eira, would have produced a wave previous to the
retreat. Nor could the motion of the waters at Madeira have been caused
by a different local earthquake; for the shock travelled from Lisbon to
Madeira in two hours, which agrees with the time which it required to
reach other places equally distant.[686]

The following is another solution of the problem, which has been
offered:--Suppose a portion of the bed of the sea to be suddenly
upheaved; the first effect will be to raise over the elevated part a
body of water, the momentum of which will carry it much above the level
it will afterwards assume, causing a draught or receding of the water
from the neighboring coasts, followed immediately by the return of the
displaced water, which will also be impelled by its momentum much
farther and higher on the coast than its former level.[687]

Mr. Darwin, when alluding to similar waves on the coast of Chili, states
his opinion, that "the whole phenomenon is due to a common undulation in
the water, proceeding from a line or point of disturbance some little
way distant. If the waves," he says, "sent off from the paddles of a
steam-vessel be watched breaking on the sloping shore of a still river,
the water will be seen first to retire two or three feet, and then to
return in little breakers, precisely analogous to those consequent on an
earthquake." He also adds, that "the earthquake-wave occurs some time
after the shock, the water at first retiring both from the shores of the
mainland and of outlying islands, and then returning in mountainous
breakers. Their size is modified by the form of the neighboring coast;
for it is ascertained in South America, that places situated at the
head of shoaling bays have suffered most, whereas towns like Valparaiso,
seated close on the border of a profound ocean, have never been
inundated, though severely shaken by earthquakes."[688]

More recently (February, 1846), Mr. Mallet, in his memoir above cited
(p. 475), has endeavored to bring to bear on this difficult subject the
more advanced knowledge obtained of late years respecting the true
theory of waves. He conceives that when the origin of the shock is
beneath the deep ocean, one wave is propagated through the land, and
another moving with inferior velocity is formed on the surface of the
ocean. This last rolls in upon the land long after the earth-wave has
arrived and spent itself. However irreconcilable it may be to our common
notions of soli