Dr. C. Davison on Earthquake- Sounds, 55 



While the sound-area may thus attain such great dimen- 

 sions in European and American earthquakes, the case is very 

 different in those felt in Japan. When Prof. Milne remarks 

 that u earthquake sounds only travel a few miles from their 

 origin " *, or "the very fact that the sound . . . was heard 

 indicates . . . that those who heard the noise were not many 

 miles removed from the origin " f, he is referring to his 

 experience of Japanese earthquakes in the first instance, and 

 probably relying on it in the second. With the detailed 

 studies on which this conclusion is founded I am not ac- 

 quainted, except in one case, that of the earthquake of February 

 22, 1880, in which the shock was sensible for a distance of 

 120 miles, while sound-records come as a rule from places 

 not more than 11 miles from the epicentre {. More decisive 

 evidence is, however, afforded by the percentage of audibility 

 of the Japanese submarine earthquakes. In the earthquakes 

 which originated beneath land, during the years 1885-1892, 

 the percentage of those accompanied by sound is, as we have 

 seen, 26*5. For the submarine shocks for the same period 

 the percentage is only 84, and none of the earthquakes 

 studied originated at a greater distance than 40 or 50 miles 

 from the shore, while the epicentres of 93 per cent, of the 

 total number were not more than 10 miles distant. 



The comparatively poor powers of the Japanese people for 

 hearing deep sounds are of course responsible for the small 

 dimensions of the sound-areas. The limited sound-area of 

 the Neapolitan earthquake is perhaps due to want of observa- 

 tions, Mallet's inquiries having been confined principally to 

 the epicentral region. In the Charleston earthquake, we 

 have a large sound-area with an apparently low percentage of 

 audibility ; but the smallness of the latter quantity is probably 

 owing to the imperfection of the records, for the percentage 

 in the central State of South Carolina is less than in six more 

 distant States. For the more recent British earthquakes 

 observations are numerous, and here we have a high per- 

 centage of audibility occurring with comparatively large 

 sound-areas. 



(3) In a few earthquakes the disturbed area and sound- 

 area practically coincide, the boundaries of both being deter- 

 mined by the same places of observation. The shocks are 

 generally of moderate strength, the disturbed areas ranging 

 from about 400 to 2000 square miles, and in one case, that 

 of the Inverness earthquake of 1890, to 4340 square miles. 



* Japan Seisinol. Journ. vol. iv. 1895, p. xviii. 



t ' Daily Telegraph,' Dec. 18, 1896. 



| Japan Seismol. Soc. Trans, vol. i. pt. 2, 1880, pp. 53, 56-57, 105. 



