4 Discussion on individual clusters

All the clusters studied in this work have been classified as belonging to types I or II of Ruprecht (1966). There exist previous results only about Hogg15 and Melotte 105, which can be compared with the present ones. Precisely, the latter cluster was selected not only because of its compactness but also because it was observed by Santos & Bica (1993) at La Silla (Chile) using the same spectroscopic technique, so that it can be used as control cluster. The five remaining clusters BH132, BH217, Lyngå11, Pismis21 and Ruprecht 144 have not been previously studied and their reddenings and ages have been determined here for the first time. Lyngå11, Pismis21 and Ruprecht144 are mentioned in early catalogation or morphological classification works (Pismis 1959; Lyngå  1965; Ruprecht 1966).

4.1 Ruprecht144

The integrated spectrum of Ruprecht144, corrected for E(B-V) = 0.25, is compared to that of the Y3A template (100 Myr) in Fig. 1. Notice that this template presents TiO bands not observed in Ruprecht144. Notice also that at $\approx$ 100 Myr very populous LMC clusters may present strong TiO bands. This phenomenon appears to be associated to AGB stars (BA87, Bica et al. 1990). Owing to statistical effects in non-massive clusters, the effect is not expected in Galactic open clusters. Indeed, the best match is found for the Y3B template (100 Myr) adopting E(B-V) = 0.32 $\pm$ 0.02 (Fig. 1). The age derived from the Balmer lines is slightly larger (Table 5). CaII triplet EWs yield [$Z/Z_\odot$] = -0.1 $\pm$ 0.2.

4.2 Melotte105

We have simultaneously determined the age and reddening for Melotte105, also known as Collinder 246 (Collinder 1931), using the template Y3B (100 Myr). The best template match yields E(B-V) = 0.31 $\pm$ 0.02. Figure 2 shows the observed cluster spectrum, the spectrum corrected for the derived interstellar reddening E(B-V) = 0.31, and the template Y3B which best matches the cluster spectrum. An alternative match using the template Y2 (50 Myr) is also shown. The derived metallicity from CaII triplet EWs is [$Z/Z_\odot$] = 0.0 $\pm$ 0.2. The age and reddening here derived are in excellent agreement with the values found by Santos & Bica (1993) using the same technique. On the other hand, using UBV photoelectric photometry Sher (1965) estimated the same age ($\approx$ 100 Myr) and a somewhat larger mean reddening E(B-V) = 0.38. The non-uniform reddening determined photoelectrically by Sher (1965) may be reflecting the existence of internal dust associated to the cluster. If this is the case, the spectroscopic reddening should be lower than the photometric one since the less reddened stars of a given spectral type should contribute to the integrated light with larger fluxes in comparison with the most reddened stars of the same spectral type. In other words, as stated by Santos & Bica (1993), it appears highly probable that the lower reddening derived from the integrated spectrum indicates an internal reddening, in the present case, of $\approx$ 0.07 mag.

Based on CCD UBV data Kjeldsen & Frandsen (1991) derived an age of $\approx$ 150 Myr and a quite larger reddening E(B-V) = 0.52. The difference found in the cluster reddening may be explained by the fact that Kjeldsen & Frandsen (B-V) colours are shifted by approximately 0.1 mag with respect to those of Sher (1965). Finally, Balona & Laney (1995) obtained E(b-y) = 0.32 from uvby photometry, equivalent to E(B-V) = 0.43, if the relation E(B-V) = 1.35 E(b-y) derived by Crawford (1978) is used. This reddening differs scarcely by 0.05 mag from that of Sher (1965), while the age estimated by Balona & Laney (250 Myr) is somewhat larger than the present one.

4.3 BH132

BH132 was first recognized as an open cluster by van den Bergh & Hagen (1975). Figure 3 shows the integrated cluster spectrum, corrected for E(B-V) = 0.28, compared to the Y4 template (500 Myr). Note that although a general resemblance between these two spectra is apparent, neither the Balmer jump nor the H$\alpha$ line show a good agreement. Besides, the CaII K line in BH132 appears to be more diluted, which cannot be a metallicity effect since CaII triplet near 8600 Å appears to be similar in both the template and BH132 spectra. This difference would be explained if BH132 were younger than the Y4 template. The best match is found for the Y3B template (100 Myr), using E(B-V) = 0.60 (Fig. 3). Note that the spectral properties of this cluster are very similar to those of Ruprecht 144 and Melotte105. Equivalent widths of TiO and CaII triplet features (Table 4) are compatible with the cluster having a nearly solar metal content.

4.4 Hogg15

Hogg15, also known as BH139 (van den Bergh & Hagen 1975) or ESO95SC15 (Lauberts 1982), was first recognized as an open cluster by Hogg (1965a, 1965b). Photoelectric UBV photometry by Moffat (1974) yields E(B-V) = 1.16 $\pm$ 0.03, a distance of 4.2 kpc, and an age $\leq$ 8 Myr. Therefore, Hogg15 is one of the few clusters known to lie in the inner arm -II, much further behind the Coalsack than the adjacent cluster NGC4609, which is located at about 1.3 kpc from the Sun and is reddened by E(B-V) = 0.36 (Feinstein & Marraco 1971). The above reddening and age of Hogg15 are in very good agreement with the results obtained from integrated spectroscopy. We show in Fig. 4 two template matches. The template match with YB.LMC (6-9 Myr) gives E(B-V) = 0.70 $\pm$ 0.05, whereas YA.LMC (3-6 Myr) yields E(B-V) = 1.05 $\pm$ 0.05. The latter is clearly the best match so that Hogg15 is confirmed as a highly reddened young open cluster. We point out that we also dispose of the near-IR spectrum for Hogg15, but we limit our comparison in Fig. 4 to the range in common with the available template. The derived metallicity is [$Z/Z_\odot$] = -0.2 $\pm$ 0.2. Moffat (1974) and Smith et al. (1994) have discussed the possibility that the WN6 star HDE311884 could be a cluster member. The cluster age is clearly compatible with such possibility (cf. Paczinski 1973).

Figure 8: Reddening histogram in the Galactic sector centered at $l = 270^\circ$

Figure 9: Same as Fig. 8 for $l = 0^\circ$

Figure 10: Age histogram in the Galactic sector centered at $l = 270^\circ$

Figure 11: Same as Fig. 10 for $l = 0^\circ$

4.5 Pismis21

Pismis21 was first recognized as an open cluster by Pismis (1959). Notice the extreme absorption effects in the observed spectrum (Fig. 5). We also show the foreground reddening corrected spectrum for E(B-V) = 1.50 $\pm$ 0.03 and the Y2 template (50 Myr) which best matches the spectrum. Although the age from Balmer lines is slightly greater (Table 5), Pismis21 is found to be a moderately young highly reddened open cluster. The derived metallicity is [$Z/Z_\odot$] = 0.0 $\pm$ 0.2.

4.6 Lyngå11

We show in Fig. 6 two template matches. The template match with Y3B (100 Myr) gives E(B-V) = 0.17 $\pm$ 0.03, whereas Y4 (500 Myr) yields E(B-V) = 0.12 $\pm$ 0.03. An inspection of the spectra reveals that the incipient 4000-Å break favours Y4 as a match, as well as the general spectral distribution. More details on the Balmer jump and 4000-Å break as a function of age are given in Bica et al. (1994). Since the Balmer-line method also gives an age closer to Y4 (Table 5), we adopt the latter match as the best solution. The derived metallicity from CaII triplet and TiO features is [$Z/Z_\odot$] = +0.1 $\pm$ 0.2.

4.7 BH217

This cluster was first recognized by van den Bergh & Hagen (1975), and is also known as ESO333SC2 (Lauberts 1982). The solution for BH217 yields an age of about 50 Myr and E(B-V) = 0.80 $\pm$ 0.03 by template match (Fig. 7). The age derived from the Balmer lines is only slightly lower (Table 5) and no metallicity could be derived for this cluster. The spectral features of BH217 are quite similar to those of the more reddened cluster Pismis21 (Fig. 5).