32 REPORTS ON THE STATE OP SCIENCE. 



II. Double and Multiple Earthquakes. 

 Attention has frequently been drawn to the fact that an earthquake 

 as it radiates may cause a collapse of strata which are in an unstable 

 condition and thus give rise to one or more secondary disturbances. 1 

 The great earthquake of Lisbon in 1755 gave rise to secondary shocks 

 in England and Ireland, and probably in many other countries. In the 

 volume containing Physical Observations made in the Antarctic Eegions 

 in 1902-03, published under the superintendence of the Royal Society, 

 page 92, I gave illustrations of secondary earthquakes the genesis of 

 which corresponded in time to the arrival of certain phases of pri- 

 mary disturbances. That the large waves of a seismic disturbance as 

 they travel round the world causing the crust of the same to rise and 

 fall like a raft on an ocean swell should give rise to one or more secon- 

 dary disturbances is not surprising. Further than this, any of the latter 

 which may be greater or less than their parent may in turn become the 

 originator of further settlements. One megaseism may therefore cause 

 a relief of seismic strain throughout the world. An indication of this is 

 seen in the fact that large earthquakes originating in widely separated 

 districts frequently occur in groups. This idea I wish to extend to the 

 possibility of secondary earthquakes originating in consequence of mass 

 displacement or ' push ' exerted in a hypofocal region, or on the 

 arrival of waves of the type P x and P : , the speeds of which are relatively 

 about four times and twice those of P 3 . This means that an earth- 

 quake originating at A might result in reliefs of strain in distant 

 localities B, 0, D, &c, on the arrival of Pi which radiated from A. 

 The seismograms obtained at stations near to B, C, D would on 

 account of the differences in the times of arrival of Pi at these places 

 coinciding with what we should expect, be attributed to the primary 

 impulse originating at A and not to impulses which had been brought 

 into existence in the neighbourhood of B, C, and D. Generally this 

 supposition is correct, but instances occur where it fails to explain 

 the amplitudes of movement and the times of arrival of P a or the 

 maximum movement recorded at these latter stations. It has been 

 shown that stations at great distances from the origin of a megaseism 

 may record movements which have travelled to them in opposite direc- 

 tions round the world. This might, at a distant station, give rise to 

 at least two maxima and a lengthening of the duration of motion. These 

 phenomena may also find a partial explanation in the hypothesis of 

 reflections within our world, or the echoes from mountain roots. 2 

 What I now suggest is that these unexplained characteristics of certain 

 seismograms may partly be the outcome of secondary disturbances the 

 existence and importance of which has hitherto been unrecognised. 



Example 1. — Guatemala Earthquake, April 19, 1902. 



On April 19, 1902, a violent earthquake took place in Guatemala. 

 It partly obliterated Quezaltenango, Amititlan, and bady damaged many 

 other places. It was accompanied by fires. If it could be shown that 

 the fires broke out before the earthquake certain fire insurance com- 

 panies were liable. On the other hand, if the fires occurred with or 



1 See Earthquakes, ' International Science Series,' p. 248. 



2 See Brit. Assoc. Reports, 1899, p. 227 ; 1900, p. 71 



