60 ON THE OKIGIN AND MIGEATIONS OF THE POLYNESIAN NATION. 



deity. The walls are of solid stonework, in height 12 feet. 

 They are even and regularly paved at the top. On the top of 

 the walls (which in some places were 10 or 12 feet thick) 

 the warriors kept watch and slept. Their houses were built 

 within, and it was considered sufficiently large to contain the 

 whole of the population. There were four principal openings 

 in the wall, afc regular distances from each other, that in the west 

 being called the King's road. They were designed for ingress 

 and egress ; but during a siege were built up with loose stones, 

 when it was considered d.pari haabuca, an impregnable fortress."* 



Considering that the normal state of the South Sea Islands 

 has from time immemorial been that of civil or rather inter- 

 necine war, there is no point of comparison between the 

 Polynesians generally and the Indo-Americans more interesting 

 than that of their fortifications. Those of the Indo-Americans 

 appear to have been generally formed of mounds of earth — a 

 mode of formation well adapted for such localities as the alluvial 

 banks of the Ohio, the dead levels near the lakes of Canada, or 

 the elevated plains of Central America, but not at all adapted 

 for the South Sea Islands. My talented townsman, the late 

 John G-alt, Esq., of G-reenock, Scotland, the author of a whole 

 series of popular works of fiction about half a century ago, and 

 father of the late Premier in Canada, has told me that he 

 had seen the remains of an Indian fort on the summit of a preci- 

 pitous ridge near Lake Simcoe, in Upper Canada. It consisted 

 of a mound of earth, enclosing a considerable extent of ground ; 

 but on the banks of the Miamis River, much farther to the 

 southward, the Indian forts had been constructed of stone. 



iS'ay, the march of ancient civilisation among the Indo- 

 Americans may even be traced, in some measure, by those most 

 interesting remains. In South America I have not heard of their 

 being found to the eastward of the Andes. The gloomy forests 

 of Guiana and the Brazils were e^idently unfavorable for the pres- 

 ervation of Indo-^A merican civilisation ; and the portion of the 

 race that wandered into these vast solitudes was necessarily 

 broken up, at an early period, into an infinity of insignificant 

 tribes that could hold little or no communication with each other, 

 and that, consequently, very soon sunk irrecoverably beneath the 

 level of the rest of their nation. But the regions of Central 

 America, the elevated plains of Bogota and Cundinamarca, the 

 open valleys of Peru, and the lofty and secluded but highly fertile 

 tracts of Chili, were much more favorable for the formation of 

 powerful states and empires ; and it is, accordingly in these por- 

 tions of the continent of South America that the ruins of ancient 

 cities and of extensive fortifications are found. In the Xorth 

 American continent, the course of the Mississippi and its tribu- 



* Ellis : Polyuesiau Researches iv, 459. 



