In the center of galaxy clusters the various types of dwarf galaxies differ in properties, such as their spatial distribution and luminosity function. Early-type dwarf ellipticals, especially nucleated ones, are the most clustered, whereas late-type dwarfs are preferentially found at the outskirts of clusters (e.g. review by Ferguson & Binggeli 1994, and references therein). The different environmental properties probably reflect the result of the dynamical procresses during the formation and evolution epoch of the clusters.
Among low mass galaxies, a striking but rare subgroup are the compact ellipticals (cEs), or M 32-type galaxies. It is under discussion, whether environmental effects such as tidal stripping have shaped their present appearance (e.g. Faber 1973) or whether these galaxies just represent the low-luminosity end of the giant ellipticals (Nieto & Prugniel 1987). E.g., M 32 itself is a companion of the Andromeda galaxy (M 31). The non-existence of any globular cluster (GC) in this galaxy suggests that previously present GCs might have been captured by M 31 during past close passages. Whether the interaction with M 31 also has formed the compact shape of M 32 remains unclear.
At the distance of the nearest galaxy clusters, the classification of compact objects such as M 32 just by morphological properties is nearly impossible. Their high central surface brightness and de Vaucouleurs surface brightness profile make them indistinguishable from background ellipticals. In the Fornax Cluster Catalog (FCC) of possible members and likely background galaxies, Ferguson (1989) lists 131 candidate cEs. Recently, Drinkwater et al. (1997, see also Drinkwater & Gregg 1998) have measured radial velocities of 67 of them. No one is member of the Fornax cluster. The fact that very few cEs are found indicates that tidal stripping does not play a major role in their formation in rich environments. However, the member and background galaxy catalogs of Ferguson are limited in the angular diameter of the objects. Especially, galaxies with diameters comparable to M 32 itself and smaller compact ellipticals are beyond the resolution limit of Ferguson's sample.
A second type of objects in the galaxy cluster populations that could easily be missed by morphological classifications are nuclei of tidally disrupted nucleated dwarf ellipticals (dE, Ns). Numerical simulations by Bassino et al. (1994) reveal that such nuclei can survive the dissolution in the gravitational field even during the entire lifetime of the universe and would appear as luminous globular cluster-like objects. The nuclear magnitudes of all Virgo dE, Ns (Binggeli & Cameron 1991), for example, fall indeed in the magnitude - surface brightness sequence defined by the globular clusters (e.g. Binggeli 1994). At the distance of the Fornax cluster, these objects would hardly be resolved and can only be uncovered by spectroscopic observations.
In the first paper (Hilker et al. 1998, hereafter Paper I) a galaxy catalog with photometric properties and surface brightness profiles for galaxies in selected fields of the central Fornax cluster has been presented. An excess population of galaxies near NGC 1399, the central galaxy of the cluster, has been found as compared to the other Fornax fields and to absolute background fields. The photometric analysis has shown that most of the excess galaxies have sizes and surface brightnesses which are more typical for background spirals and ellipticals than for dwarf ellipticals. However, as discussed before, photometric properties alone are not sufficient to distinguish between background galaxies and high surface brightness dwarf galaxies in the Fornax cluster. This paper presents redshift determinations of a bright sub-sample of our photometric catalog (Paper I) to investigate the nature of the mentioned excess galaxies. Furthermore, line indices for the objects that have been identified as Fornax members were measured.
In the following the expression "radial velocity" has been used for the measurement of cz instead of redshift, being aware of the fact that the true radial velocity for high z differs from cz depending on the applied cosmological model.
Previous radial velocity measurements of galaxies in the Fornax cluster
brighter than mag
were presented by Jones & Jones (1980), Lauberts
(1982) and Richter & Sadler (1985).
They are compiled in the Fornax Cluster Catalog (FCC) by
Ferguson (1989).
Except for the giant galaxies, there are only two galaxies that overlap with
our sample: NGC 1396 and FCC 222, two bright dE, Ns.
Held & Mould (1994) took spectra of 10 dE, Ns in
the Fornax cluster; one of these is in common with our sample.
Sections 2, 3, and 4 give a detailed description of the observations, data reduction, and velocity determination. The resulting radial velocities and the analysis of individual objects are presented in Sect. 5. The main results are summarized in Sect. 6.
Copyright The European Southern Observatory (ESO)