6 ETHNOGRAPHY. 



attributing them to the hani, or spirits. The general opinion of 

 foreigners who have seen them seems to be that they were the work 

 of another race than that which now occupies this group. There is, 

 however, no occasion for having recourse to this hypothesis. On the 

 is'and of Ualau, three hundred miles east of Banabe, similar struc- 

 tures are in use at the present day. According to the accounts of 

 Lessou, D'Urville, and Liitke, as quoted by Rieuri, all the principal 

 chiefs of Ualau, with great part of the population, have their residence 

 on a small low islet, called Leilei or Lele, situated off the eastern 

 shore of the large island, and about four miles in circumference. 

 D'Urville *ays, "in approaching the shores of Leilei, a new scene 

 presented itself to our eyes, fine houses surrounded by high walls, 

 streets well paved, &c." . . . And in another place, " the streets 

 were bordered by enormous walls of rock, which prove that these 

 natives, slight and feeble in appearance, are nevertheless capable of 



undergoing severe labour At the end of the street, a wall 



still more considerable than those which we had seen excited my ad- 

 miration. It was not less than twenty feet high by ten or twelve in 

 thickness, and forty or fifty on each side. One cannot well conceive 

 how these people, without the aid of any machine, can transport 

 blocks so ponderous as those which enter into these constructions, 

 some of which must weigh many thousands (plusieurs milliers). It 

 is still more difficult to imagine what can be the utility of these huge 

 masses. All that I could discover was that the residences of the 

 chiefs were always accompanied by these enormous walls, which 

 seemed to be one of the attributes of their dignity, like the ramparts 

 and trenches which surrounded the castles of the nobility in the 

 middle ages." 



It seems evident that the constructions at Ualau and Banabe are of 

 the same kind, and built for the same purpose. But it is also clear 

 that at the time those of the latter were raised, the islet on which they 

 stand was in a different condition from what it now is. At present 

 they are actually in the water ; what were once paths, are now pas- 

 sages for canoes, and O'Connell says, " where the walls are broken 

 down, the water enters the enclosures." This change can only have 

 proceeded from two causes. Either the sea must have risen, or the 

 land have sunk since the walls were erected. That the sea has risen, 

 or, in other words, that the level of the entire ocean has been altered, 

 will not be supposed. But that the land, or the whole group of Ba- 

 nabe, and perhaps all the neighbouring groups, have undergone a 



