--- In cybalist@..., "Piotr Gasiorowski" <gpiotr@...> wrote:
> I didn't mean _all_ sonorants. I think it's arguable that RUKI
operated normally before nasals (e.g. before the *-n- of verb
suffixes). But does *re^xU really reflect *re:ksm? I'd expect *I, not
*U from final syllabic *m. A secondary thematic ending *-om is
ususally reconstructed there (*re:k-s-om), older than the sustitution
of *re:k-s-s and *re:k-s-t with *rek-e-s and *rek-e-t (or maybe
restituted *rek-e for *re: after the loss of the final consonants).
>
Sorry, it was just a hasty thought of my.
I realize that *x- in anlaut is problematic by itself (<*sk-?
expressive? onomatopoeia?), but still what about *xromU 'lame,gammy'
and (if not of merely onomatopoeic origin) *xro,stU 'crunch;sticks,
falen deadwood etc', *xre,stjI 'cartilage;beetle etc'.
Sergei