4 Comments on individual sources

In the following, a description of the morphology of each source is given:

0023-263

This radio source has been identified with a galaxy of m_v=19.5 by WP85 at a redshift of 0.322 (Tadhunter et al. 1993; di Serego Alighieri et al. 1994). The VLBI images at 2.3 and 8.4 GHz are shown in Fig. 3.

At both VLBI frequencies this radio galaxy shows a basic double structure with a separation of $\sim$650 mas ($\sim$1.9 kpc) in PA - $34^\circ$ between the peak brightness of each lobe. Both radio emitting regions are complex and extended structures are evident. In particular, the south-eastern component is resolved into two sub-components embedded into more diffuse emission. Thus, it is not possible to fit simple Gaussian components, and sizes for each lobe are estimated from the images. Both lobes are less than $\sim$100 mas ($\sim$0.3 kpc) in extent and the south-eastern one has an overall spectral index ( $\alpha=-1.3$) steeper than the north-western component ( $\alpha=-0.7$), possibly also because it is more resolved.

The flux density detected in the VLBI images accounts for only about 40% of the total source flux density at each frequency. This indicates that there is undetected extended scale structures in the range 0.1 to 1 arcsec. The same fraction of flux density is undetected at both frequencies, suggesting that the underlying diffuse structure has similar spectral index ($\sim$-0.9) to the components visible in the images. No flat spectrum component indicative of a core is detected between the lobes at the detection limits (5$\sigma$) of $\sim$20 mJy/beam at 2.3 GHz and $\sim$10 mJy/beam at 8.4 GHz. We decided to use 5 $\sigma$ as detection limit for the cores (in this and in the following sources), given that, the complexity of the radio emission in relation to the limited uv-sampling, increases the incidence of clean artifacts at lower levels.

The MERLIN image at 5 GHz (Fig. 4) with much lower resolution shows the basic double-lobe structure of the source and the extended emission surrounding the lobes. About 90% of he total flux density is detected (3.4 Jy out of 3.8 Jy) indicating little extended emission at arcsecond scales.

0252-712

This object has been identified with a galaxy of m_V=20.9 and redshift z=0.566 (di Serego Alighieri et al. 1994). The VLBI image at 2.3 GHz (Fig. 5) shows a North-South double structure quite symmetric; the peak regions are separated by $\sim$145 mas ($\sim$0.54 kpc). Both components appear to be resolved and account for more than 95% of the whole flux density of 3.7 Jy at this frequency. No "core'' component is detected between the lobes (5$\sigma=40$ mJy) and no spectral information is available on the detected components.

1151-348

This object has been identified as a quasar with m_v=17.8 from WP85 and z=0.258 (Jauncey et al. 1978). Both 2.3 and 8.4 GHz images are available (Fig. 6), showing two components with separation of 91 mas ($\sim$0.23 kpc) and PA = $72^\circ$ and possibly some extended emission between them. The peak separation and orientation are identical (to within 0.5 mas) at both frequencies, unambiguously registering the two components.

Both lobes are less than $\sim$85 mas ($\sim$0.2 kpc) in extent and the south-western one has an overall spectral index ( $\alpha=-1.0$) steeper than the north-eastern component ( $\alpha=-0.6$). About 70% of the total flux density is detected in the VLBI components and this fraction is the same at both frequencies. Thus the underlying undetected flux density has a spectral index ($\sim$-0.7) similar to the two aforementioned components, possibly indicating that this emission is simply resolved, diffuse extension of these lobes.

Figure 5: PKS 0252-712 at 2291 MHz from the SHEVE array. The peak level is 1.14 Jy/beam and contours are shown at -1.5, 1.5, 3, 6, 12, 24, 36, 48, 72, 96% of the peak.

Figure 6: ( Left) PKS 1151-348 at 2291 MHz from the SHEVE array. The peak level is 1.57 Jy/beam and contours are shown at -0.75, 0.75, 1.5, 3, 6, 12, 18, 35, 50, 65, 80% of the peak. ( Right) PKS 1151-348 at 8421 MHz from the SHEVE array. The peak level is 0.44 Jy/beam and contours are shown at -1.5, 1.5, 3, 6, 12, 18, 35, 50, 65, 80% of the peak.

No flat spectrum component, possibly indicative of a core, is detected between the lobes, at the detection limit (5$\sigma$) of $\sim$20 mJy/beam.

1306-095

It is identified with a galaxy of magnitude m_V=20.5 and redshift z=0.464 (di Serego Alighieri et al. 1994).

This source is at low absolute declination and the uv coverage is rather poor, which implies high sidelobe levels ($\sim$50%) and a low dynamic range image. The radio emission is dominated by two distinct components separated by about 373 mas ($\sim$1.3 kpc) in PA $\sim$ $-40^\circ$(Fig. 7). The southern component is much stronger, accounting for more than 80% of the detected flux density. Only about 30% of the total flux density is detected by the VLBI components, indicating the presence of much diffuse emission completely resolved out by the VLBI baselines. It is likely that in Fig. 7 only the hot-spots of two mini-lobes are visible.

Figure 7: PKS 1306-095 at 2291 MHz from the SHEVE array. The peak level is 0.30 Jy/beam and contours are shown at -5, 5, 10, 20, 40, 80% of the peak.

1814-637

It is identified with a galaxy of m_V=18.0 with z=0.063 by WP85. The VLBI image at 2.3 GHZ shows the most complex structure of all sources in this sample (Fig. 8). The source still shows a basic double-lobed structure aligned almost North-South. However, the southern region is dominated by two components with similar brightness and much extended emission. The northern region shows a prominent bright component and extended emission, with possible North-South symmetrical extensions, but the overall extent of this component is still less than $\sim$90 mas ($\sim$70 pc).

Just over 50% of the total flux density of the source is detected, indicating the presence of further, much more diffuse extended component. The overall source size is still less than about 0.5 arcsec and no arcsecond-scale components are detected with ATCA observations.

There is also a weak component between the two major lobes at the 5% level, which corresponds to a 15$\sigma$ detection. This can be just a peak in the underlying diffuse component or could indicate the presence of a core between the two lobes. VLBI observations at another frequency are needed to determine the spectrum of this component and decide whether it is the core.

1934-63

This compact radio source has been identified as a galaxy with m_V=18.4 by Kellerman (1966) and has redshift z=0.183 (Penston & Fosbury 1978). A VLBI image at 2.3 GHz from observations carried out in 1982 has been presented in Tzioumis et al. (1989) and preliminary images at 2.3, 4.8 and 8.4 GHz have also been presented elsewhere (King 1994; Tzioumis et al. 1996). The knowledge about the pc-scale structure of this source is summarised in Fig. 10. The source appears as a very compact double, with component separation of 42 mas ($\sim$0.084 kpc) in PA 89$^\circ$, and this separation has not changed for over 15 years. As reported in Tzioumis et al. (1996), this yield to an upper limit to the expansion velocity of 0.2c. The total flux density of the source is detected by the VLBI components and there is no definite detection of a flat-spectrum core component at a level of a few % of the peak at any frequency.

2135-209

This source is a galaxy with m_v=19.4 and z=0.635 (WP85 and di Serego Alighieri et al. 1994).

No VLBI image exists for this source and here we present a MERLIN 5 GHz image (Fig. 9). The source shows a double lobed very asymmetric structure; the resolution of MERLIN at this southern declination is poor ( $163 \times 34$ mas in the case of 2135-209) to allow further consideration on the morphology. Because of the limited resolution compared to the source extension, the component parameters were derived by means of Gaussian fitting. The North-East dominant component is marginally resolved and accounts for more than 80% of the detected flux density. The total flux density of about 1.5 Jy was detected in the MERLIN image, indicating no extra extended components exist.

From one-baseline VLBI observations at 2.3 GHz it is possible to fit models to the double structure. The strong north-east component is unresolved while the south-west component is extended. The separation between the peaks in the two lobes is well determined at 167 mas ($\sim$0.65 kpc), at a position angle of 52$^\circ$, close to the position angle and separation determined from the MERLIN image. About 85% of the total flux density is detected on this VLBI baseline.

Figure 8: PKS 1814-637 at 2291 MHz from the SHEVE array. The peak level is 1.7 Jy/beam and contours are shown at -1.5, 1.5, 3, 6, 12, 18, 35, 50, 65, 80% of the peak.

Figure 9: PKS 2135-209 at 4993 MHz from the MERLIN array. The peak level is 1.02 Jy/beam and contours are shown at -0.75, 0.75, 1.5, 3, 6, 12, 18, 35, 50, 65, 80, 95% of the peak.