Dark matter has long been inferred around spiral galaxies from their flat HI rotation curves; however, in early-type galaxies (ellipticals and S0s) we do not have this luxury, and progress in finding their total masses has been slow. So not only has a fundamental component of the CDM paradigm remained largely unverified—that there should be similarly extended, massive dark halos around ellipticals—but predictions about the detailed halo properties have not been testable (cf. the halo core issues in late-type galaxies discussed at length in this volume). There is actually an advantage to studying elliptical halos: if one can observe at comparable physical radii, the compact nature of ellipticals implies that this is in a more dark-matter dominated regime than in spirals. Thus messy baryonic physics should have been less important; and it may be easier to disentangle the luminous and dark components' relative mass contributions.
The nominal tracer in elliptical halos is their integrated stellar light. To adequately constrain their dynamics, kinematical measurements must be of sufficient quality to obtain such higher-order moments as the Gauss-Hermite h4. But the drop-off in surface brightness makes this approach nonviable outside an elliptical's central parts. The best survey so far is of 21 bright, round ellipticals (see O. Gerhard, this volume), which found the circular velocity profiles vc(r) to be roughly constant out to 1–2 Reff (5–10 kpc), and ruled out a constant mass-to-light ratio (M / L) for 3 of the galaxies.
Alternative probes of elliptical halos include globular clusters (GCs), which are handy as bright objects spread out to larger radii than the galaxy light (Côté et al. 2001, 2003) — although they are a disjoint population with different properties to the galaxy. X-ray emission from thermalized hot gas filling the halo potential is also useful (D. Buote, this volume); but because the total emission correlates strongly with optical luminosity (LX ∝ LB2−3; O'Sullivan et al. 2003), and only the brightest sources are within easy reach of X-ray telescopes, the findings are biased toward the more massive systems. Similarly, strong gravitational lenses can also be used to probe into galaxy halos (P. Schneider, this volume), but any galaxies with massive, centrally-concentrated halos are systematically most likely to be detected as lenses. Rings and disks of HI and Hα-emitting gas are also sometimes found at large radii (M. Arnaboldi et al., this volume); but these are rare, and they may not trace a typical population of ellipticals. Satellite kinematics (e.g., Prada et al. 2003) and weak gravitational lensing (e.g., H. Hoekstra, this volume) can probe the outermost parts of galaxies, but their constraints are statistical: they provide limited information about mass variations with radius and with other galaxy properties.
Planetary nebulae (PNe) are arguably the ideal probes because of their simple connection to the main stellar population of the galaxy (e.g., Peng et al. 2002). Also, they are not affected by any mass-dust degeneracy (Baes & Dejonghe 2001), and their 5007 Å emission lines readily provide precise velocities.
Various studies with the above methods have found dark matter around individual elliptical galaxies, but as discussed, the selection effects may be severe. Weak lensing and satellite studies also imply massive halos around typical L* ellipticals, but further cross-checks are needed. Indeed, dynamical studies have suggested that some galaxies are much less dark matter-dominated than others (Bertin et al. 1994; Gerhard et al. 2001; N. Napolitano et al., this volume).
Besides searching for dark matter, by probing into elliptical halos we can also test other key properties against predictions of galaxy formation models. These include angular momentum and the distribution of stellar and GC orbits. In the rest of this paper, we present PN kinematical studies of five elliptical galaxies, and some implications for galaxy structure and formation.