R. D. Oldham — The Periodicity of Earthquakes. 449 



anything we know to the contrary the nearest rivei'-water may give 

 no higher figure ; the Mahanuddy, emptying itself into the Bay of 

 Bengal, has only a chlorine content of "17 per 100,000 after passing 

 over 440 miles of its course (Nicholson, Journ. Chem. Soc, 1873, 

 p. 229). 



IV. — The Periobioitt of Earthquakes. 

 By R. D. Oldham, Superintendent of the Geological Survey of India. 



MANY are the attempts that have been made to discern some law 

 in the occurrence of earthquakes, and to trace the influence 

 of the sun, the moon, or even of the planets as a cause, if partial, 

 of their origin. Many patient investigators have discovered, or 

 thought they have discovered, periods of fluctuating seismic activity, 

 varying in length from semi-diurnal to annual or even longer, 

 but so conflicting have been their conclusions that little weight 

 can be, or has been, attached to the results of their calculations ; 

 and one of the most industrious of all these investigators, the 

 Commandante de Montessus de Ballore, has declared his conviction 

 that no periodicity can be detected, and that the causes of earth- 

 quakes are pm-ely terrestrial and in no way affected by any celestial 

 body. Yet, in spite of this, the attempts and the calculations go 

 on, and one of the most recent of these is a discussion by Herr M. 

 Becke of some three hundred earthquakes recorded in the region 

 round Karlsbad between 24th October and 25th November, 1897.^ 



Tabulating these, according to the hour of occurrence, he finds 

 that there are two well-defined maxima at about three hours on 

 either side of midnight, while the minimum is at midday with 

 a minor one at midnight. He rightly observes that if this be due 

 to the influence of the sun a similar but much more marked relation 

 should be observed in the case of the moon, and after dividing 

 each lunar day into twenty-four hours and tabulating the earth- 

 quakes according to this lunar time he finds that the curve of 

 frequency so obtained does not correspond to that deduced from 

 solar times. From this he concludes that the apparent maxima 

 do not represent real maxima of occurrence of earthquakes, but 

 merely maxima of record, due to the fact that slight earthquakes, 

 which would be noticed and recorded in the morning and evening 

 hours of repose, would be overlooked in the active prosecution of 

 daily avocations, or in the slumber of the night. To this the obvious 

 objection may be raised, that persons who are asleep by midnight 

 are not likely to be awake at 3 o'clock in the morning. 



Apart from this, there seems to be a more serious flaw in the 

 argument. Herr Becke appears to have expected that the curve 

 of frequency by lunar time should show maxima at about three 

 hours on either side of the time of lower meridian passage of the 

 moon. This again depends on the assumption, tacitly made by 

 every calculator with whose work I am acquainted, that the 

 frequency may be expected to be a function of the hour angle. 



1 M. Becke, " Bericht iiber das Graslitzer Erdbeben, 24 October bis 25 November, 

 1897 " : Sitzber. k. Aktid. Wiss. Wien, 1898, cvii, Abth. 1, pp. 789-959. 



DECADE IV. VOL. VIII. — NO. X, 29 



