In the case of a horizontal shock passing through a bed of ooze or water-bearing strata, the elastic wave will tend to pack up the water during the forward motion to such an extent that it will flow or spout up through any aperture communicating with the surface. By the repetition of these movements causing ejections, sand or mud cones, like those produced by a volcanic eruption, may be formed, and by a similar action water may be shot violently up out of wells, as was the case in Jamaica in 1692.

If an emergent wave acts through a water-bearing bed upon a superincumbent layer of impervious material, this upper layer is, during the upward motion, by its inertia suddenly pressed down upon the latter.

This pressure is equal to that which would raise the upper layer to a height equal to the amplitude of the motion of an earth particle, and with a velocity at least equal to the mean velocity of the earth particle resolved in the vertical direction.

For a moment the water-bearing strata receive an enormous squeeze, and the water or mud starts up through any crevice which may be formed leading to the surface.

From this we see that liquids may rise far beyond the level due to hydrostatic pressure.[51]

Volger has attributed the origin of lights or flames appearing above fissures to the friction which must take place between various rocky materials at the time when the fissures are opened. As confirmatory of this he refers to instances where similar phenomena have been observed at the time of landslips. At the time of these landslips the heat developed by friction has been sufficiently intense to convert water into steam, the tension of which threw mud and earth into the air like the explosion of a mine.[52]

The gas eruptions which occasionally take place with earthquakes are probably due to the opening of fissures communicating with reservoirs or strata charged with products of natural distillation, or chemical action, which previously had accumulated beneath impervious strata. Of the existence of such gases we have abundant evidence. In coal mines we have fire damp which escapes in increased quantities with a lowering of the barometrical pressure. In volcanic regions we have many examples of natural springs of carbon dioxide.

These various gases sometimes escape in quantity, or erupt without the occurrence of earthquakes. Rossi mentions an instance where a few years ago quantities of fish were killed by the eruption of gas in the Tiber, near Rome. Another instance is one which occurred at Follonica on April 6, 1874. On the morning of that day many of the streets and roads were covered with the dead bodies of rats and mice. It seemed as if it had rained rats. From the facts that the bodies of the creatures seemed healthy, that the destruction had happened suddenly, and not come on gradually like an epidemic, it was supposed that they had been destroyed by an emanation of carbon dioxide. The fact that many of them lay in long lines suggested the idea that they had been endeavouring to escape at the time of the eruption.[53] If we can suppose sudden developments of gas like this to have occasionally accompanied earthquakes, we may sometimes have the means of accounting for the sickness which has been felt.

Disturbances in lakes.—It has often been observed that, at the time of large earthquakes, lakes have been thrown into violent agitation, and their waters have been raised or lowered. At the time of the great Lisbon earthquake, not only were the waters of European lakes thrown into a state of oscillation, but similar effects were produced in the great lakes of North America. In some instances, as in the case of small ponds, these movements may be produced by the horizontal backward and forward motion of the ground. At other times they are probably due to an actual tipping of a portion of their basins. Movements like these latter will be again referred to in the chapter on Earth Pulsations. On January 27, 1856, there was a shock of earthquake at Bailyborough, Ireland, which occasioned an adjacent lough to overflow its banks and rush into the town with great impetuosity. In returning it swept away two men, leaving behind a great quantity of pike and eels of a prodigious growth.[54]

Disturbances in rivers.—Just as lakes have been disturbed, so also have there been sudden disturbances in rivers. Sometimes these have overflowed their banks, whilst at other times they have been suddenly dried up. In certain cases the reason that a portion of a river should have become dry has been very apparent, as, for instance, at the time of the Zenkoji earthquake in Japan in 1847, when the Shikuma-gawa became partly dry in consequence of the large masses of earth which had been shaken down from overhanging cliffs damming a portion of its course, and thus forming, first, lakes, and subsequently, new water-courses. As another example, out of the many which might be quoted, may be mentioned the sudden drying up of the river Aboat, a tributary of the Magat, in the Philippine Islands, on July 27, 1881, shortly after a severe shock of earthquake. The water of this river ceased to flow for two hours, after which it reappeared with considerable increase of volume and of a reddish colour. Signor E. A. Casariego, who describes this, remarks that the phenomenon could easily be explained through the slipping down of the steep banks in narrow parts of its upper valley, by which means its flow had been obstructed until the water had time to accumulate and pass over or demolish the obstruction.