CORAL REEFS. APPENDIX. 217 



first brought up by upheaval to or above the sea-level. Lat- 

 eral extensions or wings grow out on either side, so as to 

 ultimately form a horse-shoe reef. Such a reef presents its 

 convexity against the prevailing surface currents, to which in 

 truth it owes its shape " {loc. cit., p. 900 ; this view of the forma- 

 tion of atolions or horse-shoe reefs is further elaborated in Mr. 

 Guppy's paper " Preliminary Note on Keeling Atoll," Nature, 

 Jan. 3, 1889). Such are seemingly the conditions that we find 

 on Santa Anna Island, but the examination of the 100-fathom 

 contour line, which closely conforms to the actual bounda- 

 ries of the island, even to the indentation of the 17-fathom 

 Port Mary — concerning which Mr. Guppy expresses himself 

 as having " been unable to obtain any satisfactory explana- 

 tion " ("Solomon Islands," p. 117) — ^proves conclusively, I be- 

 lieve, that the surface exposed above water is'merely the cor- 

 respondent of that which is below it, in other words, the island 

 has grown up on a base of its own form, which base is seem- 

 ingly a breached crateral cone of a volcano. It repeals on a 

 larger scale what is still presented by its own highest elevation, 

 the eastern volcanic cone, which carries "a small circular hol- 

 low, between 100 and 150 yards across and 35 or 40 feet in 

 depth. There was a time in its history, when the present sum- 

 mit alone appeared at the surface of the sea as a tiny ring of 

 coral reef, capping a submerged volcanic peak, the remains of 

 which still exist in' the shallow basin on the highest part of 

 the island " (op. cit., p. 113). I think we are well justified from 

 this evidence in assuming that the large breached-ring is sim- 

 ilarly only an upgrowth from a larger crateral border, upon 

 which the small cone is perched. Mr. Darwin early recognized 

 the possibility of such a structure, and he guardedly affirmed 

 his belief that under suitable conditions a " reef like a perfectly 

 characterized atoll " might be formed over the rim of a crater 

 (" Structure and Distribution of Coral-Reefs," 1842, p. 89). 



It is therefore in no way surprising that the thickness of the 

 coral-made rock on this island should be comparatively sligh t, 

 and nowhere exceeding 150 feet. 



