July 1, 1895.] 



KNOWLEDGE 



161 



Diarii of a Jnurnefi thfonqh Mongolia and Tibet. Bv William 

 Woodville Rockhill. (SmithsoEiian Institution.) 



Dairif Bacterioloiitf By Dr. Ed. Vou Freudenreich. Translated 

 by J. E. Ainsnorth Davis, B A , &c. (Methuen & Co.) 29. 6d. 



The Ti.ne Machine. By H. G. Wells. (Heinpraann.) Is. 6d 



Prti-oin/i/ for Students'. By .Vlfved Harker, M.A., &c. "s. Od. 



A Texi-Book- of Zoorjeographii. By Frank E. Beddard, M.A., &e. 

 (Cambridge University Press.) 6s. 



Inde.res to the Literatures of Cerium and Lanthanum. By W. H. 

 Magee, Pli.D. (Smitbsonian Institution.) 



Memoirs of the British Astronomical Association ; Beport of the 

 Section for the Ohsereation of Mars. By E. Walter ilaunder, F.R.-V.S. 



ON THE CAUSE OF EARTHQUAKES. 



By Prof. J. LoG.vx Lobley, F.G.S., Ac. 



IN a recent number of Knowledge I adduced evidence 

 in support of the conclusion that the general 

 climatal conditions of the globe in the Cambrian 

 period were similar to those that now prevail on 

 the surface of the earth.'' But although this 

 evidence is so abundant and cogent that the conclusion is 

 inevitable and indisputable, yet its consequences are very 

 generally overlooked, and it is frequently altogether ignored 

 in the discussion of questions on which it has a direct 

 bearing. Notably has this been the case in the discussion 

 on the cause of earthquakes. 



A connection between the cause of earthquakes and that 

 of volcanoes is very generally assumed, and Mallet's 

 dictum, that an earthquake is but an uncompleted volcano, 

 is often quoted with tacit, if not expressed, acquiescence. 

 This would seem to imply that both the cause of volcanic 

 action and the cause of seismic action had been satis- 

 factorily determined, and yet this is far from being the 

 result of the long controversy on these two important 

 questions. 



Text-books usually give several theories to account for 

 volcanic action, and while one hypothesis is on the whole 

 favoured by one author, another receives the guarded 

 assent of a second. The cause, or causes, of seismic 

 phenomena are stated still more doubtfully, notwithstanding 

 the mass of facts obtained by the laborious and prolonged 

 investigations of Mallet, and the very valuable and more 

 recent work of Prof. Milne in the seismic land of .Tapan. 



There seems, however, to be a very prevalent opinion 

 that a shrinkage of the so-called " earth's crust," conse- 

 quent upon the secular cooling of the globe, is the primary 

 cause of both earthquake and volcanic phenomena. Mallet 

 not only attributed local earth-movements generally to 

 the consequences of a gradual cooling of the globe, but 

 derives volcanic heat also from the tangential pressure of 

 the rocks of the crust by contraction following planetary 

 cooling. What is precisely meant by the " earth's crust," 

 and what the amount and rate of the cooling assumed, are 

 not stated, and so the whole matter is left in a very vague 

 and unsatisfactory position. 



To the same shrinkage from planetary cooling is also 

 ascribed the folding and contortion of rocks of all kinds 

 and all ages, except those associated with intrusive igneous 

 rocks, as well as the elevation of mountain chains and the 

 vertical uprise and subsidence of areas both large and 

 small. So repeatedly and so confidently is contraction 

 from cooling stated as being the cause of such earth- 

 movements, that it is generally accepted without question. 

 Indeed, a shrinkage of the bulk of the earth is commonly 

 regarded as required to account for the foldings and con- 

 tortions of the surface rocks, and thus prove the point. 



*"0n tbe Climate of tlie Cambrian Period," Kxowi.edoe for 

 November, 1894, p. 260. 



More than fifty years ago. Sir Henry de la Beche 

 wrote: "If we adopt the theory of a cooling globe, and 

 the necessity of the solidified crust of one period, with its 

 covering of sedimentary deposits, conforming to the reduced 

 size of the earth at another, this solid crust, with its 

 detrital covering, would be broken up, or wrinkled, or both, 

 to conform to the new adjustment of parts." So confident 

 did this truly great geologist appear to be that the theory 

 of a cooling globe, with consequent shrinkage, was sound, 

 that he did not think it necessary to state or suggest any 

 other. 



And although a thin hard crust with a great central 

 fused mass has since been shown, by Lord Kelvin and 

 other physicists and astronomers, to be incompatible with 

 the proved rigidity of the globe, a settling down and 

 accommodation of the crust to a shrunken central mass is 

 still most confidently assumed. In a recent important 

 work, the emission of lava is ascribed to its exudation from 

 a central fused mass consequent upon the pressure of an 

 exterior hard crust, t and in a still more recent text-book, 

 earthquakes are attributed to, with other causes, " the snap 

 of rocks that can no longer resist the strain to which, by 

 the cooling and consequent contraction of the inner hot 

 nucleus, they have been subjected within the earth's 

 crust."! 



If, however, the general temperature at the surface of 

 the globe was in Cambrian times similar to that of the 

 present day, there can have been no appreciable amount of 

 planetary cooling during the intervening period, and con- 

 sequently no appreciable amount of contraction of the bulk 

 of the globe, notwithstanding the enormous duration of 

 the time that has elapsed since the Cambrian epoch. If, 

 furthermore, there has been no appreciable contraction 

 during this vast period of time, there cannot have been 

 any contraction in a small unit of time, say a century 

 to cause dislocation of surface rocks. But there is not 

 merely an earthquake once a century, but without any 

 exaggeration it may be said that, in one part of the world 

 or another, there is at least one every week. 



The report of the British Association on earthquakes 

 (1851 to 1858) contains a catalogue of recorded earthquakes 

 from B.C. 160(i to a.d. 1842, which, with the catalogue of 

 Prof. Perry, of Dijon, from 1842 to the year 1850, gave 

 between 6000 and 7000 earthquakes as having been recorded 

 in 3456 years. But the following digest will clearly show 

 that only in very recent times do the records of earthquakes 

 at all approximately correspond with the number of 



occurrences : — 



Annual Ratio. • 



From B.C. 2000 to b.c. 1000 0004 



„ B.C. 1001 ,, Christian era ... 0054 



„ A.D. 1 „ A.D. 1000 0-222 



„ A.D. 1001 ,, A.D. 1850 7 740 



„ A.D. 1551 „ -K.o. 1850 17-370 



„ A.D. 1701 „ A.D. 1850 35-310 



When it is borne in mind that a great portion of the 

 surface of the earth, to take only the land surface which is 

 merely one-fourth of the whole, is sparsely peopled and 

 without observers to record natural phenomena, it will 

 be readily admitted that it is quite safe to conclude that 

 the annual number of earthquakes between 1701 and 1850 

 as stated above is much below the fact. An earthquake a 

 week may, therefore, be confidently assumed. Indeed, 

 exactly double this number (104) were actually recorded 

 by Prof. Fuchs as having occurred in 1876. In 

 addition, however, to those violent disturbances that are 



-j- " Geologv," bv Prof. Prestwich, p. 216. 

 : Geikie's " Class-book of Geology " (1883), p. 110, 



