PRELIMINARY REPORT. 31 
account of it,’ evidently given at second-hand, I expected to find an island 
somewhat like Viti Levu on a very much smaller scale. But as we 
steamed up to it from the east there could be no mistaking the magnificent 
face of nearly vertical limestone cliffs forming the whole eastern face of the 
island, and at points rising to over a thousand feet in height. At all pro- 
jecting points lines of terraces were plainly marked: at the northern point 
three could be followed, and at the southern extremity five, with traces of 
a sixth perhaps. 
Upon rounding the southern extremity of the island we could see that 
the island was composed of two ridges, running north, separated by a deep 
valley, the western ridge being much lower than the eastern, only rising 
to a height a little over 500 feet. The western ridge is also composed of 
limestone, and at the headlands we could trace three terraces. There is a 
narrow shore platform along the western face, at many points of which 
there are blow-holes where the sea throws up spray to a considerable 
height, but these blow-holes are best seen off Cook Point, the southern 
extremity of Tongatabu. 
As we steamed along the western face of Eua Island we could see 
the higher ridges of the eastern side rising above the lower crest of the 
western ridge, the slopes indicating a valley of considerable size running 
between them. We anchored at English Roads opposite the outlet of 
the drainage of the interior basin, where a small river has cut its way 
through a depression in the shore terrace. On landing we followed 
the crest of the western ridge for a few miles and could see the whole 
valley forming the basin of the island lying between the two ridges, at 
our feet; the slopes leading to the bottom are quite gentle, and the 
valley dips very gradually northward back of the outlet on the western 
shore. Nothing could show more clearly that such an island was not an 
elevated atoll, but a plateau which has been eroded and denuded for a 
long period of time by atmospheric and other agencies, and in which 
a deep basin-shaped valley with gentle slopes has been gouged out,—a 
plateau originally similar to that of Tongatébu Island and of Vavau, 
but of greater height and less extent. 
1 Corals and Coral Islands, 3d ed., p. 373. 
2 The western slope of the eastern ridge is partly volcanic. 
