The kinematics of the Galactic globular clusters and of local galaxies are of great interest with regard to a number of issues related to the formation history of the Galaxy, its dark matter content, and the dynamical evolution of stellar systems (for a more detailed discussion see e.g. Majewski & Cudworth 1993). We have shown that the use of full-scanned Schmidt plates allows the direct and accurate measurement of the cluster proper motions against the absolute reference frame represented by large numbers of galaxies. From APM scans of Tautenburg Schmidt plates with 25 years baseline we have obtained mean absolute proper motions of the Galactic globular clusters M 3 and M 92 with an accuracy of 0.5 mas/year and used the results for the determination of Galactic orbits (Scholz, Odenkirchen & Irwin 1993, 1994). Extending our work to the dSphs in Draco and Ursa Minor with very large distances ( 70 kpc) we combined APM measurements of POSS 1 glass copies, second epoch Palomar and third epoch Tautenburg Schmidt plates (Scholz & Irwin 1994). In a continuation of our programme to investigate the proper motion of Galactic halo globular clusters, we have used the APM facility to directly determine the tangential motion of the globular clusters M 5, M 12 and M 15 from Palomar, Tautenburg and UKST plates (Scholz et al. 1996).
For the very interesting case of M 5 we obtained a smaller absolute proper motion still leading to a high space velocity but not to the extremely large value known till now (cf. Cudworth & Hanson 1993). Whereas the presently observed Galactocentric distances of M 3, M 92, M 12 and M 15 are typical for their calculated orbits, we conclude from our results that M 5 is apparently an outer halo cluster only briefly visiting the nearer regions.
With a heliocentric distance of 21.8 kpc (Harris 1996) the globular cluster Pal 5 is one of the more distant clusters within our programme. There is a special interest in Pal 5 since it has such a low space motion according to the results of Schweitzer, Cudworth & Majewski (1993) who obtained an absolute proper motion of the cluster with respect to a reference frame of only 20 background galaxies. As the corresponding space velocity vector is exactly perpendicular to its radius vector to the Galactic center, Schweitzer, Cudworth & Majewski conclude a rather eccentric orbit with the present locus of Pal 5 likely at apogalacticon.
Both the sparse appearance of the cluster Pal 5 and its stellar mass function (Smith et al. 1986) are likely explained by advanced destruction due to Galactic tidal forces. Owing to its orbital parameters, it seems possible that Pal 5 is on one of its last orbits before total dissolution. Therefore, Pal 5 is considered as an ideal example for detailed dynamical studies of such a destruction process (Smith & Miller 1995). Accurate space velocity data are necessary ingredients for such investigations.
Moreover, the understanding of the dynamical history, and therewith the accurate knowledge of the orbital parameters, are also relevant to the chemical evolution of globular clusters: Enrichment of cluster stars by accretion of the ejecta from the intermediate-mass stars (Bell et al. 1981) depends on the strength of cluster winds and, therefore, on the evolution of the depth of the gravitational potential. The extremely low mass of the cluster Pal 5 makes it an important test for the hypothesis of such self-enrichment. If Pal 5 was not much more massive in the past than at present accretion is not a likely origin for CN-enhanced stars in this cluster (Smith 1985, 1996).
Following Zinn's (1993) classification of globular clusters Pal 5 belongs, with [Fe/H] -1.4 and a an anomalously red horizontal branch (Smith 1985), to the "younger halo" class. According to the halo formation scenario by Searle & Zinn (1978) these objects have been suggested to be formed in separate fragments which did not participate in the initial halo collapse (e.g. Lin & Richer 1992; Majewski 1993, and references therein) but were captured more recently by the Galaxy. The recent discovery of the disrupting Sagittarius dwarf galaxy (Ibata, Gilmore & Irwin 1994) has strongly reinforced this point of view. Globular cluster streams perhaps indicating such satellite accretion have been suggested by Lynden-Bell & Lynden-Bell (1995).
At the Donald Lynden-Bell 60th birthday conference Pal 5 has been suggested as a candidate stripped globular cluster from the Sagittarius dwarf galaxy (Lin 1996). But although the position of Pal 5 is near the orbital plane suggested for the Sagittarius dwarf by Ibata et al. (1997) the stripping scenario seems highly unlikely for Pal 5 with respect to its low space motion.
So, we decided that it would be useful to verify the Schweitzer, Cudworth & Majewski (1993) result by other observations and reduction methods. In our opinion measuring large numbers ( 1000) of reference galaxies, cluster stars and field stars in one measuring process yields a better estimate and correction of systematic errors and should, therefore provide the most reliable results.
© European Southern Observatory (ESO) 1998
Online publication: April 20, 1998