Annu. Rev. Astron. Astrophys. 1997. 35: 637-675 Copyright © 1997 by Annual Reviews. All rights reserved

2.2. M 33 (NGC 598)

The stellar population of M 33 was reviewed by van den Bergh (1991a), to which the reader is referred for details. We discuss the significant developments since then concerning the existence and nature of the stellar halo and bulge.

M 33 shows photometric evidence for nondisk light, in particular in the central regions. However, the nature of this light remains uncertain, as does whether or not there is a central bulge component that is distinct from the stellar halo.

Attempts to fit optical and IR data for the central regions with an R^1/4 law generally agree with a "bulge-to-disk" ratio of only ~ 2%, or M_V,bulge fainter than ~ -15 (Bothun 1992, Regan & Vogel 1994). Regan & Vogel emphasize that a single R^1/4 provided the best fit to their data. Some evidence is given from ground-based H-band imaging (Minniti et al 1993) and from HST V-I/I CMD data (Mighell & Rich 1995) for asymptotic giant branch (AGB) stars in the central regions in excess of the number predicted by a simple extrapolation from the outer disk; these stars have been ascribed to a rather young centrally concentrated bulge. However, McLean & Liu (1996) contend that their JHK photometry, after removal of crowded regions, shows no resolved bulge population distinct from the smooth continuation of the inner disk.

Is the R^1/4 component metal-poor or metal-rich? The giant branch of the HST CMD data is consistent with a broad range of metallicity, ranging from M 15-like to 47 Tuc-like, some 1.5 dex in metallicity. The low end of this metallicity range is consistent with that estimated earlier from ground-based CMD data for fields in the outer "halo," [Fe/H] ~ -2.2 (Mould & Kristian 1986). These outer fields showed a narrow giant branch, which is consistent with a small dispersion in metallicity, and thus the two datasets together are suggestive of a gradient in the mean metallicity and metallicity dispersion. This may be interpreted as evidence for a centrally concentrated more metal-rich component, albeit following the same density profile as the metal-poor stars. Pritchet (1988) reported a preliminary detection of RR Lyrae stars in M 33, again evidence for old, probably metal-poor, stars.

The semistellar nucleus of M 33 has a luminosity similar to that of the brightest Galactic globular clusters, M_V~-10, and a diameter of ~6 pc. Analysis of its spectrum (Schmidt et al 1990) demonstrated that its blue color reflects the presence of young stars (age less than 1 Gyr) rather than extremely low metallicity; old and intermediate-age stars with metallicity greater than 0.1 of the solar value dominate. The relation of this nucleus to the "bulge," if any, is unclear.

The only kinematic data for nondisk tracers in M 33 are for a subset of its ~ 200 "large clusters of concentrated morphology" (Christian 1993), of which perhaps 10% have the colors of the classical old globular clusters of the Milky Way. Of these clusters, 14 have kinematics that are suggestive of halo objects, in that they define a system with little net rotation and with a "hot" velocity dispersion of order 1 / 2 times the amplitude of the HI rotation curve (Schommer et al 1991, Schommer 1993). Estimates of the metallicities and ages of the "populous" clusters, based on spectrophotometry, suggest a wide range of each, with even the "globular clusters" spanning perhaps ~-2 dex to just under solar metallicity (Christian 1993). Improved estimates from better data are possible and desirable. M 33 has a very large number of globular clusters per unit field halo light, but the meaning of this is unclear.

In summary, M 33 has a low luminosity halo, which is at least in part old and metal-poor. There is no convincing evidence for the existence of a bulge in addition to this halo.