2006A&A...449..519M

Using 6 colours and 4 Lick line-indices we derive two-component models of the populations of ellipticals, involving a ``primary'' and a ``juvenile'' population. The first component is defined by the regressions of indices against the central velocity dispersion found in Papers I and II for the Nop sample of non-peculiar objects. The second one is approximated by an SSP, and the modeling derives its age A, metallicity Z and fractional V-luminosity q_V, the fractional mass q_M being found therefrom. The model is designed for ``blueish'' peculiar galaxies, i.e. the Pec sample and NGC 2865 family in the terminology of Paper I. The morphological peculiarities and the population anomaly are then believed to involve the same event, i.e. a merger plus starburst. It is possible to improve the models in a few cases by introducing diffuse dust (as suggested by far IR data), and/or by taking into account the fact that Lick- and colour indices do not relate to identical galaxy volumes. In most of the cases, the mass ratio of young stars q<sub>M</sub> seems too small for the product of a recent major merger: the events under consideration might be minor mergers bringing ``the final touch'' to the build-up of the structure of the E-type object. The same modeling has been successfully applied to blueish galaxies of the Nop sample, without morphological peculiarities however, to support the occurence of a distinct perturbing event. A few reddish objects of the Pec sample (NGC 3923 family) and of the Nop sample are also modeled, in terms of an excess of high metallicity stars, or diffuse dust, or both, but the results are inconclusive.