ITT. John Milne. 389 
index first moved in a straight line away from the origin. After this 
it was suddenly deflected and yielded a diagram in which the move- 
ment was elliptical or in the form of the figure 8, the two sets of 
vibrations being then compounded. When, however, two bracket 
seismographs of the Ewing type were installed, one to record the 
movement in the direction of the source, and the other in the 
perpendicular direction, the vibrations of the two types were kept 
apart, and the variations of each could be studied. Near the origin 
the amplitude of the condensational vibration was the greater, but 
died out more rapidly than that of the others with increasing distance 
from the origin. In both types of vibration the period increased as 
the waves diverged. In the velocity of the waves Milne found many 
variations. It increased, for instance, with the strength of the 
disturbance, and, in any one artificial earthquake, it diminished 
as the waves spread outwards. In the condensational vibrations 
the mean velocity was 438 feet per second; in the distortional 
vibrations, 357 feet per second. The effect of inequalities in the 
ground were also studied. Small hills seemed to interpose but little 
obstacle to the waves; excavations, such as ponds, checked the 
passage of both types of vibration. 
A matter of some practical consequence brought out by these 
experiments is that two points of ground only a few feet apart did 
not always synchronize in their movements. This result led Milne 
to make some further observations vue earthquakes, and he was 
able to show that points only 3 or 4 feet apart experienced 
slightly different motions. He also made a seismic survey of about 
9 acres in the grounds of the Imperial College of Engineering at 
Tokyo. On the west side the ground is somewhat marshy ; else- 
where it is dry and hard. He found that, as is now well known, the 
maximum acceleration (which measures the intensity of the shock) 
was greater on the soft than on the hard ground. An unexpected 
result was that the acceleration at the bottom of a pit 10 feet deep 
was much less than at the surface.1 
Catalogues of Japanese Earthquakes.—During his residence abroad 
Milne compiled two catalogues of J apanese earth quakes, one showing 
the distribution of seismic activity in past time, the other the 
distribution throughout the country within a limited period. The 
earlier catalogue is less well known than it deserves. It contains 
a list of 366 great earthquakes, from 295 B.c. to a.D. 1872, and gives 
for each shock the date, the intensity (according to a rough scale 
of two degrees), and the district chiefly affected. One result obtained 
from this catalogue i is worth noticing. The increase in the number 
of entries towards the present day, which is so characteristic a feature 
of all other catalogues extending over many centuries, is not observ- 
able here from the seventh “century onwards. The probable 
explanation, according to Milne, is that Japan has remained through- 
1 Trans., vol. 10, 1887, pp. 1-36; vol. 12, 1888, pp. 63-6, 67-75; vol. 13, 
1890, pp. 41-89. 
