H ON THE EARTHQUAKE PHENOMENA OF JAPAN. 367 
ratio of the speed of these two kinds of motion is not constant. The 
softer and less elastic rocks are, the nearer do these velocities approach 
each other. That the ratio of the speed of normal and transverse motions 
- is not constant is shown from a table of these velocities calculated for 
different rocks from their moduli of elasticity. 
IX. Miscellaneous. 
1. At the time of an earth disturbance, currents are produced in 
telegraph lines. : : ' ‘ 
2. The exceedingly rapid decrease in the intensity of a disturbance in 
the immediate neighbourhood of the epicentrum has been illustrated by 
a diagram. ; : 
3. For the duration of a disturbance due to a given impulse in 
_ different kinds of ground, reference must be made to the detailed descrip- — 
tions of the first four sets of experiments. 
X. The Simultaneous Observation of Earthquakes at several Stations in 
Electrical Connection. 
In my last report to the Association I gave the results of observations 
carried on at three stations, which were indicated by the numerals I., II., 
and III. These three stations I now call A, B, and C, and have added on 
_ to them other stations in the same area called D, E, F, G, H, and J. 
G refers to the records taken by an instrument placed inside a house 
specially constructed to mitigate the effects of earthquake motion. H is 
@ station ina pit 10 feet deep. To these records I shall make special 
reference. J is a station where there are now two instruments. It is 
between the house and the pit, about 20 or 30 feet from each. F is a 
station on the edge of flat marshy ground, at the foot of a slope running 
nearly north and south. At a level of about 10 feet above this are the 
stations C and D on hard ground. The relative position of these stations 
will be understood from the following sketch (fig. 1). A line from A to 
MOAT 
B bears § 15° 30’ W. The dotted line shows a gentle slope dividing the 
soft ground from the hard ground. The softest ground is in the middle 
of the triangle B, E, F, where it is quite marshy. The following five 
tables embody the principal results which have been obtained. For 
stations G, H, and J the number of observations are too few to give average 
results. I is the starting station. 
