3 Observations and data reduction

The V-band CCD images were obtained in two runs, on 1994 December 30-31, 1995 January 1-2, and 1995 May 26-29, at the 1m Jacobus Kapteyn Telescope at the Observatorio del Roque de los Muchachos (La Palma, Canary Islands). The chip used was a TEK with $1024\times1024$ pixels, providing a field of view at the f/15 focus of the JKT of $5.6\times5.6$arcmin². Typically two 900 s exposures were taken for each object in order to diminish the effects of cosmic-ray hits, saturation or guiding problems. Each couple of frames was then coadded after data reduction. Typical seeing was in the range 1-1.5 arcsec. Photometric standard stars were observed all nights in order to flux calibrate the images. The photometry zero-point errors were in the range 0.02 - 0.1 being in average of $\sim 0.05$ mag. Atmospheric extinction coefficients were around 0.15 mag airmass^-1.

During some of the nights we obtained short exposures of pairs of bright stars from the PPM catalogue (Roeser & Bastian [1988]) in order to measure the pixel scale and orientation of the chip. The resulting scale of the frames derived from 3 pairs of stars was $0.3301\pm0.0003$ arcsec/pixel and the position of the North was $0\hbox{$.\!\!^\circ$ }8 ~(\pm 0.01)$ clockwise from the frame Y axis. The preliminary reduction (bias subtraction, flat-field correction) was done using standard routines from the IRAF package. No special processing was done for the flat-fielding and we used sky flats. Flat-field correction was better than $0.7\%$.

The sky background of the images was calculated using two methods. The first was the simple technique in which the mean sky level is measured in regions free of interfering objects, evenly distributed accross the images. A second method consisted on computing the histogram of intensities on the frames and, after applying a clipping algorithm, estimate the mean value from the histogram. The sky background estimations obtained with the two methods typically agreed within $\sim 0.5$%. For the cases where a bright star produced a large extended emission, this emission was modelled and substracted from the original frame.

B2 1457+29 was observed during a service night at the WHT 4.2 m telescope using the ISIS double spectrograph to measure its redshift. A slit width of 2 arcsec was used and the dispersion was of 121 Å/mm providing a spectral resolution of $\sim 7$ Å. Standard procedures to calibrate the spectra were carried out. The absortion lines NaD, MgB, Fe1 and Ca1 were used and we obtained a redshift of $z=0.146\pm 0.003$.

Figure 1 shows V-band contour maps of the galaxies. The galaxies are identified in these maps, except for the obvious cases. The images were median-filtered before the contour maps were obtained. The sky brightness ( $\mu_{\rm sky}$), the first contour ( $\mu_{\rm first}$), the contour interval ($\Delta$mag), the signal-to-noise ratio for the first contour ($\sigma$ level), and the seeing are indicated for each map in Table 2. The magnitudes have been corrected for Galactic extinction (Burstein & Heiles [1982]). On each figure we give the scale in kpc for the adopted cosmology.

For five of the sources in Table 1 the identified optical galaxy is different from that given in the series of papers by the Bologna group (C75, F78, F87 and references therein) or we have selected one of the tentative identifications given in these papers. In addition, for three radio sources in Table 1 the association with the optical galaxies is uncertain or wrong. The identification for these eight cases is described below.

  

Table 3: Visual magnitudes of the B2 galaxies

 

Table 3: continued

B2 0838+32. The optical counterpart given by F87 consists of a galaxy and a fainter companion at 11 arcsec SE (see Fig. 1). The brighter galaxy has a better spatial coincidence with the radio emisison in the map presented at F86, and we therefore quote the position of this galaxy. The optical position listed by F87 was slightly shifted towards the fainter galaxy.

B2 0843+31. de Ruiter et al. ([1986]) and F87 note that the identification with the optical galaxy is uncertain, since the VLA radio map shows a bright unresolved component towards the south, which could be related to a background object. The CCD image does not show an optical counterpart of this source down to a limiting magnitude $V \sim 23-23.5$mag.

B2 0916+33. We noted a large offset between the position of the irregular galaxy and that quoted by C75. There are no published radio maps of this source, but its position is included in the VLA FIRST Survey (White et al. [1997]). The FIRST image shows a faint double source coincident with the position of the optical galaxy. The optical galaxy has a double nucleus, each of them coincident with one of the radio emission peaks. 43 arcsec SW of the galaxy there is a brighter pointlike radio source (R.A.(1950) = $09^{\rm h} 16^{\rm m} 49\hbox{$.\!\!^{\rm s}$ }2$ and Dec.(1950) = $+33^\circ 07'~54''$), coincident with a bright pointlike object (R.A.(1950) = $09^{\rm h} 16^{\rm m} 48\hbox{$.\!\!^{\rm s}$ }7$ and Dec.(1950) = $+33^\circ 07'~54''$), with V=18.45 mag on our image. The better similarity of this position with that given by C75 and the agreement in radio flux indicates that the pointlike optical object is likely the counterpart of the B2 source. In fact C75 quote S₁₄₁₅=100 mJy for the B2 source and FIRST and NRAO/VLA Sky Survey (NVSS, Condon et al. [1998]) quote respectively 89 and 93 mJy for the SW radio source and 22 mJy (similar value for both surveys) for the irregular galaxy. The new identification is indicated with an arrow in Fig. 1.

B2 1106+26. C75 list two optical galaxies for this radio source. The northern galaxy is the elliptical NGC 3534 and shows a better agreement with the radio position of the source in Douglas et al. ([1996], Texas Survey).

B2 1141+37. The optical counterpart given at F87 is the most sourthern one of a tight chain of four galaxies. We selected the galaxy marked in Fig. 1 as counterpart because it shows a better spatial coincidence with the central emission on the radio map presented by de Ruiter et al. ([1986]). This is also the optical counterpart given by Giovannini et al. ([1988]), who detected the core at 5 GHz at the position of this galaxy.

B2 1441+26. This source was identified with a spiral galaxy located at the center of the two lobes of the radio source. The 20 cm FIRST data and a 6 cm unpublished map (Parma, private communication) show that the core of the radio source is embedded in the eastern lobe and is located $\sim 36$ arcsec East of the spiral galaxy. The core is spatially coincident with a point-like object with V = 18.88 mag and coordinates R.A.(1950) = $14^{\rm h} 41^{\rm m} 56\hbox{$.\!\!^{\rm s}$ }3$ and Dec.(1950) = $26^\circ 14'~ 06\hbox{$.\!\!^{\prime\prime}$ }1$. This source is indicated with an arrow in the contour map in Fig. 1.

B2 1502+26, 3C 310. This source was originally identified as a galaxy pair in F78. The high resolution radio map obtained by van Breugel & Jägers ([1982]) shows that the western optical galaxy has a better spatial coincidence with the radio core. Smith & Heckman ([1989a]) also identify the western galaxy as optical counterpart.

B2 1511+26, 3C 315. This source was originally identified as a galaxy pair in F78 and F87. The northern galaxy shows a better positional coincidence with the radio core in the high resolution radio map by F87 and we therefore select this source as optical counterpart. The same identification is given by Smith & Heckman ([1989a]).

Figure 1: V-band contour maps of the observed B2 galaxies. Orientation is the standard (North up and East to the left) although the North is exactly $0\hbox{$.\!\!^\circ$ }8$ clockwise from the vertical axis. The contour levels, corrected from Galactic reddening, are indicated in Table 2