The publication of the Hipparcos Catalogue (ESA 1997) allows us to take a new look on the nature of some controversial galactic open clusters. The accurate proper motions and parallaxes of Hipparcos allow to distinguish between true star clusters and chance alignments of stars. In addition, the Hipparcos Catalogue provides new information to separate members from non-members for the confirmed clusters.
However, the catalogue also has some limitations. For example, the motions of the stars were followed only for a few years. The measured proper motions therefore reflect only the mean motion of the stars during this time interval and may differ from the longterm motion if the star is a member of a multiple system (for a full discussion of this effect see Wielen 1997). Hence if there are indications that a star is double, the astrometric solutions of Hipparcos should be treated with caution. Furthermore, even apparently single stars may have undetected companions, which perturb their motion. This possible deviation is particularly important if one tries to disprove the existence of a cluster, since discrepant proper motions in the Hipparcos Catalogue may not necessarily be a sign of real (longterm) differences in the space motion of the studied stars. Another drawback is the increasing incompleteness at fainter magnitudes.
We have therefore combined the Hipparcos Catalogue with proper motions from the ACT Reference Catalogue (Urban et al. 1997). The ACT contains 988,758 positions and proper motions of stars and was obtained by combining new reductions of the Astrographic Catalogue (AC 2000, Urban et al. 1998) with the Tycho Catalogue (ESA 1997). It is well suited for our purposes, since the mean epoch of an AC plate is as early as 1907, so that the proper motions were derived with a time difference of more than 80 years and reflect, unlike the Hipparcos proper motions, the longterm motion of the stars. In addition, the proper motions of the ACT have a mean error of only 3 mas/yr and were reduced to the Hipparcos system, which allows for a direct comparison.
Throughout this work we will neglect perspective effects and assume that all cluster stars have similar proper motions. This is justified by the relative large distances and resulting small angular diameters of the clusters studied. For the same reason we did not try to employ convergent point methods (like e.g. De Bruijne et al. 1997).
© European Southern Observatory (ESO) 1998
Online publication: November 9, 1998