There was been some recent discussion about the origin of the runes
over at Old Norse Net, where Tor Gjerde posted this link to his essay
on the topic [ http://old.no/runes/origin.pdf ].

Another link you might find interesting [
http://irs.ub.rug.nl/ppn/163895791 ], JH Looijenga's doctoral
dissertation: "Runes around the North Sea and on the Continent AD
150-700." See especially Chapter 3. As you can see, not everyone
agrees that all names ending in -o in the early inscriptions are
necessarily female.

> The linguistic evidence of the example you quote is perfectly correct,

being gmc long /o/ originated by long Indo-european /a/, which is the
feminine ending in many ie languages even today.

Yes, normally PIE /a:/ became /o:/ in Germanic. But at the end of a
word, in the language of the early runic inscriptions found in
Scandinavia, Proto-Germanic /o:/ is represented by the u-rune, e.g.
'laþu' "invitation" > Old Norse 'löð'. Thus, in the language of the
early inscriptions, -u was the nominative singular ending in the
declension that corresponds to the Indo-European a-stems. This sound
change, final /o:/ > /u:/ > /u/, was shared by the Germanic dialects
ancestral to Old English and Old Norse.

Names like Wagnijo and Niþijo are usually interpreted as ôn-stems.
The /o:/ was preserved because it was 'covered' (protected) by a final
/n/ which existed at the time of the change /o:/ > /u:/, but had been
lost in the nominative singular by the time of the inscriptions. The
/n/ survived though in oblique cases in the early inscriptions, as in
the genitive Agilamu(n)don (Rosseland stone); I(n)gijon (Stenstad
stone). This declension went on to be used exclusively for feminine
nouns in Old Norse, but in West Germanic the ôn-stem declension was
used for masculine nouns.