From: Mark Odegard Message: 5531 Date: 2001-01-15
>Truth to tell, I think *nep(o)t- was probably unanalysable in PIE. > > Piotr