fifteen to twenty seconds: the second shock, at right angles to the first, began before the first had died away, and so gave to most people the impression of a twisting motion; it was of undefined duration, but more marked in most of its effects than the first shock. This agrees remarkably with the result presently arrived at.

Rangiora.—The pendulum of the town clock vibrated in a plane W. by S. to E. by N. Other clocks, at right angles to this direction, also stopped. Mr. T. W. Rowe has been good enough to ascertain this fact for me.

The two movements, the normal and transverse, are clearly distinguished at Greymouth, Christchurch, Rangiora—in the last two places the stopping of the clocks gives a direction at right angles (or nearly so) to the direction said to have been noted. Most of the given directions can be explained only either on the hypothesis of error in observation, or on the hypothesis that it was the transverse motion whose direction was noted. The transverse motion seems to have been more marked everywhere; and, even were it not so, it is natural that people should mark its direction more carefully than the movements taking place in the first moments of surprise. The time 7·37 or 7·38 at Christchurch is on any hypothesis of the origin inconsistent with the best time given, 7h. 36min. 43sec. a.m., New Zealand mean time, at Lyttelton. Most of the times are approximate, being to the nearest five minutes only.