From: P&G
Message: 26251
Date: 2003-10-05
>There is almost a conspiracy on the part of Indo-Iranian to not reflectYes. As for h2o, it's a shame that the first syllables in
>roots of the structure *oC- (presumably mostly *h3eC-, although instances
>of *h1oC-, *h2oC- or *h3oC- cannot be excluded).
>If we look in Pokorny, weNor for *kob fit, *gol branch, *k'olemos rod, *kol(e)y lime, *kom beside,
>see no Indo-Iranian reflexes for ...
>and we have only theAnd of *sol(e)wo safe (Skt sarva)
>closed-syllable variants of ...