In explosive earthquakes, which, as just explained, occur at areas almost immediately above the point where the disturbance starts, the force is, generally speaking, the greatest. In earthquakes of this character the force is sometimes sufficiently great to throw large bodies high up into the air. In the case of the great Riobamba earthquake of 1797, the force was not only sufficiently great to fracture the earth in various places, but also to throw bodies lying on the surface great distances into the air. Bodies of men were thrown several hundred feet into the air and were afterwards found on the other side of a broad river or high up on the side of a hill.
It is possible that Humboldt did not inquire with as much care as he should have done into these reports. They were probably greatly exaggerated, since it is difficult to understand how a force great as this would have failed to detach the soil at these places, and hurl it after the people. This much, however, can be accepted, that the upward force was very great.
In the great Calabria earthquake of March, 1783, Dolomieu states that the tops of the granite hills of Calabria were distinctly seen to rise and fall. In some cases houses were suddenly raised a great distance in the air, and were afterwards brought down again to a position of rest, at a higher level without any damage occurring to them. In a similar manner during the Caracas earthquake of March, 1812, the ground was seen to rise and fall in a nearly vertical direction. But, perhaps, one of the most terrible earthquakes of this character was the earthquake that destroyed the greater part of Jamaica in June, 1793. During this earthquake the entire surface of the ground at Port Royal assumed the appearance of a rolling sea. Houses were shifted from their old sites. Many of the inhabitants who had succeeded in escaping from the city to the neighboring country were thrown great distances into the air. Some of these, by good fortune, fell into the harbor, from which, in some cases, they escaped with their lives. Here again the projectile force was probably greatly exaggerated.
Vertical movements characterized the great earthquake of Lisbon, on November 1st, 1755, the city appearing to have been not far from the point of origin.
The commonest type of earthquakes is the horizontal, where the waves emerge at the surface in a direction either horizontal or parallel to the general surface, or at least inclined to it at a very small angle. Where the materials of the earth's crust, through which the waves spread, are of the same kind and of the same density in all directions, the area shaken is approximately circular, but where the materials of the crust are more or less dense in some directions than in others, the area of disturbance is of course oblong or elliptical.
In some cases earthquakes of the horizontal type are limited almost entirely to a single direction. This is especially the case with earthquakes that occur in mountainous districts. These earthquakes are known as linear earthquakes, since they spread almost in a single line.
When earthquake waves pass from one medium to another, that is, from one kind of rock to another, the greater portion of the waves is refracted or bent out of their straight direction as they pass into the new medium; a part of the waves, however, are reflected. It is these reflected waves that probably cause rotary earthquakes.
The speed with which the surface waves move outwards in all directions, varies not only with the force of the wave, but also with the kind of material through which they pass. This velocity may be in the neighborhood of twenty miles per second, while in others the velocity is as great as 140 miles per second.
Naturally, one would suppose that the most severe earthquakes are those in which the waves move the most rapidly. On the contrary, however, the comparatively feeble shocks are sent through the earth with greater velocity.