Astron. Astrophys. 322, 730-746 (1997)

2. The KLUN sample

The galaxy sample that we use is basically the same that we have collected and investigated in a few previous papers of the "Extragalactic Data Base" (Paturel et al. 1989, Bottinelli et al. 1990, Paturel et al. 1991a, 1991b, 1994a, Bottinelli et al. 1995) and "Kinematics of the Local Universe" series (Paturel et al. 1994b, Bottinelli et al. 1992, 1993, Di Nella et al. 1996, Theureau et al. 1997). In totality it consists of 5171 spiral galaxies covering the type range Sa - Sdm (or 1 - 8). Each galaxy has a photometric diameter (), a 21-cm line width and a radial velocity measurement. 4577 of these objects have also B - magnitudes.

Any derivation of using such a data base, depends crucially on our knowledge of the statistical properties of the sample, including its incompleteness and the quality of the measured parameters. The KLUN sample has its origin in a merger of several galaxy catalogues, with the original photometric measurements made in different ways and accuracies. This makes it necessary to practice special care when the formal methods are applied to the sample.

2.1. Diameters

Diameters are expressed in units of 0'.1, with an accuracy of 0'.01, following the convention of the RC2 (de Vaucouleurs et al. 1976). They are extracted from LEDA and have been reduced to a standard common system according to Paturel et al. (1991a): isophotal diameters at the limiting surface brightness of 25 B -mag (arc sec)^-2. For most of the sample, the standard error of does not exceed . Diameters were corrected for galactic extinction (Fouqué and Paturel, 1985) and inclination effect (Bottinelli et al. 1995).

The sample can be considered complete in observed diameter down to (for details, see Paturel et al. 1994b). The diameter is measured in most cases in the disc only (Theureau et al. 1997). Hence, the diameter limit is roughly the same for all morphological types.

2.2. Magnitudes

The reduction of apparent B -magnitudes to the RC3 (de Vaucouleurs et al. 1991) system has been studied recently by Paturel et al. (1996). The mean is an average with the weights derived for each source of magnitude. The final uncertainty is derived from the total weight. In most of the sample, the standard error is less than 0.5 mag.

Apparent magnitudes are corrected for galactic extinction (), inclination and redshift effects, according to Bottinelli et al. (1995). We adopted for the RC2 system (see Sect. 7.2.).

The whole sample is complete below mag. It is important to note, that because the sample is diameter-limited and the magnitudes are correlated with the size of the bulge component for a fixed intrinsic diameter, the magnitude limit will be different for different galaxy types. In fact, for a given completeness limit in diameter , one obtains, from the exponential law, the following expression:

where is the bulge to disc luminosity ratio for a given type, as given by Simien & de Vaucouleurs (1986), and is the central brightness of a pure disc , Bosma & Freeman 1993, Theureau et al. 1997). One expects then that increases slowly from early to late types with a total shift of 0.6 mag (value obtained with and , see Sect. 5).

2.3. 21-cm line measurements

All raw 21-cm line widths, collected from LEDA, have been reduced to a common system of widths at 20 % and 50 % of the peak (Bottinelli et al. 1990). They are corrected for velocity resolution effect and tested for homogeneity. Large part of the data (almost 50 %) is based on our own observations at the Nançay Radiotelescope. The detection rate at Nançay was excellent (roughly 90 %). The correction for internal velocity dispersion was made according to Tully & Fouqué (1985). Finally,

where I is the inclination of the galaxy. The uncertainties on are typically around 0.08, as calculated according to Bottinelli et al. (1983).

2.4. Radial velocities and kinematical distances

Heliocentric radial velocities from optical or radio measurements are corrected to the centroid of the Local Group according to Yahil et al. (1977). The original optical compilation was made for the preparation of the RC3 catalogue (Fouqué et al. 1992). Radio velocities come essentially from 21-cm line measurements; they have been homogenized and tested for scale, zero-point errors, and reference effect (Bottinelli et al. 1990).

Radial velocities are used to define a kinematical distance scale, , with an infall component in the direction of the Virgo Cluster (). Similarly as in BGPT86, we applied Peebles's (1976) linear infall model. In most of the calculations we used 980 km s (Mould et al. 1980) for the Virgo mean velocity and 150 km s^-1 as in Bottinelli et al. (1995) for the LG infall.

2.5. Morphological types

Even after attempts to characterize rigorously the elements of the Hubble sequence, morphological type still remains a rather subjective datum, and experience shows that its definition depends on astronomer. We adopted the RC3 system as a standard, and converted to it the morphological types from other catalogues (see Paturel et al. 1997).

2.6. Adopted restrictions

The sample was restricted according to the following criteria:

• , to exclude objects near the Galactic plane having poorly defined magnitudes; this concerns 420 galaxies of our sample
• to exclude face-on galaxies for which the calculated is uncertain. This concerns 501 galaxies of our sample
• galaxies belonging to the triple value region around the center of the Virgo cluster have been excluded (see BGPT86). This concerns 107 galaxies in the case of our standard velocity field
• very close galaxies (sometimes with negative radial velocity) are excluded because of the importance of their peculiar motions

After these restrictions, we are left with 4164 and 3622 galaxies for the diameter and magnitude TF relations, respectively.

© European Southern Observatory (ESO) 1997

Online publication: June 5, 1998

helpdesk.link@springer.de