the shore _f_, and the barrier reef _g_, is now much
greater than originally between the shore _c_ and the reef _d_, and the
longer the subsidence continues the farther will the coast of the
mainland recede.

When the first edition of this work appeared in 1831, several years
before Mr. Darwin had investigated the facts on which his theory is
founded, I had come to the opinion that the land was subsiding at the
bottom of those parts of the Pacific where atolls are numerous, although
I failed to perceive that such a subsidence, if conceded, would equally
solve the enigma as to the form both of annular and barrier reefs.

I shall cite the passage referred to, as published by me in 1831:--"It
is a remarkable circumstance that there should be so vast an area in
Eastern Oceanica, studded with minute islands, without one single spot
where there is a wider extent of land than belongs to such islands as
Otaheite, Owhyhee, and a few others, which either have been or are still
the seats of active volcanoes. If an equilibrium only were maintained
between the upheaving and depressing force of earthquakes, large islands
would very soon be formed in the Pacific; for, in that case, the growth
of limestone, the flowing of lava, and the ejection of volcanic ashes,
would combine with the upheaving force to form new land.

"Suppose a shoal, 600 miles in length, to sink fifteen feet, and then to
remain unmoved for a thousand years; during that interval the growing
coral may again approach the surface. Then let the mass be re-elevated
fifteen feet, so that the original reef is restored to its former
position: in this case, the new coral formed since the first subsidence
will constitute an island 600 miles long. An analogous result would have
occurred if a lava-current fifteen feet thick had overflowed the
submerged reef. The absence, therefore, of more extensive tracts of land
in the Pacific, seems to show that the amount of subsidence by
earthquakes exceeds, in that quarter of the globe, at present, the
elevation due to the same cause."[1127]

Another proof also of subsidence derived from the structure of atolls,
was pointed out by me in the following passage in all former editions.
"The low coral islands of the Pacific," says Captain Beechey, "follow
one general rule in having their windward side higher and more perfect
than the other. At Gambia and Matilda islands this inequality is very
conspicuous, the weather side of both being woo