POLYNESIAN AND MALAYAN. 169 
small and quite important group. A few final vowels exhibit a muta- 
tion to diphthongs which is wholly anomalous when referred to the 
Polynesian phonetic. All the examples which present this diphthongal 
mutation are exhibited in the following list: 
a-ay tifa tipay i-oy afi gapoy 
e—ai ate gatai 0-ao lano danao 
e—-ay fale balay malino linao 
fohe bogsay o-au lango langau 
mate matay u—hoy kau cahoy 
I can see an explanation of this movement which seems valid so far 
as it goes. It will serve excellently to account for the foregoing in- 
stances; the objection will lie in the fact that it offers no explanation of 
the many instancesin which the final vowel undergoes no such mutation. 
The Polynesian languages are under an inexorable movement toward 
open syllables. In every one, the words invariably end at their present 
stage in a vowel. But as we work backward along the track of their 
migration we find convincing proof that this compulsion is modern; it 
has become effective only since their arrival in their new Pacific homes. 
In my late studies of Rapanui I have been able to establish as fact that 
the migration which eventually settled upon Easter Island left Nuclear 
Polynesia at a period when the Proto-Samoan still retained in use its 
finalconsonants. In Polynesian loan material held by Melanesian lan- 
guages we find not only final consonants, but we find distinct evidence 
that stems ending in a vowel were abraded to establish a preferred form 
with a final consonant and that this in turn has been abraded when the 
speech fashion turned toward the open syllable; and thus we have 
exposed as final a vowel originally medial in the primal stem. On the 
other hand the Malayan languages have an equal desire for closed stems; 
we encounter many vocables whose primal open stem is now closed 
by the addition of a consonant in deference to this movement. As I 
take it, these ten words of the open stem were held by their Malayan 
borrowers as (for some reason to us incomprehensible) exempt from 
their own inclination to add some final consonant. Therefore the minds 
of the speakers were under stress to avoid the easy final consonant, to 
accentuate the fact that the final sound was a vowel, accordingly to 
reproduce this mental stress by making the final vowel more vocalic 
than it was intended to be. Why this motive has left no trace in other 
stems of open type we may not now attempt to explain. 
Thus we are naturally directed to the general treatment of the 
Polynesian content by the several groups of Malayans which possess it 
to a greater or less degree. We have just seen examples of assiduity in 
its pronunciation, evidence that the material was at least subconsciously 
felt to be alien. In the same way we find that the Polynesian content 
is held uncontrolled by the ordinary rules of Malayan grammar; it is 
almost wholly free from the incidence of the customary Malayan infix- 
ature. In the foregoing synopsis (item 83) I have pointed out the 
