Recent observations have shown that globular clusters contain a significant binary population. This is a dramatic change from the conventional view of even a decade ago, which held that globular clusters formed without any binaries at all, since the observed X-ray binaries were understood to be formed through dynamical capture. Over the last few years, a number of different observational techniques have resulted in the detection of a substantial number of binaries most of which are believed to be primordial. When the many selection effects are taken into account, these detections translate into a binary abundance in globular clusters that may be somewhat smaller than those in the Galactic disk and halo, but not by a large factor. Within the current uncertainties, it is even possible that the primordial binary abundance in globular clusters is comparable to that in the Galactic disk. We discuss different successful optical search techniques, based on radial-velocity variables, photometric variables, and the positions of stars in the color-magnitude diagram. In addition, we review searches in other wavelengths, which have turned up low-mass X-ray binaries and more recently a variety of radio 981

We have used the Space Telescope Imaging Spectrograph (STIS) on the Hubble Space Telescope (HST) to obtain high spatial resolution spectroscopy of the central region of the dense globular cluster M15. The observational strategy and data reduction were described in Paper I (van der Marel et al. 2002). Here we analyze the extracted spectra with a cross-correlation technique to determine the line-of-sight velocities of individual stars. Our final STIS velocity sample contains 64 stars, two-thirds of which have their velocity measured for the first time. The new data set triples the number of stars with measured velocities in the central projected R ≤ 1 ′′ of M15 and doubles the number in the central R ≤ 2 ′′ . We combine our data with existing ground-based data to obtain non-parametric estimates of the radial profiles of the projected rotation velocity, velocity dispersion, and RMS velocity σ RMS . The results differ from earlier work in the central few arcsec in that we find that σ RMS rises to ∼ 14 km s −1 , somewhat higher than the values of 10-12 km s −1 inferred previously from ground-based data.To interpret the results we construct dynamical models based on the Jeans equation for a spherical system. If the velocity distribution is isotropic, then M15 must have a central concentration of non-luminous material. If this is due to a single black hole, then a fit to the full velocity information as function of radius implies that its mass is M BH = (3.9 ± 2.2) × 10 3 M ⊙ . The existence of intermediate-mass black holes in globular clusters is consistent with several scenarios for globular cluster evolution proposed in the literature. The inferred mass for M15 is consistent with the extrapolation of the relation between M BH and σ RMS that has been established for galaxies. Therefore, these results may have important implications for our understanding of the evolution of globular clusters, the growth of black holes, the connection between globular cluster and galaxy formation, and the nature of the recently discovered 'ultra-luminous' X-ray sources in nearby galaxies. Instead of a single intermediate-mass black hole, M15 could have a central concentration of dark remnants (e.g., neutron stars) due to mass segregation. However, we argue that the best-fitting Fokker-Planck models that have previously been constructed for M15 do not predict a central mass concentration that is sufficient to explain the observed kinematics. To fit the M15 data without any central dark mass concentration one must assume that the velocity distribution is significantly radially anisotropic near the center, which contradicts predictions from both Fokker-Planck and N -body calculations.

No abstract

Using WFPC2 on the Hubble Space T elescope, we have isolated a sample of 258 white dwarfs (WDs) in the Galactic globular cluster M4. Fields at three radial distances from the cluster center were observed, and sizable WD populations were found in all three. The location of these WDs in the colormagnitude diagram, their mean mass of 0.51(^0.03) and their luminosity function conÐrm basic M _ , tenets of stellar evolution theory and support the results from current WD cooling theory. The WDs are used to extend the cluster main-sequence mass function upward to stars that have already completed their nuclear evolution. The WD/red dwarf binary frequency in M4 is investigated and is found to be at most a few percent of all the main-sequence stars. The most ancient WDs found are D9 Gyr old, a level that is set solely by the photometric limits of our data. Even though this is less than the age of M4, we discuss how these cooling WDs can eventually be used to check the turno † ages of globular clusters and hence constrain the age of the universe.

We have used HST imaging of the central regions of the globular cluster 47 Tucanae (= NGC 104), taken with the WFPC2 and ACS cameras between 1995 and 2002, to derive proper motions and U-and V -band magnitudes for 14,366 stars within 100 ′′ (about 5 core radii) of the cluster center. This represents the largest set of member velocities collected for any globular cluster. The stars involved range in brightness from just fainter than the horizontal branch of the cluster, to more than 2.5 mag below the main-sequence turn-off. In the course of obtaining these kinematical data, we also use a recent set of ACS images to define a list of astrometrically calibrated positions (and F475W magnitudes) for nearly 130,000 stars in a larger, ≃ 3 ′ × 3 ′ central area. We describe our data-reduction procedures in some detail and provide the full position, photometry, and velocity data in the form of downloadable electronic tables. We have used the star counts to obtain a new estimate for the position of the cluster center and to define the density profile of main-sequence turn-off and giant-branch stars into essentially zero radius, thus constraining the global spatial structure of the cluster better than before. A single-mass, isotropic King-model fit to it is then used as a rough point of reference against which to compare the gross characteristics of our proper-motion data. We search in particular for any evidence of very fastmoving stars, in significantly greater numbers than expected for the extreme tails of the velocity distribution in a sample of our size. We find that likely fewer than 0.1%, and no more than about 0.3%, of stars with measured proper motions have total speeds above the nominal central escape velocity of the cluster. At lower speeds, the proper-motion velocity distribution very closely matches that of a regular King model (which is itself nearly Gaussian given the high stellar density) at all observed radii. Considerations of only the velocity dispersion then lead to a number of results. (1) Blue stragglers in the core of 47 Tuc have a velocity dispersion σ µ smaller than that of the cluster giants by a factor of √ 2, consistent with the former being on average twice as massive as normal, main-sequence turn-off stars. (2) The velocity distribution in the inner five core radii of the cluster is essentially isotropic, and the detailed dependence of σ µ on R for the brighter stars suggests that heavy remnants contribute only a fraction of a percent to the total cluster mass. Both of these results are in keeping with earlier, more realistic multimass and anisotropic models of 47 Tuc. (3) Using a sample of 419 line-of-sight velocities measured for bright giants within R ≤ 105 ′′ , we obtain a kinematic distance to the cluster: D = 4.0 ± 0.35 kpc, formally some 10%-20% lower than recent estimates based on standard CMD fitting, and more consistent with the value implied by fitting to the white-dwarf cooling sequence. And (4) by fitting simple models of isotropic, single-mass stellar clusters with central point mas...