THE GREAT EARTHQUAKE OF S. E. JAPAN 145 



threatened by severe after-shoclvs. After 5 minutes from the beginning 

 I stood np and went over to see tlie instruments. 



0. Instrument observations. None of tlie instruments provided 

 with magTiification index liigher than 5 times registered even the 

 preliminary tremors completely, but they gave important data as to 

 the direction of the initial phase. Two seismographs writing the 

 earth-movements, one in two-time magnification and the other in 

 natural size, registered the preliminary tremors completely, but their 

 writing index went out within 6 seconds after the beginning of the 

 principal portion, except the E. W. component of the former which 

 kept working throughout. Consequently it will be most convenient 

 to trace the earth-nnn-ements according to this instrument, and if 

 necessary, to su])plement these results with the data given by the rest 

 of the instruments. (PI. XIII.) 



The time of commencement at Hongo was llh 58m 4.4s. This is 

 the mean value of the times given by all the available instruments 

 in the institute. Later on, I will give the time of commencement at 

 the origin. 



The initial phase of the earth movements Avas equally perfectly 

 observed by all the instruments. It consisted of a displacement of 

 1.5 ram. towards N26"E and 0.5 mm. upwards in 2.3 sec, indicating 

 that the origine was in the ojiposite direction, that is, in the directiou 

 S26°W. Further, the duration of the preliminary tremors was indicated 

 to have been equal to 12.4 sec, from whicli we obtain the focal 

 distance in km., if we multiply this l)y the coefficient 7.4. 



Let us now go on to describe the principal ])ortion of the quake, 

 which consisted chiefly of, or was at least preceded by, rigid or trans- 

 verse waves. In the seismogram, its beginning was clearly shown by 

 a very big displacement marked with the letter a. Comparing botli 

 components, the N. fS. on(^ was found to be more conspicuous than the 

 other. It indicated three distinct waves ahc, cde and efg each with 

 a period of 1.35 sec Ji.st at the end of these vibrations, the writing 

 index went out of the record -receiver. The annexed diagram (PI. XIV.) 

 shows the resultant horizontal earth-movement corres|)onding to these 

 three waves. The displacement fij is equal to 8.86 cm., from \vhich 

 tlie intensity of the earth vibration comes out to have been about 

 1/10 of the acceleration of gravity ; that is to say, every construction 

 standing on the ground of the University was shaken at that moment 



