The magic angle: a solved mystery.

We resolve the long-standing mysterious discrepancy between the experimental magic angle in EELS--approximately 2theta(E)--and the quantum mechanical prediction of approximately 4theta(E). A relativistic approach surpassing the usually applied kinematic correction yields a magic angle close to the experimental value. The reason is that the relativistic correction of the inelastic scattering cross section in anisotropic systems is significantly higher than in isotropic ones.