346 ON THE FORMATION OF CORAL ISLANDS. 



perfect ; while the least degree of elevation, as now 

 described, is precisely what happens where earthquakes 

 attend the eruptions of volcanoes, situated often at 

 considerable distances. 



It is possible that the volcanic power may, in the 

 particular case above named, have been exerted under 

 Eeooa itself: the nature of the summit of the hill is 

 not described by Cookor by Mariner, as both have 

 neglected to notice volcanic rocks now known to 

 exist. But whether this be the case or not, the 

 presence of a volcano in this group is established. 

 Toofooa contains one which is always burning, at a 

 distance of only seventy miles from Tongataboo. And 

 as the small island Kao, about three miles from Too- 

 fooa, is a cone ; it is probably of the same nature. 



It is indeed plain that a volcanic force has been 

 exerted very extensively in this part of the Pacific 

 Ocean. Of the Friendly Isles, of Cook's arrangement, 

 thirty-five are hilly; and so are Otaheite, Bolabola 

 and Eimeo. Though he has not mentioned volcanic 

 rocks, they occur in many places: there are three 

 burning volcanoes in the Friendly Isles: and Eap, 

 which lies to the west of the Caroline Islands, is a seat 

 of volcanic force. According to Kotzebue, earth- 

 quakes are here frequent and violent. When Ulea 

 trembles, all the Coral reefs in the vicinity are shaken. 

 In the North Pacific also, coral is found on Owhyhee, 

 far above the sea ; and in this island, Mouna Roa is 

 a volcano, as all the rest of this lofty mountainous 

 group is formed of volcanic rocks. 



These facts complete the chain of evidence in a 

 manner so satisfactory, that it is unnecessary to enlarge 

 on it. But the elevation of volcanic islands in other 

 seas illustrates and confirms these reasonings, as does 

 the history of the Italian alluvia, which, itself, receives 



