en would such an
atoll at the depth of half a mile, or 2640 feet, have a diameter of two
miles. Hence it would appear that there must be a tendency in every
atoll to grow smaller, except in those cases where oscillations of level
enlarge the base on which the coral grows by throwing down a talus of
detrital matter all round the original cone of limestone.

Bow Island is described by Captain Beechey as seventy miles in
circumference, and thirty in its greatest diameter, but we have seen
that some of the Maldives are much larger.

As the shore of an island or continent which is subsiding will recede
from a coral reef at a slow or rapid rate according as the surface of
the land has a steep or gentle slope, we cannot measure the thickness of
the coral by its distance from the coast; yet, as a general rule, those
reefs which are farthest from the land imply the greatest amount of
subsidence. We learn from Flinders, that the barrier reef of
north-eastern Australia is in some places seventy miles from the
mainland, and it should seem that a calcareous formation is there in
progress 1000 miles long from north to south, with a breadth varying
from twenty to seventy miles. It may not, indeed, be continuous over
this vast area, for doubtless innumerable islands have been submerged
one after another between the reef and mainland, like some which still
remain, as, for example, Murray's Islands, lat. 9 degrees 54 minutes S.
We are also told that some parts of the gulf inclosed within a barrier
are 400 feet deep, so that the efficient rock-building corals cannot be
growing there, and in other parts of it islands appear encircled by
reefs.

It will follow as one of the consequences of the theory already
explained that, provided the bottom of the sea does not sink too fast to
allow the zoophytes to build upwards at the same pace, the thickness of
coral will be great in proportion to the rapidity of subsidence, so that
if one area sinks two feet while another sinks one, the mass of coral in
the first area will be double that in the second. But the downward
movement must in general have been very slow and uniform, or where
intermittent, must have consisted of a great number of depressions, each
of slight amount, otherwise the bottom of the sea would have been
carried down faster than the corals could build upwards, and the island
or continent would be permanently submerged, having reached a depth of
120 or 150 feet, at which the effect