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

Next Contents Previous


3.1. Bulges and Ellipticals

In the most simplified picture of galaxies, a galaxy consists of a bulge that follows an R1/4 profile and an exponential disk, whereas elliptical galaxies are simply the extension of bulges in the limit of bulge-to-disk ratio tending to infinity.

The picture has been complicated by the discovery that most intermediate luminosity ellipticals (as classified from photographic plates) have significant disks (e.g. Bender et al 1988, Rix & White 1990). These disks can be very difficult to detect, especially when seen face-on. Kormendy & Bender (1996) have recently proposed that ellipticals with "disky" isophotes, which tend to be of lower luminosity than those with "boxy" isophotes, are the natural extension of the Hubble sequence of disk galaxies.

Futhermore, many ellipticals show nuclear disks, either from their kinematics or high-resolution imaging (e.g. review of de Zeeuw & Franx 1991). These disks are very concentrated towards the center and are therefore different from the extended disks in normal spiral galaxies. Sometimes these disks have an angular momentum vector opposite to that of the bulge (e.g. IC 1459, Franx & Illingworth 1988), implying that the gas that formed the disk did not have its genesis in the stars of the bulge but was accreted from elsewhere. Notice, however, that some spiral galaxies also show evidence for these "nuclear disks," including the Milky Way (Genzel et al 1996) and the Sombrero galaxy (Emsellem et al 1996).

HST observations confirm the similarity in some aspects of low-luminosity ellipticals and bulges. Most of these systems have power-law profiles in their inner parts, with steep profile indexes (e.g. Faber et al 1997). In contrast, most high-luminosity ellipticals show "breaks" in their surface brightness distribution within 1kpc or less from the center, i.e. relatively sudden changes where the intensity profiles flatten. It is not clear yet what formation processes have caused these variations, although it has been suggested that the dynamical effects of massive black holes may be responsible (Faber et al 1997). HST imaging of large samples of spirals is needed to determine better the structure of their bulges. Preliminary results (pre-refurbishment) indicate that a significant fraction of bulges in early-type spirals have power-law profiles in their inner parts, while late-type spirals have shallower inner profiles and often an unresolved nucleus (e.g. Phillips et al 1996).

These results suggest caution in the analysis of other data, as bulges are not necessarily the only important component near the center and as the formation histories of the centers of different galaxies may have been quite different from each other. Indeed, the central 1 kpc or so of most, if not all, galaxies clearly contain something unusual - even without the benefit of detailed HST images (e.g. note NGC 4314 in the Hubble Atlas, which is a barred galaxy that has spiral arms in the center of the bar).

Beyond the very central regions, a systematic variation of surface brightness profile with bulge luminosity has been established, in that bulges in late-type spiral galaxies are better fit by exponential profiles than by the de Vaucouleurs profile, which is appropriate for early-type spirals (e.g. Andredakis et al 1995, de Jong 1995, Courteau et al 1996). HST imaging of late-type spirals is needed to better determine the structure of their bulges. Preliminary results indicate that a significant fraction of bulges in late-type spirals have power-law profiles in their inner parts (e.g. Phillips et al 1996).

Much recent research into the properties of elliptical galaxies has demonstrated the existence of a "fundamental plane" that characterizes their dynamical state (e.g. review of Kormendy & Djorgovski 1989, Bender et al 1993). The bulges of disk galaxies in the range S0-Sc (T0-T5) have also recently been demonstrated to occupy the same general locus in this plane (Jablonka et al 1996). Furthermore, these bulges have a similar Mg2 line strength-velocity dispersion relationship to that of ellipticals, but the bulges are offset slightly to lower line strengths. This offset may be due to bulges having lower metallicity or lower age. Contamination by disk light can produce a similar effect. Jablonka et al argue in favor of a close connection between ellipticals and bulges. Balcells & Peletier (1994) find that bulges follow a color-magnitude relationship similar to that of ellipticals but that bulges have a larger scatter. Furthermore, they find that bulges and ellipticals of the same luminosity do not have the same colors and that bulges are bluer. The offset is similar to that seen by Jablonka et al in the strength of the magnesium index, but Balcells & Peletier interpret it as indicating a real, though complex, difference between bulges and ellipticals. In addition to the data noted above on the central parts of bulges, Balcells & Peletier (1994) find that the amplitude of radial color gradients also varies systematically with bulge luminosity. They interpret their results as consistent with bright bulges (MR < -20) being similar to ellipticals (despite the color zero-point offset), whereas faint bulges are perhaps associated with disks.

The potential well of the outer regions of disk galaxies is clearly dominated by dark matter, whereas the properties of dark matter haloes around elliptical galaxies are less well known (e.g. de Zeeuw 1995). How do properties of bulges scale with dark haloes? Figure 4 shows the ratio of bulge dispersion divided by the circular velocity of the halo (derived from rotation of tracers in the disk) against bulge-to-disk ratios. The square on the right represents elliptical galaxies, derived from models by Franx (1993), which assume a flat rotation curve. The triangle on the left corresponds to the inner regions of pure disks, as derived for a sample of Sa-Sc galaxies by Bottema (1993) (it should be noted that the inner regions of disks are not cold, but warm). Bulges may be seen to lie on a rather smooth sequence between these two extreme points. This suggests that the bulges in galaxies with low bulge-to-disk ratios may have been formed at the same time as the disk, whereas bulges in galaxies with large bulge-to-disk ratios are so much hotter than the disk that it is more likely that they formed separately. More and better data would be valuable to improve the diagram.

Figure 4

Figure 4. (a) The central velocity dispersion of stellar tracers, sigma, against dark halo circular velocity, Vc. Open symbols are bulges; closed symbols are ellipticals. Circular velocities for the ellipticals are derived from models, as described by Franx (1993). (b) The ratio of velocity dispersion in the bulge to dark halo circular velocity, sigma / Vc, taken from Franx (1993), plotted as a function of bulge-to-total luminosity (B/T) ratio, for the entire range of Hubble Type. The triangle at left is valid for the inner regions of pure disks, the square at right for ellipticals. Note that systems with low B/T have kinematics almost equal to those of inner disks.

Next Contents Previous