2. When the length of the wave is not less than a thousand times the depth of the water, the velocity of the wave depends (sensibly) only on the depth, and is proportional to the square root of the depth. It is, in fact, the same as the velocity which a free body would acquire by falling from rest under the action of gravity through a height equal to half the depth of the water. _Rollers_ are of the nature of a violent _ground swell_, and possibly the worst of them may be due to the propagation of an earthquake wave. They come with little notice, and rarely last long. All the small islands in the Mid-Atlantic experience them, and they are frequent on the African coast in the calm season. They are also not unknown in the other oceans. In discussing the meteorology of the equatorial district of the Atlantic, extending from lat. 20 deg. to 10 deg. S, Captain Toynbee observes that "swells of the sea are not always caused by the prevailing wind of the neighborhood. For instance, during the northern winter and spring months, northwesterly swells abound. They are sometimes long and heavy, and extend to the most southern limit of the district. Again, during the southern winter and spring months, southerly and southwesterly swells abound, extending at times to the most northern limit of the district. They are frequently very heavy and long." The great _forced sea waves_, due to earthquakes, and generally to subterranean and volcanic action, have been known to attain the enormous height of 60 feet or more, and sweep to destruction whole towns situated on the shores where they have broken--as for example Lisbon and places on the west coast of America and in the island of Java. Though so destructive when they come in toward the land, and begin to feel the shelving sea bottom, it is not probable that, in the open ocean, this wave would do more than appear as a long rolling swell. It has, however, been observed that "a wave with a gentle front has probably been produced by gentle rise or fall of a part of the sea bottom, while a wave with a steep front has probably been due to a somewhat sudden elevation or depression. Waves of complicated surface form again would indicate violent oscillations of the bottom." The altitude and volume of the great sea wave resulting from an earthquake depend upon the suddenness and extent of the originating disturbance and upon the depth of water at its origin. Its velocity of translation at the surface of the sea varies with the depth of the sea at any given point, and its form and dimensions depend upon this also, as well as upon the sort of sea room it has to move in. In deep ocean water, one of these waves may be so long and low as to pass under a ship without being observed, but, as it approaches a sloping shore, its advancing slope becomes steeper, and when the depth of water becomes less than the altitude of the wave, it topples over, and comes ashore as an enormous and overwhelming breaker.