C/C metathesis applies when otherwise a nasal would occur in postvocalic position.
rukka 'house' rukka-kma
to:t5e 'deer, meat' to:t5e-kma

It does not occur after consonant-final stems.
ru:k 'string' ru:k-mak *ru:k-kma
hu:s 'nose' hu:s-mak *hu:s-kma
It does not apply when non-nasal-initial suffixes are added to vowel-final stems.
?ama 'body, person' ?ama-tka *?ama-kta
pire 'world, land' pire-tka *pire-kta

C/V metathesis occurs after vowel-final stems:
?ama 'body, person' ?ama-tka
pire 'world, land' pire-tka


It does not apply after consonant-final stems.
?urkan 'mortar' ?urkan-tak *?urkan-tka
lo:t 'mud' lo:t-tak *lo:t-tka

The observation that [k] occurs in both prevocalic and preconsonantal position in the morphemes above suggests that the driving force behind metathesis is not due to the potential ill-formedness of [k] in either of these positions. This is supported by the fact that the distribution of [k] is not restricted in the language more generally. It occurs word-initially, [kutra] 'belt', word-finally, [murt5umak] 'blackish ones', and can both precede and follow a contiguous consonant, [hutSekins] 'dog', [men?amatka] 'on your body'. The wide distribution of [k] may be attributed to the observation that the stop, like all stops in the language, is aspirated before a consonant, before a vowel and word-finally; that is in all positions. The observation that [k] is released even in preconsonantal position is of significance as it indicates that important contextual cues to place and manner are present in the release burst, albeit more weakly before a consonant than a vowel.

The distribution of /m/, on the other hand, is much more restricted. /m/ occurs most frequently in prevocalic position, either intervocalically or following a consonant. In postvocalic position, on the other hand, the distribution of /m/ is limited: [m] rarely occurs in preconsonantal, postvocalic position, except when followed by a homorganic consonant, e.g. [t5a:kampi] 'bring', or is part of a geminate, e.g. [t5amman] 'other side'. The limited distribution of /m/ in preconsonantal position is not limited to the labial nasal. Of the two other nasals in the language, the palatal nasal never occurs in preconsonantal position, while the apical /n/ tends to assimilate in place of articulation to a following consonant

The lack of place contrasts among nasals in postvocalic, preconsonantal position is consistent with the observation that the internal cues to place of articulation in nasals are fairly weak. As Wright (1996) notes, this factor contributes to the observation that place neutralization among nasals in preconsonantal position is not uncommon cross-linguistically. It is therefore not surprising that /m/ rarely occurs in preconsonantal position in Mutsun except when part of a geminate or preceding a homorganic consonant; that is, in positions where the increased duration of the labial gesture can serve to enhance the perceptibility of the nasal's place of articulation. With respect to contextual cues, the most powerful cues for place of articulation of the nasal are frequency and durational information included in the vowel formant transition (Kurowsky & Blumstein 1984, Malecot 1956, Repp 1986). Under the assumption that transitions are generally more amplified going into a vowel, the observation that place contrasts among Mutsun nasals are observed in prevocalic position yet not in postvocalic, preconsonantal position is consistent with this view. Thus, metathesis results in positioning the nasal consonant with vulnerable cues in a position where its perceptibility is enhanced (Hume 1997).

In both the nominal thematic and locative suffixes, the final consonant and vowel metathesize; the CCV alternant occurs after vowel-final stems while the CVC alternant occurs after consonant-final stems. Consonant/vowel metathesis appears to be prosodically motivated as means of ensuring a closed penultimate syllable (Hume 1997, for discussion relating CV metathesis see Blevins & Garrett 1997).