the grounds
on which it was formerly embraced. In the first place, it had been
remarked that there were many active volcanoes in the coral region of
the Pacific, and that in some places, as in Gambier's group, rocks
composed of porous lava rise up in a lagoon bordered by a circular reef,
just as the two cones of eruption called the Kamenis have made their
appearance in the times of history within the circular gulf of
Santorin.[1123] It was also observed that, as in S. Shetland, Barren
Island, and others of volcanic origin, there is one narrow breach in the
walls of the outer cone by which ships may enter a circular gulf, so in
like manner there is often a single deep passage leading into the lagoon
of a coral island, the lagoon itself seeming to represent the hollow or
gulf just as the ring of dry coral recalls to our minds the rim of a
volcanic crater. More lately, indeed, Mr. Darwin has shown that the
numerous volcanic craters of the Galapagos Archipelago in the Pacific
have all of them their southern sides the lowest, or in many cases quite
broken down, so that if they were submerged and incrusted with coral,
they would resemble true atolls in shape.[1124]

Another argument which I adduced when formerly defending this doctrine
was derived from Ehrenberg's statement, that some banks of coral in the
Red Sea were square, while many others were ribbon-like strips, with
flat tops, and without lagoons. Since, therefore, all the genera and
many of the species of zoophytes in the Red Sea agreed with those which
elsewhere construct lagoon islands, it followed that the stone-making
zoophytes are not guided by their own instinct in the formation of
annular reefs, but that this peculiar shape and the position of such
reefs in the midst of a deep ocean must depend on the outline of the
submarine bottom, which resembles nothing else in nature but the crater
of a lofty submerged volcanic cone. The enormous size, it is true, of
some atolls, made it necessary for me to ascribe to the craters of many
submarine volcanoes a magnitude which was startling, and which had often
been appealed to as a serious objection to the volcanic theory. That so
many of them were of the same height, or just level with the water, did
not present a difficulty so long as we remained ignorant of the fact
that the reef-building species do not grow at greater depths than
twenty-five fathoms.

_May be explained by subsidence._--Mr. Darwin, after examining a va