Even rough examination of CMDs presented in Figs. 1-3 leads to
the conclusion that Be 18 is an old open cluster. Its main-sequence
terminates at 
and the group of stars located at
and 
can be identified with the cluster
red giant branch. The data obtained during run #3 were used to
produce a CMD statistically cleaned from the field stars.
We isolated two groups of stars from these data:
those lying within radius R=415 arcsec from the cluster center,
and those lying in an outer ring, between 619 and 809 arcsec. The inner
circle and the observed part of an outer ring cover equal areas on the
sky. The V vs. V-I CMDs for both selected subfields are presented in
Fig. 6 (click here). As we demonstrated above, the radius of Be 18 is about 800 arcsec.
Figure 6:
The CMD's for the field covering central part of Be 18 (left)
and for the ``comparison" field (right)
Therefore the CMD for the outer ring contains some cluster
members. Nevertheless the field stars are dominating group in this
sample (see Fig. 5). Therefore the outer ring served as a ``comparison
field". For each star from this field a nearest match in the CMD of the
inner ring was located. Subsequently, a pair of stars with the lowest
separation was removed from both CMDs. This procedure was
continued as long as it was impossible to locate pairs with a
separation in magnitude 
and a separation in color
. The resulting ``cleaned" CMD for the inner region
of Be 18 is shown in Fig. 7 (click here). It shows a well populated subgiant branch
and a rich clump of helium burning giants. A group of stars forming
bright extension of the cluster main-sequence contains candidates
for blue/yellow stragglers. The CMD shown in Fig. 7 contains about
25 horizontal branch stars. The only known old open cluster
with a comparably well populated horizontal branch is NGC 6791
possessing 22 stars in its red giant clump (Kaluzny & Udalski 1992).
The CMD for the whole field of Be 18 (see Fig. 2)
contains about
35 candidates for the helium burning giants. Examination of Fig. 6b shows
that field stars give negligible contribution to this region of the
cluster CMD.
The presented data are not sufficient for an accurate determination of
such parameters of the cluster as its age, distance and metallicity.
We can obtain only some limits on these parameters.
Ages of old open clusters can be estimated with a
good accuracy using the 
parameter, where 
is
the magnitude difference between the turnoff and the clump of
red giants. 
depends very weakly on the cluster
metallicity and is almost independent of the reddening (e.g.
Cassisi et al. 1994;
Carraro & Chiosi 1994). We note that slightly different working
definitions of 
parameter are used by different
groups (e.g. Phelps et al. 1994 vs. Paper I).
Using definition from Paper I we get for Be 18 
.
Based on the data published by Montgomery et al. (1994) we obtained
for the well studied open cluster M 67.
It may be concluded that Be 18 is slightly younger than M 67.
Current estimates of the age of M 67 group around 5 Gyr (Carraro
& Chiosi 1994; Hobbs & Thorburn 1991;
Nissen et al. 1987).
Carraro and Chiosi derived the following relation between age, metallicity
and 
:
where 
is age in billions of years. Assuming that M 67 and Be 18
have similar metallicities we arrive at the conclusion that the age of the
later is about 4.3 Gyr. Being a distant cluster located in the
galactic anticenter Be 18 has probably lower metallicity than
M 67. In such a case the age of Be 18 would be lower than 4.3 Gyr.
Figure 7:
Field-star corrected CMD for the central part of Be 18
According to photometry published by Montgomery et al. (1994) the
turnoff of M 67 occurs at (B-V)=0.56 which corresponds to
if we adopt E(B-V)=0.05 for this cluster.
The turnoff of Be 18 is observed at 
.
Assuming for a moment that Be 18 and M 67 have the same age
and metallicity we arrive at the conclusion that the reddening of the
former cluster is 
.
In fact Be 18 is probably younger and metal poorer (note its location
in the galactic anticenter) than M 67.
Therefore, the turnoff of Be 18 is probably bluer than the turnoff of
M 67.
In such a case E(B-V)=0.46 can be treated just as a conservative
lower limit on the reddening of Be 18. A similar result can be obtained
based on the analysis of the V vs. V-I CMD of the cluster.
For clusters with the solar metallicity the lower boundary of
the red giant clump
remains around 
for the range of ages
0.5-5 Gyr (e.g. Cassisi et al. 1994). For Be 18 the faint
end of the red horizontal branch is observed at 
.
Hence, for the assumed solar metallicity of the cluster, we
can estimate its apparent distance modulus at
. Furthermore, adopting E(B-V)=0.46,
we obtain D=5.8 kpc for the cluster distance.
As it was noted above, the metallicity of Be 18 is likely to be lower
than solar. This implies a higher reddening of the cluster but at the
same time it leads to a higher value of 
(
becomes brighter for lower values of metallicity).
Consequently, the above given estimate of the cluster distance is
relatively insensitive to the adopted metallicity of the Be 18.
The galactocentric distance of the cluster is about 12.4 kpc and its
height above the galactic plane is about 520 pc.
Concluding this section we note a large linear extension of Be 18.
For the adopted distance of 5.4 kpc and the angular size of about
26 arcmin we derive 44 pc for the linear diameter of the cluster.
An interesting feature of the CMD of Be 18 is thre morphology of its horizontal branch. This branch is tilted and shows a relatively large extension - both verticaly and horizontaly - in comparison with other well studied old open cluster like M 67 or NGC 6791. Its width reaches 0.14 in the B-V color and 0.12 in V-I. Particularly well defined horizontal branch is visible on the V vs. (V-I) CMD shown in Fig. 3. We note that the current models do not reproduce the observed extension of the red horizontal branch of Be 18. Another interesting feature which can be noted in Figs. 1-3 and Fig. 6 is a presence of an apparent sequence of stars located about 0.5 magnitude above the horizontal branch.