We consider the compact (≲ 2–4 asec) radio sources in the 3CR catalogue with normal (α ≲ 0.5) straight spectrum and low frequency (νc≲ 200 MHz) turnover. The curvature in the spectrum, if due to synchrotron self-absorption, implies structures of the order of tenths of an arcsec, while its straightness at high frequencies suggests that no dominant very compact (∼1 mas) feature is present. The radio sources matching the above requirements are listed in the Table. The sample is probably not complete but it is representative of the class.

Quite naturally so, VLBI'ers tend to be somewhat far-sighted: because of the very high resolutions, and poor UV-coverages, obtained with current VLBI networks only distant objects can be suitably studied. Unfortunately one thus tends to oversee some basic properties of a very popular class of objects, namely the compact, steep-spectrum radio cores. I would like to put these objects in their proper perspective.

Symmetric, two-sided morphology seems to argue against relativistic effects dominating compact radio emission. This kind of structure has been reported for a number of sources (Phillips and Mutel 1982; Pearson 1983) based on maps made at one frequency. Various arguments, all indirect, can be made for these sources being (1) Twin regions formed at the ends of jets which emerge from an invisible core, or (2) misidentified core-jet sources wherein the core and an unusually bright knot are wrongly taken to be a “double.” A telling test of both hypotheses is to map the sources in question over an octave or so of frequency. Proponents of view (1) would predict that the two double components will show nearly identical spectral indices and that weak central cores with flat or rising spectra might even be revealed. Champions of view (2) would predict that one end or the other will dominate at high frequencies (the core!) or that complex bridges of emission (the jet!) will be revealed between the components at low frequencies. We have followed our initial discovery of 5 symmetric compact doubles by (A) attempting to enlarge the sample of symmetric sources available to study, and (B) by investigating at 5 GHz those doubles for which the best maps exist at 1.7 GHz.

With the aim of studying the spatial structures of a complete sample of radio sources found at 6 cm wavelength we selected 13 sources from the “S5–survey” (Kühr et al., 1981) which fulfil the following criteria

1) δ > 70°, |bII| > 10°

2) S5 GHz > 1 Jy (at the epoch of the S5–survey)

3) α5 GHz2.7 GHz > − 1.5 (S ∝ να)

6 of these sources are identified by means of optical spectroscopy with QSOs, 6 are BL Lac-type objects, and the identification of 1 source is presently unknown. Observations at frequencies ranging from radio to X-ray have been reported (Biermann et al., 1981; Biermann et al., 1982; Eckart et al., 1982).

The Southern Hemisphere VLBI Experiment (or SHEVE) was a joint US-Australian-South African venture with both astronomy and geodesy goals. The principle astronomy goal was to make models or maps of the following sources: at 2.3 GHz (with six antennas and 9 usable baselines) – Centaurus A (the nearest active galaxy), Circinus X-1 (a flaring binary), the VELA pulsar, and 26 other active galactic nuclei and quasars; at 8.4 GHz (only one baseline) – Centaurus A and the galactic center.

The analysis is based on VLA 1465-MHz observations of GB and GB2 sources (Machalski et al. 1982; Machalski and Condon 1983a, 1983b). From those flux-density limited samples compact sources were selected. The following definition of source “compactness” has been used Subjects of analysis were: (a) fraction of compact sources among radio galaxies, quasars, and “empty field” sources, (b) identification rate vs. high-frequency spectral index for compact sources, (c) spectral-index distributions for compact identified and/or EF sources, (d) flux-density ratio for double compact sources, and (e) correlation between radio and optical luminosities for compact sources.

We review the important properties of active nuclei, in particular (i) the optical polarization and its relation to the jets found by VLBI (ii) the energy distribution and the temporal variations of the continuum spectrum and (iii) the distribution of the matter in the broad line region.

The compact objects which are the subject of this talk are essentially quasars and Seyfert galaxies; I will only briefly mention a couple of results about BL Lac objects. Before describing the X-ray properties of these objects, it is useful to introduce a “working” definition of radio-quiet and radio-loud quasars: I will call radio-loud quasars all the objects which have been detected at radio frequencies and have a spectral index between radio (5 GHz) and optical frequencies (2500 A) greater than 0.35 (Zamorani et al. 1981); all the other objects will be considered radio-quiet. Note that this definition is independent of distance and is a function only of the relative importance of radio and optical emission.

The observational status of superluminal radio sources is reviewed, and the results for one well-studied source (3C 345) are discussed in detail. We discuss the relativistic-beam model of superluminal sources, and concentrate on one aspect, namely the geometric constraints provided by combined X-ray and proper-motion measurements.

Monitoring of the radio structure in the core of 3C 273 has yielded a series of hybrid maps in which substantial source evolution can be seen. We discuss the superluminal motion of components in the jet of 3C 273, and the evolution of their flux densities and spectra. For 3C 279 we identify a new epoch of superluminal expansion with v/c = 6.5.

VLBI observations of 3C 345 at 10.8 GHz and 22.2 GHz show that the position angle of the new component is increasing as it separates from the core. Also, the apparent velocity of the component is increasing. This is the first clear evidence for non-radial motion and acceleration of an individual component in an extragalactic radio source.

The quasar 3C147 was observed in October 1982 with a ten-station VLBI interferometer in an effort to search for possible superluminal motion (Simon et al. 1983) in the core. Previous observations, in 1976, had revealed that the core consisted of two compact components embedded in a larger halo. The new high quality map reveals that this double structure has disappeared and it is not possible to interpret the structural changes unambiguously in terms of an expansion or contraction.

From five sets of VLBI observations spaced between 1972 and 1981, we estimated the positions of components of the superluminal quasar 3C 345 relative to the position of the single component of the quasar NRAO 512. The relative proper motion of the easternmost component of 3C 345, believed to be the “core”, was found to be 0.02±0.02 mas/yr. This result is consistent with the “core” being stationary and the “jet” components moving with respect to NRAO 512.

We report here VLBI observations at four epochs of the source BL Lacertae made with the U.S. VLBI Network (plus Bonn) at 10.65 GHz which show clear evidence of uniform superluminal expansion at v/c = 4.4 ± 0.2 (H0 = 55 km s−1 mpc−1). The maps, which were made every three months from 1981.76 to 1982.75 (Figures 1, 2), were part of an ongoing program of VLBI observations of BL Lacertae at 5 and 10.65 GHz starting in 1980.4, near the beginning of a series of violent flux outbursts (Aller, Hodge, and Aller, 1983; Aller and Aller, 1984). Previous results from earlier epochs have been reported by Mutel, Aller, and Phillips (1981) and Phillips and Mutel (1982).

The behavior of the linear polarization of BL Lacertae during the recent series of radio bursts indicates the presence of Faraday rotation in the vicinity of the source and the formation of an axial compression in the radio jet.

Superluminal motions were first seen in 3C 120 between 1972.5 and 1974.4 (Seielstad et al. 1979 and references therein). Between 1975 and 1980, the source was monitored along with 3C 273 and 3C 345 by the Caltech group. One reasonably clear episode of expansion was seen in 1979 (Walker et al. 1982) but, for most of the time, the source evolution was so rapid that it was difficult to relate the structures seen at successive epochs.

We present VLBI and VLA maps of the superluminal radio source 3C 120 (z = 0.03). The 18 cm VLBI maps shown in Figures 1 and 2c were constructed from a 14 station VLBI observation on 10, 11 October 1982. The map in Figure 1 is at full spatial resolution. The Figure 2c map was made from a uv-tapered data set and shows the extended emission. The VLBI maps show a jet ~175 pc long with distinct knots and bends (H0 = 55 kms−1 Mpc−1). The 18 cm jet smoothly connects to the 6 cm superluminal jet (Figure 2d). The VLBI jet remains well collimated through several bends and only begins to spread out 100 pc beyond the core. At 175 pc from the core, the jet is wide and weak, and disappears into the under-sampled region of the uv-plane (< 2(10)6 λ's).

This is a report on an ongoing project designed to study structural variations on the milliarcsecond scale in four BL Lac-type objects. The observations are made at λ6 cm with a global VLBI network consisting of the combined US and European networks and with some additional antennae like one in South Africa and one in Finland. A typical experiment involves 7–9 stations and produces maps with a resolution (FWHM) of 1–1.5 m.a.s depending on the declination of the source. Seven experiments have been made since the start in 1978, resulting in hybrid maps at three epochs for each of the four main sources: A0 0235+164,0735+178,Mk 421, and 1749+701. The same restoring beam has been used for each source at all epochs. The epoch March 1979 is an exception though, the maps from then were restored with a larger beam corresponding to the resolution on the European baselines only. The results from the epochs 1978–1979 have already been published (Bååth et al. 1981), and papers describing the results from the later epochs are in preparation.

Five objects mapped as part of a VLBI survey have been re-observed at 5 GHz, and four of them have also been observed at 10 GHz. Three of the objects show no substantial structural variations: an upper limit of 2c can be placed on the apparent relative velocities of the components. One object (0711+356) shows structural variations which are mostly simply described in terms of a superluminal contraction. The remaining object (3C371, 1807+698) shows substantial structural variations which suggest that it probably is a superluminal source. The source 0710+439 is especially interesting as it consists of a central flat-spectrum core component straddled by two compact steep-spectrum components.

The quasar 4C39.25, whose nuclear radio structure was stable for many years, now shows a contraction in the brightness distribution. Possible explanations are discussed.