EARTHQUAKE MOTION TO GREAT DISTANCES. 141 



which the earthquake was experienced corresponded therefore to about 20' 1 5 5 '5* 

 Greenwich time. 



Allowing for a rate of propagation of 3 kiloms. per second, the earthquake would 

 have taken about 2 m to reach Santiago, arid 6 m to reach Buenos Ayres. Making 

 these deductions, the time of origin becomes 20 h 50 and 20 h 49 < 5 m respectively. As 

 greater value should be attached to the Santiago record, both because it was automatic 

 and because of its greater proximity to the centre, we may take the time of origin 

 as 20 h 50" 0-5 m . 



The records abstracted below, except in the case of Tokio, are taken from the 

 ' Bollettino d. Ufficio Centrale Met. e Geodyn.,' for October, 1894. 



Pavia. 21 h 46 ra 55 s commencement, 21 h 57 52* end, on the Brassart seismograph. 



Siena. 21 h 12 m commencement on the Vicentini microseismograph, slow undula- 

 tions distinct at 21 h 52 m , which continue conspicuous till 22 h 8 m . 



Rome. 21 h 7 35* commencement on both components of the grande sismom. 

 On the N.W. S.E. component these reach a maximum of 3'5 millims. at 21 k 15" 10* ; 

 towards 21 h 40 m slow undulations l)egin on both components. Maximum at 

 21 h 55 ra 40', end about 23 h . On the other seismograph the commencement is not till 

 21 h 40 m , maximum 21 h 55 m 40 s , end 22 h 14 15*. 



Ischia. 21 k 33 commencement on the Brassart seismograph ; 22 h 8 m maximum, 

 end at 22 h 40". 



Rocca di Papa. 21 h 49 35" commencement, 22 k 10 end, on the grande sismom. 



Nicolaiew. Commencement 21 h 12 6', a sudden increase at 22 k 17" 6 , curve 

 disappears from 21 h 24 6' to 21 h 32" 6 s , fresh disappearance till 22 h 2" 6*, end O h 37" 6 1 

 of following day. 



Charkoiv. Commencement 2l h 8 ra 36", the curve disappeared for an hour, end at 

 O h 10 m 54 s of the following day. 



Tokio. In a letter by Dr. E. VON REBEUR PASCHWITZ to ' Nature ' (vol. 52, p. 55, 

 1895), a letter from Professor MILNE is referred to, in which it is stated that the 

 earthquake was recorded at Tokio by three instruments at 18 h O m , 18 h 5", 17 h 4 1", 

 respectively. The times were measured from a time signal at about the previous 

 noon, but the record of these was lost by fire. In the case of the first 'instrument the 

 time signal was always impressed within about \ m or 1"' of noon, consequently 

 the error cannot exceed a few minutes. These times, recorded also in ' Brit. Assoc. 

 Rep.,' 1895, p. 147, where the last is marked as evidently wrong, correspond to 

 Greenwich mean times 21 h 0", 21 h 5 m , and 20 h 41 ra respectively. 



The times, given in minutes after 2 1 1 ' Greenwich mean time, may be classified as 

 follows : 



