5. THE SIZES OF THE EMITTING REGIONS

If the massive black hole models are valid, the most direct and accessible evidence about conditions near the hole (maybe r 10 r_s) could come from the optical continuum of BL Lac type objects. The continuum from all quasars might emanate from an equally compact region, but there is then a greater chance of obscuration or absorption by the gas responsible for the emission lines. This gas itself may occupy a volume 1 light week across, the line widths being comparable with the virial or escape velocity from the region. It is unclear whether the apparently variable X-ray emission sometimes observed is thermal or non-thermal: it could be Compton or synchrotron radiation from very near the hole, or thermal bremsstrahlung from shock-heated gas with dimensions comparable to the emission line region [34]. X-rays have been detected from a number of galactic nuclei, but it is still unknown whether any compact extragalactic X-ray sources display line emission; this would obviously be the most decisive evidence on the mechanism. If a broad (possibly redshifted) and rapidly varying X-ray Fe line were to be discovered, this could do more than almost any other single observation to pin down the nature of the central energy source. The random velocities inferred from the optical line-widths could in principle generate post-shock temperatures up to T_virial 1 MeV. Still higher gas temperatures are attainable closer to the hole, but the energy may then be radiated predominantly by synchrotron-type mechanisms.

The smallest radio components in quasars revealed by VLBI have scales 10¹⁹ cm. There are thus perhaps as many orders of magnitude between their size and the central energy source as there are between the compact (VLBI) and the most extended ( 1 Mpc) radio structures observed. Even though these compact components may well be energized by a direct flux of plasma originating near the hole, one would not expect powerful radio emission from a region << 10¹⁸ cm unless a coherent mechanism operated. The lower-power compact radio sources in the nuclei of some nearby galaxies (e.g., M81, M87 [40, 41] are smaller, and can be attributed to accretion onto a defunct quasar. (Indeed the arguments leading to eq. (7) suggest that low values of / _crit should yield predominantly non-thermal emission.) The radio emission sets indirect constraints on the density and disposition of the gas in the nucleus, since it implies that relativistic particles can reach the emission region without being braked, and that free-free absorption does not prevent the radio emission from reaching us.

In this connection, one would much like to know: (1) whether there are any systematic differences between the optical properties of radio and radio-quiet quasars; and (2) how many radio-quiet lineless "Lacertids" exists.

Some general constraints on the sizes of the emitting region, which apply to almost any quasar model, follow from straight-forward considerations of brightness temperature and opacity.

Whereas the brightness temperature is a serious constraint on models for the radio continuum, the problem is much less acute in the optical, ultraviolet and X-ray band. At optical wave-lengths, the production of ~ 10⁴⁶ erg s^-1 of continuum radiation from a region even as small as ~ 3 x 10¹³ cm (~ r_s for a 10⁸ M hole) demands a brightness temperature of 10¹⁰ K - well below the limit possible for an incoherent synchrotron- type process. The high radiant energy density necessitates a strong (but not implausible) magnetic field if the "Compton catastrophe" is to be avoided. The fact that inferred particle lifetimes may be less than the light travel time across the emitting volume causes no problems if the acceleration occurs throughout the relevant volume.

If gamma-ray emission were convincingly detected from a quasar, a lower limit to the size of the emission region would follow from the requirement that the photons be able to escape without interacting with a softer photon to produce an e⁺-e^- pair.

The polarization of the continuum obviously provides clues to the emission mechanism and the magnetic field structure. The existence of high radio polarization from compact components (implying that internal Faraday rotation is not large) strongly suggests that the relevant relativistic plasma cannot coexist with a significant density of thermal electrons.