>I said "think all KsK > sK". I think kYsYkY > sYkY is likely,
> On 2008-01-22 04:38, stlatos wrote:
> > Mainly, why not *puksk^o+ > *pukk^o+ > ? first since, despite your
> > questions about it in Gmc., a very sim. rule is known in Indic. If
> > you think all KsK > sK first, I disagree, though I'm willing to ask:
> > do you have any more examples?
> *pr.k^-sk^e/o- > Skt. pr.ccHati is an obvious example.