300 EEPORT— 1889. 



From the above tables, wbich give the period and area shaken by each 

 of these shocks, we see that the shocks which disturbed the whole of 

 Tokio each had on the average at the same time shaken a much larger 

 area than those which were only noticed on the high ground. 



Further, those which were felfc by the residents on the low ground had 

 on the average a much shorter period than those which were only felt on 

 the high ground. 



This latter observation may explain why so many shocks are not re- 

 corded on the low ground. 



Another explanation is that in many instances a vibratory motion 

 passing beneath Tokio may only reach the surface where the super- 

 incumbent soft materials are thin — that is, upon the hills — the relatively 

 thick deposit of soft material on the low ground absorbing the motion 

 like a buffer. 



There were 19 shocks which the instruments at the Imperial Ob- 

 servatory did not record. Out of the 19, however, 8 disturbances had 

 only been felt by one observer, and therefore we cannot say with cer- 

 tainty that there were more than 11 shocks which the Central Observa- 

 tory failed to observe. On the other hand, there were 10 shocks recorded 

 at the Observatory which were not observed by any of the 134 observers 

 in the city. The most probable reason why 11 earthquakes were unre- 

 corded at the Observatory is because these disturbances were too limited 

 in area to reach the district where it is situated. One conclusion we 

 arrive at is, that a set of seismographs located at an Observatory in a city 

 like Tokio, no matter how carefully they may be looked after, cannot be 

 expected to record more than 80 per cent, of the total number of earth- 

 quakes felt in that city. 



Another conclusion resulting from these observations is that residents 

 on the high ground upon the western and northern sides of Tokio feel 

 more earthquakes than residents who live upon the low ground towards 

 the south and east. One explanation for this is that the movement upon 

 the low ground is slower than that on the high ground ; but to place this 

 explanation on a more certain foundation, it is necessary to make instru- 

 mental observations. 



A certain number of earthquakes, however, appear to have originated 

 beneath the high ground in the Kojimachi-Akasaka districts, and do not 

 appear ever to have extended to the low ground. This fact will always 

 make the high-ground disturbances more numerous than those felt upon 

 the low ground. When I was resident within the area of local dis- 

 turbances near Toranomon I came to the conclusion that these local 

 shocks might in many instances be recognised by their character, which 

 is that of a small but sudden little tip from beneath, the vibrations, which 

 only continue two or three seconds, causing hanging lamps to oscillate 

 vertically. 



So far as safety is concerned, notwithstanding the above observations, 

 I am yet of opinion that the high and hard ground is better than the 

 low soft ground, on which earthquake motion is always greater than it 

 is upon the high ground and where destruction has almost always been 

 relatively excessive. 



