I'm beginning to think that it takes very little heating to get vapour on the suction side of the pumps. Probably especially with auto gas. The lower the pressure, the lower the boiling point, right?

Finn (still have not tried returning fuel from the "unused" pump to the tank)

Ed Anderson wrote:

 In Paul's case, I am not certain it was vapor lock since he mentioned he found the his header tank cool to the touch.  Certainly not a definitive temperature sensor - but a reasonable one.  Certainly not to be discounted as a possible source of the problem and it possibility needs to be confirmed or eliminated.