544 Transactions. — Miscellaneous. 



HUA, mllh ; GAHU, damp, moist; Malay, susu, Diilk. If we 

 acknowledge that u means in a general sense to ooze, to he wet, 

 to drip, we have then to find the a of UA. "We shall be accused 

 of a poetical flight perhaps if we assert that it is probably a 

 worn-down form of eangi, the shy (Marquesan, ani), and that 

 UA may have formerly been ubangi, " heaven-milJc " (or " sky- 

 drip'"?). The comparatives in the Malay Archipelago cer- 

 tainly point in that direction. The Malay form of eangi, the 

 heavens, is lani, and we find inCajeli, ulani, ra^?^; Camarian, 

 ulani, rain; Teor, hurani, rain.; Baju, huban, rain; Wa,hai, 

 ULAN, rain; Bisaya, ulan, rain; Gah, uan, rain; Api, ua, 

 rain. Here we can trace geographically at the present time 

 each form of the word still in existence — u-lani, u-lan, u-an, 

 u-a, ua — a very extraordinary thing; and this abrasion of im- 

 portant words leads me to the consideration of the second and 

 briefer part of my paper. 



Paet II. A Possible Eeconsteuction of Maoei. 



While searching for compa,ratives of Maori w'ords in the other 

 dialects, I took each dialect separately, and, in going through 

 it, thoroughly indoctrinated myself into the systematic letter- 

 change common to that particular dialect and to Maori. Thus, 

 in reading Marquesan, wherein the letter r is almost wanting, 

 it is necessary to mentally prefix an r to any word beginning 

 with a vowel, or to insert an r between two vowels standing 

 together, on the chance that r may here be a missing letter. 

 Thus, hae, a house, with r inserted between a and e, becomes 

 HAEE, evidently the equivalent of the Maori whaee, a house ; 

 iMU, sea-moss, is the Maori eimu, moss ; &c. In Hawaiian the 

 k is lost, and, seeing pue, a hill, I know it for the Maori 

 PUKE, a hill: Ai, tJie neck, is the Maori kaki (Icaki), the 

 neck. (This in addition to the change of k for t which makes 

 the Hawaiian kakaa, to roZ^ = Maori tataka, &c.) The Tahitiau 

 has a double loss ; both k and ng are gone, and one has to read 

 in either or both : thus the Tahitian Ao, fat, is the Maori ngako, 

 fat; the Tahitian aau, the iozrc/s, is Maori ngakau, the boicels ; 

 AA, to insiilt, is the Maori kanga, to curse. So accustomed does 

 one grow^ to the reading-in of missing consonants that the eye 

 is apt to play one false, and it is difficult to write the diale(!tic 

 word without making a mistake. But reading these conson- 

 ants into the Polynesian words, and having before me con- 

 stantly the idea that the original state of language was that 

 " of open syllables of one consonant followed by a single vowel, 

 or of a single vowel," I have unavoidably come to the con- 

 clusion that, although the Maori of New Zealand is by far the 

 best preserved of all Polynesian dialects in its conservation of 

 consonants, yet it is almost impossible for one trained in com- 



