in the front
(+) and back (-) lobes can then be calculated in terms of their
projected separations s, the angle 
_0 of the
source to the plane of the sky, and the index m (see Section 3.1). To take the simplest example, with
m=1,
The RM dispersion in the front
(+) and back (-) lobes can then be calculated in terms of their
projected separations s, the angle _0 of the
source to the plane of the sky, and the index m (see Section 3.1). To take the simplest example, with
m=1,
(_+ / _-)
= (s_- / s_+) 
[(1 - sin
_0) /
(1 + sin _0)]. (1)
Even for a typical angle of 30 degrees, a Q value of 3
(in the correct sense) is needed to give a depolarization asymmetry in
the wrong sense. Most Q values are more modest, and the
depolarization asymmetry is most marked for quasars, which are not
expected to have _0 near zero,
so that the depolarization asymmetry will be dominated by orientation
rather than environmental influences which change the structure of the
source.
The correlation of depolarization with arm-length depends on the source orientation. As noted above, the depolarization asymmetry in quasars will tend to be dominated by orientation effects. However, for sources close to the plane of the sky, the arm-length asymmetry will be relatively more important. For example, if s_+/s_- is 0.8, sources within 13 degrees of the plane of the sky will have their depolarization asymmetry dominated by the arm-length asymmetry rather than orientation. The correlation of depolarization with arm-length (Pedelty et al. 1989) is present in galaxies but not in quasars (Garrington & Conway 1991). This provides independent support for schemes unifying radio galaxies and quasars (Barthel 1989) where the radio galaxies are aligned close to the plane of the sky and quasars are seen end on.
The above considered the depolarization asymmetry as simply an arm-length effect, as the depolarization depends on distance from the source centre (Strom & Conway 1985; Strom & Jägers 1988). However, the asymmetry could be due to more material being present on one side of the source. The lack of correlation between the depolarization asymmetry and the optical line emission (GCL) shows that the optical line emitting gas is not responsible for the depolarization, favouring a model where the depolarization is due to the general intracluster medium whereas the structural and optical line asymmetries are due to the jet encountering an isolated cloud.