In regard to the rate at which the corallines build their cells there is some diversity of opinion--some asserting that the process is imperceptible, while others state as positively that it is rapid. There can be no doubt that some localities and positions are more favourable to the growth of coral than others. Dr Allan, while at Madagascar, made several experiments to test this. He selected several masses of coral, each weighing about ten pounds, and of different species. These he placed three feet below the surface of the sea, and staked them in to prevent removal. In a little more than six months they were found to have risen nearly to the surface, and to have attached themselves to the solid rock. There is also a case mentioned of a ship in the Persian Gulf which, in the course of twenty months, had her copper encased with living coral to the thickness of two feet. On the other hand, it is asserted, and we doubt not with equal truth, that many reefs do not seem to increase in size in the course of many years. When a coral reef has reached the surface, the formation of an island instantly begins; but it necessarily takes a long time ere this island becomes habitable by man. Among the first plants that raise their heads to the sea-breeze is the graceful cocoa-nut palm. This tree is exceedingly hardy, and is found growing on reefs which are so low that at a distance the trees seem to be standing on the surface of the water. Indeed many of them spring out of the pure white sand, and their roots are washed perpetually by the salt spray. Nevertheless, the fruit of such trees is sweet and good. Coral islands of the kind we have just described seldom rise more than a few feet above the level of the sea; but most of them are clothed with luxuriant vegetation. We might easily fill a volume on the subject of the ocean's inhabitants, small and great; but we think the few to which we have made reference is sufficient for the purpose of showing that one set of creatures accounts for that strange luminosity of the ocean which is seen at times in all marine parts of the globe, while another set accounts not only for the sudden appearance of coral islands in the sea where no such islands existed in days of old, but also, partly, for that circulation of the waters of the ocean which is absolutely necessary to the wellbeing of all the creatures on this earth. There are other animals in the sea, besides medusae, which assist in giving luminosity to its waters; and there are other insects, besides corallines, which extract its lime, destroy its equilibrium, and assist in causing its perpetual motion; but the two species which we have described are the best types of the respective classes to which they belong. CHAPTER SIXTEEN. VOLCANIC ISLANDS--OPINIONS OF THE ANCIENTS--"ATLANTIS"--INSTANCE OF THE FORMATION OF A VOLCANIC ISLAND--CONCLUSION.