Description of the catalogue

We have arbitrarily defined a quasar as a starlike object, or an object with a starlike nucleus with broad emission lines, brighter than absolute magnitude M$_{\rm B}$ = -23; only objects brighter than this limit appear in table 1.

In table 2, we list all confirmed, probable or possible BL Lac objects with or without a measured redshift, without consideration of their absolute magnitude. As better spectra are becoming available, broad emission lines have been detected in a number of objects formerly classified as BL Lac; they have usually been moved to table 1 (Véron-Cetty & Véron, 2000).

The Seyfert 1 galaxies have broad Balmer and other permitted lines; the Seyfert 2 galaxies have Balmer and forbidden lines of the same width. Osterbrock (1977; 1981) has divided the Seyfert galaxies showing broad Balmer lines (Seyfert 1) into five subgroups : Seyfert 1.0, 1.2, 1.5, 1.8 and 1.9 on the basis of the appearance of the Balmer lines. Seyfert 1.0s are ``typical" members of the class, as described by Khachikian & Weedman (1971; 1974), while Seyfert 1.5s are objects intermediate between typical Seyfert 1 and Seyfert 2 galaxies, with an easily apparent narrow H${\beta}$ profile superimposed on broad wings. The classes Seyfert 1.2 or 1.8 are used to describe objects with relatively weaker and stronger narrow H${\beta}$ components, intermediate between Seyfert 1.0 and 1.5 and Seyfert 1.5 and 2 respectively. In Seyfert 1.9, although the broad H${\alpha}$ emission is clearly evident, broad H${\beta}$ cannot be detected with certainty by mere visual inspection of the spectra.

We have adopted the more quantitative classification introduced by Winkler (1992) :

				 
     S1.0    5.0   < R 

     S1.2    2.0   < R < 5.0 		

     S1.5    0.333 < R < 2.0 		

     S1.8            R < 0.333 broad component visible in H${\alpha}$ and H${\beta}$
     S1.9                      broad component visible in H${\alpha}$ but not in H${\beta}$ 
     S2                        no broad component visible 

where R is the ratio of the total H${\beta}$ to the [OIII]$\lambda$5007 fluxes.

Several objects have been found to show extreme spectral variability, changing from Seyfert 1.8 or 1.9 to Seyfert 1. In some cases these changes are consistent with changes in the reddening to the BLR while, in others, they are probably due to real changes in ionizing flux (Goodrich, 1989a; 1995; Tran et al. 1992b). In some Seyfert 2 galaxies, a broad Pa ${\beta}$ line has been detected, indicating the presence of a highly reddened broad line region (Goodrich et al. 1994); we call these objects S1i. A number of Seyfert 2 galaxies have, in polarized light, the spectra of Seyfert 1s (Antonucci & Miller, 1985; Miller & Goodrich 1990; Tran et al. 1992a); we call them S1h. Typical full widths at half-maximum of the Balmer lines in Seyfert 1 galaxies lie in the range 2000-6000 km s^-1. However, there is a group of active galactic nuclei with all the properties of Seyfert 1 galaxies, but with unusually narrow Balmer lines (Osterbrock & Pogge, 1985; Goodrich, 1989b); they are defined as having the broad component of the Balmer lines narrower than 2000 km s^-1 FWHM (Osterbrock, 1987); we call them S1n.

Table 3 lists ``active galaxies" : Seyfert 1, 1.5, 1.8, 1.9, 1h, 1i, 1n, 2, and Liners (as defined by Heckman, 1980) fainter than the above quoted absolute magnitude. A number of galaxies with a nuclear HII region denoting a burst of star formation are also included, the reason being that they have been called Seyfert in the past and later reclassified; we consider it useful to keep trace of these reclassifications to avoid further confusion.

Table 4 lists the objects which once have been believed to be quasars or BL lac objects and are now known to be either stars or normal galaxies.

Table 1 contains 13$\,$214 objects, table 2, 462 , table 3, 4$\,$428 and table 4, 55.

The catalogue is believed to be complete for quasars, BL Lac objects and Seyfert 1 galaxies.

Since the discovery in 1979 by Walsh et al. of the first gravitationally lensed quasar, Q 0957+561, a number of such objects (40) and of physical pairs with separation less than 10" (10) have been found. They are listed in table 5. Mortlock et al. (1999) have stressed the difficulty sometimes encountered in distinguishing lensed quasars from physical pairs.

Table 6 gives the list of all the Markarian objects appearing in the catalogue :

    col. 1 :  Markarian number
    col. 2 :  Name under which the galaxy appears in the present catalogue when it is not the
              Markarian number
    col. 3 :  Short position
    col. 4 :  Table in which the object appears. 

Tables 1, 2 and 3 give :

1) The most common name of the object. For the meaning and the sources of the designations see Hewitt & Burbidge (1987), Fernandez et al. (1983) and Kesteven & Bridle (1977). For the sources discovered by the ROSAT X-ray satellite, we have used the following acronyms : RXS for the sources appearing in the All-Sky Bright Source Catalogue (Voges et al. 1999), 1WGA for the sources published in the WGACAT catalogue (White et al. 1994) and RX for the others.

When the name is preceded by an *, the object has not been detected as a radio source, either because it has not been observed or because it is fainter than the sensitivity limit of the radio telescope used; there is no uniform upper limit to the flux density of the undetected objects.

2) The best available J2000 optical or radio coordinates; the J2000 positions have been converted from the B1950 positions using the matrix given by Aoki et al. (1983). No references are given for the sources of these positions. An O or an R following the coordinates means that the position given is either an optical or a radio position measured with an accuracy better than one arc sec. An A means that it is only an approximate position which may be wrong by several arc minutes. The availability of the Digitized Sky Survey (DSS) allows quick measurements of the optical position of any object brighter than $\approx$19.5 mag. It has already been used to measure the position of several hundred QSOs (Schneider et al. 1992; Bowen et al. 1994; Kirhakos et al. 1994; Véron-Cetty & Véron, 1996). Optical positions with an accuracy better than 2" have also been measured for the 19$\,$369 galaxies in the Zwicky catalogue (Falco et al. 1999) and for the 12$\,$921 UGC galaxies (Cotton et al. 1999).

3) The 6 and 11 cm flux densities (in Jy) with references to the literature. When several measurements are available we took arbitrarily one of them. When a reference is given for the 6 cm flux density but the value of the flux density itself is left blank and there is an * in column 1, only an upper limit is available and this upper limit is not much greater than 1 mJy; in case there is no * in column 1, the reference refers to a detection but at a wavelength other than 6 cm.

4) The redshift as published. An * in front of the redshift means that it has been estimated from a low dispersion slitless spectrum and is of lesser accuracy or even plainly wrong as the emission lines may easily be misidentified. We have given only those values which are described as probable in the original sources and not the possible values.

5) In this column an attempt has been made to classify the objects as follows :

    S1          Seyfert 1 spectrum with broad permitted hydrogen emission lines, without  sufficent
                information for a more detail classification.
    S1.0, S1.2, S1.5, S1.8, S1.9, S1i, S1h, S1n (see above). 		
    S2          Seyfert 2 spectrum defined as having a strong [O III]$\lambda$ 5007 line compared with H${\beta}$,		 		 		 together with a [NII]$\lambda$6584 line of strength comparable with that of H${\alpha}$.		 		
    S3          Seyfert 3 or Liner as defined by Heckman (1980). Broad Balmer lines are observed
                in some Liners; they are called S3b.
    S or S?     These objects are probably  or possibly  Seyfert galaxies, but available  data are
                insufficient for a more detailed classification.
    H2          Nuclear HII region, i.e. nucleus of a galaxy with an emission line spectrum similar
                to that of an emission nebula ionized by hot stars.
    HP          A high degree of optical polarization (> 3%) has been measured (see for instance : 
                Impey et al. 1991; Moore & Stockman, 1984). 

Low redshift quasars are classified as S1 when a good spectrum shows that they are similar to Seyfert 1 galaxies.

In table 2, we find in this column :

    BL      for a confirmed BL Lac object.
    blank   for a probable BL Lac.
    ?       for a possible BL Lac. 

6) The V, B-V and U-B photoelectric or photographic magnitude and colours, when available. (the survey of the literature for photographic colours may be incomplete) (an * in front of the magnitude indicates that the colours and the magnitude are photographic, while an R or an I indicates a red or an infrared magnitude). Maoz et al. (1993) have measured homogeneous V magnitudes for 354 QSOs with an accuracy of $\pm$0.1 mag; they have been included. For a few objects, the O magnitude, measured on the blue Palomar Sky Survey plates, or the UK Science Research Council SRC-J Survey plates, believed to be accurate within $\pm$ 0.2 mag., has been extracted from the APS database (Pennington et al. 1993). For a number of objects we give the O magnitude, extracted from the USNO-A2 catalogue (Monet et al. 1996) or the Cambridge Automated Plate Measuring Machine (APM) catalogue (Irwin et al. 1994), recalibrated by E. Flesch (private communication); these magnitudes are flagged with an O. The O and Johnson B magnitudes are related by B-O = -(0.27$\pm$0.06) (B-V) (Evans, 1989).

In the other cases, the magnitude given is an estimate as found in the original publications; these magnitudes are generally quite inaccurate and inhomogeneous; they are most often m$_{\rm pg}$ or B magnitudes instead of the Johnson V magnitude. Much care should be taken when using them for any purpose. Anyway, even when a photoelectric V magnitude is given, it is not very meaningful as most quasars are variable. On the other hand, the colours of quasars vary little, so the listed colours should be accurate. Again, it should be noted that some of the colours listed are photographic and, therefore, less accurate; moreover, in each catalogue of photoelectric measurements, the faintest objects measured are affected by relatively large errors; this too should not be overlooked. For the galaxies, in table 3, we have chosen the magnitudes and colours measured in the smallest possible diaphragm (preferentially 16 arcsec) as we are interested in the nucleus rather than in the galaxy itself.

7) The absolute magnitude M$_{\rm B}$ computed by assuming H₀=50 km s^-1 Mpc^-1, q₀=0, and an optical spectral index ${\alpha}$ equal to 0.3 (defined as S $\propto$ $\nu ^{-\alpha}$) (Francis et al. 1991).

The absolute magnitude is computed as follows : M = m + 5 - 5 log D - k + $\Delta$ m(z) where m is the B magnitude, D = c/H₀ A, with A the photometric distance (Terrell, 1977) :

$$A=z[1+\frac{z(1-q_{0})}{(1+2q_{0}z)^{0.5}+1+q_{0}z}]$$

z is the redshift taken from column 5; k=-2.5 log(1+z)$^{1-\alpha}$ is the k correction, $\Delta$m (z) is a correction to k taking into account the fact that the spectrum of quasars is not strictly a power law of the form S $\propto$ $\nu ^{-\alpha}$, but is affected by emission lines and by the Ly ${\alpha}$ forest depleting the continuum to the blue of Ly ${\alpha}$. Assuming that the spectrum is a power law with ${\alpha}$=0.3 may not give the best possible estimate of the k correction (Wisotzki 2000).

The column labelled ``V" gives the V magnitude when B-V is also given in which case we have used B = V+(B-V). When B-V is not given, this column usually gives the B magnitude, unless it is preceded by an R or an I; the R magnitudes have been transformed into the B system by using an average <B-R>=0.57 and the I magnitudes by using <B-I>=1.1 for low z QSOs. When the reference for the magnitude is Maoz et al. (1993), the magnitude is V and we have used <B-V>=0.40.

We list in table 1 those objects which have an absolute magnitude M$_{\rm B}$ brighter than -23; clearly some objects would move from table 1 to table 3 and vice versa if other values for q₀ and the spectral index were used or if an accurate B apparent magnitude was available for all objects.The variability may have a similar effect, as well as the size of the diaphragm used for the measurement as the contribution of the underlying galaxy for weak quasars may not be negligible.

8) The next three columns give the reference for the finding chart, the photometry and the redshift respectively. In many cases, the last reference in table 3 is that of the classification of the object (as a Seyfert galaxy or otherwise); in these cases the redshift can usually be found in Palumbo et al. (1983).

9) The B1950 position.