a depth of twenty or thirty fathoms or more, from the nearest
land; and when this land is an island, the reef surrounds it like a low
wall, and the sea between the reef and the land is, as it were, a moat
inside this wall. Such reefs as these are called "encircling" when they
surround an island; and "barrier" reefs, when they stretch parallel with
the coast of a continent. In both these cases there is ordinary dry land
inside the reef, and separated from it only by a narrower or a wider,
a shallower or a deeper, space of sea, which is called a "lagoon,"
or "inner passage." But there is a third kind of reef, of very common
occurrence in the Pacific and Indian Oceans, which goes by the name
of "atoll." This is, to all intents and purposes, an encircling reef,
without anything to encircle; or, in other words, without an island
in the middle of its lagoon. The atoll has exactly the appearance of a
vast, irregularly oval, or circular, breakwater, enclosing smooth water
in its midst. The depth of the water in the lagoon rarely exceeds twenty
or thirty fathoms, but, outside the reef, it deepens with great rapidity
to two hundred or three hundred fathoms. The depth immediately outside
the barrier, or encircling, reefs, may also be very considerable; but,
at the outer edge of a fringing reef, it does not amount usually to more
than twenty or twenty-five fathoms; in other words, from one hundred and
twenty to one hundred and fifty feet.

Thus, if the water of the ocean should be suddenly drained away, we
should see the atolls rising from the sea-bed like vast truncated cones,
and resembling so many volcanic craters, except that their sides
would be steeper than those of an ordinary volcano. In the case of the
encircling reefs, the cone, with the enclosed island, would look like
Vesuvius with Monte Nuovo within the old crater of Somma;[121] while,
finally, the island with a fringing reef would have the appearance of an
ordinary hill, or mountain, girded by a vast parapet, within which would
lie a shallow moat. And the dry bed of the Pacific might afford grounds
for an inhabitant of the moon to speculate upon the extraordinary
subterranean activity to which these vast and numerous "craters" bore
witness!

When the structure of a fringing reef is investigated, the bottom of the
lagoon is found to be covered with fine whitish mud, which results from
the breaking up of the dead corals. Upon this muddy floor there lie,
here and there,