>that the second site uses a 'hacked' font (LAMCOUR) for phonetic
> symbols, making them obscure), you'll see that forms in -i are
widespread.
>I must admit that the Friulian forms make a stronger case for
>derivation from *dui:.
>Richard.
It's not sufficient to see/to show that there are -i-s in some
forms of 'two' in some Italic dialect.
I can show you a H in Romanian and a H in Spanish but this doesn't
mean at all that this H represent a Common Latin Evolution.
You need to show us that this is really a Common Latin Evolution
(inscriptions, texts or common phonetic rule but in this case you
need to fix the areal and the timeframe) so is not sufficient to show
us an -i in some Italic dialects and an -i in Romanian.
Best Regards,
Marius