Scope of i-umlaut in Old English

I have a question about the scope of i-umlaut in Old
English. I'm afraid it is very basic, but the historical
grammars aren't very clear on the subject. (Just as they are
generally not concerned, it seems, with the relative
chronology of sound changes.)

Basically i-umlaut applies only to stressed vowels, and to
post-tonic *o: in *lo:ko:jan > *lo:ke:jQn > _locian_ etc.
But what about other post-tonic vowels, and pre-tonic
vowels? There are both _fyr(r)_ and _feor(r)_. Would these
be a (secondary?) stressed and unstressed variant
respectively?

BTW does the unstressed final *-an really go to *-Qn
and then back to -an, or is it only a spelling issue.
In any case it must be the nasal that prevents fronting
of the *a.

--
/BP 8^)>
--
Benct Philip Jonsson -- melroch at melroch dot se

"Maybe" is a strange word. When mum or dad says it
it means "yes", but when my big brothers say it it
means "no"!

(Philip Jonsson jr, age 7)