Thanks, but I'm not convinced by these examples. The way I see it:
The Equator has very sharp dihedral and no sponsons. It means it can swing easily to the left or the right of its thrust line. I've estimated the angle between the wing and the sea to be approximately 6 degrees. When the sea gets rough, the wing can go flat, resulting in the other wing going way up. And because there are no floats at the wing tips, there is potentially nothing to prevent the wing from being submerged. Also, once the aircraft is way off balance, the engines, which are relatively central, add to the swinging.
The Seabee had floats, however small. Not only the discrepancy between the floatation line of the fuselage and that of the floats is minor (I've estimated it at 3 degrees), but the very presence of the float, however small, keeps the wing from moving any further down in case of lateral swing.
The Be-103 has a near gull-wing design. The part that is closer to the fuselage on either side is flat and acts in the same way as the Dornier sponsons. It keeps the aircraft upright in a rough sea. Because of the cranked wing, even if the wingtip came close to the water, the flat part would maintain the aircraft afloat anyway. The wing could not be submerged. The engines being well apart on either side of the fuselage, there wouldn't be such a massive weight to offset the aircraft's balance when it swings to one side, as the engine on the other side would act as a counterweight.
As I said, this is my understanding of it, but I'm no engineer and I'm open to comments and criticism, and willing to be proven wrong if I am!