soever,—not only on the hill-sides, but also on the sea-shore, and even, as has been already stated, on the coral reef itself, where the soil, if we may use the name, was nothing better than loose sand mingled with broken shells and coral rock. So near to the sea, too, did this useful tree grow, that in many places its roots were washed by the spray from the breakers. Yet we found the trees growing thus on the sands to be quite as luxuriant as those growing in the valleys, and the fruit as good and refreshing also. Besides this, I noticed that, on the summit of the high mountain, which we once more ascended at a different point from our first ascent, were found abundance of shells and broken coral formations, which Jack and I agreed proved either that this island must have once been under the sea, or that the sea must once have been above the island. In other words, that as shells and coral could not possibly climb to the mountain top, they must have been washed upon it while the mountain top was on a level with the sea. We pondered this very much; and we put to ourselves the question, "What raised the island to its present height above the sea?" But to this we could by no means give to ourselves a satisfactory reply. Jack thought it might have been blown up by a volcano; and Peterkin said he thought it must have jumped up of its own accord! We also noticed what had escaped us before, that the solid rocks of which the island was formed were quite different from the live coral rocks on the shore, where the wonderful little insects were continually working. They seemed, indeed, to be of the same material,—a substance like limestone; but, while the coral rocks were quite full of minute cells in which the insects lived, the other rocks inland were hard and solid, without the ap-
