Ivi PROCEEDINGS OF THE GEOLOGICAL SOCIETY. 



2. Their frequency increases at the perigee (when the moon is 

 nearest to us) and diminishes at the apogee (when she is furthest 

 removed). 



3. The shocks of earthquakes are more frequent when the moon 

 is on the meridian than when she is 90° removed from it. 



Mr. Mallet has re-examined and confirmed these results. He pre- 

 sents a table of no less than 5879 earthquakes in the northern hemi- 

 sphere and 223 in the southern hemisphere, distributed in months. 

 The northern observations are thus classed in months and in three 

 months : — 



January 627 1 



February 539 1 1669 



March 503] 



April 489 ] 



May 438 1 1355 



Jime 428 J 



July 415 1 



August 488 1 1366 



September 463 J 



October 516 1 



November 473 I 1489 



December 500 J 



5879 



The predominance is clearly in the month of January, the minimum 

 occurs in the end of June or the beginning of July, The six extreme 

 months give 3158, and the six median months give only 2701. The 

 three consecutive months having most shocks are January, February, 

 and March, the three which have the fewest are May, June, and July. 



Those for the southern hemisphere are perhajDS too few to be 

 employed with much confidence in a similar way. They seem to 

 give a principal maximum in November, and a second in May, wdth 

 two minima in March and August. The first three months give 42, 

 the next 56, the third 47, and the last 78. If 56 be taken as the 

 mean, we have 



56—14 for the first period of three months, 

 — for the second, 

 + 9 for the third, 

 -\-22 for the fourth. 



These extremely curious results, coinciding as they do with the 

 results of many recent researches, which show the reality of the 

 lunar influence on terrestrial magnetism and the temperature and 

 pressure of the atmosphere, seem to point to an influence exerted by 

 the moon's attraction on the condition of the interior masses of the 

 earth. Whether, however, we admit this influence to be sensible on 

 a fluid interior, or effective in generating electrical or other actions, 

 capable of transformation into mechanical force, and however much 

 force we ascribe to such influence, it can never be regarded as a cause 



