1902.] R. D. Oldham — Tidal Periodicity in Earthquakes of Assam, 153 



forces are sufficient to give rise to the tides, and that the difference 

 between the spring tides and the neaps is due to the forces whose 

 effect has been seai'ched for in this paper, it is quite conceivable that 

 they should not be without effect in determining the moment at which 

 a gradually increasing strain becomes too great for the resistance, and 

 the fracture is produced which gives rise to an earthquake. 



IV. — Conclusions. 



From what has gone before we may draw the following conclusions. 



1. That there was a very large variation in the diurnal distribu- 

 tion of earthquakes in Assam during the years 1897-1901, shocks being 

 most frequent between 10 and 11 p.m., and again between 6 and 7 a.m. 

 This greater frequency is a real one and not merely due to a larger 

 number of shocks happening to be recorded at those times. 



No satisfactory cause can be assigned for this irregularity of 

 distribution, which must for the present be accepted as a fact true for 

 a limited period and area. 



2. Superimposed on this large and unexplained variation in 

 frequency, there is a smaller variation which has the appearance of 

 being due to the tidal stresses set up by the attraction of the sun. 



3. If this smaller variation is really due to tidal stress, 

 then the horizontal stress is much more efficient than the vertical stress, 

 and the effect is less due to the amount of the stress than to the rate 

 and range of its variation. 



4. That these conclusions must be taken as purely provisional and 

 require verification from a more extended series of observations. For 

 their verification we require an instrumental record from some station 

 within or near the tropics, where earthquakes are fairly frequent, and 

 extending over 19 or 20 years. 



J. it. 20 



