RE: Dihedral
It's a little hard to see without a demonstration, but I'll give it a try. Imagine an airplane rolling to the left. In this configuration the right wing is producing more lift than the right since the ailerons are changing the camber of both wings differently. As you know, increasing lift always comes with a penalty in the form of increasing drag. This concept of adverse yaw is important since it is causing a right yaw (opposite the turn). This yawing increases the angle of attack on the low wing, effectively creating positive stability about the longitudinal axis. Does this make sense?
Yes, the transports loose some anhedral as the wings take on aerodynamic load. Most still retain some because the sweepback and high wing is effectively creating dihedral. Without adding some anhedral pilots would be fighting the aircraft just to bank it.
P.S. Campgems, are you referring to the "single" lift vector splitting into a horizontal and vertical component in a bank? Yes, the vertical component of lift decreases as bank angle is increased and this explains why elevator (to increase AOA) is needed in turns. I don't see where the components of lift and the horizon comes into play when discussing dihedral. AOA is not a measurement relative to the horizon or ground, but a measurement relative to the angle between the relative wind and the chord line of the wing.