| A robust and general method of testing for the presence of an obscured quasar inside a radio galaxy is to look for high mid-IR luminosity from the nuclear obscuring material. We conducted a survey of 3CR FR-II radio galaxies and quasars using the Keck I telescope Long Wavelength Spectrometer. The sample is flux-limited and selected on an isotropic property (low-frequency radio lobe flux). We also observed Cygnus A and two FR-I radio galaxies, Centaurus A, and M87.; For the FR-I sources, we f0ind that the nuclear mid-IR source in M87 is consistent with pure synchrotron emission from the base of the jet. This result establishes the existence of "nonthermal" narrow line radio galaxies wherein the energetics are dominated by jet kinetic luminosity rather than thermal accretion. However, Cen A does contain a modest quasar-like nucleus, so the central engines of FR-I radio galaxies are heterogeneous in nature.; ? showed that for 3CR FR-II sources with Z < 0.5, there appears to be an excess number of low-luminosity Narrow Line Radio Galaxies with small projected linear size relative to the simplest orientation models. The simplest explanation is that there is a population of relatively small, low power FR-IIs that lack hidden quasars. An alternative hypothesis is possible if the opening angle of the dusty torus scales with luminosity and the radio power decreases with time (?). Our observations test these hypotheses.; For our FR-II sample, the mid-IR flux is positively correlated with projected linear size of the radio lobes at 99.8% confidence. Many of the smaller radio galaxies have small mid-IR flux (upper limit 3sigma < 1.5 mJy) which indicates they probably lack a powerful thermal accretion flow. It seems likely that many FR-II radio sources are predominantly powered by nonthermal means, while the largest and most powerful FR-II radio galaxies show, strong mid-IR emission indicating that they harbor a hidden quasar. This is consistent with the fragmentary optical spectropolarimetric data which are available. Thus we conclude that Singal, rather than Gopal-Krishna et al. has the correct interpretation of the radio source statistics. |
