A couple points to add to this discussion...

1) How does the use of rudder affect the g-loading on the wings? A
typical snap roll, in competition, anyway, is initiated by the rudder
after the pilot has loaded up the wings. This loading is less than that
required to produce a stall.

2) Snap rolls cause twisting loads and gyroscopic effects that do not
register on a g-meter. Are those loads significant? How would one
determine what those twisting limits are?

3) Reducing the aircraft's weight will reduce the load on the wings, but
it does not affect the load on non-lift-bearing parts of the aircraft,
like the propeller, engine mounts, and pilot seat.

Bottom line for me is to read the aircraft flight manual and reduce the
published limits by a percentage equal to the square root of the age of
the airplane + the cube root of the age of the pilot.

If there are no published limits, do a good preflight on your parachute
and be sure you will have an easy egress.


--
Dennis Yugo
http://www.worldpassage.net/~dyugo