ON SEISMOLOGICAL INVESTIGATION. 24& 



The effects on magnetic needles at various observatories have, however, 

 been very different. 



At magnetic observatories in England, Pola, Vienna, Copenhagen, 

 Toronto, and other observatories only the slightest perturbations are 

 noted, and these only rarely. On the contrary, at Utrecht, Potsdam, 

 and Wilhelmshaven movements of magnetic needles and the occurrence of 

 seismic waves are comparatively of frequent occurrence. 



One explanation for this marked difference in the behaviour of mag- 

 netic needles at different stations rests on the fact that these three latter 

 stations stand on the vast plain of alluvial drift which stretches from 

 Holland eastwards across Northern Germany into Russia, and it may be 

 assumed that the seismic waves on this ocean-like expanse of soft materials 

 are slower and larger than those exhibited in the harder materials on or 

 near to which other observatories are situated. Exceptions to such an 

 explanation are, however, found in the records from Copenhagen and 

 Zikawei, and before we can say with certainty that there are great 

 differences in the character of the mechanical movements at different 

 stations, seismometric records must be obtained from the same. 



Other reasons which may be adduced to explain why at one set of 

 stations magnetic needles are disturbed, whilst at another set they are 

 practically quiescent may be as follows : — 



First, we may assume that at one set of stations the needles have 

 periodic movements which more neai'ly synchronise with the period of the 

 earth waves than those of needles at stations where magnetometer 

 disturbances are rare. 



The only notes hitherto collected which bear on this point are con- 

 tained in the following table, in which the times given are the intervals 

 in seconds taken to complete a double or back-and-forth circular vibration 

 of declination (unifilar) and horizontal force (bifilar) magnetometers : — 



Unifilar, Bifilar. 



s. s. 



Stonyhurst H-30 13-20 



IJombav 5-33 S'OO 



Greenwich 50-00 42-00 



There is also an upper declination magnet 



(for eye observations), with a period for 



double swing of 61" approximately. 



Vienna S-S.-S 15-3G 



Pola 7-98 



Potsdam 1000 800 



Wilhelmshaven 15 9 16-7 



Kew 105 13-6 



Falmouth 170 18-8 



Antwerp ....... 11-4: 18-6 



Copenhagen 71 4-9 



The magnet for horizontal force at Antwerp is kept at right angles to 

 the meridian by deflection magnets. 



From the table it will be noted that the periods given for Kew, where 

 disturbances at the time of earthquakes are rare, are not very different 

 for those given for Potsdam, where disturbances are frequent. 



A very much more important point, however, is that all large earth- 

 quakes commence with a series of short-period waves. Five-second 

 periods are marked. These are followed with others having periods of 

 10 sees., whilst later there may be waves with periods of 20 and even 



