But what had destroyed their vegetation, if it ever existed?
Were they not, too, the mere remnants of a submerged and destroyed land, connected now only by the coral shoals?
So it seemed to us, as we ran out past the magnificent harbour at the back of Virgin Gorda, where, in the old war times, the merchantmen of all the West Indies used to collect, to be conveyed homeward by the naval squadron, and across a shallow sea white with coral beds.

We passed to leeward of the island, or rather reef, of Anegada, so low that it could only be discerned, at a few miles' distance, by the breaking surf and a few bushes; and then plunged, as it were, suddenly out of shallow white water into deep azure ocean.

An upheaval of only forty fathoms would, I believe, join all these islands to each other, and to the great mountain island of Porto Rico to the west.

The same upheaval would connect with each other Anguilla, St.Martin, and St.Bartholomew, to the east.

But Santa Cruz, though so near St.Thomas's, and the Virgin Gordas to the south, would still be parted from them by a gulf nearly two thousand fathoms deep--a gulf which marks still, probably, the separation of two ancient continents, or at least two archipelagoes.
Much light has been thrown on this curious problem since our return, by an American naturalist, Mr.Bland, in a paper read before the American Philosophical Society, on 'The Geology and Physical Geography of the West Indies, with reference to the distribution of Mollusca.' It is plain that of all animals, land-shells and reptiles give the surest tokens of any former connection of islands, being neither able to swim nor fly from one to another, and very unlikely to be carried by birds or currents.