In section 11.2 of my outline of Central American Island Carib (IJAL 22.35) I pointed out that the prefix h(31) and the suffix -i -o -u (359), markers of the nominal absolute which today are different morphemes (since either, neither or both may be present in the absolute form of a noun), must have arisen from the splitting of a suffix that Breton, in his description of the early Dominican dialect, wrote -heu (= my /ho/); thus: halao < alaho seat (bench, stool, chair, etc.) and hitao < itaho blood, whose dependent forms were and remain, respectively, -ala and -ita in both dialects. So far as my own data are concerned, the modern suffix occurs only after final /e/ or /a/ of the complete stem (including other derivational suffixes), with which it constitutes one of the falling diphthongs /ei/, /ao/ or, more rarely, /eu/ or /au/. However, according to Fr. John J. Stochl, S. J., founder of the Carib Language Group of students at St. John's College, Belize, it also occurs after stem-final /i/, /u/ (and /o/?) as an element lengthening these vowels; so, e.g., he writes (using y to indicate [i']): eb6gy -ebegi cost, where I heard no difference between the final vowel sounds of absolute and possessed forms. Now while dissyllabic, geminate vowel clusters do occur in this dialect (aliiha to read has four syllables, the second /i/ being heard as an echo vowel; v. IJAL 21.235), I did not find vowel length to be distinctive. It may well be that Stochl's ebegy is phonologically eb6gii; and that the second element of a geminate vowel cluster loses its syllabicity in word-final position. Or the lengthening may be 'theoretical' rather than 'actual'; as also nasalization seems to be in some forms (v. IJAL 22.7-8, 33-34, and 24.40). But the presence of the suffix -i -o -u in these positions is at all events historically justified if we may judge by the occurrence of its correspondent in the recent Dominican dialect. So, word-lists of the latter show: uihi -uini meat (flesh or fish or personal consumption) corresponding to Stochl's uii ~ -ui (my spelling) and to my own di ~ -ii or -di, abuhu -abu bone(s) corresponding to both Stochl's and my abu -bu, and halaho -ala seat (v. first paragraph above) corresponding to both his and my halao -ala. It seems clear that Stochl's uii (uwy) meat and nui (nuin) my meat are more 'correct' than my ui and nui (which latter I heard more frequently than nii); and while I have native testimony as to the 'equivalence' of the nii / nui variation, failure to record uii instead of (or as well as) ui may possibly be due to mishearing on my part. But I strongly suspect that we have here an instance of change in progress; and that both nasalization of the final vowel in possessed nouns like nui and its gemination (or lengthening) in their absolute forms (like uii) are as much 'on their way out' as is the apical spirant of Spanish words in -ado (amado, hallaco, etc.), though like the latter they are still capable of what J. H. D. Allen Jr. has recently called 'reconstitution' in emphatic or very deliberate speech (v. WORD 12.251 ff.). Recent Dominican halaho seat, as compared with earlier alaho and with modern Central American halao, suggests that initial aspiration of such absolute nouns preceded loss of /h/ from the suffix. And Rat's recent Dominican hita blood, as compared with earlier itaho and with modern Central American hitao, at least shows that this loss did sometimes occur in the former dialect. It also shows-if this word really ended in /a/ (and not in a diphthong)how what at first had been a stem-alternant, hita-ita, became a complex word in which htook over the function of the lost suffix. A still more curious instance of 'distance metathesis' with displacement of an intervocalic aspirate is recent Dominican NO. 4 323