Migrations in the Pacifie Ocean. 331 
Tangiia, or Tahitian party, and may at that time, have preserved some 
peculiarities of their original tongue which were afterward lost in Raro- 
tonga, on a more complete intermixture wiih the Ngati-Karika, or 
Samoan party. 
“In the foregoing traditions, the existence of cannibalism at a very 
early period will be noticed, as also the custom with conquered chiefs, 
of betaking themselves to the open sea to escape the vengeance of their 
adversaries. It was in this manner that Temoe or Crescent Island, a 
coral islet which lies about thirty miles to the southeast of the Gambier 
group, was peopled between sixty and seventy years ago. A defeated 
party fleeing from Mangareva, were drifted to this island, and remained 
there with their descendants, till the arrival of the Catholic missionaries, 
who, hearing of their situation, sent for them and restored them to their 
original homes. This well authenticated fact shows the manner in 
which most of the South Sea Islands have probably received their first 
inhabitants.” 
We have followed Mr. Hale sufficiently far to show the char- 
acter of his argument, and shall but briefly notice the remaining 
pages. At the Paumotus, there is the same evidence with regard 
to the origin of the present inhabitants. Yet he remarks that 
they have an element in their language which indicates a com- 
mingling with some different race. At New Zealand, it is the 
traditionary belief that the first nations came from Havaiki, a 
country to the east, and the names of the canoes and first emi- 
grants are retained by them. From the variations in the dialect 
from the original type, it is inferred approximatively that the set- 
tlement of New Zealand took place nearly at the same time with 
Tahiti, or probably not less than 3000 years ago. The names 
for the months in different dialects, for the winds, and for nu- 
merals, affords confirmation of the general conclusions adduced. 
With regard to the Tonga Islands, he finds reason for tracing 
them direct from the East Indies through the Feejee group, and 
attributes some of the characteristics of the Feejee natives to a par- 
tial mixture of Malaisian with the original Papuan of those isl- 
ands. Much that is interesting on this point might be cited. 
The Kingsmill or Tarawan Islands, both by tradition and lan- 
guage are connected with the Samoan on one side and with Ba- 
nabe, one of the Philippine Islands on the other. 
These various conclusions are laid down on the chart, (a reduc- 
tion of Mr. Hale’s,) which accompanies this article. All the vari- 
ous groups are seen to trace their origin more or less directly to 
the Samoan or Tonga Islands. 
