8.1.2 Dependence of Quasar Fraction on Redshift
In his original unification paper, Barthel (1989) used the ratio of quasars to radio galaxies in the 3CR catalog (essentially SSRQ to FR II) to determine a critical angle separating the two classes. However, Singal (1993b) found that the quasar fraction in the 3CR appears to increase with redshift (18) and noted that Barthel's result depended on choosing the redshift interval 0.5 < z < 1. Given the small numbers involved, Singal's number ratios are actually within 2 of no redshift dependence assuming Poisson statistics (Gehrels 1986). Moreover, once low-excitation 3CR FR II galaxies are excluded (Sec. 5.2), the quasar/radio-galaxy fraction is independent of redshift (Laing et al. 1994).
The number ratios of quasars to radio galaxies binned by redshift imply a set of critical angles which in turn correspond to predicted ratios of mean linear sizes for the two classes. These size ratios agree with the observed values to within ~ 2 , contrary to Singal's conclusions, once the associated uncertainties in the predicted values are taken into account (Saikia and Kulkarni 1994).
Even taking Singal's numbers at face value, any inconsistencies can be
explained easily by allowing for a moderate misalignment (~ 20° -
30°) between the radio axis and the axis of the optically thick torus
hiding the broad emission line region in radio galaxies
noted, a modest dependence of quasar fraction on
redshift might be explained by an evolution in the opening angle of the
obscuring torus, as it might by anything causing a correlation between opening
angle and luminosity
18 In a related study with a largely overlapping sample, Lawrence (1991) found the fraction of broad-line objects increased with radio power and possibly decreased with redshift.