Key points are not available for this paper at this time.
Theory shows that to explain the polarization of light collisionally redistributed from the far line wings of an atomic transition, one must consider correlated events in which absorption during a collision, and propagation to the final Zeeman-state superposition at the end of the collision is important. Polarizations of up to about 40% have been measured in the far line wings, substantially confirming this prediction, and showing that scattering experiments cannot just be characterized by simple absorption or emission profiles.
Thomann et al. (Mon,) studied this question.