Thread:User talk:CodeCat/PGmc prepositions and adverbs

I was reading Guus Kroonen's work on Proto-Germanic n-stems, more specifically the section on prepositions and the effects of Kluge's law:


 * Preposition *in, with directional adverbs in the locative, allative, and ablative: *innai, *inn, and *innanê, respectively. (I assume the final vowel is overlong, am I correct?)


 * Preposition *ub, with *uppai, *upp and *ubanê; with variants between the forms and with Gothic showing a full grade variant in *eup(p)-.


 * Preposition *uz, with *ūtai, *ūt and *ūtanê; with variants in *ut- and *utt-.

The ablatives of *utanê and *innanê he says were originally *uttê and *innê, before they took *-anê by analogy from what was originally *upóneh₁.

With allatives *upp and *inn being back-formed from *uppai and *innai.

He also mentions that West Germanic *fana/*fanê could go back to *h₂póneh₁, a zero-grade derivative of *h₂epó; mentioning also PGmc *þerh and *þurh being full and zero-grades.

I'm not sufficiently knowledgeable in this area, so I was wondering if you could share your thoughts on these?