Miguel & dentis

I researched a bit about that ominos PIE radical propposed by
Miguel "g^ombh" .There is no one. At least not in my Pokorny-s
list.
But is one there how I said that it should be , with "e" not
with "o". So we have the radical "g^embh"
And in this case the game changes. Because if "o" nevery
diftongued to "jo" but to "oa" , here Miguel was indeed right,
the "e" diftongued to "y" and here i get my g^>z in zimtz.
It remains only the "ts" hard to explain because bh>b in
romanian substrate here.
g^embh>zimb>zimtz ( where ts is the weak point).

From the same radical seems to comme the rumanian word for "to
smile"= a zâmbi which in this case works with n o w e a k p
o i n t
gêmbh>zimb>zâmbi
The DEX means, romanian "a zâmbi=to smile" and the substantive
"zâmbet=smile" come from bg. "zAbja se" or from slavic
"zOnbu=teeth"
Such a good lock that in slavic was an "n" there for
explaining the "m" there.. hehe.
regards

a. moeller