The time that F and G dwarf stars spend on the main-sequence span a range from 109 to several times 1010 years. This means that such stars may be used as tracers of the chemical and dynamical evolution of the Galaxy; in fact, a combination of chemical and kinematical data is a very powerful tool for studying the galactic chemical evolution, cf. e.g. Edvardsson et al. (1993a) and Wyse & Gilmore (1995).
Spectroscopic abundance analysis of stars have now become accurate
enough to admit determinations of rather small (
dex) relative
abundance differences in differential studies of stellar samples of
large size. These developments make it possible to explore the
galactic chemical evolution in considerable detail.
Edvardsson et al. (1993a) analysed 189 F and G dwarf stars, with -1.1 dex 
dex. Accurate abundances were determined for a number of
key elements, O, Na, Mg, Al, Si, Ca, Ti, Fe and Ni, as well as a number
of s-process elements. The abundance results combined with accurate
velocity data enabled a detailed study of the chemical evolution of
kinematically distinct populations. Extensions of this study were
made by Tomkin et al. (1995) and Woolf et al. (1995)
who measured carbon and
europium abundances, respectively, for about half of the stars in the
Edvardsson et al. (1993a) sample.
The study by Edvardsson et al. (1993a) raised a number of new
questions concerning the most metal-rich stars in the galactic disk,
not the least concerning the build-up of sodium, magnesium and
aluminium. Sodium, aluminium and, possibly, magnesium relative iron vs.
[Fe/H] showed an increase for 
dex (cf.
Edvardsson et al. 1993a Figs. 15a-l). Are these "upturns'' real? Large star-to-star
scatter was also encountered for the abundance of certain elements:
magnesium, aluminium and titanium, relative to iron at a given
[Fe/H]. Could this scatter be reduced by using more/better abundance
criteria or is the scatter intrinsic to the stellar population? One
suggestion was that "upturns'' and scatter could be due to a mixing
of populations with different ages and with different mean distances
from the galactic centre, e.g. a mixture of old metal-rich stars, more
concentrated to the centre and young Extreme Population I stars on
solar like orbits. We have therefore studied a sample of 47 metal-rich
dwarf stars, with photometric metallicities 
dex, chosen to represent different mean perigalactic distances and
presumably different ages.
The paper is organized as follows: in Sects. 2 and 3, we describe the selection criteria of the stellar sample, the observations and reductions, Sect. 4 contains a description of the analysis while the errors are discussed in detail in Sect. 5. Abundance results and their interpretation in terms of models of galactic chemical evolution are presented in Sect. 6 and, finally, Sect. 7 contains a summary and discussion.