Figure 12: Luminosity function of NGC 6530 (solid line) compared to the
theoretical one for a cluster of 3 million years by Fletcher &
Stahler (1994) (dashed line). a) Luminosity function binned in
0.2 
steps. b) Incremental luminosity function
45. This star shows large (
) photometric
variations, as well as probably some H
emission and a strong infrared
excess. In the Hertzsprung-Russell diagram (HRD) it is located close to the
main sequence, near the pre-main sequence evolutionary track of a
star, confirming its membership to NGC 6530. We conclude
that this cluster member is a good new Herbig Be candidate, which is near the
end of its pre-main sequence phase.
53. Although in our CCD spectrum we recognize the spectral
characteristics of an early B-type star, the slope of this spectrum seems
to suggest a much later spectral type. However, the NaID lines
are very deep, indicating that this is due to a very large amount of
interstellar reddening, which is confirmed by the extreme value of E(B-V).
The high probability of membership (78%) of this star in the proper motion
study by van Altena & Jones (1972) makes it unlikely that this
star is a background supergiant. Furthermore, luminosity class I
characteristics should be recognizable in our spectrum in that case. When we
look at the position of this star in the HR-diagram, we also note that it is
located slightly above Palla & Stahler's birthline for an accretion rate of
. This suggests that this star does indeed
belong to the cluster, but might have already evolved off the main sequence.
According to the evolutionary tracks by Maeder & Meynet
(1988), an object at this position in the HRD would be a 
star with an age of about 8 million years. In that case the extreme
amount of reddening could be due to circumstellar material, possibly ejected
by the central star.
74. At the position of this star, there seem to be two infrared sources.
We identified the weaker one of the two in the H band (1.6 
m) and at
longer wavelengths as the optical star, whereas the other could be an embedded
source, not unlike the ones found in NGC 6611 (Hillenbrand et al.
1993). Furthermore, the optical component has 
,
much smaller than the interstellar one to the cluster, suggesting it is a
foreground object. When we look at this star's position in the HRD however,
we conclude that its position agrees well with that of most NGC 6530
members. Also, it has a high (70%) probability of cluster membership in the
proper motion study by van Altena & Jones (1972). Therefore,
we conclude that maybe the optical colours of this star also contain a
contribution of the embedded source, which results in an erroneous value of
E(B-V). More observations are necessary to clarify this issue.
79. This star is located to the right of the birthline in the cluster's HRD. Although the membership probability of this star in the proper motion study by van Altena & Jones (1972) is relatively high (69%), we conclude that most probably this star is a background giant. However, the possibility that it is a post-main sequence cluster member cannot be completely excluded.
114. LkH
108. In our IDS spectrum this star shows strong
emission, whereas all other lines are in absorption.
Furthermore, it shows large (
) photometric variations and a strong
infrared excess. Volk & Cohen (1989) present an IRAS LRS
spectrum for LkH
108 in which some weak silicate absorption
features, as well as some PAH emission can be seen on a very red continuum.
However, when we compare the flux level in this LRS spectrum with our SED,
there seems to be too much flux in it to be due to LkH
108 alone.
The LRS spectrum is probably contaminated with other nearby infrared
sources, making it difficult to draw any definite conclusions from it. If
the PAH and silicate features are due to LkH
118, this would point
to it being a young stellar object, however. In the cluster's HRD
LkH
108 is located between the main sequence and the birthline for
an accretion rate of 
, close to the
evolutionary track of a 
star. Since this star fulfills all of
the criteria for membership of the Herbig Ae/Be stellar group as originally
outlined by Herbig
(1960) as well as all tested secondary criteria from
Thé et al.
(1994), we conclude that LkH
108 must be a new Herbig Be
star, which is probably physically located in NGC 6530 and younger than
10
years.
118. The status of this object remains unclear. The star's position in the HRD, as well as the inconclusive membership probability (47%), seem to suggest that this star is a foreground object. However, its E(B-V) of 120 indicates that it is a background object. It could be that this star is an embedded foreground object or a multiple system.
W9. HD 164816. This star is located slightly to the right of the
main-sequence, in a region well above the birthline in the cluster's HRD.
Since this O-type star is seen to ionize part of its surrounding hydrogen
gas, this star is without any doubt a cluster member. Therefore, this star
must already have evolved off the main-sequence. Further support for this
view comes from the observation by Böhm-Vitense et al. (1984)
that the 1549 Å CIV line of W9 has a moderate P Cygni profile,
indicative of mass loss. By comparing its
position in the HRD with the post-main sequence evolutionary tracks by
Maeder & Meynet (1988), we conclude that HD 164816 is a 
star with an age of about 
years.
151. This star shows some moderate (
05) photometric
variations, as well as probably some H
emission and a moderate, but
clearly visible, infrared excess. In the cluster's HR-diagram it is located
close to the main sequence, near the evolutionary track of a 
star. Therefore we conclude that this star is a cluster member as well as a
good new Herbig Be candidate, near the end of its pre-main sequence phase
(age 
years).
172. This star is located to the right of the birthline in the cluster's HRD, provided that the star is at the distance of NGC 6530. This, combined with the low (19%) membership probability in the proper motion study by van Altena & Jones (1972), indicates that this object is in fact not a cluster member. However, its E(B-V) of 035 in Table 5 suggests that it is a cluster member. Since the equivalent width of its NaID lines indicate much larger amounts of gaseous material than those seen in most cluster members, we conclude that most probably this star is a background giant.
180. LkH
112. This star is a well-known Herbig Be star, with
moderate (
05) photometric variations (Boesono et al.
1987; Shevchenko et al. 1993). Light curves and
colour-magnitude diagrams of this star using data from literature are shown
in Figs. 13 (click here)a and 14 (click here)a. In the light curve we can see
pulse-like shapes very similar to the ones observed in the well-known
Herbig
Ae star HR 5999 (Pérez et al. 1992), which we understand now
as being indicative of random instabilities in the line of sight due to a
combination of phenomena such as extinction changes, bursts and clumpy
accretion. From Fig. 14 (click here)a we also note that there seems to be a weak
correlation between V and the U-B, B-V and V-R colours, suggesting
that the observed variations are at least partly due to, presumably
circumstellar, variable extinction. Our IDS spectrum shows very strong
emission in H
and H
, whereas all other lines are in
absorption. Furthermore, it displays a moderate infrared excess. In the
cluster's HRD the star is located between the birthline for an accretion
rate of 
and the main sequence, close to the evolutionary track
of a 
star. We conclude that this cluster member is indeed a
Herbig Be star, with an age of about 
years.
Figure 13: a) Light curves of LkH
112 in the Johnson
photometric system. b) The same for LkH
115
Figure 14: a) Colour-magnitude diagrams of V versus the Johnson
U-B, B-V and V-R colour indices for LkH
112.
b) The same for LkH
115
199. The JHK magnitudes of this star seem to match its visual magnitudes poorly. The star does not appear to be variable, however. Therefore perhaps another star was measured in the near-IR observations.
201. HD 315031. When we look at the position of this star in the
Hertzsprung-Russell diagram, we note that it is located above Palla &
Stahler's birthline for an accretion rate of 
, which would suggest that this star is not a cluster
member. The membership probability by van Altena & Jones for this star
(48%) is not really conclusive, but the fact that this star has a similar
polarization as other stars in NGC 6530 (McCall et al. 1990)
suggest that it is in fact a cluster member. The deviating position of HD
310531 in the HRD is probably due to the fact that it is a binary with a
secondary of the same spectral type as the primary (Boggs &
Böhm-Vitense 1989).
223. HD 315021. Böhm-Vitense et al. (1984) discovered
both in- and outflowing material in the 1549 Å CIV and
1640 Å HeII line profiles of this star. In our HR-diagram it is
located close to the birthline for an accretion rate of 
, at the evolutionary track of a 
star.
Therefore we conclude that most probably this star is a post-main sequence
star, in which case its age would be about 
years according to
the evolutionary tracks of Maeder & Meynet (1988).
245. LkH
115. This star is a well-known Herbig Be star, with
moderate (
04) photometric variations (Boesono et al.
1987; Shevchenko et al. 1993). Light curves and
colour-magnitude diagrams of this star using data from literature are shown
in Figs. 13 (click here)b and 14 (click here)b, respectively. From these figures we
note the star definitely shows variations in both brightness and all
colours, but there is no obvious correlation between magnitude and colour.
Therefore, the observed variations cannot be due to variable circumstellar
extinction. Since the colours also vary, the variations are probably also
not due to an eclipsing companion. Perhaps the best possible explanation for
the observed variability are variations in the accretion rate, not unlike
those seen in many other Herbig and T Tauri stars (e.g. Herbst et al.
1994). Our IDS spectrum of LkH
115 shows very strong
emission in H
, as well as some weak emission in H
, whereas
all other lines are in absorption. Furthermore, the SED displays a moderate
infrared excess. In the cluster's HRD it is located between the main
sequence and the birthline for an accretion rate of 
, close to the evolutionary track of a 
star. Therefore we conclude that 
115 is indeed a Herbig Be
cluster member, with an age of about 
years.
261. This star is located to the right of the birthline in the cluster's HRD. Although the membership probability of this star in the proper motion study by van Altena & Jones (1972) is high (75%), we conclude that most probably this star is a background giant.
270. HD 165052. According to its position in Fig. 11 (click here), this
should be a very massive (
) star, which has evolved slightly
off the main-sequence. This view is again supported by the detection of
outflowing material in the CIV line at 1549 Å (Böhm-Vitense et
al. 1984). According to the post-main sequence evolutionary tracks by
Maeder & Meynet (1988), its age should be about 
years in this case. However, since this is a binary system (Morrison &
Conti 1978), some deviations of the star's position in the HRD can
be expected, making this age rather uncertain.
304. This star is located to the far right in the cluster's HRD,
making it impossible for this star to be a pre-main sequence cluster member.
At present it is not clear whether this star is a late-type background giant
or a post-main sequence cluster member with circumstellar material, as seems
to be the case for star No. 338. If the latter is the case, it would
probably make it a post-AGB star with a mass of about 
and an
age of about 
years.
315. The JHK magnitudes of this star in the SED seem to match its visual magnitudes very poorly. The star does not appear to be variable, however. Therefore perhaps the wrong star was measured in the near-IR.
330. This star is located to the right of the birthline in the cluster's HRD. The membership probability by van Altena & Jones for this star (48%) is not really conclusive. However, in our IDS spectrum of this star we can see B-type characteristics as well as traces of a cooler object. Therefore we conclude that this object is most probably a close binary, possibly located in NGC 6530.
338. This star is located to the far right in the cluster's HRD,
making it impossible for this star to be a pre-main sequence cluster member.
On first view this star seems to be a late-type background giant. However, in
view of the high membership probability (69%) of this star in the proper
motion study by van Altena & Jones (1972), the fact that this
star has a similar polarization as other stars in NGC 6530 (McCall et al.
1990) and the fact that this star was marked as variable by
Sagar &
Joshi (1978), the possibility of it being a post-main sequence
cluster member with considerable amounts of circumstellar material should be
seriously considered. Comparing this star's position in the HRD with the
post-main sequence evolutionary tracks by Maeder & Meynet
(1988), we find that in such a case this star would be a 
star that has just evolved off the Asymptotic Giant Branch (AGB).
Its age would be about about 
years. In view of the age of most stars in NGC 6530 this is fairly
old, but still possible. If this scenario is correct, this would make it one
of the youngest and most massive Post-AGB stars known. New observations,
especially in the infrared, remain necessary to get more certainty on the
true nature of this interesting object.