>Have experiments been done to show that it is only a local minimum >that is reached by soap bubbles and not a global minimum or is this >just the party line?
You can make some wire frames and see for yourself. It's not too hard to create something that's obviously bistable and can be "flipped" from one local optimum to another by blowing. It's not hard to bend such a case to make one of the optima obviously better than the other. E.g. start with a cube.
