We have observed 14 radio galaxies with either the Australia Telescope Compact Array (ATCA) or the Very Large Array (VLA) depending on their declination. One exception is 1938-15, which was observed with both the instruments. The general characteristics of the observed objects are listed in Table 1. The total power, calculated at 4.8 GHz, and the spectral index, estimated between 2.7 and 4.8 GHz ( $\alpha^{4.8\, {\rm GHz}}_{2.3\, {\rm GHz}}$ , defined as $S \propto \nu^{-\alpha}$ ), have been taken from Wall & Peacock (1985).

Large angular (LAS) and linear (LLS) sizes (taken from the lower resolution images available in literature, see notes to the sources) and measured from the lower contour, are also listed.

$\begin{figure}\begin{tabular}{cc} {\psfig{figure=ds1732f2a.eps,angle=-90,width=8cm}\psfig{figure=ds1732f2b.eps,angle=-90,width=8cm} } \end{tabular}\end{figure}$

Figure 2: Image of 0039-44 at 6-cm (Left) and 3-cm (Right) lower resolution with superimposed vectors indicating the projected electric field direction. The vectors are proportional in length to the fractional polarisation (1 arcsec = 0.08 ratio for the 6 cm). The contour levels are: $0.8 \times -1$ , 1, 2, 4, 8, 16, 32, 64, 128, 256 mJy beam^-1. The peak flux at 6 cm is 587.8 mJy beam^-1 and at 3 cm is 305.0 mJy beam^-1

$\begin{figure}{\psfig{figure=ds1732f3.eps,angle=-90,width=15cm} } \end{figure}$

Figure 3: Image of 0043-42 at 6-cm (full resolution). The contour levels are: $1.4 \times -1$ , 1, 1.5, 2, 3, 4, 6, 8, 16, 32, 64, 128 mJy beam^-1 The peak flux is 291.0 mJy beam^-1. The two inserts show the images of the hot-spots at 3 cm (full resolution). The contour levels are: $0.6 \times -1$ , 1, 2, 3, 4, 6, 8, 10,12, 16, 32, 64, 128, 256 mJy beam^-1 for the southern hot-spot and $0.8 \times -1$ , 1, 2, 3, 4, 8, 12, 16, 20, 24, 30, 40, 50, 60, 80, 100, 150, 175, 200, 225 mJy beam^-1 for the northern one. The peak flux is 155.0 mJy beam^-1

2.1 ATCA observations

The 3 cm observations (with the 6km configuration) are essential to improve the previously available radio images and to be able to detect, e.g., jet structures (if any) close to the core. However, for the three larger sources (0043-42, 0625-53 and 2058-28) we have also obtained observations with the 1.5 km configuration in order to better image the extended emission. The data from the two configurations have been combined to obtain the best images. For two objects (0043-42 and 2058-28) the resolution of our combined 3 cm data was still too high to properly image the extended emission. Only the 6 cm data will be presented here for 2058-28, while for 0043-42 we only show an image of the hot-spots at 3 cm.

Each source was observed in scans of about 20 min for a total of about 6h in each configuration. The scans were spread throughout a 12 hour observing period in order to optimise the (u,v) coverage within the available integration time. The ATCA observing parameters are summarised in Table 2.

The ATCA data were calibrated by using the MIRIAD package (Sault et al. 1995), which is necessary for the polarisation calibration of ATCA data. The flux density scale was calibrated against observations of PKS 1934-638, assumed to be 5.83 Jy at 4.8 GHz and 2.84 Jy at 8.6 GHz according to the latest analysis by Reynolds (1996). A few iterations of phase-only self-calibration were then applied to each source. Together with the total intensity images we have also obtained the images for the Stokes parameters (Q, U), the polarised intensity ( P=(Q²+U²)^1/2) and position-angle ( $\chi= 0.5~{\arctan}~(U/Q)$ ) images. The polarised intensity and the fractional polarisation (m=P/I) were estimated only for the pixels for which $P>3\sigma_{QU}$ .

In order to compare the 8-GHz and the 5-GHz data (being taken with the same array they have different resolution), we have generated 8- and 5-GHz Stokes I, Q and U images at the best 5-GHz resolution (see Table 2). This was done by convolving the 8-GHz visibilities with the appropriate Gaussian during imaging, thus weighting down the longer spacing. The derived depolarization is defined as the $DP^6_3 = m_{6\, {\rm cm}}/m_{3\, {\rm cm}}$ . We also used the position angles at the two frequencies to derive the two-point Faraday rotation (RM). The RM is defined as $\chi(\lambda^2) = \chi_{{\rm intr}}+RM \lambda^2$ where $\chi_{{\rm intr}}$ is the intrinsic position angle and $\chi$ the apparent position angle at the $\lambda$ of the observations. Since the position angles are ambiguous by $n\pi$ , an ambiguity affects also the RM calculated from only two frequencies. Thus, the RM presented here will have to be confirmed by observing the objects at at least one other frequency.

2.2 VLA observations

VLA snapshot observations were obtained for 8 sources in the declination range north of $-30\hbox{$^\circ$ }$ using the A-array configuration on the 26 Oct. 96. The VLA observations are summarised in Table 3. The observations were carried out using the standard 6 cm (4.8 GHz) continuum mode, that is, with two 50 MHz-wide channels at bandwidth-separated frequencies (4.835 & 4.885 GHz). Each source was observed in two scans of about 15 minutes.

Data reduction followed standard procedures using the NRAO software package AIPS. The flux scale was calibrated against observations of the flux standards 3C 48 & 3C 286, adopting the standard scaling of Baars (1977). A few iterations of phase-only self-calibration were applied to each source. Again, together with the total intensity images we have also obtained the images for the Stokes parameters, the polarised intensity (P) and position-angle ( $\chi$ ) images.

Table 2: ATCA observations
Object Conf. $\lambda$ Resolution $\sigma_{\rm I}$ $\sigma_{\rm P}$

cm arcsec arcsec deg mJy/beam mJy/beam

0039-44 6A 6 2.26 1.55 -6.2 0.26 0.19

3 1.29 0.91 -6.7 0.18 0.15

0043-42 6A+1.5D 6 2.19 1.61 -4.8 0.48 ...

6* 4.44 3.15 -3.0 0.60 0.15

3 1.19 0.88 -5.1 0.20 ...

0409-75 6A 6 2.02 1.25 -14.6 0.81 0.35

3 1.17 0.73 -14.7 0.28 0.11

0625-53 6A+1.5D 6 2.03 1.64 27.3 0.71 0.16

3 1.10 0.92 28.9 0.28 0.10

1938-15 6A 6 5.89 1.47 -2.4 0.60 0.12

3 3.41 0.88 -1.8 0.40 0.11

1954-55 6A 6 2.43 1.29 78.1 0.53 0.15

3 1.63 0.76 78.6 0.42 0.10

2058-28 6A+1.5D 6 3.17 1.67 -10.7 0.29 0.15

6* 7.13 3.56 -9.3 0.47 0.16

**Table 2:** ATCA observations
Object	Conf.	$\lambda$	Resolution	$\sigma_{\rm I}$	$\sigma_{\rm P}$
		cm	arcsec	arcsec	deg	mJy/beam	mJy/beam

0039-44	6A	6	2.26	1.55	-6.2	0.26	0.19
		3	1.29	0.91	-6.7	0.18	0.15
0043-42	6A+1.5D	6	2.19	1.61	-4.8	0.48	...
		6*	4.44	3.15	-3.0	0.60	0.15
		3	1.19	0.88	-5.1	0.20	...
0409-75	6A	6	2.02	1.25	-14.6	0.81	0.35
		3	1.17	0.73	-14.7	0.28	0.11
0625-53	6A+1.5D	6	2.03	1.64	27.3	0.71	0.16
		3	1.10	0.92	28.9	0.28	0.10
1938-15	6A	6	5.89	1.47	-2.4	0.60	0.12
		3	3.41	0.88	-1.8	0.40	0.11
1954-55	6A	6	2.43	1.29	78.1	0.53	0.15
		3	1.63	0.76	78.6	0.42	0.10
2058-28	6A+1.5D	6	3.17	1.67	-10.7	0.29	0.15
		6*	7.13	3.56	-9.3	0.47	0.16

^* Values derived from the lower resolution map (3 km maximum baseline).

Table 3: VLA observations
Object Conf. $\lambda$ Resolution $\sigma_{\rm I}$ $\sigma_{\rm P}$

cm arcsec arcsec deg mJy beam^-1 mJy beam^-1

0034-01 A 6 0.41 0.38 -33 0.04 0.03

0035-02 A 6 0.42 0.37 -18 0.19 0.075

0117-15 A 6 0.46 0.31 1 0.11 0.06

0442-28 A 6 1.14 0.44 -22 0.09 0.032

0453-20 A 6 0.89 0.43 -22 0.30 0.08

1602+01 A 6 0.39 0.36 -38 0.12 0.085

1938-15 A 6 0.53 0.29 -20 0.65 0.21

2314+03 A 6 0.40 0.36 -42 0.09 0.029

**Table 3:** VLA observations
Object	Conf.	$\lambda$	Resolution	$\sigma_{\rm I}$	$\sigma_{\rm P}$
		cm	arcsec	arcsec	deg	mJy beam^-1	mJy beam^-1

0034-01	A	6	0.41	0.38	-33	0.04	0.03
0035-02	A	6	0.42	0.37	-18	0.19	0.075
0117-15	A	6	0.46	0.31	1	0.11	0.06
0442-28	A	6	1.14	0.44	-22	0.09	0.032
0453-20	A	6	0.89	0.43	-22	0.30	0.08
1602+01	A	6	0.39	0.36	-38	0.12	0.085
1938-15	A	6	0.53	0.29	-20	0.65	0.21
2314+03	A	6	0.40	0.36	-42	0.09	0.029

$\begin{figure}{\psfig{figure=ds1732f4.eps,angle=0,width=8.8cm} } \end{figure}$

Figure 4: Image of 0043-42 at 6-cm (low resolution) with superimposed vectors indicating the projected electric field direction. The vectors are proportional in length to the fractional polarisation (1 arcsec = 0.14 ratio). The contour levels are: $1.8 \times -1$ , 1, 2, 3, 4, 6, 8, 16, 32, 64, 128, 256 mJy beam^-1. The peak flux is 545.8 mJy beam^-1. The cross indicates the position (from di Serego et al. 1994) of the optical galaxy

$\begin{figure}{\psfig{figure=ds1732f5.eps,angle=0,width=8.8cm} } \end{figure}$

Figure 5: Image of 0409-75 at 3-cm (full resolution) with superimposed vectors indicating the projected electric field direction. The vectors are proportional in length to the fractional polarisation (1 arcsec = 0.22 ratio). The contour levels are: $2.0 \times -1$ , 1, 2, 4, 8, 16, 32, 64, 128, 256 mJy beam^-1. The peak flux is 1.255 Jy beam^-1. The cross indicates the position (from di Serego et al. 1994) of the optical galaxy

$\begin{figure}{\psfig{figure=ds1732f6.eps,angle=-90,width=8.8cm} } \end{figure}$

Figure 6: Image of 0409-75 at 6-cm with superimposed vectors indicating the projected electric field direction. The vectors are proportional in length to the fractional polarisation (1 arcsec = 0.15 ratio). The contour levels are: $4.0 \times -1$ , 1, 2, 4, 8, 16, 32, 64, 128, 256 mJy beam^-1. The peak flux is 2.349 Jy beam^-1

$\begin{figure}{\psfig{figure=ds1732f7.eps,angle=-90,width=8.8cm} } \end{figure}$

Figure 7: Image of 0625-53 at 6-cm with superimposed vectors indicating the projected electric field direction. The vectors are proportional in length to the fractional polarisation (1 arcsec = 0.07 ratio). The contour levels are: $1.5 \times -1$ , 1, 2, 3, 4, 6,8 mJy beam^-1. The peak flux is 22.0 mJy beam^-1

$\begin{figure}{\psfig{figure=ds1732f8.eps,angle=-90,width=8.8cm} } \end{figure}$

Figure 8: Image of 0625-53 at 6-cm with superimposed boxes proportional to the rotation measure (RM, 1 arcsec = 543 rad m^-2). Filled squares represent positive numbers for the RM while empty squares represent negative numbers

$\begin{figure}{\psfig{figure=ds1732f9.eps,angle=-90,width=8.8cm} } \end{figure}$

Figure 9: Image of 1938-15 at 6-cm (VLA data) with superimposed vectors indicating the projected electric field direction. The vectors are proportional in length to the fractional polarisation (1 arcsec = 1.05 ratio). The contour levels are: $5.0 \times -1$ , 1, 2, 4, 8, 16, 32, 64, 128 mJy beam^-1. The peak flux is 1.160 Jy beam^-1

2 Observations

2.1 ATCA observations

2.2 VLA observations