This island group covers a surface area of 370 square leagues, and it's made up of some sixty subgroups, among which we noted the Gambier group, which is a French protectorate. These islands are coral formations. Thanks to the work of polyps, a slow but steady upheaval will someday connect these islands to each other. Later on, this new island will be fused to its neighboring island groups, and a fifth continent will stretch from New Zealand and New Caledonia as far as the Marquesas Islands.

The day I expounded this theory to Captain Nemo, he answered me coldly:

"The earth doesn't need new continents, but new men!"

Sailors' luck led the Nautilus straight to Reao Island, one of the most unusual in this group, which was discovered in 1822 by Captain Bell aboard the Minerva. So I was able to study the madreporic process that has created the islands in this ocean.

Madrepores, which one must guard against confusing with precious coral, clothe their tissue in a limestone crust, and their variations in structure have led my famous mentor Professor Milne-Edwards to classify them into five divisions. The tiny microscopic animals that secrete this polypary live by the billions in the depths of their cells. Their limestone deposits build up into rocks, reefs, islets, islands. In some places, they form atolls, a circular ring surrounding a lagoon or small inner lake that gaps place in contact with the sea. Elsewhere, they take the shape of barrier reefs, such as those that exist along the coasts of New Caledonia and several of the Tuamotu Islands. In still other localities, such as Réunion Island and the island of Mauritius, they build fringing reefs, high, straight walls next to which the ocean's depth is considerable.

While cruising along only a few cable lengths from the underpinning of Reao Island, I marveled at the gigantic piece of work accomplished by these microscopic laborers. These walls were the express achievements of madrepores known by the names fire coral, finger coral, star coral, and stony coral. These polyps grow exclusively in the agitated strata at the surface of the sea, and so it's in the upper reaches that they begin these substructures, which sink little by little together with the secreted rubble binding them. This, at least, is the theory of Mr. Charles Darwin, who thus explains the formation of atolls--a theory superior, in my view, to the one that says these madreporic edifices sit on the summits of mountains or volcanoes submerged a few feet below sea level.

I could observe these strange walls quite closely: our sounding lines indicated that they dropped perpendicularly for more than 300 meters, and our electric beams made the bright limestone positively sparkle.

In reply to a question Conseil asked me about the growth rate of these colossal barriers, I thoroughly amazed him by saying that scientists put it at an eighth of an inch per biennium.

"Therefore," he said to me, "to build these walls, it took . . . ?"

"192,000 years, my gallant Conseil, which significantly extends the biblical Days of Creation. What's more, the formation of coal-- in other words, the petrification of forests swallowed by floods-- and the cooling of basaltic rocks likewise call for a much longer period of time. I might add that those 'days' in the Bible must represent whole epochs and not literally the lapse of time between two sunrises, because according to the Bible itself, the sun doesn't date from the first day of Creation."

When the Nautilus returned to the surface of the ocean, I could take in Reao Island over its whole flat, wooded expanse. Obviously its madreporic rocks had been made fertile by tornadoes and thunderstorms. One day, carried off by a hurricane from neighboring shores, some seed fell onto these limestone beds, mixing with decomposed particles of fish and marine plants to form vegetable humus. Propelled by the waves, a coconut arrived on this new coast. Its germ took root. Its tree grew tall, catching steam off the water. A brook was born.