226 



DR. CHARLES DAVISON OX THE 



[May 1904, 



distinct sounds, accompanying each part of the shock, and sepa- 

 rated by a brief interval of rest and quiet. Few observers, 

 however, noted the relative intensity of both parts of the sound and 

 shock. The first part of both sound and shock was the more intense 

 according to 7 observers, the second part according to 9, while 

 1 considered them to be approximately equal in intensity. Again, 

 8 observers state that the first part of the shock was the stronger 

 and that no sound was heard with the second part ; and 1 that the 

 second part of the shock was the stronger, while no sound accom- 

 panied the first. Thus, all observers agree in connecting the louder 

 part of the sound with the stronger part of the shock. 



Time-Eelations of the Sound and Shock. 



In the following table, the letters p, c, and /indicate the number 

 of records per cent, in which the beginning or end of the sound 

 preceded, coincided with, or followed, the corresponding epoch of 

 the shock ; the letters g, e, and I indicate the number of records 

 percent, in which the duration of the sound was greater than, equal 

 to, or less than, that of the shock. 1 The last line of the table 

 contains the average percentages for four strong earthquakes, 

 namely, the Pembroke earthquakes of 1S92 and 1893, the Hereford 

 earthquake of 1896, and the Inverness earthquake of 1901. 



Beginning. 



P 



End. 



/ 



Within isoseismal 7 03 34 



Between isoseisntals 7 and G ."3 40 



,, „ and 5 ... 55 37 



5 and the 



boundary of the sound-area H ' 



Whole sound-area 57 37 



Average for strong earthquakes. 76 15 



20 



17 

 23 



24 



21 



19 



53 



52 



G2 



1 



27 



31 

 15 



6 



55 \ 24 



28 



56 



Relative 

 Duration. 



9 



59 



48 

 39 



31 



46 



I 



37 



45 

 49 



69 



46 

 21 



7 

 12 



A comparison of the last two lines of the table shows that, in the 

 Derby earthquake, there was a closer approach than usual to coin- 

 cidence in both terminal epochs and therefore to equality in duration. 

 Moreover, this tendency was almost as marked in the central region 

 as in the outer zones, from which we may infer : (1) that the sound- 

 waves travelled with the same, or very nearly the same, velocity as 

 those of larger amplitude and longer period, and (2) that the mar- 

 ginal regions of the foci were of comparatively-small dimensions in 

 a horizontal direction. 



1 The number of records from the outer zone (that between the isoseismal 5 

 and the boundary of the sound-area) is muoh less than from the others, and 

 the corresponding percentages are therefore of inferior value. 



