On the opposite coast of Calabria, the sea-waves were higher and the results more serious. At Reggio the streets were flooded up to the first floor of some of the houses, and many ruined buildings along the coast (especially to the south of Reggio) were swept away. How far along the Calabrian coasts the sea-wave was noticed is still unknown. In Sicily it was observed along the north coast to Termini, and along the east coast to Syracuse. It must, however, have spread far beyond these limits, for at Malta, about three hours after the earthquake, a wave swept into the harbour, disturbed the smaller boats, and, rushing over the shore embankments, washed into the low-lying houses. When the records of the tide-gauges in the more distant harbours of Italy, in France, Spain, and elsewhere are examined, it will probably be found that the waves did not entirely vanish before reaching the farthest corners of the Mediterranean Sea.

The mere existence of the sea-waves points to a considerable dis

placement of the ocean-bed. Reports of any measurable variation in depth must, for the present, be received with caution, but a careful survey will probably reveal some change of level along a linear band, the position of which may be ascertained most readily by that of the fractures in the telegraph cables which connect Sicily with the mainland. Along both shores there seem to have been some changes of level, but it is not yet clear that they exceed those which would be caused by a sliding of the superficial beds. At Messina parts of the harbour works have sunk to the level of the water. At Reggio the quarter by the sea has disappeared, and large tracts of land to the south of the town are now lying under water. More trustworthy evidences of crustal movements on land are to be found in the twisted railwaylines and distorted surface features in the neighbourhood of Reggio.

That the Messina earthquake belongs to the class known as ' worldshaking' is evident from the distance of the stations at which it was recorded by seismographs, the farthest at present reported being Perth, in West Australia. At several Italian stations the oscillations were so extensive that they injured the instruments. At Laibach (about forty miles north-east of Trieste) eleven out of twelve seismographs were too sensitive to register the whole of the movement. Even at Kew, the range of motion exceeded the limits of the Milne seismograph. The first tremors were recorded at this observatory at 4 hours 23.6 minutes a.m., Greenwich mean time being an hour earlier than that of Italy; while the two largest oscillations occurred at 4 hours 31.1 minutes and 4 hours 32-7 minutes respectively. The four principal movements were also indicated by blurred interruptions of the curve of the declination magnet at almost the same instants as those of the seismographic record, thus showing that the disturbances of the magnet were of mechanical origin.

Soon after the earthquake was over, the usual train of after-shocks began, slighter far than the original shock, but strong enough to alarm the survivors and occasionally to bring down some of the shattered but still standing walls. No detailed record has or could have been kept of these shocks, but in the neighbourhood of Messina they seem to have been less numerous than is generally the case. We hear of several shocks having been felt during one night, but not of that almost incessant quivering of the ground which follows a great earthquake in the central district. Around Reggio they were more frequent. A week after the earthquake they occurred at intervals of about twenty minutes. One of the most interesting of the after-shocks was that felt in Stromboli (one of the Lipari Isles) on January 3. This lasted about half a minute, and was strong enough to damage many houses. The volcano, one of the few active vents in Europe, also sent out streams of lava, while prolonged subterranean explosions were heard. This is especially noteworthy, for at the time of the great earthquake Etna, Stromboli, and Vulcano remained quiescent.