48 



Dr. C. Davison on Earthquake-Sounds. 



varies considerably in different districts, and has endeavoured 

 to trace the connexion between the frequency of sound- 

 accompanied earthquakes and the geological structure of the 

 district. I shall refer to these conclusions in the next 

 section. 



It is clear, however, that the number of earthquakes ac- 

 companied by sound must be a function of the total number 

 of earthquakes, and, therefore, that a better method would be 

 to study the variation in the ratio of the former number to 

 the latter. I have tried to do this for the whole of Japan, 

 but, great as is the number of recorded shocks, it is not 

 sufficient for the purpose ; though a map of a smaller district, 

 in which earthquakes were exceedingly numerous, is given 

 in fig. 2. Taking only those rectangles which contain not 

 less than 20 epicentres during the years 1885-1892, the per- 

 centage of the total number of earthquakes that were accom- 

 panied by sound may be as high as 77, and as low as zero. 

 In two rectangles, chiefly affected by the after-shocks of the 

 Kumamoto earthquake of 1889, the percentage is only 7. 



A point of great importance is the relation between the 

 audibility of an earthquake and its disturbed area. This is 

 shown in Table V. 



Table V. 





Disturbed Area in square miles. 



<100 



100- 



500 



500- 

 1000 



1000- 

 2000 



2000- 



5000 



5000- 

 10,000 



> 10,000 



Number of earthquakes . 



Number of earthquakes 

 accompanied by sound 



Percentage of earthquakes 

 accompanied by sound 



1848 

 223 

 12-1 



2169 

 556 

 25-6 



672 

 163 

 24-3 



403 

 160 



39-7 



308 195 

 158 132 



51-3 67-7 



1 



218 

 153 

 70-2 



If we omit the years 1890-1892, on account of the large 

 numbers of aftershocks of the Kumamoto and Mino-Owari 

 earthquakes that were unaccompanied by sound, the corre- 

 sponding percentages are 



25-9, 24-4, 41-7, 45*5, 62-9, 69*8, and 59*7. 



Thus we see that, with some comparatively unimportant 

 exceptions, the audibility of an earthquake increases with its 

 disturbed area, that is, roughly, with its intensity. But the 



