would result an appearance as if this windward bulwark had really been erected by the 'instinctive foresight' of the animalculæ. If the opinion we have thus hazarded be correct, it is evident that, in different situations, there will be coral rocks, exhibiting very different characters; and thus, from the charts given of the coral islands which lie in the Indian and Chinese seas, in the regions of the six months' monsoons, it may be inferred that every side is equally advanced in formation. Chamisso says, that the larger species of coral insects, which form blocks measuring several fathoms in thickness, seem to prefer the violent surf on the external edge of the reef; which, with the obstacles opposed to the continuation of their life in the middle of a broad reef, by the amassing of shells and fragments of coral, easily account for the outer edge of the reef first approaching the surface; and the same circumstances must also contribute to the circular character of these reefs, or groups of reefs. Whether those islands, which have greater length than breadth, are opposed in their greatest extent to the winds and waves, appears to depend on the size and arrangement of their submarine supports. The arguments of Quoy and Gaimard, who say that the species which form the most considerable banks, such as the Meandrinæ, certain Caryophylleæ, and especially the Astreæ, require the influence of light to perfect them, and consequently cannot be developed at a depth of from ten to twelve hundred feet, are only applicable to those species; and if the species of the genus Astrea are alone capable of covering immense extents of surface, and do not commence their operations at a greater depth than twenty-five or thirty feet, why may not the branched madrepores, which do live at considerable depths, have formed the platform for their reception, just as we see the marine algæ distributed in different zones or depths of the sea[1]?
In the uncultivated tracts of our latitude, vegetation generally commences by the appearance of pulverulent lichens, which are succeeded by foliaceous plants of a more perfect organization, by mosses, and finally, varying with the soil and situation, by monocotyledonous or dicotyledonous plants, which gradually make their appearance; but the coral islands of the Pacific, not adapted to support plants requiring a depth of soil, often first afford a basis to high trees provided with fibrous roots, as the Pisonia, Cordia Sebastiana, Morinda citrifolia, and Pandanus odoratissimus, which, at a distance, give to these small islands the form of a hill. The loose dry stones of the first ridge are penetrated by the hard roots of the Tifano, which expands its branches into a tail, spreading tree, and is attended by the fragrant Suriana, and the sweet-scented Tournefortia, in the shelter of whose foliage the Achyranthus and Lepidium seem to thrive the best. Beyond the first
