Hi PeteJD,

To answer your question the fluctuation cycle of the coolant raising to the top of the rad, falling several inches and then raising again to the top of the rad is roughly 15-20 seconds. Bear in mind this happens with and without a thermostat.

To bluebirds point it could be air pockets, however on other applications I have only experienced trapped air in a cooling system when the rad filler neck was not at the highest point of the cooling system. That's not to say that there's not air in the system but I think it would work it's self out fairly easy considering the configuration of the cooling system on this application.

These cooling system are very simple, I have never experienced this anomaly on any other application.

I would ask any member that has a similar application, Pontiac 350.
Do the following on thier own vehicle and report back the findings.

Remove the cap from the rad measure the approximate distance from neck of the rad to the fluid level, start the engine observe/record any significant changes from ambient to steady state. Record the steady state temp observe and record coolant level several for minutes. Place the cap on the rad squeeze the top rad hose record inches of squeeze to full pinch.
Let run for 10 min periodically squeezing the hose and recording.

Just to be clear I fully understand the expansion and contraction of fluids as they heat and cool, this is not the question of this thread, rather the the erratic fluctuating coolant level.

As reference on any vehicle I have that doesn't have a coolant overflow tank I do a simple calculation to provide the appropriate air compensation to prevent overflow dV(volume change)= V(volume of coolant system) B(volumetric temperature expansion coefficient)((T1(final temp)-T0(initial temp). So for this application dv= .016088M^3*.00057(104C-21C)=.0007m^3 or 26 ounces. So in this case with engine off I filled the system complete to the top of the rad then opened the petcock at the bottom of the rad and pulled off a little less than a quart, made a mark on the rad to note the proper level for the system once burped, started the engine and this is when I noticed the issue with the fluctuating.