Annu. Rev. Astron. Astrophys. 1979. 17:
135-87 Copyright © 1979 by Annual Reviews. All rights reserved |
2.2 Mass of the Milky Way
Historically the mass of the Milky Way has been determined from the rotation curve. This involves two distinct but interrelated observational problems: finding the shape of the rotation curve interior and exterior to the solar radius R0, usually from 21-cm HI studies (Kerr & Westerhout 1965, Burton 1974), and setting the scale of the rotation curve by estimating the circular velocity at the sun, V0. The latter measurement is difficult due to lack of a suitable inertial reference frame. One approach is to use extreme Pop II objects as a reference; this technique yields a lower limit, since the amount of rotation of the Pop II spheroid is unknown.
Using this method, Oort (1965) showed that V0 190±30 km sec-1, in good agreement with Hartwick & Sargent's (1978) value of 220 km sec-1 based on velocities of globular clusters and dwarf spheroidal galaxies. On the other hand, a best-fit solution of 300 km sec-1 is obtained from the dynamics of the Local Group (see Section 6.4). Between these two extremes is the officially adopted I.A.U. value of 250 km sec-1 for a solar radius of 10 kpc.
A fresh attack on the determination of V0 has been made recently by Gunn et al. (1979). They combine observations of H I interior to the sun with the requirement that the rotation curve join smoothly to their suggested flat rotation curve exterior to the sun. Their reasoning is too complex to detail here, but their preferred value of V0 is 220 km sec-1. In our opinion the uncertainties are large, but the method does minimally require V0 260, in contrast to V0 = 300 found from Local Group dynamics.
Once V0 and R0 are known, the mass in the Milky Way interior to the sun can be obtained by a variety of modelling techniques (see, e.g., Schmidt 1965). However, these results have lately diminished in significance in the face of mounting evidence for large amounts of nonluminous matter far beyond the sun's orbit. From stellar motions, Fitzgerald et al. (1978) found that the rotation curve stays flat outside the sun for several kiloparsecs. Hartwick & Sargent (1978) analyzed the distribution of radial velocities of globular clusters and nearby dwarf spheroidal galaxies, which they took to be bound to the Milky Way. The outermost sample tests the potential at an effective radius of about 60 kpc and gives an interior mass of 8 x 1011 M for an isotropic distribution of velocity components.
Finally, using the globular cluster system as a probe of the galactic potential and tidal fields, Webbink (in preparation) has mapped the mass distribution out to a radius of ~ 100 kpc. Independent estimates of galactic mass were obtained from the tidal radii and radial velocities of 126 globular clusters and 7 dwarf spheroidal companions of the galaxy. The deduced mass distribution of the Milky Way based on radial bin averages is shown in Figure 1. The present tidally-limited radius of the galaxy due to M31 is ~ 200 kpc. Within this assumed radius, Webbink derives a total mass of 1.4±0.3 x 1012 M from tidal effects on globular clusters and 1.4±0.8 x 1012 M from globular cluster radial velocities. The consistency between Webbink's two completely independent determinations of the mass distribution is strong empirical evidence for the existence of a dark envelope around the Milky Way.
To obtain M / LB for the galaxy, we use Sandage & Tammann's (1976) calibration of LB versus rotation velocity to estimate the luminosity of the galaxy. The result is 2.0 x 1010 L. With Webbink's mass estimate, we obtain M / LB 70±20 on our mass-to-light system. This value is an upper limit, since the mass may not extend as far as the assumed tidal cutoff of 200 kpc.