One more vote for the airstream and resulting low pressure above the wing sucking the fuel out.
This just goes to show we don’t really know how lift works.
Like this:
My real life experiment was in a PA28 when I only had about 55 hours of flying time. The Fireman in France who filled me up did not replace the cap properly he just laid it back in the hole without twisting it. My fault completely of course for not checking. I did a 30 minute flight with absolutely no fuel cap and I dis not lose any (significant) fuel but he only filled me to tabs. I suppose it could have fallen off when landing – I never found the cap at either Airport.
It was a few years ago so my memory has faded but I still can’t believe how stupid I was and also why I did not notice in flight. I certainly check the fuel caps properly now!
(BTW – I stood right next to him, looked in the tank to check the fuel was at tabs and watched him put the cap back – what I did not check was he twisted the cap shut. I would instantly spot that now as the cap not be lined up but I had only owned the plane for about 10 hours so was not that familiar with it).
A Maule will fly quite happily with no fuel cap, without loss of fuel. Don’t ask how I know, but I guess old BD designed his bushplane for the kind of numbskull who forgets… Fortunately, the caps are cheap as chips.
The fuel system of a C-172:
LeSving wrote:
The fuel system of a C-172:
I reread peter’s original post, and the caps where on and leaking. My guess is the fuel vent was blocked, and the tank pressurised, and the fuel was pushed out the stand pipe.
There is an inspection for the vent, if it is blocked and the your caps don’t leak then you might get a leak somewhere else… 
https://support.cessna.com/custsupt/contacts/pubs/ourpdf.pdf?as_id=17285
[ local copy ]
The problem with the obstructed check valve(s) was presumably that sea-level air pressure is maintained which means an over-pressure condition as you climb….I don’t think this has anything to do with fuel being forced out of the caps….it might however cause a rupture in the tank….
I believe the situation for Cessnas, or any aircraft with a BOTH fuel selector settiing is that if a cap is left off the reduced pressure in that tank will cause a cross-feed from the other tank….when the tank has then filled completely fuel will start over-boarding….thus you will eventually drain the other tank completely empty…the tank with the cap missing however should retain most of its fuel… ie the siphon will be broken once there is no more supply from the other tank…
The solution is to use LEFT or RIGHT settings instead of BOTH…of course this is not possible on the C150/152…
So that explains Jacko’s and Archer-181’s observations… sorry Ted…you’re wrong…
I am still looking for a decent explanation of the physics behind the fuel “pouring out” of a missing or leaky filler cap, out of a tank which has the fuel level say 10cm below the hole.
In that case I would not be expecting significant turbulence inside the tank due to the airflow above the hole, and obviously any air coming in elsewhere (the only possible source of which is the fuel tank vent) will be able to easily escape via the filler hole.
Depending on the fuel tank geometry and g-forces on the aircraft, there should be a good amount of sloshing.
