SIMBAD references

We have studied the dissolution of initially mass-segregated and unsegregated star clusters due to two-body relaxation in external tidal fields, using Aarseth's collisional N-body code NBODY4 on GRAPE6 special-purpose computers. When extrapolating results of initially non-mass-segregated models to globular clusters, we obtain a correlation between the time until destruction and the slope of the mass function, in the sense that globular clusters that are closer to dissolution are more strongly depleted in low-mass stars. This correlation fits observed mass functions of most globular clusters. The mass functions of several globular clusters are, however, more strongly depleted in low-mass stars than is suggested by these models. Such strongly depleted mass functions can be explained if globular clusters started initially mass segregated. Primordial mass segregation also explains the correlation between the slope of the stellar mass function and the cluster concentration that was recently discovered by De Marchi and coworkers. In this case, it is possible that all globular clusters started with a mass function similar to that seen in young open clusters in the present-day universe, at least for stars below m=0.8 M_☉. This argues for a near universality of the mass function for different star formation environments and metallicities in the range -2<[Fe/H]<0. We finally describe a novel algorithm that can initialize stationary mass-segregated clusters with an arbitrary density profile and amount of mass segregation.