Low luminosity elliptical galaxies

Abstract

Long-slit spectra for the photometric axes of a sample of 14 elliptical galaxies, predominantly low-luminosity ellipticals, in the Virgo cluster and in nearby groups, are studied to investigate the galaxy kinematical structure and stellar evolutionary history. To determine the galaxy kinematical structure, the shape of the line-of-sight velocity distribution (hereafter LOSVD) is measured using the Fourier Correlation Quotient method of Bender (1990), adopting the parametrisation of the LOSVD due to van der Marel and Franx (1993). This parametrisation enables the asymmetrical and symmetrical deviations of the LOSVD from a Gaussian function to be measured by the amplitudes H(_3) and H(_4) of the Gauss-Hermite series respectively. Rotation, velocity dispersion (σ), H(_3) and H(_4) are determined as a function of radius for both the major and minor axes of our sample. To summarise, LOSVD asymmetries were measured for the major axes of 12 galaxies which in the majority of cases have been interpreted as evidence for central disk-like components; evidence of both radial and tangential anisotropy were found from the measurement of H(_4); central decreases in σ are measured for 3 galaxies, which is interpreted as evidence that they have undergone some form of merger or interaction. On the basis of their measurements, galaxies are classified into 3 classes: types 1, 2 and 3. "Type 1" galaxies show strong evidence for both disk and bulge components and have the greatest measured values of H(_3) for our sample. Galaxies of "type 2" show strong evidence for embedded disk components and most (3 of 4) are measured to have central decreases in a. "Type 3" galaxies have kinematically-decoupled cores. Other galaxies, not classified, are NGC 3379 and NGC 4468.Measurements of the line-strength indices Mg(_b), Mg(_2), Hβ, Fe5270, Fe5335 and <Fe> are determined as a function of radius for all spectra and established to the Lick/IDS scale. The relations Mg’(_b)-Mg(_2), Mg’(_b), - log(σ), Mg(_b)- <Fe> and Hβ-[Mg(_b) <Fe>] are then studied. The Mg’(_b),-Mg(_2) relation of Wegner et al. (1998) and the calibrations of Worthey (1994) are compared with measurements here: this is used as a check of our calibration of Mg’(b), and Mg(_2). Measurements in the Mg’(_b)-log(σ) plane are considered separately for each galaxy and compared with the central relation of Colless et al. (1998). Measurements for most galaxies are found to be in good agreement with this relation. Measurements of Mg(_b) and <Fe> are similarly considered for each galaxy and compared with the predictions of the models of Worthey (1994). For the majority of galaxies, measurements are clearly offset from the Mg(_b)-<Fe> model grid of Worthey (1994), representing an [(^Mg)-(_Fe)] overabundance. This is an important result which shows that the [(^Mg)-(_Fe)] overabundance detected previously for the giant ellipticals similarly exists for ellipticals of the low luminosities studied here. For most galaxies this overabundance is found to be a constant function of radius. For the "type 2" galaxy NGC 3605, and NGC 4468, measurements are consistent with solar abundance ratios. Finally, measurements of Hβ and [Mg(_b) <Fe>] are compared with the models of Worthey (1994) to distinguish gradients in both age and metallicity. For all galaxies, gradients in metallicity are found with metallicity decreasing as a function of radius. For 5 galaxies (NGC 4564, NGC 3377, NGC 4478, NGC 4339 and NGC 3605) age gradients are also detected, with the galaxy centre shown to be younger than the surrounding galaxy. Interpreting these results together, different formation scenarios are proposed for the different galaxy types. For galaxies of "type 1", formation by homogenous, dissipational collapse is proposed. Galaxies of "type 2" show evidence for a less homogenous evolutionary history involving dissipationless collapse. No conclusive scenario is proposed for galaxies of "type 3".