« AnteriorContinua »
moon than at half-moon, and when she is near the earth than when she is far off. They are also more frequent at the equinoxes than at the solstices. In these cases the difference is, however, slight, but the result is of importance as confirming the generally accepted view that the sun and moon produce in the solid crust of the earth a small tidelike effect. Other connections have also been detected, such as that with temperature, magnetic phenomena, the aurora, barometric variations, seasons of the year and times of the day and night, all of which have been shown to be more or less marked. However, the one which is of special interest for us in the present consideration is the close, though not very obvious connection which has been shown to exist between certain motions of the earth’s poles and the frequency of earthquakes.
In many writings on this subject the earth is treated as if it were a stationary globe in space, and as if the only causes which ought to be considered were purely local ones. Such a manner of approaching the question would appear to be, at the very outset, destined to failure, for it leaves out of account one of the most real and no doubt very important elements of the problem—the rotation of the earth round its axis. When one brings to mind the fact that, owing to this cause the surface of our globe is moving, near the equator, with a velocity of about eighteen miles a minute, it is evident that the omission of such a factor should a priori be expected to prevent a full solution of 80 complex a problem.
Since the regularity of a rotating body depends on the even and permanent distribution of its mass about the axis of revolution, it is clear that any sudden or considerable modification of the distribution of this mass must necessarily affect, to a greater or lesser degree, the position of the earth's poles; and similarly any sudden movement of the poles from their normal path must produce a strain throughout the whole mass of the rotating body, and, if continued for a sufficiently long period, cause some disturbance in the material composing it, if there be any portion capable of being displaced. In order that such an effect might be produced, it is not necessary to postulate a fluid interior for our globe, but that at certain localities there are masses of movable matter of some sort, which will yield under the application of great forces.
* Referring to the interior condition of the earth, Lord Kelvin wrote, many years ago, the following passage, which he has never modified : ‘In the honeycombed solid and liquid mass thus formed, there must be a continual tendency for the liquid, in consequence of its less specific gravity, to work its way up; whether by masses of solid falling from the roofs of vesicles or tunnels, and causing earthquake shocks, or by the roof breaking quite through when very thin, so as to cause two such hollows to unite, or the liquid of any of them to flow out freely over the outer surface of the earth ; or by the gradual subsidence of the solid, owing to the thermo-cynamic melting which portions of it, under intense stress, must experience, according to the views recently published by Professor James Thomson (1861). The results which Vol. LXIII-No. 371
These movements may be illustrated by the uneven running of a machine when the fly-wheel is not well balanced, or even better by the erroneous path of a 'bowl,' due to the bias. Some interesting calculations have been made of the amount of pole displacement which could be produced by modifications in the present distribution of sea and land on the earth's surface, and Sir G. H. Darwin finds that the pole might be moved through several degrees. Lord Kelvin calculated that an elevation through 600 feet, of a portion of the earth's surface having an area of 1000 square miles, and ten feet thick, would alter the position of the earth's axis by 0.3" (about thirty-four feet). At first sight these may seem small quantities in comparison with other astronomical numbers, but the mass of the earth which is shaken is very great, so that the amount of energy involved is enormous. We shall see presently that in the case of large earthquakes very much greater masses may be in question.
Some twenty years ago Professor Milne called attention to the relationship that appeared to exist between the frequency of earthquakes and irregular movements of the poles. He has given much attention to the examination of these phenomena, and, as the result of a very careful and exhaustive investigation, has arrived at the conclusion that the years of greatest pole movements are also years of maximum earthquake frequency, and conversely, that great seismic activity seems to be followed by more marked pole displacements. Sir G. H. Darwin suggested that earthquakes tend to adjust the figure of the earth to one of equilibrium about its instantaneous axis. This result is not surprising, for the amount of material put suddenly into motion during a 'world-shaking ' earthquake is often very great. In a paper read at the recent meeting of the British Association at York, Professor Milne gave some exceedingly interesting figures as to the probable quantity of matter thus disturbed. From observations made after the great earthquake of Assam it has been calculated that it may have resulted from the collapse of 500,000 square miles, while from other considerations it would appear to be proved that these disturbances often reach a depth of from twenty to thirty miles.
The inference to be drawn from these figures is that the mass displacement for a very large earthquake may as a maximum reach ten million cubic miles (500,000 20 miles). This material may be moved vertically through a range of ten feet. Even if we regard this estimate as twenty times greater than those corresponding to average megaseismic efforts, it will be admitted that the molar displacements here are of very great magnitude. Further, as they occur frequently, we have to look for cumulative effects which, under certain conditions, may possibly be appreciable.
must follow from this tendency seem sufficiently great and various to account for all that we see at present, and all that we learn from geological investigation of earth. quakes, of upheavals, and subsidences of solid, and of eruptions of melted rock.' (Thomson and Tate, vol. i. pt. 2, p. 484.)
may be interesting to note that as the volume of the earth is estimated to be 260,000,000,000 cubic miles, it appears from the figures just given that the large fraction of yooo of the earth's volume is capable of being disturbed by the forces which give rise to earthquakes. Hence it is not strange that there should be a notable connection between earthquakes and pole displacements. A few years ago Dr. R. von Kovesligethy of Budapest calculated that during the years 1895–1902 each of the 200 world-shaking earthquakes recorded caused an average displacement of the pole through 0·00275".
therefore look on it as an established fact that great earthquakes cause a more or less considerable 'wobble on the earth's rotation.
Now, reasoning from the two facts just referred to, that a sudden disturbance of a portion of the mass of the earth may cause a displacement of the poles, and that conversely a movement of the poles may result in an earthquake, it appears logical to conclude that an earthquake in one locality might, under favourable circumstances, give rise to a corresponding earthquake shock or volcanic eruption elsewhere. In the first place, as has been shown, the effect of earthquakes on the earth’s axis is a measurable quantity; that a motion of the axis may affect movable matter is proved by the rinsing of a cup. The principles here referred to have been explained by the present writer and illustrated by means of some simple experiments with a little top or tee-to-tum. This was a disc-shaped hollow brass top, about an inch in diameter and ß inch deep; it was turned with great care, and could be spun by the fingers on a smooth surface. When spun on a piece of smoked glass it left a fine clear-cut trace, and spun perfectly truly. A small smooth steel ball was now dropped in; at once the top began to wobble, and now left a trace like a series of small e's in handwriting, and the ball could be heard rolling along the inner edge of the top. A second ball was added, of the same size : for an instant there was heard a slight movement of the balls, then they became quiet, and the top was spinning as evenly and as truly as it did before the addition of the first ball. The balls had arranged themselves automatically at the opposite ends of a common diameter. When a third ball was added to the two already in the top, there was, for a moment, a movement of the balls, and then all became quiet again; and the three balls were found to have placed themselves at equal intervals round the circumference of the tee-to-tum. The experiment could be varied by attaching a fixed weight to a point on the edge of the top, and dropping in a ball of the proper size, in which case the balance was restored as before. Instead of balls, water or some other liquid was sometimes used with satisfactory results. These experiments are all illustrations of the well-known laws of ' centrifugal force,' and need no further explanation. The general · Proc. Royal Dublin Soc. xi. no. 11, 1906.
theorem illustrated by them may be stated in the following words : ' A rotating body which contains matter capable of shifting its position tends automatically to restore itself to its original state of rotation, when any cause has disturbed its motion of revolution.' So much is this the case that it was evident from the experiments that a body containing loose matter free to move spins much more evenly than a rigid one, for the liquid &c. inside it makes up for slight departures from perfect balance in its construction.
The following simple means may be used to test and show the results just explained. A little paraffin wax is placed in the tee-totum, along with the two, three, or more balls as the case may be. The whole is then warmed until the wax melts. A light cover is screwed on, and the tee-to-tum is set spinning. The wax has time to become hard before the motion ceases. The balls are found to have automatically taken up the positions indicated by theory. Various modifications of the experiment may be made with this simple apparatus illustrating other phenomena, such, for example, as the relation between the distribution of land and water on the earth's surface.
From the considerations we have already dealt with it is clear that the principles thus illustrated may be applied to our globe. The sudden disturbance of a large tract of the earth's surface corresponds to the addition of a ball to the hollow top, and the same causes that tend to send the other ball to the opposite side would, in the case of the earth, cause disturbances in other places symmetrically placed with regard to the original earthquake; these remarks apply particularly to places near the equator. It must be remembered that the case of the earth is not as simple as that of the top, since gravitation must be taken into account, and also for the fact that the total mass remains constant in the same circumstances. However, mathematical investigations show that a subsidence for example at one place will tend to produce a counter disturbance in a corresponding locality. The simplest case would be that of two disturbances at opposite ends of a diameter following closely on each other. A more unusual occur. rence would be three disturbances in places equally distant. These views were first suggested with reference to the latter phenomenon.
The month of April 1906 was remarkable for three great disturbances. A serious eruption of Vesuvius took place on the 8th; on the 14th a very severe earthquake occurred in Formosa ; while on the 18th San Francisco was destroyed. These three places are not only on almost the same parallel of latitude, but are as nearly as possible equally distant. These circumstances, together with the fact that they all took place within a few days of each other, suggested the possibility of there existing, among them, some such connection as that already pointed out. In a word, the position of San Francisco is such that, in this view, an earthquake in that quarter of the globe
was an event to be anticipated with some degree of probability. That such an event could be foretold is not suggested, but, as we shall see presently, a probability of a disturbance in that part of North America could have been scientifically indicated.
In the meeting of the British Association already mentioned, Professor Milne dealt at some length with the theory set forth above, and stated the result of an examination of the Shide (I. of W.) records made by him to test the evidence afforded by past great earthquakes in favour of that view. His researches seem to indicate a very remarkable agreement between facts and theory. His conclusion had best be given in his own words :
To test whether the members of the groups exhibit some symmetrical distribution in space corresponding to that proposed, the earthquakes which have originated in districts separated from each other by 180 degrees in longitude, but on the same latitude, have been compared with each other. . . . In 1899 and 1905, which are years when the geographical distribution of origins exhibited marked differences, 126 earthquakes were recorded. Twenty of these appeared in ten pairs, each member of a pair being in symmetrically located districts. ... The average interval between the occurrence of earthquakes such as are here considered has, during the last six and a half years, been seventy-two hours, and nearly all have originated from the ten districts. One inference from this is that the distribution in time and space of the above ten pairs may not be anything more than chance. Whether this is to be accepted as generally true remains to be determined by a more complete and extensive analysis of registers. Not only should large earthquakes be compared with their kind, but also with small earthquakes and volcanic eruptions.
A few cases of triplets could also be pointed out. From these facts it follows that in the districts mentioned there would have been some foundation for the belief that an earthquake in the corresponding district might occur ; arguing merely from the figures just given the chances against such an event were about ten to one. No doubt if smaller shocks had been examined as well as volcanic eruptions many more instances would be found. It is interesting to note that the earthquake at Valparaiso on the 17th of August in the same year 1906 was followed a few days later by one in North-west Australia ; and a shock in the Dutch West Indies on the 27th of September by one at Calcutta on the 29th. These places are exactly opposite each other, and on the same parallel. The districts examined by Professor Milne include the chief earthquake centres; they are as follows: (1) West of Alaska ; (2) Central America and West Indies ; (3) West Coast of South America ; (4) Central North America, with the corresponding
* The calamity which saddened England in the January of last year, when Jamaica was desolated, is too fresh in our memories to need mention. But England was not alone in her sorrow, for Holland too had to bewail a similar disaster in her fertile possessions in the Dutch East Indies. The fact that these calamities took place within a week of each other, that the places are situated almost diametrically opposite each other, and that they are on almost the same parallel of latitude is significant in *he light of the views here set forth. Other cases might also bo cited.