108 REPORT— 1900. 



the times at which large deviations in latitude occui". In April, when 

 these deviations were comparatively small, large earthquakes did not 

 occur. 



When considering the possibility of any relationship between earth- 

 quakes and these extremely frequent and practically oscillatory changes 

 in latitude, there are two points of importance to be remembered. 



The first is that with each of these earthquakes there is a sudden 

 shifting of a large mass of material at a seismic origin. The molar dis- 

 placement for the Indian earthquake of June 12, 1897, is estimated by 

 Mr. R. D. Oldham by an area of 6,000 or 7,000 square miles, and it is 

 not improbable that earthquakes which have caused the Pacific Ocean to 

 oscillate for a period of twenty-four hours were accompanied by displace- 

 ments of larger magnitude. 



The second consideration is that each of the large earthquakes here 

 considered has been accompanied by surface or distortional waves 

 Avhich in many instances afiect the whole surface of the globe. These 

 waves, so far as we can infer from their velocity, period, and maximum 

 angle of inclination, vary between twenty and seventy miles in length, 

 and are from a few inched to two or three feet in height. If they attain 

 the magnitudes here given (see p. 83) they seem certainly sufiicient to 

 relieve a district in erogenic strain. 



A further test of the suggestion that slight nutational efiects may 

 result from earthquakes would be to compare observations indicating 

 small changes in latitude made before and after the times of large earth- 

 quakes referred to in the report, the more important of which are as 

 follows : 



The times given are the approximate times at the origin. These are 

 expressed in Greenwich mean time (civil). or 24 hrs.= midnight. The 

 times at which the large waves reached any distant station may be calcu- 

 lated by the application of Curve II« or 116 in the table on p. 67. 



VI. Selection of a Fault and Locality suitable for Observations on 

 Earth-Moveinents. By Clement Reid. 



The selection of a favourable site for observations upon differential 

 movement between the two sides of a fault presents many difficulties, and 

 the locality we have chosen is more to be regarded as the best available 

 than as ideally perfect. Leaving out of account for the present considera- 

 tions other tlaan geological, there are certain conditions, most of which 

 must be complied with if the observations are to be of real value. 



The fault selected must be : 



1. Of considerable magnitude, and not be merely a branch fault which 

 the next earth -movement may easily leave unafiected. 



2. It should be of known date, and belong to a recent geological 

 period. This consideration is important, for a Tertiary movement is far 



