1902.] R. D. Oldham — Tidal Periodicity in Earthquakes of Assam. 145 



value in the table approximates to 3 hours the greater is that rate of 

 change, and the closer it lies to h. or 12 h. the less is the rate of 

 variation of the tide- producing forces. 



The passage of the circles of maximum vertical force is not subject 

 to the same changes as that of the other circles, and never varies more 

 than 1 h. 22 m. from six o'clock ; the effect of this force must therefore be 

 looked for about that time in the morning and evening or somewhat later. 



Finally, it is necessary to notice one objectiou, which might be 

 raised to the preceding passages, that the effect is not necessarily to be 

 looked for at any fixed time before or after the meridian passage of the 

 satellite, but that, for each place, there will be something equivalent to 

 what is known as the "establishment" of a port in the case of marine tides, 

 The objection, however, is not valid, for in this case we have not to do 

 with free travelling waves, like that of the tides, which take a greater 

 or less time to travel from the place where they originate to the place 

 where they are felt, but with the direct effect of the stresses which 

 produce the waves. These depend solely on the latitude of the place 

 and the declination of the satellite, and for them there is nothing in 

 any way analagous to the " establishment " to be considered. 



Ill, Discussion of the Data. 



After this preliminary exposition of what is to be looked for, we 

 may pass on to a consideration of the results obtained. In the record 

 discussed there are contained 1274 distinct shocks, and, on counting these, 

 it was found that, in each hour of the twenty-four, the number of shocks 

 recorded was as given in the tabular statement No. Ill, where all shocks 

 recorded from h. to h. 59 m. are placed under 0, those between I h. 

 and 1 h. 59 m. under ] and so on. 



The most casual inspection of this table shows that the shocks are 

 not at all uniformly distributed during the twenty-four hours, and 

 that there is, a great preponderance during the hours preceding mid- 

 night, with a lesser increase towards 6 a.m. It may also be noted that 

 the night shocks seem more numerous when the sun is more that 9° S 

 and the day shocks when it is more than 9° N, but no proper compari- 

 son is possible on account of the difference in the total number of 

 shocks in each line. Tor comparison they must be brought all to the 

 same ratio, and this may be done, either by calculating the percentage 

 of the total number of shocks recorded in each hour, or more simply 

 by dividing each figure by the mean value for the line ; this gives a 

 result showing the proportion of the number of shocks recorded in each 

 hour to the average number for one hour. In this way we get the 

 result shown in the next tabular statement. 

 J. II. 19 



