*gwh in Gmc.

[David L. White]

        I agree with the various people (about three of them, I think) who
have written that the difference between /n/ and /m/ before /f/ is probably
not significant.  I would add that it is possible that at some point the
vowels in question were merely nasalized (in phonetic output, I mean, not
phonemically), more or less as in modern English "mountain", which in
ordinary speech does not have a real first [n] in it, just as nasalized
[au].  That might explain how the effects of a stressed vowel, as seen in
Verner's Law, might operate across a phonemically (but not phonetically)
present nasal, a phenomenon which otherwise seems a bit odd.  Then again,
maybe not.
        /pw/ does not seem likely.  Labialized labials are disfavored for
fairly obvious phonetic reasons.  Even pharyngealized labials are
disfavored, merely because the acoustic effect of pharyngealization is
somewhat similar to lablialization.  In nearly two hundred years of IE
linguistics no need has been perceived to posit labialized labials.  It
could be objected that /pw/ is not a labialized labial but a sequence, but
more or less the same phonetic considerations apply.  Note that in modern
English we permit dentals and velars before /w/, for example "twelve",
"dwarf", "thwart", and "queen", but not labials, save in very recent
non-native acquisitions like Swahili "bwana".

Dr. David L. White