Star formation history of early-type galaxies in low density environments IV. What do we learn from nuclear line-strength indices?

In this paper we analyze the line-strength indices in the Lick-system measu red by Longhetti et al. (1998a, b) for a sample of 51 early-type galaxies l ocated in low density environments (LDE) and showing signatures of fine str uctures and/or interactions. The sample contains 21 shell-galaxies and 30 m embers of interacting pairs. Firstly we perform a preliminary comparison between three different sources of calibrations of the line strength indices, namely Buzzoni et al. (1992, 1994), Worthey (1992), Worthey et al. (1994) and Idiart et al. (1995), der ived from stars with different effective temperature, gravity, and metallic ity. Looking at the three indices in common, i.e. Mg2, Fe5270, and HP, the calibrations by Buzzoni et al. (1992, 1994), Worthey (1992) and Worthey et al. (1994) lead to mutually consistent results. The calibration of H beta b y Idiart et al. (1995) can be compared with the previous ones only for a li mited range of ages, in which good agreement is found. Mg2 and Mgb indices predicted by the Idiart's et al. (1995) fitting functions result to be syst ematically lower than those obtained from using Worthey (1992) calibrations . Secondly, we discuss the properties of the galaxies in our sample by compar ing them both with theoretical Single Stellar Populations (SSPs) and the no rmal galaxies of the Gonzalez (1993: G93) sample. The analysis is performed by means of several diagnostic planes. In the sigma, Mg2, Fe5270 and Fe5335 space, normal, shell- and pair-galaxie s have a different behavior. First of all, normal and pair-galaxies follow the universal a vs. Mg2 relation, whereas shell-galaxies lie above it; seco ndly the Fe versus Mg2 relation of normal, shell- and pair-galaxies is flat ter than the theoretical expectation. This fact hints for enhancement of al pha-elements with respect to solar partition in galaxies with strong Fe ind ices and/or high velocity dispersion, mass and luminosity in turn. In the sigma vs. H beta plane normal galaxies seem to follow a nice relatio n suggesting that objects with shallow gravitational potential have strong H beta values (youth signature?), whereas shell-and pair-galaxies scatter a ll over the plane. A group of galaxies with deep gravitational potential an d strong HP is found. Is this a signature of recent star formation? In the H beta vs. [MgFe] plane,(1) which is perhaps best suited to infer th e age of the stellar populations, the peculiar galaxies in our sample show nearly the same distribution of the normal galaxies in the G93 sample. Ther e is however a number of peculiar galaxies with much stronger H beta. Does this mean that the scatter in the H beta vs. [MgFe] plane, of normal, shell - and pair galaxies has a common origin, perhaps a secondary episode of sta r formation? We suggest that, owing to their apparent youth, shell- and pai r-galaxies should have experienced at least one interaction event after the ir formation. The explanation comes natural for shell- and pair-galaxies wh ere the signatures of interactions are evident. It is more intrigued in nor mal galaxies (perhaps other causes may concur). Noteworthy, the distribution in the H beta vs. [MgFe] plane of normal, shel l- and pair-galaxies is confined within a narrow strip that runs significan tly steeper than the path followed by aging SSPs. This feature is explained as due to metal enrichment always accompanying star formation. Shell-galaxies encompass the whole range of ages inferred from the HP vs, [ MgFe] plane, indicating that among them recent and old interaction/acquisit ion events are equally probable. If shells are formed at the same time at w hich the rejuvenating event took place, shells ought to be long lasting phe nomena.