8.1.3 Absence of Superluminal Motion in Radio Galaxies
The subrelativistic kiloparsec-scale jet velocities in FR I galaxies, typically of the order of 1000-10,000 km s-1 (Bicknell et al. 1990), used to represent a problem for the FR I/BL Lac unification, which requires relativistic speeds at least on small scales. It was not clear that FR Is were relativistic even on the smallest scales, and if they were, it was not clear how they were decelerated or what the observational signature of that deceleration would be. Now, new observational (Venturi et al. 1993; Feretti et al. 1993; Giovannini et al. 1994) and theoretical results (Laing 1994; Bicknell 1994) supporting the presence of relativistic motion on parsec scales in these sources have changed our understanding of FR Is. Physically reasonable models have also been proposed for the required deceleration (Laing 1994; Bicknell 1994).
If FR Is are relativistic on small scales, they should show superluminal motion, even near the plane of the sky. A jet with a Lorentz factor of 5 would have apparent transverse speeds a 2.3 and 1 for viewing angles of 45 and 90 degrees, respectively [Eq. (A4)]. The few measurements of jet proper motions with VLBI, however, suggest that presently studied FR I radio galaxies display relativistic but still subluminal speeds. Using the data compiled by Vermeulen and Cohen (1994), we obtain an average value < a > ~ 0.5 (converting to our adopted value of H0 = 50 km s-1 Mpc-1) for four FR Is (NGC 315, M87, Centaurus A, and NGC 6251). The speeds for FR IIs seem to be no different: a 0.5-1.0 for Cygnus A (Carilli et al. 1994).
These results may imply that radio galaxies have smaller Lorentz factors than BL Lacs and radio quasars. It is important to remember, however, that detection of VLBI components in radio galaxies is hampered by relativistic deamplification and dilution by unbeamed emission. For = 5, for example, jets are deamplified for orientation angles 35° (Appendix A), which would include basically all radio galaxies (see Table 3); for larger Lorentz factors, the angle is even smaller and the deamplification larger. (The jet-to-counterjet ratio remains significantly larger than unity even at large angles; Fig. 22.) Similarly, at these angles the ratio of transverse jet luminosity to unbeamed luminosity is 10-2 for = 5 and f = 0.01 [Eq. (C6)] and equals ~ 10-4 at 90 ° [Eq. (C7)]. Thus the bulk of the emission may well appear to be stationary even if a transverse relativistic jet is present. Another consideration is that significant beaming can be reconciled with subluminal motion of knots if these are reverse shocks advected by the jet; their motion would then give a misleading indication of the flow velocity (Bicknell 1994).
|Figure 22. The jet to counter-jet ratio, J, versus angle to the line of sight for p = 2. Different curves correspond to different Lorentz factors: from the top down, = 15, 10, 5, 2. Note that the ratio is essentially independent of at large angles.|
In the end, either FR Is are intrinsically different from BL Lacs or they will exhibit superluminal motion on small scales. Surveys of FR Is with the Very Long Baseline Array (VLBA) should help decide this question. It is extremely interesting that in one nearby FR I galaxy, M87, which has been extremely well mapped in the radio, superluminal motion has been detected on kiloparsec scales, with a up to 2.5 (Biretta et al. 1995). At least one FR I, then, must have an appreciable bulk flow on both parsec and kiloparsec scales.