Re: [tied] Early PAlb Depalatisations of k', g' > k, g

elmeras2000 wrote:

> Is it really possibly to get Gk. kri^ and kri:thé: out of something
> with a syllabic r?

I'd prefer a derivation from full-grade *g^HrezdH- with the sporadic
raising of *e before some clusters (you know the showcase examples). How
else can one connect it with <hordeum> and Germanic *gersto: (the latter
presumably with a new e-grade)?

Piotr