192 The East Anglian Earthquake.
of these last observations it is essential to bear in mind that
the value of the observation decreases with the length of the
interval that is allowed to elapse between the disturbance and
the recording of the direction of swing, as it is well known
that most objects when set swinging gradually shift their
plane of vibration owing to the friction of their supports. In
other words, it rarely happens that the direction of the initial
impulse coincides with the direction of least friction between
the suspending hook, &c, and the loop or other support from
which the latter is suspended.
Observations respecting the projection of bodies have also
to be interpreted with caution. The mere overthrow of an
object is useless in determining direction, not only because
the particular phase of the movement which causes any
particular object to topple over may not coincide in direction
with the general direction of propagation, but likewise
because the mode of support or attachment, the form,
position of centre of gravity, &c, of the body all conspire in
determining the direction in which it falls.101 It is only
when a body is distinctly projected103 that any weight can be
101 See Prof. Perry's " Note on the rocking of a Column "; Trans. Seism.
Soc. Japan, vol. iii., p. 103. The old form of seismometer, consisting of
a number of upright cylinders of gradually diminishing size standing in
two rows at right angles to each other, has been found by Milne to be
practically useless for determining direction, the same earthquake having
caused the columns to fall in various directions. Trans. Seism. Soc.
Japan, vol. iii., pp. 46—52 ; and Phil. Mag., Nov. 1881, p. 357. See also
Ewing's ' Earthquake Measurement,' p. 67.
102 The determinations of the maximum velocity of displacement of an
earthquake, by means of measurements of the projection and overthrow
of bodies, are based upon the erroneous assumption that the movement
is of the nature of a sudden jerk, and are therefore of but little value
(See Ewing's 'Earthquake Measurement,' p. 75). In his latest Report to
the British Association on Japanese Earthquakes, Milne states that—
" (1) An earth particle usually reaches its maximum velocity during the
first inward movement. A high velocity is, however, sometimes attained
in the first outward semi-oscillation. (2) The intensity of an earthquake
is best measured by its destructive power in overturning, shattering, or
projecting various bodies. (3) The value,—
used by Mallet and other seismologists to express the velocity of shock as