|Annu. Rev. Astron. Astrophys. 1980. 18:
Copyright © 1980 by . All rights reserved
Life is the art of drawing conclusions from insufficient premises.
Samuel Butler, Notebooks
The enormous amount of structural data gathered in recent years suggests the following description of the radio-source phenomenon. A "machine" imbedded in the nucleus of a galaxy ejects a collimated flux of energy in two opposite directions along its angular momentum axis (Section 4.3.3) in the form of a recurrent spluttering of plasmons or a quasi-continuous "beam" of relativistic particles. As a fairly direct manifestation of this nuclear activity we see radio and optical ("quasar") continuum cores and broad line emission regions (Sections 4.3 and 5.2). Provided the nuclear activity is strong enough, prolonged enough, and/or stable enough in direction, an extended radio source is built up. The overall morphology of the extended lobes is determined by the degree of activity, as well as the amount of translational and/or rotational motion of the nuclear machine and/or of the surrounding medium (Sections 2.2, 3 and 4).
What is the nuclear machine? There are several pointers that it may be a rotating black hole that derives its energy from accretion (Lynden-Bell 1969). First, one can cite the elephantine memory for direction discussed in Sections 4.3 and 5.2. Second, photometric and spectroscopic studies of M 87 provide some direct evidence for the presence of a black hole in this, one of the closest radio galaxies (Sargent et al. 1978, Young et al. 1978). Third, collimated extragalactic radio sources have several properties in common with the nonthermal radio emission associated with some galactic X-ray stars (e.g. Braes & Miley 1973). For example the triple source associated with Sco X-1 is uncannily like a miniature edge-brightened extragalactic radio source. SS 433 also shows extended radio emission symmetrically distributed with respect to an X-ray star (Spencer 1979; W. Gilmore and E.R. Seaquist, in preparation; R. Hjellming, K.J. Johnson, and G.K. Miley, in preparation). The only energy sources that have been put forward as possible power houses for both X-ray stars and galactic nuclei are accreting black holes. Fourth, the energy budget is consistent with the appetite of a not too hungry black hole. The most energetic radio sources require total energies in excess of 1060 ergs supplied over a period probably longer than 108 yr (an age derived by combining the most plausible outflow speeds in Section 4.4 with the typical source sizes for Section 2.3). The resultant energy needs of 10-2 M per year could well be supplied by an accreting black hole buried in a dense galactic nucleus.
Although the above scenario is certainly very oversimplified and possibly conceptually wrong, it is, in my view, the most plausible unified picture consistent with currently available data.
In the near future we shall certainly see further progress on the observational side. The completed VLA will map the detailed morphologies of the jets and hot spots, while moresophisticated optical detectors will provide visual information. The resultant statistical studies should enlarge our knowledge of the energy transport and particle acceleration processes. Second, and perhaps more important, there is the possibility of tackling one of the most intriguing remaining problems - the nature of the nuclear energy conversion and collimation processes. There must be a fundamental mechanism which works over a vast range of physical conditions which can produce collimated radio sources with sizes spanning more than a factor of 107 and energies more than a factor of 1020. During the next few years we may expect to see an attack on this problem from several separate wavelength regions. More sensitive VLBI measurements of radio cores of both galactic and extragalactic sources will become available together with the Einstein Observatory's X-ray data. In addition high resolution optical observations from the space telescope should provide considerable information about the dynamics of the inner regions of active galaxies. A third fruitful avenue of research promises to be a continuation of the work described in Section 5.3, carrying out more detailed comparisons between the morphologies of radio sources and the structures and dynamics of their parent galaxies.
Nearly three decades ago an historic encounter took place in which Baade wagered Minkowski a bottle of premium whiskey that the visible object identified with Cygnus A was a colliding pair of galaxies. Baade apparently lost the bet. More recently it has been suggested that the hypothetical nuclear black holes that power extended extragalactic radio sources are fueled by infall of material to the nuclei (see, for example, Rees 1978b). The rate of infall may well be influenced by galaxy merging, a process now thought to play an important role in the evolution of galaxies (e.g. Silk & Norman 1979). It is ironic that the wheel of theory shows signs of moving almost full circle. Surely Baade posthumously deserves at least half a bottle of medium quality whiskey?
Review articles inevitably reflect the personal preferences and prejudices of the author. I am grateful to the many friends and colleagues who over the years were instrumental in forming my present conditioned outlook. Without risk of incrimination I also wish to thank Wil van Breugel, Ed Fomalont, Dan Harris, Harry van der Laan, Colin Norman, Richard Strom, Tony Willis, and Lorenzo Zaninetti for critically reading the manuscript and Lenore for patiently typing it. I acknowledge the receipt of a NATO research grant (no. 1828).