From: stlatos
Message: 67783
Date: 2011-06-14
>Baltic treated e and o differently. An e changed to o by xW behaved like any other e. Therefore, e and o had different tones in Baltic. This means different tones on PIE a caused it > e or o. Since vez^ìmas requires original ó-i-ó , o>e by Y but it retained its tone just as any other e/o/a change due to C.
> W dniu 2011-06-14 21:44, stlatos pisze:
>
> > (w opt. o>e by gYH , etc., as L genu: ; G gónu ),
>
> This is one of the showcase examples of Shindler's acrostatic neuters
> with *o in the strong cases and *e in the weak ones. As the same
> alternation is found in other environments (*doru-/*deru-,
> *h2oju-/*h2aju-), it certainly has nothing to do with the fact that
> *g^onu-/*g^enu- happens to have an initial *g^.
>