190 F. OMORT AND Iv. HIRATA : 



to be generally felt without any instrumental aid. The values 

 of the constant k, 2;iven in the 6th column of the Table vary 

 within rather wide limits, namely, 4 and 28, but it would be 

 impossible to expect any very perfect agreement between several 

 cases, as, besides the difficulty of accurately measuring the 

 duration, D, the value of the area of disturbance, A, is very 

 apt to be subjected to accidental modifications, the origins 

 being, in the present cases, mostly submarine. The average 

 value of k is 1.3.6, so that the above equation becomes, for 

 Miyako, 



i>=0.024x-^- 



Further discussions of this question shall be made in the 

 report on the Tokyo seismometrical observations to be soon 

 published. 



39. The duration of the vertical motion. 



The relation of the duration of the vertical motion, measuied 

 in 14 cases, to that of the horizontal, was as follows. 



In 5 cases, the durations of the two components were equal 

 to each other ; in 1 case, the V.M. lasted longer than the H.M. 

 in the ratio of 1.2 : 1 ; and in the remaining 8 cases, the 

 duration of the V.M. was shorter than that of tlie H.M., the 

 ratios being respectively 0.6 : 1, 0.8:1, 0.8:1, 0.4:1, 0,5:1, 

 0.7 : 1, 0.6 : 1 and 0.6 : 1. Taking the average of these 14 

 cases, the durations of the vertical and the horizontal components 

 were in the ratio of 0.8 : 1. 



40. The duration of the pri7icipal portion of an earthquake, 

 which is much shorter than that of the whole of it, varied in 

 the present cases, as given in the oth column of the Table, 

 between 0.7 and 26.0 seconds. The principal jDortion is generally 



