46 



Dr. C. Davison on Earthquake- Sounds. 



The proportion of observers who hear the sound is generally 

 very great in British earthquakes, as will be seen from 

 Table III. With one exception the percentage is higher 

 for slight than for strong shocks. This of course is partly 

 due to the sound-area in the latter cases being somewhat less 

 than the disturbed area, but not to any great extent, for the 

 number of observers outside the sound-area is always small. 

 It is rather due, I think, to the sound being by far the most 

 prominent feature in a weak earthquake. 



Italy. — The organization for the study of earthquakes is 

 more complete in this country than in any other except 

 Japan. The results are published by the Central Office of 

 Meteorology and Geodynamics at Rome *. Owing to the 

 uncertainty in some of the time-records, it is not always 

 possible to group together the observations of the same 

 earthquake, and a small error may thus enter into the 

 estimates. In the two years 1895-1896 the total number of 

 earthquakes (excluding those of distant origin, and others 

 registered only by instruments) is 692, and of these 217, or 

 3T4 per cent., are recorded as accompanied by sound. In 

 different earthquakes the percentage of records in which 

 sound-phenomena are mentioned is very variable, ranging 

 from 18 for the earthquake of June 10, 1895, to 47 for that 

 of April 13, 1895. In the Verona earthquake of June 7, 

 1891, 34 per cent, of the observers heard the sound. Out of 

 274 after-shocks of this earthquake, only 25, or 9 per cent., 

 were attended by sound ; and of one of the strongest of these 

 shocks, that of Aug. 31, only five out of 101 observers were 

 recorded as auditors of the sound f. 



Admirable as are the observations collected by the Central 

 Office, it is important, from the present point of view, to 

 notice how large a number (more than 60 per cent.) consist 

 of records from single places, though there were often several 

 observers at each place mentioned. The next Table shows 

 the relation between the number of records of each earth- 

 quake and the audibility of the accompanying sound- 

 phenomena. 



This Table shows very clearly how the audibility of the 

 sound increases with the number of records of the earthquake. 

 It might of course be argued from this that the number of 

 records depends on the extent of the disturbed area, and this 

 again on the strength of the shock ; in other words, that the 

 sound is more frequently audible with strong, than with 



* In supplements to the Boll, della Soc. Sismol. Ital. 

 t M. Baratla, Annali dell 1 Uff. Centr. Meteor, e Geod. vol. xi. parte 3 

 (1892). 



