22 Bcniicc P. Bishop Museum — Bulletin 



Table 6 reveals the fact that Vavau is more closely tied to Tongatabu 

 than is Haapai, whether the index is considered as based on total names 

 or on shared names only. The index differences are not large, being two 

 in the first instance and one in the second. They would seem to indicate 

 that intercourse between Tongatabu and Vavau has been fully as great 

 or slightly greater than between Tongatabu and Haapai. The names ex- 

 clusively shared by Tongatabu and Vavau total 212, by Tongatabu and 

 Haapai 193 ; whereas the names exclusively shared by these three groups 

 total only 132. Perhaps the explanation of this lies in the supreme poli- 

 tical position of Tongatabu, from which the emissaries of the Tui Tonga 

 were continually going forth to the other groups to collect tribute and to 

 carry out many other missions. On the other hand the fact must not be 

 overlooked that the high index figures which connect Haapai, \'avau, and 

 Niuafoou most closely with Tongatabu are to some extent due to the 

 great number of place names in Tongatabu (2264) increasing the chances 

 for a high percentage of shared names. The only group that fails in this 

 regard is Niuatoputapu, which by both reckonings (in Table 6) has most 

 in common with Vavau ; Tongatabu runs a close second — only two index 

 units below Vavau — whether it is considered from the standpoint of all 

 names or shared names only. Apparently the propinquity of Niuatoputapu 

 and Vavau may be adduced as the explanation. 



While speaking of Niuatoputapu in relation to Vavau, it might be well 

 to note the aloofness that appears to exist between Niuatoputapu and Niu- 

 afoou shown in Table 6 by the two lowest index figures, 10 and 22.5. 

 Apparently the two Niuas were concerned more with Vavau, Haapai, and 

 Tongatabu than with each other. This again would seem likely to be 

 correlated with the larger population and greater political importance of 

 the three main groups of the archipelago. Of the three main groups 

 Tongatabu and Vavau have more in common with the outlying Niuas than 

 has Haapai. 



The position of centrally located Haapai is anomalous. It has been 

 already noted that she has less in common with Tongatabu than has Vavau. 

 Just above it has been noted that she has less in common with the two 

 Niuas than have Tongatabu and Vavau. Now comes the third negative 

 correlation and by far the most significant of all: Her lack of relation- 

 ship with the remote Niuas is quite eclipsed by the aloofness she displays 

 towards her northern neighbor Vavau. Whether the indexes based on all 

 names is used or that on shared names only, the situation is the same and 

 glaringly apparent — in both Haapai has relatively less in common with 

 Vavau than with the other four groups of the kingdom, thus violating 

 expectancy based on the law of chanre and expectancy based on geo- 



