Now that I'm back home with a full toolset...the plot thickens.

I hooked up a timing light and tried to crank it -- got a few flashes
and the engine sputtered a bit, then no flashes.   So it looks like
my gas-station-parking lot conclusion that I "had spark" was wrong --
spark is, at best, intermittent, and may be nonexistent most of the
time.

Fiddling with the distributor, i realize it's loose and I may have
messed up the timing.   I mechanically set it back to TDC according
to Bently, but something doesn't look right.   To get the timing set
to TDC on #1, I have to basically put the distributor at one end of
its range of motion.    Suddenly a lot of things start making
sense.   Why does the (basically new) cap show a lot of burning &
arcing, and why does the arcing on the rotor look like it's at the
leading edge instead of in the middle?   Why does it run so poorly at
idle, but fine at higher rpms?  I pull the distributor and realize
that the PO (or some PO's mechanic) installed the distributor drive
gear wrong!   The slot is pointing at the distributor lock down stud,
NOT at the stud about 30 degrees counter clockwise.

So my new theory is this:  with the distributor gear mis-aligned, the
only way to get the timing right was to have the distributor rotated
to the end of it's travel.   This, however means that the rotor is
not actually lined up with the cap when the spark happens, requiring
the spark to jump inside the cap, giving me erratic spark at idle.
At higher RPMs the mechanical and vacuum advances kick in, and the
rotor becomes aligned properly with the cap, resulting in better
spark pathway.

This doesn't explain why the car suddenly died on the freeway.
Perhaps the loose distributor timing slipped?  Or maybe all the
arcing inside the cap & rotor has killed the hall sensor and it's a
interactive problem?   In any case, now that I've got the distributor
out I'm going to see if I can find a rebuilt one as this one looks
like hell.

Stay tuned, and thanks for the help!