5th. It is possible, as remarked by Humboldt, that interference may, though very rarely act as a protecting condition, in diminishing the energy of shock at some local point; but it is certain that places are frequently situated at nodal points, where the simultaneous arrival, of the main wave along with a subordinate one, increases the severity of injury, &c.
It is barely within the limits of probability, that two shocks shall arrive at the one spot, with such coincidence in time, that the opposite semiphases of the two respective waves, shall partly or wholly neutralize each other; but it is a different affair, and of much simpler conditions, that two or more separate waves should simultaneously arrive at the one spot, i.e., should overtake or intersect at it.
There is no doubt that this actually did occur at Montemurro, at Marsico Nuovo, and at Vignola, where the conditions producing it will be evident, from Map A; and most probably it took place also, to some extent, at Avigliano, and the cities of the Melfi group, and possibly also at Canosa.
6th. The effects of formation, and of the change from one formation to another; of situation in deep clay plains, or those consisting of loose material, or on solid rock, in locally and suddenly modifying the energy of the shock, and its effects, on towns, &c., have been already made sufficiently clear, and the principles have been elucidated by several examples, in Part II.
Much pains has been devoted by earthquake describers, (Hamilton, Dolomieu, Spallanzani, and others,) to the question whether towns situated upon the plain on loose
