1897.] on Recent Advances in Seismology. 331 



maximum for both hemispheres, is attributed by Dr. C. G. Knott * 

 to the fact that in winter we have large accumulations of snow and 

 steeper barometric gradients than in suiumer, and it is these inequa- 

 lities of stress of long continuance which cause yieldings to be more 

 frequent at one season rather than at another. 



The most important feature in the Japanese records, which gives 

 to them a value greater than those of any other country, is the fact 

 that the various shocks may be classified according to the district 

 from which they originated, and at the same time a value or 

 weight can be given to each, according to the area it disturbed, whilst 

 primary and secondary shocks can be readily separated from each 

 other. 



The advantage of such tables, when, for example, we seek for a 

 possible connection between certain lunar influences or the rising of 

 the tide upon a coast, because such influences are at a maximum in 

 different districts at different hours, is at once apparent, whilst all 

 surprise that investigators who have only had at their disposal tables 

 of earthquakes the origins of which have been in widely separated 

 districts have failed in establishing laws, which we might anticipate, 

 at once disappears. 



Thanks to the liberality and foresight of the Japanese Govern- 

 ment, we are now in a position to make investigations hitherto 

 impossible, and to confirm or disprove very many of the results of 

 previous investigators. Dr. Knott, who is engaged upon these volu- 

 minous statistics, finds a confirmation of the law of Perry that there 

 is a maximum in earthquake frequency near the time of perigee ; 

 that there are maxima associated with the moon's declination ; its con- 

 junction with the sun ; the time of the moon's meridian passage ; 

 and the ebb and flow of tides. Until these investigations have been 

 completed and published, their importance cannot be fairly estimated. 

 Dr. F. Omori has pointed out the existence of diurnal and semi- 

 diurnal periodicities, and that the frequency of after-shocks follows 

 fairly definite laws ; the former of which investigations has by rigid 

 treatment been emphasised and extended by Dr. C. Davison. 



Many investigations have been made to discover a relationship 

 between seismic phenomena and those of an electric or magnetic 

 character, but the only certain result is to show that the artificial or 

 actual shaking of the ground near to an earth plate may be accom- 

 panied by temporary currents, whilst the displacement of large bodies 

 of strata, as for example those which accompanied or caused the earth- 

 quake of 1891, may result, as pointed out by Prof. Tanakadate, in a 

 permanent readjustment in the relative position of the isomagnetics 

 in a district. 



After this earthquake, the cause of which was attributable to the 

 sudden fracturing of rocks, the line of v/hich is traceable on the 



* Trans. Seis. Soc. vol. iv. pt. 1, " Earthquake Frequency," C. G. Knott, 

 F.R.S.E. 



