From: Miguel Carrasquer
Message: 27172
Date: 2003-11-14
>As I already posted to Piotr, I think *okto:u is the source of a _later_I did consider analogy from *ok^tó:(w), but I just didn't feel that was
>analogical
>spread of accent to the second syllable in *septm. We can conceive of "7"
>and "8" as forming pairs when counting and "9" and "10" do too. Miguel is
>too caught up on numerals ending in *-m to see this natural counting
>pairing,
>and the quite apparent cause of accent shift.