164 J. HOPKINSON 



heat, and it is so converted when it meets with a certain amount 

 of resistance. It is known that electric currents are constantly 

 traversing the earth at no great distance from the surface, and 

 it does not seem imlikely that such ciurents may meet wdth 

 resistance and so may generate heat having an expansive force 

 sufficient to cause at least a tremor on the surface. In fact, it 

 seems to be within the boimds of possibility that the meeting of 

 currents of positive and negative electricity may cause a sub- 

 terranean thunderstorm. The frequent occurrence during 

 earthquakes of electric storms of various kinds, such as 

 thunderstorms, auroral displays, "flashes as of lightning," and 

 the appearance of " a sea of fire," tend to show such a reaction 

 between the heavens and the earth as may warrant the sup- 

 position of a common origin for the disturbed state of the 

 atmosphere, especially acute at St. Albans, and the slight 

 earth tremors with accompanying noise at Barnet and Hadley, 

 on the 15th of March, 1905. 



Earthquakes have occurred in Hertfordshire, it will be seen 

 from the preceding record, in every month from September to 

 May, but not in any of the three summer months, June to 

 August. This fact is not without significance, for whether we 

 take the British Isles only, or the whole of the northern 

 hemisphere, it will be foimd that they are considerably more 

 numerous in winter than in summer, and those which occur in 

 winter are usually more severe than those which occur in 

 summer ; in other words there is in the colder season greater 

 seismic activity both as to frequency and force. David Milne 

 gave a table (op. cit., p. 53) in which he showed that of the 255 

 British earthquakes of known dates recorded up to October, 

 1839, there occurred in the first quarter of the year 74, in the 

 second 44, in the third 58, and in the fourth 79. He called 

 these periods winter, spring, smnmer, and autumn. Calculating 

 from liis table the distribution in each season of the year, 

 commencing the spring with March and ending the winter with 

 February, the usual meteorological division of the seasons, it 

 appears that there were 53, or 21 °/o, in the spring ; 44, or 17 °/o, 

 in the summer ; 84, or 33 °/o, in the autumn ; and 74, or 29 %, 

 in the winter. Robert Mallet, in his foui'th and last Report to 

 the British Association " On the Facts and Theory of Earth- 

 quake Phenomena," gave a table showing the monthly dis- 

 tribution of aU earthquakes of known dates up to the end of the 

 year 1850, from which it appears that there were 5879 in the 

 northern hemisphere thus distributed over the seasons of the 

 year : spring, 1430, or 24|^ °/q ; summer, 1331, or 23 % ! autumn, 

 1452, or 24^ °/q ; and winter, 1666, or 28 °l^ ; summer thus 

 having much the smallest number, and winter much the largest. 

 So few earthquakes have been recorded in the southern hemi- 

 sphere that their inclusion would not make any appreciable 

 difference to the seasonal distribution, but it is evident that 

 they ought not to be included, as the seasons there are reversed. 



