Very little is known about this quasar from previous studies. We do not detect any extension even after deconvolution and PSF subtraction. Note however that the background is quite noisy, especially to the South-East (Fig. 1); this prevents a reliable determination of the PSF at faint levels, making the analysis of the presence of faint extensions very uncertain.

According to Hutchings & Neff (1990), PKS 0812+020 is a one-sided lobe radio source, with a very close neighbour only 7 kpc away (in probable tidal interaction) and 16 galaxies within a distance of $\pm\, 30''$ .

The J band image shows structures similar to what is seen in the optical by Wyckoff et al. (1981), in their search for the underlying galaxy. The radio lobe 10'' to the North-West, found to be coincident with a diffuse optical emission region (Wyckoff et al. 1983), corresponds to an object (#17) of magnitude 18.8 in J and 17.7 in K'. We detect extensions to the East and South out to about 3'' and 2.5'' respectively from the quasar, and a very close object 3.5'' to the North (object A). The deconvolved image (Fig. 2) shows that the counterpart of the radio lobe (object #17) has elliptical contours; however Ellingson et al. (1991a) have not measured its redshift.

The HST image (Fig. 17) shows that the extension seen to the east in the J band image is an independent object lying 2'' from the quasar (object B). The optical counterpart of the radio lobe (#17) seems to be a spiral galaxy. The extension to the South is not visible in this image. From the HST image we can also say that objects #14, #19 and #20 are galaxies.

In Table 2, four objects taken from Ellingson et al. (1991a) are quoted with redshifts between 0.30 and 0.408. Out of these, only objects #11 and #19 are inside the field shown in Fig. 2. We do not detect galaxy 7 in Ellingson et al. (1991a), and their object 12 (z = 0.4038) is unfortunately outside our field. Note that object #19 is exactly at the same redshift as the quasar.

$\begin{figure} { \psfig {figure=ds1608f17.ps,height=6cm} } \protect\end{figure}$

Figure 17: HST image of PKS 0812+020 (filter F675W). Contour levels are 6.5, 9, 15, 30 and 50 (in arbitrary units)

$\begin{figure} { \psfig {figure=ds1608f18.ps,height=6cm} } \protect\end{figure}$

Figure 18: HST image of PKS 0837-120 (filter F702W). Contour levels are 2, 3, 5, 10, 20 and 30 (in arbitrary units)

3.3 PKS 0837-120 (3C 206, z = 0.1976)

Our J band image (Fig. 3) shows the same structures as those seen in the optical by Véron-Cetty & Woltjer (1990), with a number of objects in the close vicinity of the radio quasar. The deconvolution of the K' image allows to resolve the extension to the S-SW (indicated with a line in Fig. 3) as an object; the host galaxy is also visible and its derived magnitude is $K'_{\rm gal} = 13.7$ .

In the HST image (Fig. 18) the closest objects are clearly resolved and appear as galaxies, in particular object #17, the object about 7'' to the South-West, and those about 11'' to the North-East which seem to be a close interacting pair (corresponding to object #20 in Table 2 and object 94 in Table I by Ellingson et al. 1989). The very small extension to the SE corresponds to a faint galaxy about 1.7'' from the QSO center.

Redshift measurements are available for 7 of the 13 objects reported as probable galaxies in Table 2, with values between 0.0828 and 0.2677, five of them being very close to that of the quasar (Ellingson et al. 1989). Only one of them, object #10 in Table 2 (with z=0.1966), is in the field of Fig. 3. Note that in Ellingson et al. (1989) there are two additional objects close to the QSO with measured redshifts that we have not included in Table 2: for that placed $\approx 7''$ West of the QSO (88 in Ellingson et al. 1989; z=0.1929), we cannot measure magnitudes due to contamination by the QSO; neither do we resolve the other one, located 10'' NE of the QSO (94 in Ellingson et al. 1989; z=0.1994).

3.4 3C 215 (z = 0.41)

This quasar appears to have a very complex radio structure (see e.g. Bridle et al. 1994); it is in probable tidal interaction with a companion galaxy only 28 kpc (about 6'') away and is surrounded by 14 galaxies within $\pm\, 30''$ of the quasar (Hutchings & Neff 1990). Extended ionized gas has been detected along $\rm PA = 190\hbox{$^\circ$}$ (Crawford & Fabian 1989). The R band profile of 3C 215.0 appears somewhat more extended than the stellar profile, and consistent with an elliptical host galaxy; a companion object is present 7'' South-East of the quasar, but with no measured redshift (Hutchings 1992).

In Fig. 4 we see a faint extension to the North-West. We list 16 objects in Table 2. Note that objects #11 and #16 have the same redshift as the quasar. Three weak objects, labeled A, B and C are found to coincide well with three features in the Hutchings et al. (1998) radio map.

$\begin{figure} { \psfig {figure=ds1608f19.ps,height=6cm} } \protect\end{figure}$

Figure 19: HST image of 3C 215 (filter F814W). Contour levels are 5, 7, 10, 15, 20 and 25 (in arbitrary units)

The HST image (Fig. 19) shows that all the objects located close to the quasar but one, that we detect in J, seem to be galaxies, object #14 having a spiral morphology (the brightest one, #15, is a star).

3.5 IRAS 09149-6206 (z = 0.057)

The properties of the underlying galaxy have been discussed by Véron-Cetty & Woltjer (1990), who claim that a spheroidal model gives a much better fit than a disk model to their i filter imaging. The parameters of the fit in the V band are given in more detail by Véron-Cetty et al. (1991), leading to an absolute magnitude $M_{V{\rm gal}} = -23.0$ .

In Fig. 5 we clearly see the host galaxy of this bright quasar, which extends about 40'' along its major axis direction (PA $\approx 62\hbox{$^\circ$}$ ). The average isophotal profile results to be clearly broader than that of the PSF, as shown in Fig. 24. We have deconvolved the image and subtracted the PSF to obtain a final continuous profile with no central hole (a "flat top profile with no hole in the center'', Aretxaga et al. 1995; Rönnback et al. 1996); the resulting profile follows quite well an r^1/4 law; the host galaxy is therefore probably an elliptical galaxy with magnitude $K'_{\rm gal} = 10.35$ . This field is very rich in objects, but no redshift is available.

3.6 PKS 1011-282 (z = 0.253)

The extension corresponding to the host galaxy is visible in our J image (see Fig. 6) and the deconvolution enhances this structure. However, we do not detect in J the diffuse arc-like feature observed in the [OIII] line image (about 20'' North-West of the quasar), implying that this is a region of ionized gas with no underlying stellar population, as already mentioned by Boisson et al. (1994) who detected no optical counterpart.

3.7 3C 275.1 (z = 0.557)

Hintzen et al. (1981) have shown that this quasar could be at the center of an elliptical galaxy belonging to a galaxy cluster. Later, extended ionized gas with a complex structure and large scale motions was reported by Hintzen & Stocke (1986); the ionized gas appears to be elongated along an angle roughly perpendicular to the radio axis (e.g. Stocke et al. 1985). Hintzen & Romanishin (1986) obtained optical images in the R and redshifted narrow [OII] $\lambda$ 372.7 bands, and found a clear extension along $\rm PA = 45\hbox{$^\circ$}$ ; once an elliptical component has been subtracted, some small extensions remain mainly along the major axis. Redshifts from Ellingson & Yee (1994) show that the quasar has a very close companion at the same redshift (object #6 in Table 2) and that it belongs to a group of galaxies, which Krempec-Krygier et al. (1998) have analyzed in detail from the physical and dynamical points of view; they have shown that the quasar is located at the bottom of the gravitational potential well of the group.

Our J band image (Fig. 7) shows the host galaxy as elliptical contours with a PA roughly consistent with that found in the optical images. There is also a small extension to the East. Deconvolution is not possible since the stars in the frame are too faint or contaminated by close objects to give an accurate estimation of the PSF. Several nearby objects are also visible in the J image, a number of these being foreground galaxies (see redshifts in Table 2).

The HST image (Fig. 20) shows that object #4 has a spiral morphology and that objects #9 and #6 are galaxies, #6 being at the same redshift as the quasar. Besides, we can note that galaxies #4 and #6 correspond to two small radio emitting features in the Hutchings et al. (1998) radio map.

$\begin{figure} { \psfig {figure=ds1608f20.ps,height=6cm} } \protect\end{figure}$

Figure 20: HST image of 3C 275.1 (filter F675W). Contour levels are 2.9, 4, 5, 7 and 10 (in arbitrary units)

$\begin{figure} { \psfig {figure=ds1608f21.ps,height=6cm} } \protect\end{figure}$

Figure 21: HST image of 3C 281 (filter F814W). Contour levels are 3.5, 5, 7 and 15 (in arbitrary units)

$\begin{figure} { \psfig {figure=ds1608f22.ps,height=6cm} } \protect\end{figure}$

Figure 22: HST image of 4C 11.50 (filter F702W). Contour levels are 13, 20 and 25 (in arbitrary units)

$\begin{figure} { \psfig {figure=ds1608f23.ps,height=6cm} } \protect\end{figure}$

Figure 23: HST image of 3C 334 (filter F675W). Contour levels are 0.8, 1, 2 and 7 (in arbitrary units)

3.8 PKS 1302-102 (z = 0.286)

Hutchings & Neff (1992) have detected two objects in the close vicinity of the quasar and related their presence to the merging state of the system; they also fit a profile which is a combination of an r^1/4 law and an exponential law. Bahcall et al. (1995) did not detect the host galaxy in their HST image. After deconvolution with the PSF however they can see two companion galaxies located about 1'' and 2'' N-NW away from the quasar center (see their Fig. 8). From a new HST image, Disney et al. (1995) detect the two companion objects with no need of PSF subtraction and the host galaxy is well fit with a r^1/4 profile. The detection of the host galaxy from HST imaging was later confirmed by Bahcall et al. (1997), who gave values for the size and morphology of the host galaxy.

$\begin{figure} { \psfig {figure=ds1608f24.ps,height=12cm,width=8.5cm} } \protect\end{figure}$

Figure 24: Isophotal profiles of quasars (crosses) that result to be clearly different from those of the corresponding PSFs (stars) which have been normalized to the QSO for direct comparison. Fluxes are in arbitrary units

Our J band image (Fig. 8), very similar to that of McLeod & Rieke (1994b) in the H band, is well suited to trace the host galaxy; its isophotal profile is clearly different from that of the PSF (see Fig. 24) as in McLeod & Rieke (1994b), and notably more extended than that obtained by Wyckoff et al. (1981). The underlying galaxy is extended along a direction roughly perpendicular to the radio axis, as already noted by Gower & Hutchings (1984). In the deconvolved image we have subtracted the quasar by using the PSF normalized to a central value that produces no hole after subtraction. This barely allows us to detect the object 2'' North of the quasar, which appears as an isophotal distorsion in the direct image, but we do not separate the quasar from the small object 1'' away. The contribution of the quasar to the total light in J results to be about 60%. The resulting profile of the host galaxy can be well fit by an r^1/4 law and its derived magnitude is $J_{\rm gal} = 14.9$ . These results are in agreement with McLeod & Rieke (1994b) concerning the relative contribution of the host galaxy (31% with $H_{\rm gal} = 14.79$ )although they fit an exponential. We note that the presence of the very close object 1'' away does not produce any significant contamination in the average profile, which is dominated by the host galaxy.

3.9 3C 281 (z = 0.599)

In our J band image (Fig. 9) an object 5'' South of the quasar is visible, together with extensions to the North and to the East. The deconvolution enhances the Northern extension and separates the Eastern one as an object.

The HST image (Fig. 21) shows that these extensions are in fact two objects located respectively 3.5'' North and 2.7'' East of the quasar center. We also see that objects #6 and #9 in Table 2, which were reported as a probable galaxy and a star respectively, are both galaxies. Object 132 in Yee et al. (1986), located 5'' North of #9 is also a galaxy, but is too weak to be detected in the J band image.

3.10 4C 20.33 (z = 0.871)

This quasar is an asymmetric radio source with a collimated one-sided jet extending from the quasar to the South (Mantovani et al. 1997). Fabian et al. (1988) detected extended ionized gas around this quasar.

Isophotes are elongated towards the South in our J band image (Fig. 10), a feature which is enhanced after deconvolution. In this case, and contrary to some of the objects previously described, the elongation appears to be along a direction roughly similar to that of the radio axis (Mantovani et al. 1992). We also see a number of objets (probably galaxies) close to the quasar that could be associated with it. Redshifts are obviously needed for this field.

3.11 4C 11.50 (z = 0.436)

The image by Stockton and MacKenty (1987) shows extended ionized gas elongated along PA $\approx 306\hbox{$^\circ$}$ . A second quasar (object #8) at much higher redshift (z=1.901, Wampler et al. 1973) is located 5'' to the East and a tight group of three galaxies (object #7) is associated with 4C 11.50 at an average distance of 10'' West of the quasar (Stockton 1978).

We observe similar structures in our J and K' band images (Figs. 11, 12). The quasar appears elongated along PA $\approx 47\hbox{$^\circ$}$ in our deconvolved K' image, where the three objects to the West reported by Stockton & MacKenty (1987) are clearly separated. The northernmost one is elongated to the West, due to the presence of a separate object as seen in the HST image (Fig. 22). The object detected 12'' South-West of the quasar (object 142 in Yee et al. 1986) is extended in the HST image, but is too weak to be detected in J. Note that the host galaxy of the quasar appears to be detected in the HST image. A weak object labeled A appears to coincide with a faint feature in the Hutchings et al. (1998) radio map.

We give R magnitudes from the literature for the objects in the field in Table 2; note that the quasar at z=1.901 (object #8) is indicated as a star. We also give redshifts for objects #7 and #9 from Ellingson et al. (1991a); both objects result to be companion galaxies to the quasar (z=0.4323 and 0.4331 respectively).

3.12 Mrk 877 (z = 0.114)

Boroson et al. (1982) detected faint emission from ionized gas extended over several arcseconds from Mrk 877. The analysis of the H band image by McLeod & Rieke (1994a) results in an quasar isophotal profile which is slightly different from that of the PSF. Our J band image (Fig. 13) has been taken in poor seeing conditions (1.4 arcsec) but good spatial sampling (0.5 arcsec/pix) and shows a definite difference between the QSO and PSF profiles (see Fig. 24). The profile from our deconvolved J band image can be well fit by an r^1/4 law, suggesting here also that the host galaxy to this quasar is an elliptical galaxy, with an estimated magnitude $J_{\rm gal}=14.95$ . A number of objects appear in the QSO field with profiles different from the PSF and sizes comparable to the QSO host and could be possible companion galaxies.

3.13 3C 334 (z = 0.555)

Our J band image (Fig. 14) shows an extension to the North-West (to about 5'' from the quasar) and two objects 10'' North and 7'' South respectively (objects #5 and #6 in Table 2). The latter corresponds to that detected in [OII] by Hes et al. (1996). A weak object, labeled A, coincides with a bright feature in the Hutchings et al. (1998) radio map and is also visible in the HST image (Fig. 23).

The deconvolved image allows to separate the small extension to the North-West as an object (indicated in Fig. 23 with a line). This object, as well as objects #5 and #6, are clearly separated from the quasar in the HST image, #6 probably being a galaxy. It would be very interesting to determine the redshift of these three objects and to search for any relation with the three weak Ly $\alpha$ absorptions observed at z = 0.5387, 0.5449 and 0.5491 (Jannuzi et al. 1998).

3.14 MC 1745+163 (z = 0.392)

Stockton & MacKenty (1987) give an image of the ionized gas in MC 1745+163, which they claim to be resolved, with an elongation to the West along PA $\approx 90\hbox{$^\circ$}$ .

Our J band image shows this extension as well; though in the direct image (Fig. 15) the major axis PA is smaller $\rm (PA =50\hbox{$^\circ$})$ ,it is somewhat larger (62 $^\circ$ ) in the deconvolved image. The isophotal profile is notably different from that of the PSF already in the direct image (Fig. 24). The resulting profile of the deconvolved and PSF subtracted image is well fit by an r^1/4 profile; the host galaxy is therefore likely to be an elliptical with $K'_{\rm gal} = 16.19$ . Many objects are detected in this field; most are stellar like, except for objects #12 and #16 which look like galaxies but for which we do not have redshifts.

3.15 4C 11.72 = PKS 2251+113 (z = 0.323)

This quasar has two companion galaxies with comparable redshifts: z = 0.3287 et 0.3240 for objects #4 and #5 in Table 2 respectively (Robinson & Wampler 1972, following observations by Gunn 1971). It is embedded in an extended ionized nebulosity (Stockton & MacKenty 1987; Hutchings & Crampton 1990), where the gas appears to be very disturbed, oppositely to the host galaxy, which Hutchings & Neff (1992) classify as an "undisturbed r^1/4 galaxy''; from integral field spectroscopy, Durret et al. (1994) have shown that the velocity field was in fact consistent with the presence of a rotating disk, on to which are superimposed several smaller blobs possibly interacting with the main envelope.

Optical images in b and v by Block & Stockton (1991) show the extended nature of the residual after the QSO subtraction (see their Fig. 5). Our J image is given in Fig. 16. Since we do not have in our field a suitable star to compute the PSF, we cannot proceed either with the deconvolution nor with a PSF subtraction to look for the nearest objects, labeled A and B by Block & Stockton (1991). Objects #9 and #10 in Table 2 correspond to objects 7 and 4 in Block & Stockton (1991). Object #7 corresponds to two objects, the northernmost one corresponding to object W in Gunn (1971).

Jannuzi et al. (1998) have observed this quasar spectroscopically with the HST and detected a strong C IV-O VI associated absorption line system at $z = 0.3256 \sim z_{\rm em}$ . It would be very interesting to obtain higher spectral resolution optical data to study the kinematics and composition of the gas, in order to investigate the possibility that it is associated either with the AGN (Petitjean et al. 1994) or with a hot phase resulting from the interaction of several objects (Durret et al. 1994).

There is an additional Ly $\alpha$ absorption line at z = 0.3236. If the latter is associated with the gaseous halo of object #5 which has the same redshift, then the radius of the halo should be larger than $95\,h^{-1}_{50}$ kpc for q₀ = 0. This radius is consistent with studies of the low-redshift Ly $\alpha$ forest (e.g. Le Brun et al. 1996). However this quasar is part of a group of galaxies where the distribution of gas into individual halos is questionable. Indeed there is no H I absorption associated with object #4 which is at a projected distance of $135\,h^{-1}_{50}$ kpc from the line of sight.

3 Description of individual objects

3.1 A 0401-350A (z = 0.22)

3.2 PKS 0812+020 (z = 0.402)