1 Introduction

The very metal-rich dwarf stars in the solar neighbourhood have historically not attracted as much attention as the more metal-poor (solar like and halo stars) stars which tell us about the early phases of the chemical evolution of our galaxy. The properties of metal-rich stars are important when we try to interpret integrated spectra from metal-rich stellar populations, such as the Bulge and giant elliptical galaxies. A small group of so called super-metal-rich (SMR) stars have played a significant role in shaping the conceptions of such populations. Famous examples are the dwarf HD 32147 (HR 1614) and the giant $\mu$ Leonis. In the review by Taylor (1996) - the latest paper in a long series started in the 1960s - SMR stars are discussed in great detail, in particular, the reality of extremely high [Fe/H]. Taylor found that no giant star fulfills the criteria for SMR-ness that he sets and only a handful of dwarf stars do, and that most of them are candidates rather than firm members in this class. $\mu$Leonis has, however, been studied by several groups using high-resolution spectroscopy, a recent example being Smith & Ruck (2000), who find that the star is indeed super-metal-rich with ${\rm [Fe/H]} = +0.29$. Thus, the question of the reality of super-metal-rich giants is still very much alive and each case has to be judged on its own.

The exact definition of super-metallicity has, as reviewed by Taylor (1996), varied. Spinrad & Taylor (1969) adopted +0.2 dex as the lower limit, based on the overall metallicity of the Hyades, which they found to be +0.2 dex. The metallicity for the Hyades has recently been revised (Taylor 1994; Cayrel de Strobel 1997) to +0.1 dex. Even values as low as 0.0 dex have been quoted. This has resulted in classes of stars that sometimes are regarded as SMR and sometimes not. Taylor rectified this unsatisfactory situation by adopting the original +0.2 dex as the threshold on the grounds that no giant stars had been shown to have a metallicity higher than this value (but see Castro et al. 1997; Smith & Ruck 2000). Taylor (1996) defines a star to be SMR if it has ${\rm [Fe/H]}> 0.20$ with 95% confidence. He also adopts [Fe/H], i.e. the iron abundance, as the measure of "metallicity'' rather than the more general [Me/H]. As an aside one may note that a second terminology is also in use - Very Strong-Lined (VSL) star. This term implies just that the star has strong lines and might therefore be a SMR candidate. This is a particularly useful term when working with low resolution spectra.

SMR stars have attracted more attention recently due to their possible connection with extra-solar planets, e.g. Gonzalez (2000 and references cited therein), Fuhrmann et al. (1997, 1998). Gonzalez (2000) has shown that the solar-type parent stars of extra-solar planets are more metal-rich on average compared to the general field star population. In particular, the very short period systems are either above the SMR limit or near it. By comparing them to the SMR stars we may gain insight as to the relationship between planets in short-period orbits and the SMR-ness of the parent star.

A few other recent studies have targeted known SMR candidates and stars with high [Me/H] (as derived from photometry): Feltzing & Gustafsson (1998); Castro et al. (1997); and McWilliam & Rich (1994). In general the abundance ratios seem to continue the trends of the disk population. However, no detailed theoretical predictions for Galactic chemical evolution exists for ${\rm [Fe/H]}>0.2$ dex, so the interpretation of the observed abundance trends for metal-rich stars is still pending.

The combination of abundance ratios with kinematical data may give us additional clues. For example, we can study stars on highly eccentric orbits which trace the evolution in the Galactic disk closer to the Galactic centre. Not much is known about these stars, but there are some very intriguing observations: Barbuy & Grenon (1990) found that dwarf stars on very eccentric orbits contained much more oxygen than what was expected from standard models of Galactic chemical evolution of the disk, and Edvardsson et al. (1993) found large spreads and "upturns'' for certain elements, Na, Si, Ti, Al, for stars with $0.0~{\rm dex} < {\rm [Fe/H]}< 0.2$ dex. The trends for Na, Si and Ti were confirmed up to $\sim$0.4 dex by Feltzing & Gustafsson (1998). They concluded that the "upturn'' in Na abundances relative to Fe is not due to a mixture of stars born at different distances from the Galactic centre.

In this paper we investigate, by means of detailed spectroscopic analyses, the metallicities as well as the abundance of several elements for 8 dwarf stars selected from the meticulous review of SMR candidates by Taylor (1996).

The paper is organized as follows: in Sects. 2 and 3 we detail the observations and the selection of program stars, as well as reductions and measurements; Sect. 4 discusses the detailed abundance analysis, Sect. 5 presents the abundances element by element, in Sect. 6 we derive ages for the stars and discuss the age-metallicity relation in the solar neighbourhood, Sect. 7 discusses the kinematics of the stars in our sample and which galactic component they belong to, Sect. 8 provides a short discussion of the SMR-planet connection and, finally, Sect. 9 summarizes our findings.