The sample of stars of this new Edition of the Catalogue has conserved several biases, and
cannot be considered as
representative of the stellar content of the solar neighbourhood.
Figure 2 (click here) represents the distribution of
observations listed in File 1 in the plane (
, 
).
Some parts of this diagramme have a higher density of observations than in an
usual HR diagramme. This correspond probably to the personal interest
of astronomers in problems concerning Am stars 
,
F dwarfs and subgiants 
,
early G solar type stars 
,
early G giants 
,
late G and K giants 
.
One can note also a lack of
G and K dwarfs. The principal reason is that these stars,
being intrinsically faint, are more difficult to observe at high resolution and high S/N
than the giants of the same 
.
It is a pity for the study of stellar evolution, because it is well known that
the
full span of ages is
still present
among low mass G and K and later type stars. M star spectra are very complicated
and are difficult to analyse in detail, and in this Edition there is still a very low number
of them.
Figure 2: 
plotted against 
for all the entries of File 1
in the temperature range 
.
In Fig. 3 (click here), we present an histogram of the 4716 [Fe/H] determinations given in File 1. The most frequent metallicity is around -0.15 dex. This value is of the same order as the mean metallicity of the solar neighborhood found from ubvy photometry by Eggen (1978). It is important to remember that the sample listed in the Catalogue is strongly biased and no statistical conclusion can be given, but one can note that the solar metallicity is on the metal rich part of the histogram, which is in the same sense than Eggen's conclusion: the Sun seems to have a higher metallicity than the majority of stars in the solar neighbourhood.
Non-solar metallicities, as shown in Fig. 3 (click here), may reflect very different astrophysical phenomena. Many stellar metallicities reflect the interstellar medium from which the star formed, and whose chemical composition is appreciably different from the Sun. This interpretation applies to most solar and lower mass stars, and it is just these, from late F or early G type, that span the full range of stellar ages present in galaxies. Abundances of these stars interest experts in chemical evolution, age and population effects in field stars and cluster stars.
The peculiar [Fe/H] values found in more massive stars reflect not the original abundances, but the alteration by physical processes (selective mass-loss, radiative diffusion, modified or not by magnetic fields, nuclear burning induced by convection and turbulent mixing). These stars are studied by experts of the physical structure of stars.
Figure 3: Distribution of the 4716 [Fe/H] determinations in the File 1
An interesting feature of the histogram in Fig. 3 (click here) is the fraction
of deficient metallicities which is much higher than the true proportion of metal-poor
stars in the solar neighbourhood.
The comparison of this histogram with the same one published
in the 1991-Edition shows that the number of deficient stars observed during the last 5 years
has considerably increased. A
large number of nearby metal deficient stars have been observed also
in several surveys of halo
and thick disk stars,
conducted either with low resolution spectroscopy
(Beers et al. 1992),
or from high resolution spectroscopy
at low S/N ratio, from high proper motion lists
(Carney et al. 1994). It
must be mentioned that
over 3000 determinations of [Fe/H] have been obtained by these authors, but only [Fe/H] obtained
by detailed analyses triggered by those lists are quoted in our Catalogue.
The interest in the first generation of stars will continue to grow
in the next years because these stars provide clues about the chemical
evolution of the Galaxy, and on nucleosynthesis processes in general.
The problem is that halo and thick disk stars are faint and difficult to
observe at high dispersion. Figure 4 (click here) shows the histogram of the visual
magnitude of the field stars of File 1. It can be seen that the proportion of
stars fainter than V=10 which have been observed at high resolution, is quite low. The dotted
histogram represents the stars of File 1 which have [Fe/H] 
. These
stars are the faintest of the sample.
Figure 4: Distribution of the visual magnitude V
for the nearly 2500 stars of File 1. The dotted
histogram represent the metal deficient stars ([Fe/H] 
)
The problem of the limiting magnitude of high resolution spectroscopy is particularly important for clusters and the Magellanic Clouds. Except for a few nearby clusters, the 84 systems listed in File 2 are very distant and the most frequent magnitude of the stars listed in this table is V=12.5. Another consequence of the large distance of clusters is that the cool dwarfs are poorly represented in File 2 by comparison to giants. Also, only 43% of the stars of the sample have a spectral type, by contrast to File 1 where 96% of the field stars have a spectral type. The wide availability of 10-m class telescopes will soon increase greatly the limiting magnitude of high resolution spectroscopy and this will have an important impact on the understanding of the chemical properties of the globular clusters and nearby galaxies.