2 Spectroscopic observations

Several spectroscopic observations of REX sources have been carried out during the period 1995/1998 using the 88'' telescope of the University of Hawaii (UH) in Mauna Kea (5 observing runs), the 2.2 m and 3.6 m telescopes of ESO in La Silla (1 run) and the UNAM 2.1 m telescope in S. Pedro Martir (3 runs). The results coming from the first two observing runs at UNAM have been presented in Wolter et al. (1998) and will not be considered here.

The instrumental configurations are summarized in Table 1. In all cases, we have used a long-slit and low dispersion (from 3.7 Å/pixel to 13.2 Å/pixel) grating that maximizes the wavelength coverage. For the data reduction we have used the IRAF longslit package. The spectra have been wavelength calibrated using an He-Ar (UNAM, ESO) or a Hg-Cd-Zn (UH) reference spectrum. The photometric standard stars used for the relative flux calibration are: Feige 34 (UNAM 96/12), HD 19445 (UH 96/01, UH96/08), LTT377 (ESO 96/12), SAO 098781 (UH 97/03), HD 84937 (UH 98/02), PG0216+032 (UH 98/10).

In general, we have two exposures for each object, except for few cases in which we have only one spectrum. The subtraction of the cosmic rays has been made manually, from the extracted spectrum.

On average, the seeing during the observing runs ranged from 0.9 $^{\prime\prime}$ (August 1996, December 1996, February 1998) to 1.5 $^{\prime\prime}$. We have consequently used, except for few cases, a slit of 1.5-1.6 $^{\prime\prime}$ to maximize the signal.

In Tables 2,  3 and  4 we present the journals of the observing runs at UH 88'', ESO 2.2 m/3.6 m and UNAM 2.1 m telescopes, respectively. For each object we report name, NVSS position (J2000), X-ray flux corrected for Galactic absorption (in units of 10^-13 erg s^-1 cm^-2 in the 0.5-2.0 keV band), NVSS integrated flux density at 1.4 GHz (in mJy), observing set-up, date of the observation and total exposure time. The objects without the prefix "REX" in the name do not belong to the final REX catalog.

The X-ray flux has been derived from the count-rate using the value of Galactic $N_{\rm H}$ at the source position (Dickey & Lockman 1990) and assuming $\alpha_{\rm X}=1$, which is the mean value expected for our sources. For some of the sources that are not in the final REX catalog, the X-ray flux has been computed from the count-rate found in the 1RXP catalog (ROSAT NEWS No. 32, 1994), converted from the 0.1-2.4 keV to 0.5-2.0 keV band. These X-ray fluxes are indicated with an asterisk in Tables 2,  3 and  4. The uncertainty on the X-ray fluxes is about 20%. The typical rms of the radio (NVSS) maps is 0.45 mJy/beam. We note that in the case of objects in cluster or in group of galaxies the X-ray flux is the sum of the active nucleus plus the extended thermal component. Given the available statistics and the spatial resolution of the PSPC, particularly poor in the external part of the field, in most cases we are not able to distinguish the two components. We have evidence, from the optical images, that 4 objects belong to a cluster/group of galaxies: we believe that in these cases the X-ray luminosity is mainly due to the diffuse intracluster gas (see Sect. 3).