148 



EASTER ISLAND. 



Table 29. 



From this table we might derive the conclusion that there was a 

 widely severed division in Southeast Polynesia when referred to the basis 

 of recognizability in other languages of the Polynesian family ; that the 

 Paumotu stood in one group in the possession of the general treasure of 

 Polynesian speech exceeding a moiety of its own speech equipment; 

 that, unevenly spaced geographically about this central archipelago, 

 four groups of islands shared the Polynesian stock to a fairly even 

 extent, that is to say to the extent of rather more than a quarter of their 

 vocabulary, yet not quite reaching so high as one- third. Thestudyof the 

 tabulations in the foregoing chapters specifically dealing with four of 

 these archipelagoes should have served to indicate the fallacy of such an 

 opinion, the impropriety of this transaction in gross. We have referred 

 once and many times again to the division of Polynesian speech between 

 the Proto-Samoan and the Tongafiti, and to the mingled yet indiscrim- 

 inate stream which has flowed over Southeast Polynesia and which we 

 have been forced to designate as general Polynesian. The only use 

 which it is permitted us to make of this table is the consideration of its 

 unexpressed member, the unidentified element. 



Here we have something positive to deal with, if we may. It is that 

 in the Paumotu slightly less than half of the speech equipment, in the 

 other subdivisions of the province a shade more than two-thirds of the 

 speech equipment, are wholly unidentifiable elsewhere in Polynesia ; so 

 far as we, or any, know, they are quite unknown anywhere in the world. 

 This factor we may consider with profit. 



Let us, in a single concrete instance, look further into this. In the 

 Rapanui we find the vocable reirei in the sense of trampling or pound- 

 ing. It lies within the unidentified two-thirds. Nowhere in Southeast 

 Polynesia is any mutation variant of reirei found to carry any such 

 signification, nowhere in the distal migrations north or south, nowhere 

 does it appear in the Tongafiti stream, nowhere in the earlier Proto- 

 Samoan stream ; collation item by item of a dozen Polynesian vocabu- 

 laries does not disclose it ; in the similar study of more than a hundred 

 Melanesian tongues not even so much as its wreck is found as drift and 

 jetsam on that traverse of Proto-Samoan fleets almost twenty centuries 

 ago. It is found but here, on this rocky islet and the orient extreme of 

 all Polynesian wandering, if Sirrah leyop&vov its singularity is so pro- 



