TRANSACTIONS OF SECTION A. 397 
9. The Distribution of Large Earthquakes in Space and Time. 
By Rev. H. V. Guu, S.J. 
In this paper the author adduced further evidence in favour of the theory he 
proposed some years ago: that under certain conditions earthquakes taking place 
at a given locality may occasion seismic and other disturbances at other places 
symmetrically situated round the earth’s circumference. This theory was 
accepted by the late Dr. Milne. The present paper was based on the analysis of 
almost a thousand earthquakes which Milne had tabulated with regard to their 
occurrence in time and space, and which was only completed last year. The 
writer deduced from this catalogue the interesting fact that nearly 20 per cent. 
of lange earthquakes takes place in groups of two or more within a few days of 
each other, occurring at distant places symmetrically situated. He explained 
why, from the principle on which he bases this theory, the number of such groups 
is not greater. He finds that these earthquakes may be divided into three general 
categories :— 
(1) Groups in which thedifferent earthquakes take place within a short interval 
of each other in distant places symmetrically situated (18°6 per cent.)*; (2) groups 
in which the different disturbances take place within a short interval of each 
other in or near the same locality (57:1 per cent.) ; (3) individual earthquakes 
having no connection in time or space with other disturbances (24°3 per cent.). 
It was pointed out that these three classes of disturbance are to be expected as 
the result of the general strain set up in the earth, treated as a rotating body, 
owing to the mass-displacements which tend to cause a deflection of the earth’s 
axis. The deflection itself may be small, but the effect produced may easily be 
sufficiently great to occasion a seismic disturbance which, owing to other causes, 
is on the point of manifesting itself. 
The accompanying table represents the result of the analysis of the earth- 
quakes marked on the British Association chart. 
The first column indicates the year. The second gives the total number of 
earthquakes which were sufficiently great to be considered world-shaking, and 
which were registered at least over a hemisphere. Column I. gives the number 
of shocks which were members of groups of two or more occurring at distant 
places symmetrically located. Column II. gives the number of those which were 
members of groups occurring at or near the same place. Column ITT. gives the 
single disturbances unconnected with others, either as regards time or space :— 
| Is Il. Iil. 
. ‘ liffer eau in s 
Year. | Total No. Gra aera S Dae ac a Single shocks. 
Aa: Per cent. ib Benuanin ; Per cent. 
1899 69 13 18:8 39 56 17 24-6 
1900 45 12 26-6 20: 44-4 13 29-0 
1901 56 8 14-4 36 64-2 12 21-4 
1902 67 18 26:8 42 63-2 7 10 
1903 81 21 25-9 40 49-4 20 24-7 | 
1904 67 10 15 37 55 20 30 | 
1905 79 9 11-4 49 62 21 26-6 | 
1906 120 23 19-2 64 53:3 33 27:5 
1907 88 14 16 53 60 21 24 
1908 89 12 13 | 51 57-4 26 29-2 
1909 128 23 18 | 80 62:5 25 19-5 
Total 889 163. aap erga Ae 
Mean 
perann. 80-81 146 18-6 46-4 57:09 19-54 24-2 
1 These numbers give the mean values for the whole period of eleven years. 
Tt is important to note that there is, on the whole, a very good agreement between 
the mean value and the values for the individual years, indicating that the dis- 
tribution of earthquakes in time and space is according to some such law. 
————- 
