What does intermediate (arcsecond) scale structure tell us about the activity deep in the VLBI cores of galaxies and quasars? There are certain key areas where a knowledge of this structure is particularly relevant. One is in defining the relationships between the various classes of objects leading to unified schemes. A second is in the understanding of jet physics. For the latter it is very important to know if the material in the jets supplied from the nucleus is heavy or light, slow or fast, ejected one side at a time or symmetrically. The optimum resolution to study jet material is on the arcsecond (≡ kiloparsec) scale where very high dynamic range maps are possible and where it becomes feasible to find out something about the physical conditions in the surrounding medium from optical and X-ray observations.

Efficient VLBI tests of relativistic models for superluminal motion require a complete sample of sources free from selection effects in orientation, coupled with an observational requirement for bright cores. Probably the best way to construct such a sample is to extract from a low frequency sample those sources with the highest radio luminosity, since these tend to have the strongest cores. One must then ask whether there are any sources included which would not be if they were differently oriented; that is, either the flux or the power falls below the limit for the sample when the core and jet flux is removed.

We have conducted a VLBI survey of a complete, flux-density limited sample of 65 extragalactic radio sources, selected at 5 GHz. We have made hybrid maps at 5 GHz of all of the sources accessible to the Mark-II system. The sources can be divided provisionally into a number of classes with different properties: central components of extended double sources, steep-spectrum compact sources, very compact (almost unresolved) sources, asymmetric sources (sometimes called “core-jet” sources), and “compact double” sources. It is not yet clear whether any of these classes is physically distinct from the others, or whether there is a continuous range of properties.

The size, and therefore the importance, of the population of compact, steep-spectrum, radio sources has only recently been recognised. While it is now clear that the extended steep-spectrum sources are powered by a pair of, originally anti-parallel, beams which transport energy to the outer lobes some 105–106 parcsec away, our understanding of the compact steep-spectrum sources is almost nil. This is largely because our radio maps have not had high enough resolution to show their structures in any detail. However 1.67 and 5 GHz MERLIN observations (resolutions 0″.25 and 0″.1) of the ~20 steep-spectrum 3CR sources whose LAS is ≲2″ have now allowed us to classify their structures, at least in broad terms. These MERLIN maps, and recent VLBI maps, show that while there is a wide range of structures - from colinear doubles to amorphous “blobs” - the “peculiar” structures are strongly concentrated in the objects whose optical counterparts are called QSO's.

3C147 is a compact (≲1″), steep spectrum radio source identified with a quasar at z = 0.545 (0″.001 = 7.4 pc; c/Ho = 6000 Mpc and qo = 0.5). The radio structure shown by VLBI observations at 18 cm (Readhead & Wilkinson, 1980; Simon et al., this volume), at 50 cm (Wilkinson et al., 1977), and at 90 cm (Simon et al., 1980 and 1983) shows a bright ‘core’ (60 pc at one end of a ‘jet’ ~0″.2 (1.5 kpc) in length oriented in p.a. ~ −130°. In this sense 3C147 is typical of the one-sided ‘core-jet’ structures commonly found in the centres of other extragalactic radio sources. However, MERLIN observations at 6 cm (Wilkinson, this vol.) and VLA observations at 2 cm (Crane & Kellermann, unpubl.; Readhead et al., 1980) show a larger elongated feature extending ~0″.5 (3.7 kpc) to the North East of the bright core in p.a. ~25° or on the opposite side to the 0″.2 jet.

Radio interferometry at 89 GHz with transcontinental baselines can probe the region of active galaxies and quasars where collimation of relativistic outflows occur on length scales of 1017–18 cm. Three experiments have been conducted recently to demonstrate the feasability and importance of this investigation. Highlights of the three pioneering experiments are discussed below. Results obtained for the luminous sources 3C84, 3C273 and 3C345 are then summarized.

Linear polarization measurements at milliarcsecond resolution have been made at λ6 cm using four stations of the U. S. VLBI Network and the Mark III recording system. Calibration of the cross-polarization contamination of the feeds was done using the cross- to parallel-hand fringe ratios of the essentially unpolarized source 0Q 208, and is believed accurate to 0.5%. Other sources, such as AO 0235+164 and 0106+013, which are nearly unresolved in the unpolarized fringes, show clear signs of resolution in their polarized fringes. A preliminary map of the BL Lacertae object OJ 287 is compatible with a weakly-polarized optically-thick core and a highly-polarized optically-thin jet about 5 mas in length.

The Nuclear Radio cores of several nearby extended radio galaxies (e.g. M87, 3C236) consist not only of optically thick (< 1 pc) components, but also of emission on somewhat larger scale. As extended radio sources associated with quasars have on average stronger and more luminous radio cores (see e.g. Miley, 1980, Ann. Rev. Astron. Astrophys. 18, 165), we have started a project to study the properties of these quasar cores.

High resolution maps of 3C205 using MERLIN and the European VLBI network (EVN), some of which are shown in Fig.1, reveal unusual characteristics in this high redshift (z = 1.534) source. The most striking of these are as follows:

1. 1) The hotspots A and B, and the core C are aligned to within 0″.05 (at the hotspot), as measured on the MERLIN 6 cm map.

2. 2) The compact feature in the southern hotspot (A1) has projected dimensions of ~0″.03 x ~0″.015 (see Fig.1d). The corresponding minimum internal energy density of this feature is ≳2.3 × 10−6 erg cm−3.

3. 3) There is a continuous zig-zag ridge in the southern component, which starts with a definite spur of emission extending in a south-east direction from the compact feature A1 (see Fig.1b). The magnetic field in A1 is in approximately the same position angle as this spur and the VLBI elongation, which both point towards a bright secondary peak in the southern lobe (1a and 1b).

The quasar 3C298 (1416+06) has a complex radio-frequency spectrum, with a spectral bend below 100 MHz and a steep slope of −1.2 in the range 150 MHz – 3 GHz. At higher frequencies the spectrum flattens with some evidence of variability.

The low-frequency turnover of a selected sample of quasars is found to be strongly correlated with their redshift (Menon 1983)). Assuming that the components which produce the low-frequency turnover are hot spots in the quasars I have derived the physical parameters of the hot spots and show that there is a continuity in the various parameters as the angular sizes of the hot spots vary from subarcseconds to milliarcseconds. Hence the low-frequency turnover can be used as a probe into the physical conditions at the earliest phases of quasar activity.

AO 0235+164 is a very compact, flat-spectrum radio source. It is identified with a BL Lac object, and has optical absorption-line systems at z = 0.524 and z = 0.852. A complex set of HI absorption lines is seen at z = 0.524 (932 MHz), and several of these lines change significantly in depth over periods of less than a year. This is the only known case of variable extragalactic absorption lines. A faint nebulosity 2 arcsec south of AO 0235 + 164 has an emission-line redshift of z = 0.52 and may be an intervening galaxy. The radio spectrum of this source (between major outbursts) is remarkably flat, with the total flux density staying between about 1 and 3 Jy over a range of at least 1000 in frequency. Such a flat spectrum would lead one to expect a complex, wavelength-dependent structure consisting of several components with different self-absorption frequencies. However, the observed radio structure of 0235+164 is about as simple as one could imagine – it is a nearly unresolved point source in VLBI experiments from 900 MHz to 22 GHz. Recent VLBI experiments at 6 and 13 cm have shown evidence for some elongation of the source in a generally NE-SW direction, but only at low contour levels (< 15% of the peak). The major portion of the flux density appears to come from a core which is unresolved in VLBI experiments over a range of ~ 25 in frequency.

The spectrum of AO 0235+164 (z>0.85) reveals the presence of absorbing gas (z=0.524) which has been studied extensively at radio and optical wavelengths (See Wolfe, Davis and Briggs, 1982 for bibliography). A program by Wolfe et al. to monitor the radio absorption line at 932 MHz showed that the profile had changed during a 5 year period. The observation led to a tightly constrained model in which variations in the source continuum were capable of driving the HI hyperfine level populations (and hence the optical depths of the features in the absorption line) by virtue of the hypothesized close proximity of the source to the clouds. A continued monitoring program has not shown the predicted correspondence between the continuum variation and the behavior of the individual features, and this model has therefore been ruled out.

The BL Lac object 1219+28 has been observed at visible wavelengths and at 2.3 GHz, 5.0 GHz and 8.3 GHz using the VLBI network.

The study of steep spectrum sources containing compact components has become increasingly popular as the number of papers on the topic in this volume indicates. I wish to discuss the properties of a number of steep spectrum sources which are dominated by compact components whose spectra remain steep to quite low frequencies. Most of the known sources of this type do not have optical counterparts to the limit of the Palomar Sky Survey prints and thus, with the implicit assumption that they are extragalactic, have been referred to as “Optically Quiet Quasars”.