166 F. OMORI AND K. HIE ATA : 



The mean of the above three angles of emergence is 11°, 

 and it will be observed that the direction of motion dipped 

 towards AV instead of towards E, where was situated the origin 

 of the disturbance. Of course it is out of question to inquire 

 into the relation between the direction of motion and the 

 position of the origin in cases like the present, where the 

 intensity of earthquake motion at the observing station is already 

 very feeble. 



6. Eqke. No. 3. The 31st August, 1896 ; 4.42.15 p. m. 

 Intensity, ürong. This earthquake was strongly felt in the 

 central and western portions of the Province of Ugo and Riku- 

 chu, and was one of the fore-shocks of the destructive earth- 

 quake, which took place on the same day at 5.9.f55 p. m. Judg- 

 ing from the isoseismal lines (fig. 4), the epicentre was situated 

 at about 100 km. S 80° W of Miyako. 



The seismographic diagram (fig. 2) shows a considerable 

 amount of horizontal and vertical motion, which consisted of 

 ripples superposed on slow undulations. The H.M. was more 

 than four times greater in tlie (EW) than in the (NSj. The 

 duration of the H.M. was 73 seconds, and that of the V.M. was 

 89 seconds. 



The preliminary tremor, whose duration was 10 seconds, 

 consisted of ripples of the following elements : — 



These ripples were superposed on slow undulations, whose 

 maximum 2a was 1.4 mm. (period 1.4 seconds) in the (EW), 

 very small in the (NS), and 0.2 mm. in the (V). 



