> An interesting set. What's the feminine plural - _treze_, _breze_?
Yes.
> I don't have smoothing by final -i in my rules, but it's not needed
> for words of Latin origin. Are we into the region of arbitrary
> application of rules? We have the example _mare_ 'sea',
> _mãri_ 'seas', but _gras_ 'fat m.s.' _gra$i_ (not *grã$i) 'fat
> m.pl.'.
zare, zãri; garã, gãri; narã, nãri; cãma$ã, cãmã$i (regionally
also cãme$ã/cãme$e, cãme$i & chime$e, chime$i); casã, cã$i ->
this one only in a colloquial+dialectal+"uneducated" style:
otherwise casã, case (masã, only mase, never *mã$i; plasã, only
plase, never *plã$i - yet exception: when plasã means "subdivision
of a judetz (county)"! Etcetera.
(Well, you have to learn 'em; the rules aren't of 100% help.
Similar correspondences in words ending in -Vscã, -V$ti + -V$te
versus -Vsce: muscã, mu$te; gâscã, gâ$te; gãlu$cã, gãlu$te
{{some native speakers in some regions, esp. in Moldavia, are
tempted to make such plurals: mu$ti (even creating homophonous
or semi-homophonous aspects: mu$ti "you bite," mu$chi "muscle"),
gâ$ti, gãlu$ti -- in some cases such -$ti plurals are pan-
Romanian & high-style Romanian, as in femeiu$cã, femeiu$ti, a
(at times pejorative) diminutive of femeie "woman" and in
the following 5:}}; mascã, mã$ti; cascã, cã$ti; pu$cã, pu$ti;
bascã, bã$ti ("Baskenmütze; beret"); pascã, pã$ti ("matzen,
matzot"); but... broascã, broa$te ("frog" & "lock"); frescã,
fresce :o))
> Actually, there may be a minor rule here - wild card searches on ?as
> and ??as yielded *no* examples of the pattern -as, -ã$i, but several
> counterexamples. I was searching DEx on-line at
> http://dexonline.ro/ .
Adjectives ending in -as are rara avis. OTOH, you'll find
plenty of them ending in -a$, -e$, -o$.
> Richard.
George