I think I may have found the problem. Being the persistent type
and double-checking that the IC is still alive I prodded around
with the scope and found that Reset was low, even though it
has a pullup. There was the minutest solder bridge to 0V under
some flux. Man, I had to look hard for it, but it's gone now and
the code flow has improved markedly. Well, it would do wouldn't
it ?

Now, Ken, should I still trust myself after working all day with a
chip that's been in flipping reset ? 8-O

BTW, if anyone does still see room for improvement in the
posted code that would be appreciated

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.