It has been known for some time that many bulges have a radial light profile
that is not an elliptical-like *r*^{1/4} law
(Andredakis & Sanders
1994;
de Jong 1995;
Courteau, de Jong, &
Broeils 1996);
instead, they are reasonably
well described by an exponential light profile. Incidentally, the bulge of
our own Milky Way also has an exponential light profile
(Binney, Gerhard, &
Spergel 1997).
Recent high-resolution investigations using data from the *Hubble Space
Telecope (HST)* have strengthened the evidence for exponential light
profiles down to the smallest scales at the end of the spheroids
luminosity sequence
(Carollo et al. 1998,
2001;
Carollo 1999).
Several studies
have used the generalized surface density profile *I*(*r*)
exp[-(*r* / *r*_{o})^{(1/n)}]
introduced by
Sérsic (1968)
to model the bulge light
(Andredakis, Peletier,
& Balcells 1995;
Graham 2001;
MacArthur, Courteau,
& Holtzman 2003).
These studies report shape-parameter values for bulges of late-type spirals
ranging between *n* = 0.1 and 2. Some of the Sérsic analyses
attribute a significant meaning to the derived precise values of *n*
(Graham 2001;
Balcells et al. 2003).
Tests based on simulated data, however, show a large dependence
of the derived parameters on, for example, the input parameters; indeed,
MacArthur et al. (2003)
stress that, on average, the underlying surface
density profile for the late-type bulges is adequately described by an
exponential distribution. The same studies show the existence of a coupling
between bulges and disks that is manifested by an almost-constant scale
lengths ratio *h*_{bulge} / *h*_{disk}
0.1 for late-type
spirals, and a similar scaling relation even for earlier-type
systems. This is interpreted to indicate a similar origin for bulges of
all sizes in hosts of any Hubble type.

For more massive, early-type bulges, ground-based studies using the
Sérsic law to describe their light distribution have found values of
*n* close to, or even in excess of the elliptical-like
de Vaucouleur's (1948)
value of *n* = 4
(Graham 2001).
However, the analysis of high-resolution
*HST* images for a sample of early-type bulges provides values of the
Sérsic shape index *n* not in excess of ~ 3
(Balcells et al. 2003).
The difference in the estimates for *n* is due to the
contribution of photometrically distinct central point sources, which
at ground-based resolution are confused for bulge light (see
Section 1.5).
Balcells et al. interpret the *n* < 3 Sérsic shape indices
in the massive bulges as an indication that even these systems, like
the smaller exponential-type bulges, are not the outcome of violent
relaxation during collisionless accretion of matter. Both in the
ground-based and *HST* analyses, a trend remains between the bulge
Sérsic shape parameter *n* and the bulge luminosity and
half-light radius; the trend is in the direction of brighter, bigger
bulges having larger *n* values
(Graham et al. 2001;
Balcells et al. 2003).

Detailed studies of the integrated stellar populations of bulges of all
Hubble types have also been pushed forward by the availability of
high-resolution multi-color images from the *HST*
(Peletier et al. 1999;
Carollo et al. 2001).
The independent analyses agree on the basic result that (1) massive
early-type bulges have very red colors, unambiguously indicating old ages
( 8 Gyr) for the
average stellar populations of these systems, and
(2) the smaller, later-type (almost) exponential bulges have on average
significantly bluer colors.

Kinematically, the Sb pseudo-bulges studied by Kormendy (1993) represent the extreme case of a general behavior shown by bulges of any Hubble type and mass: these all appear to have kinematic properties that are closer to disklike structures rather than to elliptical galaxies. Indeed, based on the comparison between minor axis radial velocity dispersions of disks and bulges, Falcón-Barroso et al. (2003) report that even the early-type, massive bulges are actually thickened disks.

The following important considerations on bulges that emerge from the above
analyses. (1)
The earlier-type bulges form a continuum with the late-type bulges
in terms of the shapes of the surface brightness profiles. The smallest
bulges
are exponential structures, and the largest appear to be intermediate cases
between the exponential and the elliptical-galaxy ones. (2) Bulges of
spirals are coupled to their host disks in a similar way along the Hubble
sequence (with a possible weak trend toward marginally higher
*h*_{bulge} / *h*_{disk} ratios for early-type
bulges). (3) There is a
spread in average stellar metallicity and ages amongst bulges, but also a
clear trend toward smaller bulges being less enriched and younger stellar
structures than the more massive, earlier-type bulges. (4) Bulges of any
size show some kinematic features that are typical of disks.