ox THE Ai'TEK-SHOCKS OF EARTHQUAKES. 



115 



are put in the same horizontal rows, may not he strictly €<3mparable 

 ^vith each other, yet the ratios in the columns (l\) and (\) are seen 

 to be in each nearly constant. This Avill show that the law of decrease 

 with time in the fre(|uency or activity of the after-sli(^cks was nearly 

 the same, at least in the days immediately succeeding the initial 

 earth<|uake, in each of the three cases. 



The dependence of the frequency ov activity on the magnitude 

 of the initial disturbance is uKjre clearly shown by the following table 

 of total numbers and activities of after-shocks during the first 30 

 days and the first 2 years : — 



Xow the tot'dl (land and .<ea) areas of disturliance of the Mino- 

 Owari. Kumamoto, and lvag<3shi]na earthquakes were respectively about 

 54000, (i500, and 5000 square /•/. The magnitudes of these earth- 

 quakes, therefore, if represented by their areas, are in the ratios of 

 ll:l.o:l approximately; i.e.. the Mino-Owai'i eartli([uake was 

 greater 1)v eight or nine times than the Kumamoto eartlKpiake, and 

 by eleven times tlian tlie Kagoshin^a earthquake. The frequency or 

 activity ïov the first earth((uake is .seen from the above tal)le to l)e 

 greater by fi)ur or five times than that for the second, and greater by 

 six times than that fov the third. If we compare together the two 

 earthquakes of Kumamoto and Kago.shima, we find that the ratio of 

 the frequencies or activities of tlieir after-.shocks for the first 30 days 

 is 1.3, which is equal to the ratio of their areas of disturbance. 



Thus the proportionality of the frec|uency or activity to the area 



