98 



REPORTS ON THE STATE OF SCIENCE. 



The numerals indicate the number of large earthquakes which occurred 

 in successive periods of 36-5 days. 



Two periods connected by brackets indicate times when the change 

 in direction of pole-movement, as shown on Albrecht's figure, was com- 

 paratively rapid (see fig. 1). 



If we compare the number of earthquakes which occurred in these 

 selected periods with the total number of earthquakes which were 

 recorded in equal intervals of time before and after the deflection periods 

 we obtain the following results : — 



Com- 

 parable 

 Totals 

 Earthquakes before deflection: 7 8 12 8 18 — 8 12 5 6 — = 60 

 during „ 9 15 17 11 14 5 16 9 10 10 11 = 87 



after „ 8 12 8 15 — 11 12 5 4 — 8 = 64 



Omitting the four cases where figures for comparison are not com- 

 plete, out of seven deflections there are six instances where the greater 

 number of earthquakes have taken place during deflection periods. 



If this result be added to that for the years 1892 to 1899 ' we see 

 that out of twenty-three deflection periods there are eighteen instances 

 where the greater number of earthquakes have been recorded for the 

 deflection period. The totals for before, during, and after comparable 

 deflection periods are respectively 167, 287, and 217, or as 1 : 1-72 : 1-29, 



A closer determination of the possible relationship between pole- 

 deflections and earthquake-frequency is obtained if the deflections between 

 successive periods are expressed in angular measure. Such measurements 

 have been made for each of the ten periods in the years 1899 to 1904. 

 Opposite to each measurement figures give the number of earthquakes 

 which occurred during the seventy-three days to which a measurement 

 refers. 



These have been divided into two groups. The first group embraces 

 observations made in the years 1892 to 1899. During this interval the 

 pole-path shows many irregularities.^ 



The second group refers to the periods between 1901 (-6) and 1905(-0). 

 During this interval the pole-path was comparatively regular (see fig. 1), 



Group I. 



' See Brit. Assoc. Rep., 1893, p. 80. 

 f Ibid., 1903, p. 80, fig. 2. 



