ON THE EAUTIIQUAKE PHENOMENA OF JAPAN. 209 



"Further, it may he added that certain semi-vibrations have been 

 described by the pointer of the seismograph moving across a movin"" 

 record receiver in the direction of its motion, which have the anomalous 

 appearance of having- been described in less than no time. 



From these observations it_ would appear that it is hardly safe for us 

 to regard the backward and forward motion of the earth as simple 

 harmonic motions, and maximum velocities and accelerations calculated 

 upon such an assumption may possibly lead to false results. 



TV. An endeavour to find out the relative extent and variation in 

 direction of the motion of an earthquake at neighbouring points in a given 

 area, the contour and geological structure of ivhich is irregular. 



To work ont this problem seven similar seismometers were distributed 

 on the hills and in the valleys near my house. The chief difficulty which 

 had to be overcome in this investigation was to obtain a type of seismo- 

 meter which, whilst magnifying the actual motion of the ground, was 

 sufficiently simple to allow of a number being employed, and which, when 

 under the same conditions, would give the same result. The results of 

 experiments to find such an instrument are given in Vol. III. of the 

 'Transactions of the Seisraological Society.' The seismometers which 

 were chosen consisted of heavy pendulums suspended in cases to shield 

 them from currents of air. Against the bobs of these pendulums, in grooves 

 at right angles to each other, slips of wood were placed. At the time of 

 an earthquake these, being pushed against the pendulum by the motion 

 of the stands on which they rested, caused pointers which were attached 

 to them by one thread of a bifilar suspension to swing round and o-ive a 

 magnified representation of the motion which had taken place. 



The so-called hills surrounding the plain of Tokio are irregular, flat- 

 topped spurs, jutting out from an elevated plateau about 100 feet hio-h 

 into the flat plain on which a great portion of the city of Tokio is situ- 

 ated. The area over which the seismometers were distributed had a 

 radius of about a quarter of a mile. Two of the seismometers were 

 placed on the top of the spurs, two'wex-e placed near together on the side 

 of a spur, whilst three were placed at diSerent points on the plain. 



The results obtained from the records of fourteen small earthquakes 

 are : — 



1st. That the maximum amplitude of motion and the direction of 

 motion at all the stations were different. 



2nd. The greatest motion was experienced upon the flat ground, and 

 the least upon the hills and their flanks. 



These results have been confirmed by observations made with other 

 instruments. For instance, one of Professor Ewing's bracket seismographs 

 at the University, situated on the flat ground of Tokio, usually records a 

 much larger amplitude of motion than seismographs constructed on similar 

 principles placed at my own house, situated on a small plateau half-way 

 up the side of a hill about a mile distant from the University. Also the 

 duration of a distui-bance is longer on the low ground than it is upon the 

 high ground. As an example, the earthquake of March 11, 1882, may 

 be quoted. At the University this disturbance lasted about 41- minutes 

 the maximum amplitude of motion being 8 millimetres; at my "house the 

 motion could only be traced upon the moving plate on which it was 

 recorded for a period of about 1^ minute, and the maximum amplitude 

 was about 3 millimetres. That there is less disturbance npou the hills 

 than in the plain at Tokio is a fact that has long been recoo-nised bv the 

 1882. p c o J 



