We built two samples of stars: one of them made up of CP2 stars and the other
of normal stars. The application of the Multiple Discriminant Analysis
gave a combination of the Strömgren-Crawford photometric indices which
was able to optimally distinguish both samples. The analysis was conducted separately
for the early, intermediate and late photometric regions defined
by Strömgren. We used both observed and reddening free indices.
The best result was obtained, for the early region, using the observed indices
as variables to characterize the unreddened populations (Eqs. (1) and (2)).
For these stars, a peculiarity level of detects 50% of the
CP2 stars with a contamination of around 2%.
Furthermore,
presented a high correlation with peculiarity
indices in other photometric systems, such as
and
.
The ability of to detect peculiar stars is smaller than
that of the
or
indices. This was not
unexpected,
because they specifically measure the depression at
.
However, the
efficiency of
was higher than that of
or
used elsewhere.
For the intermediate and late regions, even though a small separation
between peculiar and normal stars was present, the correlation of
index with
was very poor.
New observations enlarging the sample of cool peculiar stars with available
photometry are needed in order to carry out
a more accurate study of these stars.
On the other hand,
cannot discriminate between cool CP2 stars
and Am stars.
A list of 60 new early CP2 candidates among those stars with only
measurements was proposed.
Finally, the difference between the E(b-y) of CP2 stars computed from the
calibrations suitable for normal stars and the actual E(b-y)
was shown to correlate with
the peculiarity index (Eq. (7)).
Acknowledgements
This work was supported by the CICYT under contract ESP95-0180 and by the Ayudas para la utilización de recursos científicos de carácter especıfico by the DGICYT. We should like to thank X. Luri for his kind collaboration in performing the observations in September 1995. We thank Prof. C. Jaschek for his comments on an earlier draft of this paper and Dr. J.-C. Mermilliod for providing us with the latest version of Geneva's photometric catalogue. The authors also thank the referee, Dr. P. North, whose comments have contributed greatly to the paper.
The 1.52 m telescope is operated by the Observatorio Astronómico Nacional in the Calar Alto Observatory of the Centro Astronómico Hispano-Alemán. The 50 cm Danish telescope is operated by the ESO in La Silla Observatory.