PCA. *tōtawäwa he treats him so: O. tōdawād: nindōdāzu I treat myself so.

with *-e0e- I . . . . thee, independent mode; here awe gives ō:

PCA. *wentamawäwa he tells it to him (*wē- name, *-nt- action on inanimate object, *-amaw- to or for animate object): F. witamawäwa, M. wehtamowäw, O. windamawād: PCA. *kewentamōle I tell it to thee: F. kewitamōne, M. kiwēhtamun, O. kiwindamōn.

17. After consonants the semivowels are subject to various changes, some of which appear to antedate the PCA. time:

18. PCA. postconsonantal wa occurs in word-final, but this is hidden in all but F. by the loss of final vowels:

PCA. *a@ankwa star: F. anagwa, C. atāhk, M. anāh (plural anāhkuk), (. anāng.

However, in two-syllable words with short vowel in the first syllable, where all preserve the final vowels, PCA. wa appears clearly: PCA. *ihkwa louse: F. C. O. ihkwa, M. ihkuah.

In other positions it seems that postconsonantal wa had become ō in pre-Central-Algonquian, but that already in PCA. wa had in most cases been re-created by analogy. The distribution of ō and wa differs in the dialects:

PCA. *a@ānkōki stars (stem *aṇānkw-, *-aki animate plural): M. anāhkuk, O. anāngōg—analogic F. anagwagi, C. atāhkwak.

PCA. *me'tekōki trees, *me'tekōli sticks (stem *me'tekw-, *-aki animate plural, *-al inanimate plural): F. mehtegōni, O. mihtigōg— analogic C. mistikwak, mistikwa, M. me'tekwak, me'tekwan.

19. Postconsonantal ya had been changed in pre-Central-Algonquian time to yä, which appears in F. and in the O. reflex i (by §26):

PCA. *a'senyäki stones (stem *a'seny-, animate plural ending *-aki): F. asenyäni (inanimate form, PCA. *-alı), O. asinīg; analogic forms C. asiniyak, M. a'senyak.

20. Postconsonantal wā is preserved, except that M. has a falling diphthong ua, which alternates with wa as long with short.

PCA. *akwahwäwa he draws him from water by tool (*akwa- from water, *-ahw- act by tool on animate object): F. agwāhwäwa, C. akwāhwäw, M. akuahäw (h analogic for hw in certain forms of this verbal type), O. agwāhwād.

21. Postconsonantal ya lost its y before PCA. time after w. This is clearest in F., where, for instance, a suffix -ya- of descriptive inanimate verbs gives the forms:

tahkyawi it is cool (PCA. *tahk- cool),

mōwāwi it is soiled (PCA. *mōw- soil),

meckwāwi it is red (PCA. *meçkw- red).

In other positions M. has the falling diphthong ia parallel with ua and alternating with ya. C. loses the y in most positions, O. probably in all:

PCA. *-myā- smell: M. wihkimyakwat it smells good (PCA. *wenkgood taste or smell, *-ekw- undergoing, *-at- inanimate intransitive), C. wihkimāmäw he likes the smell of him (*-am- eat, smell animate object), O. pīdjimāmād he scents him (PCA. *pyät- hither).

PCA. *nyānanwi five: F. nyānanwi, C. niyānan, M. nianan, O. nānan. After tc (affricate, as in English chin) and c (abnormal sibilant, as in English shin) the y is lost in all but M.:

PCA. *nenetcyānehsa my child: F. nenītcānesa, M. ninītsianeh, O. ninīdjānis.

PCA. *icyāte if he goes there: M. isiat, O. ijād.

22. For combinations of semivowel and high back vowel there are few examples. The combination of *me'tekw- wood (§18) with *-ō0canoe shows kō for kwō:

PCA. *me'tekoci dug-out canoe: M. me'tekōs, C. mistikōsi; cf. F. wämehtegōciha Frenchman (boat-person).

23. PCA. we existed in final position, to judge by F. forms such as apehtawi-kicegwe halfway up the sky. Between consonants we does not occur; where it is morphologically postulated, u appears in its stead, and so it must have been in PCA. Thus the initial element

PCA. *pōxkw- break: F. pōhkwāwi it is broken, C. pōskwāw; C. pōskwaham he breaks it by tool, M. pühkwaham,

combining with the transitive verb final PCA. *-en- by hand (illustrated in §16) gives:

PCA. *pōxkunamwa he breaks it with his hand: F. pōhkunamwa, C. pōskunam, M. põhkunam,

and with the transitive final PCA. *-eck-aw- (example in §16) it gives:

PCA. *pōxkuckawäwa he breaks him with his foot: F. põhkuckawäwa, C. pōskuskawäw; cf. M. põhkuskam he breaks it with his foot; O. põhkuckā it is broken.

24. PCA. had postconsonantal ye in final position, to judge by F. forms such as ahkwitc-asenye on top of the stone. Elsewhere we meet i for morphologic y plus e, parallel with u for we (§23). Thus there is a local noun-suffix *-enki, as:

PCA. *wekenki in his house: F. wigegi in the house, C. wikihk in his house, M. wēkih; cf. O. adōhpōwining on the table.

Combined with the noun-stem PCA. *a'seny- stone (§19), this gives: PCA. *a'seninki on the stone: F. asenigi, M. a'senih.

25. Postconsonantal wä appears in M. as ī:

PCA. *pyätwäwäkesiwa he is bound hither with noise (*pyät- hither, *-wäwä- noise, *-ak- post-medial, *-esi- animate intransitive): F. pyätwäwägesiwa, M. pītīwäkesiw; cf. O. pīdwäwäcin he steps hither with noise (PCA. *-hc-in- fall, lie, animate).

PCA. *ixkwäwa woman: F. ihkwäwa, C. iskwäw, O. ihkwä; in M. only as non-initial stem, e.g. umä'numiniahkɩw Menomini woman.

26. PCA. postconsonantal yä loses its y in C. and appears as ī in M. and O.:

PCA. *nyäwwi four: F. nyäwi, C. näwu, M. niw, O. nīwin.

After c F. and O. also lose the y:

PCA. *acyä- backwards: F. acäyāmōwa he flees back, C. asänam he hands it back, M. asīnam, O. nindacänă I hand it back.

After tc F. loses the y, but O. goes with M. in reflecting it: PCA. *wetcyäwäwa he accompanies him: C. witcäwäw, M. witsiwäw, O. wīdjiwād; cf. F. witcäwäwa he goes along with people.

27. PCA. postconsonantal wi appears in C. as u, in M. as ɩ. Its preservation in word-final can normally be seen only in F., as:

PCA. *me'tekwi stick: F. mehtegwi, C. mistik, M. meʼtek, O. mihtig. Disyllables with short first vowel show it everywhere:

PCA. *meçkwi blood: F. meckwi, C. mihku, M. mehkih, O. mick wi. Within the word, PCA. *pōxkw- (§23) followed by connective *-i(§9) appears in F. põhkwinehkähwäwa he breaks the other's arm (PCA. *-nexkä- arm, non-initial, *-ahw- act by tool on animate object), M. pühkikatähäw he breaks the other's leg (PCA. *-kātä- leg).

PCA. *takwihcinwa he arrives where the others are (*takw- with others, *-i- connective, *-hc-in- fall, lie, animate): C. takusin, O. tagwicin; cf. F. tagwicimäwa he places him with the others (PCA. *-hc-im- lay an animate object), M. takīkāpuwiwak they stand grouped (PCA. *-kāpawistand).

The combinations twi, lwi were probably replaced by ti, l before PCA. time (§13). Hence forms like the following are probably analogic reformations, which may however in part go back to PCA. time:

F. menwisenwi it lies well; this would be PCA. *melwih@enwi; M. menihnen could be this or the phonetically demanded PCA. *melthoenwi. Similarly C. miyusin he lies well would be PCA. *melwihcinwa, while M. menīhsin could be this or the proper PCA. *melchcinwa. Forms like F. tātwineciwäcinwa he falls tearing his testicles (PCA. *tātw- tear, *-0e'ciwä- testicles) or F. pyämigatwi it comes are too plainly demanded by the analogies of the language to indicate PCA. twi.

28. PCA. postconsonantal wē appears as ī in M.:

PCA. *apweyi canoe-paddle: C. apwiy, M. pih (plural piyan), O. abwi.

PCA. *kwe@umäwa he longs for him: F. kwinumäwa, M. kīnumew. PCA. *kwerkwihcimäwa he whistles for him: M. kiskihsimäw, O. kwickwicimād.

PCA. *kwenkwēhca bird-name, after its call: M. kihkih mud-hen (plural kihkihsak), O. kwingwīci magpie (suffix *-iw- added).

We have seen that twē was replaced by ti, probably before PCA. time (§14). A further complication in the treatment of postconsonantal wē seems to be that in some surroundings it was replaced by ō; this also was probably due to a sound-shift which had occurred before PCA. time.

The words just quoted do not fall under any apparent analogy; they would seem to show that we was preserved after p and k. Examples of ō replacing postconsonantal we are furnished, e.g., by the suffix PCA. *-ewi- forming denominative verbs, as

PCA. *namä'sa fish: F. namäsa, M. namä's: PCA. *namä'sēwiwa he turns into a fish, is a fish: F. namäsiwiwa,

M. mōhkumān American (loan-word from other Alg. language): mōhkumānēwiw he is like an American, he votes.

Where this suffix is added to stems in consonant plus w the result is ō: PCA. *-axkw- wood, solid: F. cegwāhkwa pine: pagānāhkōwiwa he is a walnut tree.

PCA. *-a'@emw- dog: M. wāpeska'nem white dog (pl. -uk, by §18): mihkwa 'nemōwiw he is a useful dog.

Similarly the diminutive suffixes -eh- (F.) and *-ēns- (M. O.): F. namäsiha little fish, M. namä'sēhseh (latter with a second diminutive suffix *-ehs- superadded),

but from stems in postconsonantal w:

PCA. *a@emwa dog: M. anäm (pl. -uk): anämōhseh little dog; cf. F. anemōha dog.

PCA. *amexkwa beaver: F. amehkwa: amehkōha little beaver. PCA. *axkehkwa kettle: M. ahkäh (pl. -kuk), O. ahkihk: M. ahkähköhseh little kettle, O. ahkihkōnz.

But F. has also forms like pemitāhkwīhi collar-bone (PCA. *-āxkw-), which are perhaps to be taken as survivals; in general the habits of adding common suffixes have been regularized in the sense of always using ō.

29. PCA. postconsonantal yē did not exist; this sequence had been changed in pre-Central-Algonquian to consonant plus ī (§14).

Of these only the last three were voiced. The rest are everywhere lenes. Non-initially they are voiced in O. (b, d, g, dj, z, j,-where dj represents