At first I was too busy to raise the
gear, then I chose to leave the landing gear down to keep the CG
from moving any further aft.

As we climbed slowly to 3000 feet, I started the transfer pump to move
fuel from the wings to the nose, and headed for a nearby airport with
a longer, wider runway (5000 feet), as I wasn't sure I could safely land
where I had launched. Along the way, I explained the situation to my
passenger then had him drag my flight bag from the baggage area to his
lap, while I leaned forward over the stick. Once the flightbag (10 lbs)
was moved to the seat, the pitch oscillations completely stopped, and I
could comfortably trim the plane to fly hands-off. The flightbag moved
the CG forward no more than about 0.2". Fuel transfer might have also
contributed a little.

Bob,

 

I have been a bit busy but I am glad Mark (Speed Demon) Ravinski replied and I agree completely with him. 

 

Speed makes things better so leaving the gear down is a bit of a problem.  I thought that throwing out the gear might be something to do if you can't get out of a rear-CG exacerbated flat spin.  However, I haven't been there and I am not interested in doing that sort of maneuver.

 

Leaving the flaps in "takeoff" position (or slightly less) might be a better solution.  Since the 235/320/360's experience such a strong nose down pitching moment when the flaps are taken out of reflex, such an action might "trim out" some of the rear CG pitch instability without adding too much drag.  I have anecdotal information that this procedure seemed to add pitch stability to an overloaded rearward-CG 360 with wing extensions.

 

Perhaps someone would perform this sort of experiment on their next ill-conceived overloaded flight (Bill H. excepted).  Or, better yet, someone has already done this and will report how it worked out so that we might learn.