208 The East Anglian Earthquake.
rotate round B as an axis in the direction of the hands of a
watch, i. e., from right to left. If the direction of the shock
had been along e' G, the rotation round B would have been
in the opposite direction, i. e., from left to right. We may also
suppose G e' to be resolved into the two components, G B,
G f, of which the former causes the column to bear heavily
upon B, and the latter causes rotation round B. G e may
also be resolved into the components G B, G /', of which the
latter causes revolution. Another way of stating these facts
would be to regard the line of direction of the impulse, say
from f' to f, and the perpendicular let fall upon it from
A, as forming a couple having an arm A f, and causing
the column to rotate from left to right, and so on for other
directions.
From the foregoing considerations it will be evident that
rectilinear impulses parallel with the sides or diagonals of a
rectilinear column ought not to cause rotation. A shock
having a direction passing through any of the shaded octants
in the preceding figure would cause rotation from right to
left, whereas if the line of direction passed through one of the
unshaded octants the rotation would be in the opposite
direction. Owing to the great instability of a chimney when
tilted up on one of its corners, it is certain that a shock
having a direction parallel or nearly parallel with one of the
diagonals would cause rotation round the corner as an axis
in one direction or the other. A shock parallel with one of the
sides would probably cause displacement, either towards or
away from the origin of the impulse. The chimneys at
Colchester (p. 47) illustrate this effect.
It will now only be necessary to apply these considerations
to a specific case of rotation produced by the present earth-
quake in order to justify the conclusion to which we have
already given expression (p. 65, note 30), viz., that the
direction of rotation of chimneys is useless in determining
the direction of propagation of the shock. Mr. J. E. Wilkins's
square shaft at Wivenhoe (p. 85) was caused to rotate from
right to left, so that the shock may have come from any point
between N. & N.E., E. & S.E,, S. & S.W., or W. & N.W. It