From: Miguel Carrasquer
Message: 39857
Date: 2005-09-02
>--- In cybalist@yahoogroups.com, glen gordon <glengordon01@...> wrote:Latin ob is likely from *(h1)op(i) ~ *(h1)epi.
>>
>> Etherman23:
>> > Don't you find it just a little bit too coincidental
>> > that there's a reconstructable particle *H2ebhi?
>>
>> Shouldn't that be *h1ebHi with *h1? You're basing this
>> on the Anatolian locatives like Lycian /ebi-/, right?
>
>I just double checked and it's actually H3ebhi, as in Lat. ob.