SECULAR EVOLUTION VIA BAR-DRIVEN GAS INFLOW: RESULTS FROM BIMA SONG
Kartik Sheth, Stuart N. Vogel, Michael W. Regan, Michele D. Thornley, and Peter J. Teuben
ABSTRACT
We present an analysis of the molecular gas distributions in the 29 barred and 15 unbarred spirals in the BIMA
CO (J = 1 0) Survey of Nearby Galaxies (SONG). For galaxies that are bright in CO, we confirm the conclusion by
Sakamoto et al. that barred spirals have higher molecular gas concentrations in the central kiloparsec. The SONG
sample also includes 27 galaxies below the CO brightness limit used by Sakamoto et al. Even in these less CO-bright
galaxies we show that high central gas concentrations are more common in barred galaxies, consistent with radial
inflow driven by the bar. However, there is a significant population of early-type (Sa-Sbc) barred spirals (6 of 19) that
have no molecular gas detected in the nuclear region and have very little out to the bar corotation radius. This suggests
that in barred galaxies with gas-deficient nuclear regions, the bar has already driven most of the gas within the bar
corotation radius to the nuclear region, where it has been consumed by star formation. The median mass of nuclear
molecular gas is over 4 times higher in early-type bars than in late-type (Sc-Sdm) bars. Since previous work has
shown that the gas consumption rate is an order of magnitude higher in early-type bars, this implies that the early
types have significantly higher bar-driven inflows. The lower accretion rates in late-type bars can probably be attributed
 to the known differences in bar structure between early and late types. Despite the evidence for bar-driven
inflows in both early and late Hubble-type spirals, the data indicate that it is highly unlikely for a late-type galaxy to
evolve into an early type via bar-induced gas inflow. Nonetheless, secular evolutionary processes are undoubtedly
present, and pseudobulges are inevitable; evidence for pseudobulges is likely to be clearest in early-type galaxies
because of their high gas inflow rates and higher star formation activity.
Subject headingg: galaxies: evolution -- galaxies: nuclei -- galaxies: spiral -- galaxies: starburst --
galaxies: structure -- ISM: molecules
1. INTRODUCTION
Nearly three-quarters of all nearby disk galaxies have a bar
(Eskridge et al. 2000; Menendez-Delmestre et al. 2004); recent
studies suggest that the bar fraction remains high at z > 0:7
(Sheth et al. 2003) and is perhaps constant to a redshift of 1
(Elmegreen & Elmegreen 2004; Jogee et al. 2004). Models
show that the nonaxisymmetric bar potential induces large-scale
streaming motions in the stars and gas (Sellwood & Wilkinson
1993 and references therein; Athanassoula 1992a, 1992b). This
mixing is responsible for the shallower chemical abundance gradients
 observed in barred spirals compared to unbarred spirals
(Vila-Costas & Edmunds 1992; Zaritsky et al. 1994; Martin &
Roy 1994; Martinet & Friedli 1997). Studies of gas kinematics
in the bar indicate that molecular gas is flowing inward down
the bar dust lanes (Downes et al. 1996; Regan et al. 1999; Sheth
et al. 2000, 2002). Enhancement of the star formation activity
in the nuclei of barred spirals (e.g., Sersic & Pastoriza 1967;
Hawarden et al. 1986; Ho et al. 1997) suggests that either the
conditions at the centers of barred spirals are more favorable for
star formation or these regions have a higher gas content. We
note that the enhancement in star formation activity is primarily
found in early Hubble-type barred spirals (Ho et al. 1997). Ho
et al. divide the galaxies with the Hubble types Sa-Sbc as early
types and galaxies with type Sc-Sdm as late types; we use this
division throughout the remainder of this paper (see also x 2).
The observational results noted above provide indirect observational
 evidence that bars transport gas toward their centers.
Models also suggest that further inflow of gas by a ''bars-
within-bars'' scenario may fuel active galactic nuclei (Shlosman
et al. 1989), although the evidence for such fueling is weak
(Regan & Mulchaey 1999; Knapen et al. 2000; Laine et al. 2002;
Laurikainen et al. 2004). Other studies have suggested that barinduced
 gas inflows may lead to the formation of new (pseudo)
bulges (e.g., Kormendy 1993; see also the recent review by
Kormendy & Kennicutt 2004) and evolve late Hubble-type
spirals into early Hubble types (e.g., Friedli & Benz 1995;
Kenney 1996; Zhang 1999). The most dramatic prediction is that
bars may destroy themselves if they accumulate sufficient mass at
their centers (Pfenniger & Norman 1990; Friedli & Benz 1993;
Norman et al. 1996; Bournaud & Combes 2002; Athanassoula
2003; Shen & Sellwood 2004; see also the review by Kormendy
& Kennicutt 2004 and references therein).
Starburst activity, the formation of a pseudobulge, secular evolution
 of a galaxy along the Hubble sequence, and the destruction
of the bar are all expected to occur when gas accretes in the nuclear
region of a galaxy. Is there any direct evidence for excess gas at the
center of barred spirals? While studies of individual galaxies have
shown that at least some bars have a substantial amount of molecular
 gas near their centers (e.g., Kenney 1996; Turner 1996),
only one study has addressed the issue of whether the central
regions of barred spirals are systematically more gas-rich than
those in unbarred spirals. With a sample of 20 (10 barred and
10 unbarred) galaxies, the Nobeyama Radio Observatory-Owens
Valley Radio Observatory (NRO-OVRO) survey (Sakamoto et al.
1999a) found that the central concentration ( fcon), defined as the
ratio of the nuclear molecular gas surface density (C6nuc)tothe
total (disk-averaged) molecular gas surface density (C6tot), is significantly
 higher in barred spirals than unbarred spirals (Sakamoto
et al. 1999b). While fcon increases when gas is transported to the
center, it decreases when gas is consumed by circumnuclear star
formation. Since barred spirals have elevated circumnuclear star
formation activity (e.g., Sersic & Pastoriza 1967; Hawarden et al.
1986; Ho et al. 1997), a priori one expects lower fcon-values in
barred spirals, unless there is a fresh influx of molecular gas.
Thus, a higher fcon-value in barred spirals than unbarred spirals is
strong evidence for bar-induced gas inflow and accumulation.
Moreover, fcon is robust against biases in galaxy size, mass, and
distance. In addition, given the shallower metallicity gradients
typically found in barred spirals, variations in CO brightness with
metallicity would cause fcon to be smaller in barred than unbarred
spirals, the opposite of what is observed (Ho et al. 1997).
Although the NRO-OVRO result is statistically significant, the
sample size is modest (20 galaxies) and consists of CO-bright
galaxies. The NRO-OVRO sample has only four galaxies of
Hubble type Sc or later. Hence, it is unclear whether the observed
difference in fcon between barred and unbarred spirals is representative
 of all spiral galaxies. Does this result extend to CO-
faint spirals? How does it vary with Hubble type? Is there any
evidence for secular evolution from late to early Hubble type?
With the larger and more diverse sample of spiral galaxies in
SONG, we can address these outstanding questions. SONG contains
 3 times as many bars as the NRO-OVRO study and 50%
more unbarred spirals. There are 27 galaxies in SONG with
lower CO brightness than those in the NRO-OVRO sample because
 SONG galaxies were chosen based on an apparent optical
 magnitude instead of CO brightness. Moreover, SONG has
15 spirals of Hubble type Sc-Sd; thus, we can directly evaluate
 the influence of bars on late Hubble-type spirals. One other
advantage of SONG is that it imaged the CO emission over
the entire bar region in all but one of the barred spirals; hence,
we can compare the gas distribution in the central region to the
available gas in the bar and place the observations in the context
of secular evolution.
2. OBSERVATIONS, DATA REDUCTION, AND ANALYSIS
In SONG, the molecular gas distributions (using the CO J =
1 0 line) in 44 nearby spirals (Sa-Sd) were mapped with the
Berkeley-Illinois-Maryland Association (BIMA) interferometer
(Welch 1996) and the NRAO 12 m single-dish telescope at Kitt
Peak. The SONG sample galaxies are listed in Table 1. The details
 of the survey and properties of the sample are described in
greater detail in Helfer et al. (2003; in particular see x 2.2). In some
cases the distances and position angles used by us are different
than those noted by Helfer et al. (2003); we used the most recent
measurements from the literature. We note that we checked the
results of our study with the Helfer et al. distances and position
angles and found no systematic differences, and the results from
this study remain unchanged. The selection criteria for the SONG
sample were heliocentric velocity VHel < 2000 km sC01,decli-
nation C14>C020, inclination i < 70, and apparent magnitude
BT <11. This lead to a sample that is representative of disk galaxies
 in the local universe.
The typical data cube has a synthesized beam of 600 and a
field of view of 30. In a 10 km sC01 channel the typical noise is
C2458 mJy beamC01. The measured fluxes are accurate to 15%;
uncertainties are due to the absolute amplitude calibration. The
galaxy centers are obtained from the 2MASS Large Galaxy Survey
 (Jarrett et al. 2003).
With the BIMA SONG data, we measure the molecular gas
mass (Mnuc) and surface density (C6nuc) in the central kiloparsec
using standard routines in MIRIAD (Sault et al. 1995); we discuss
 why we chose the central kiloparsec at the end of this
section. We use the FCRAO or the NRAO 12 m SONG data for
the total molecular gas mass (Mtot) and surface density (C6tot).
The mass is calculated using
Mgas =1:36 ; 1:1 ; 10DMpcSCO; (1)
where Mgas is in solar masses and SCO is in Jy km sC01.
The factor of 1.36 is the usual correction for elements other
than hydrogen. This equation uses a CO-to-H2 conversion factor
 of 2:8 ; 1020 cmC02 (KkmsC01)C01 (Bloemen et al. 1986). The
exact value for the conversion factor is controversial and may
vary by factors of 2-4 (see the discussion in Young & Scoville
1991). C6 is then calculated by dividing this gas mass by the area.
Twelve of the 44 galaxies have relatively faint emission in the
central kiloparsec; we list a 2 C27 upper limit for these galaxies.
The results are shown in Tables 2 and 3 for the barred and
unbarred spirals, respectively. Throughout this paper, we use the
de Vaucouleurs et al. (1991, hereafter RC3) classification (SB
and SAB) for barred spirals.
As noted earlier, we classify galaxies of type Sa-Sbc as early
types and galaxies with types Sc-Sd as late types. Since a significant
 number of galaxies in SONG are type Sbc (14 of 44), we
discuss briefly the motivation for the adopted division. It is based
on the significant differences in the star formation properties
between galaxies of type Sbc and earlier as opposed to those
of later types (Ho et al. 1997). Since star formation depends
on gas content, it is reasonable to use the same boundaries as
Ho et al. (1997); otherwise comparison with previous studies
 becomes difficult. Furthermore, the division is also motivated
by prior observational and theoretical work on properties of
barred galaxies (Elmegreen & Elmegreen 1985; Combes &
Elmegreen1993; Huang et al. 1996). These studies have found a
fundamental difference in the properties of early- and late-type
bars. Early-type bars generally have a ''flat'' photometric profile,
 while late-type bars usually have an ''exponential'' profile.
Of the 14 Sbc galaxies, 11 are barred spirals. We examined the
optical/IR data for these 11 galaxies and find that 7 are clearly
flat bars (NGC 2903, NGC 4321, NGC 4303, NGC 3992, NGC
3344, NGC 3953, and NGC 4051). The optical/IR data for the
other four are not good enough to make a definitive classification,
 although the CO kinematics in NGC 5005 suggest that
it too hosts a strong bar of the type usually found in early-type
barred spirals. For these reasons, we place the Sbc galaxies in the
early-type category.
In column (2) of Tables 2 and 3 we list the CO flux from the
central kiloparsec in the combined (BIMA+NRAO 12 m single
dish) maps for the 24 galaxies for which we collected on-the-fly
(OTF) data. In column (3) we list the CO flux from the BIMA-
only data. In most cases the BIMA-only data recover nearly all
the CO flux. There are a few exceptions, in which the flux in the
combined maps is 40%-50% higher than in the BIMA-only
maps: IC342, NGC 0628, NGC 3521, and NGC 4414. In these
cases, the relative angular sizes of the beam and galaxy, the inclination,
 or the CO distribution likely have contributed to spatial
 filtering or beam-smearing effects.
We choose to use the BIMA-only maps to measure Mnuc because
 in most cases the BIMA-only fluxes are consistent with
Mnuc from the combined map (see also Fig. 53 in Helfer et al.
2003 and the discussion in Helfer et al. 2002). There is no evidence
 that barred or unbarred spirals are preferentially affected
by the lack of single-dish data, and we want to use as homogeneous
 a data set as possible. We did not correct the BIMA-only
data for beam smearing. Our typical 600 beam subtends an area of
diameter of <1 kpc for almost all galaxies in our sample, as can
be seen in Table 2 in Helfer et al. (2003), where the area subtended
 is shown in the plane of the sky. In the plane of a galaxy,
the beam may subtend a larger region (at most by cos iC01) along
the galaxy minor axis. For the galaxies in our sample this correction
 is less than a factor of 2 (i < 60), and therefore our beam
typically subtends <1 kpc along the minor axis of the galaxy
plane.
In our analysis we also plot the seven additional galaxies from
the NRO-OVRO survey that are not in common with SONG;
these include five unbarred galaxies, which are rarer in both samples.
 We confirm that the SONG data are consistent with the
NRO-OVRO data by comparing the measured fluxes for the
13 galaxies in common between the two samples; after accounting
for differences in the CO-to-H2 conversion factor and adopted
distances, the fluxes are consistent within the errors.
We derive Mtot from the global CO fluxes obtained by the
FCRAO survey (Table 3 in Young et al. 1995); these fluxes
were derived from fitting an exponential CO profile to several
discrete pointings along the major and minor axis of each galaxy.
SONG, on the other hand, measured the CO fluxes using OTF
maps for 24 galaxies and discrete pointings for the remaining
galaxies. The SONG fluxes are consistent with the FCRAO
survey (see Fig. 51 in Helfer et al. 2003). In the five cases for
which we did not have FCRAO fluxes, we used the SONG
fluxes for Mtot where available. These are also listed in Tables 2
and 3.
We measured Mnuc in the central kiloparsec because, as discussed
 in x 1, most of the evolutionary effects depend on the
accumulation of gas near the nucleus. The choice is also partially
 influenced by Sakamoto et al. (1999a, 1999b), who showed
that more than half of the galaxies in their sample have nuclear
concentrations that are well described by an exponential scale
length of less than 500 pc. Though galaxies in the SONG data
set are less strongly concentrated (many are not even centrally
peaked; see Regan et al. 2001), examining the inner kiloparsec
provides consistency with the previous study. A different measure
 of nuclear gas properties might be a dynamical length scale
such as the radius of the largest stable x2 orbit. However, locating
the largest stable x2 orbit requires precise measurements of the
rotation curve and potential (Regan & Teuben 2003). This is
difficult near the nuclei of galaxies due to the lack of highresolution
 data and the presence of streaming motions induced
by the bar. Another measure might be to choose a bulge radius
as determined from near-infrared images, but deriving a unique
bulge-disk decomposition is difficult. Yet another measure might
be to choose a distance that is a standard fraction of the galaxy
diameter. This choice is problematic because there is no evidence
 that nuclear properties of galaxies are connected with the
disk properties. In fact, star formation properties of the circumnuclear
 regions are known to be distinct from those of the disk
(Kennicutt 1998). Hence, we choose the central kiloparsec as
our fiducial diameter for the nuclear region.
3. RESULTS
3.1. Molecular Gas Masses in the Central Kiloparsec
of Barred and Unbarred Galaxies
The mass of the molecular gas in the central kiloparsec (Mnuc)
is plotted in Figure 1 against the total molecular gas mass in the
galaxy (Mtot). Barred spirals are shown with a circle and unbarred
 spirals with a triangle. SONG galaxies have filled symbols.
 The data from Sakamoto et al. (1999b) for galaxies not in
common with the SONG sample are marked with open symbols.
The diagonal lines indicate the fraction of the total gas mass in
the central kiloparsec.
Before we discuss the nuclear masses, we note that there is no
obvious difference in the distribution of Mtot between barred
and unbarred spirals. The range for Mtot spans C24108-1010 MC12.
For the entire SONG sample, the mean total molecular gas
mass is hMtotibar = (4 C6 1) ; 109 MC12 for the barred spirals and
hMtotiunbar = (6 C6 2) ; 109 MC12 for the unbarred spirals. If we
compare the red symbols (early types) to the blue symbols (late
types), there is also no obvious difference in Mtot-values between
 early and late Hubble-type galaxies. This is consistent with
large single-dish surveys of molecular gas in galaxies (e.g., Sage
1993; Young et al. 1995).
The fraction of the molecular gas that is in the central kiloparsec
 varies by almost 3 orders of magnitude, from 0.001 to
nearly 0.6. Figure 1 shows that galaxies with the highest values
of this fraction and also those with the highest Mnuc are mainly
early types, especially bars. Of the eight galaxies with nuclear
gas mass fractions above 0.1, six are barred, and five of these six
are early Hubble-type galaxies.
Another eight barred spirals have only upper limits for their
nuclear molecular gas mass. Five of these are early-type bars; not
only is no gas detected in the central kiloparsec, but none is
detected within the entire region interior to the bar. The existence
of early-type barred spirals without any nuclear gas is consistent
with bar-driven gas transport, as we discuss in the next section.
In addition, we note that these results contrast with the emphasis
of previous studies on large excesses of molecular gas in the
central kiloparsec (Sakamoto et al. 1999b; Jogee et al. 2005).
For example, Jogee et al. (2005) conclude that the central kiloparsec
 of barred spirals is significantly different from the outer
disks with molecular gas masses in excess of 2 ; 108 MC12 up to
3 ; 109 MC12. However, for the barred spirals in SONG, we find a
range of nuclear masses from 107 to 109 MC12, with a mean of
C242 ; 108 MC12. Indeed, 18 of 29 barred spirals in SONG have
Mnuc < 2 ; 108 MC12. Clearly the high Mnuc-values emphasized by
previous studies apply only to a subset of barred galaxies.
3.2. Bar-driven Gas Transport
In Figure 2 we plot the molecular gas surface density for the
central kiloparsec (C6nuc) versus that averaged over the galaxy
disk inside D25 (C6tot) for all galaxies shown in Figure 1. The
diagonal lines now indicate constant values of the central concentration
 parameter, fcon, which we discussed in x 1.
The figure immediately shows a clear difference in central
concentration ( fcon) between barred and unbarred spirals. All six
galaxies with fcon > 100 are barred spirals. At the other end, for
fcon < 10, 7 of the 10 galaxies are unbarred. The higher fconvalues
 in barred spirals thus provide the strongest and most direct
 evidence for bar-driven gas transport, consistent with the
results of Sakamoto et al. (1999b).
3.2.1. Bar Gas Transport in CO-faint Galaxies
As most of the highest fcon-values are in CO-bright galaxies
(larger symbols), we specifically address the question of whether
bars also enhance fcon in CO-faint galaxies. Figure 3 shows only
the 26 (16 barred and 10 unbarred) SONG galaxies with FCRAO
brightnesses <10 K km sC01. We see that for fcon > 10, 9 of the 13
galaxies with significant detections in both total and nuclear
surface densities are barred. Dividing the 26 CO-faint galaxies
into two bins, we see that 11 of the 13 highest fcon galaxies are
barred, compared with only 5 of the lowest 13 fcon galaxies.
Thus, even in the sample of galaxies that are not CO-bright, we
conclude that there is evidence for bar-induced gas transport.
This evidence is further strengthened upon noting that most
of the low-fcon barred galaxies have no gas at all within the
bar corotation radius. The corotation radius is believed to be at
1.2 times bar end (e.g., Athanassoula 1992b). Galaxies without
gas in the bar region can be understood as those in which bardriven
 accretion has proceeded to the point at which all the gas
presently available to the bar has already been driven to the
central region and converted into stars.
Jogee et al. (2005), in a paper that focused on the role of bars
in starbursts and secular evolution, proposed an evolutionary
sequence for bars and starbursts. In type I nonstarbursts, gas is
still distributed through the bar region and is only beginning to
accumulate in the central region. Type II nonstarbursts have
significant central concentrations, but the gas has yet to accumulate
 to the surface densities required for a starburst. In starbursts,
 the sufficient gas has accumulated to drive a starburst.
We would therefore add a category, type III nonstarbursts,
which are galaxies in the poststarburst phase. All the gas within
the radius swept by the bar has already been driven to the nucleus
and has been consumed by a starburst (or sufficient gas has been
consumed so that the current gas surface density is lower than the
critical density needed for a starburst). It would be interesting to
look for evidence of this in the stellar populations. Such galaxies
will be free of molecular gas within the bar region until some
event enables the feeding of gas through the barrier that seems to
be present near the bar corotation radius. If this happens, then
central gas accumulations and starbursts would be episodic.
3.2.2. Bar Gas Transport as a Function of Hubble Type
First we consider early Hubble-type barred spirals, which we
take to be Hubble types Sbc and earlier. There are 26 early-type
galaxies in SONG, plus 6 from the Sakamoto sample; these are
shown in Figure 4. For galaxies above the median fcon for earlytype
 galaxies, 13 of 16 are barred, compared with only 7 of 16
below the median. The median fcon for the early-type bars is 43,
compared with 14 for the unbarred early types. Clearly, the evidence
 for bar driven inflow in early-type galaxies is very strong.
Next we consider the 16 late-type spirals, shown in Figure 5.
For galaxies above the median fcon, seven of eight are barred
spirals, whereas for galaxies below the median, only three of
eight are barred. The median fcon for late-type bars is 20, 3 times
higher than that for unbarred late-type galaxies. Thus there is
also good evidence for inflow in late-type bars. We conclude that
bars transport gas inward regardless of Hubble type.
Although both early- and late-type bars have fcon-values
about 3 times higher than their unbarred counterparts, note that
the median fcon for early-type barred galaxies is about twice that
for late-type bars. Also, the median C6nuc (MC12 pcC02) for barred
early types is 400, compared to 87 for late-type bars. In part, the
higher fcon and C6nuc for early types can be understood as a consequence
 of the higher critical surface density threshold for star
formation in early types resulting from their steeper rotation
curves. This is probably the reason why even among the unbarred
 spirals in our sample, the early-type galaxies have about
a 3 times higher median fcon and C6nuc compared to those for the
late types; however, this comparison should be viewed with caution
 because of the small number of unbarred galaxies.
Ho et al. (1997) attributed the higher star formation rates in
early-type bars to the higher critical density threshold and predicted
 the higher gas surface densities that we indeed find.
However, for early-type bars to maintain these higher central mass
concentrations requires that the early-type bars have significantly
higher mass inflow rates than late-type bars. This is because the
gas consumption rate implied by the star formation rates observed
by Ho et al. are higher in early types--the median value of the
star formation rate in barred early types is 0.08 MC12 yrC01,com-
pared to 0.007 MC12 yrC01 in barred late types. That is, not only are
the median surface densities and nuclear masses a factor of 4
higher in early-type bars, but the gas is consumed at a typical
rate more than 10 times higher. Clearly, the mass inflow rates in
early types must be much higher than for late-type bars.
How can we explain this difference? We believe the reason is
related to the different type of bars that are typically found in
early and late Hubble-type galaxies: ''flat'' and ''exponential''
(Elmegreen & Elmegreen 1985, 1996; Ohta et al. 1986; Combes
& Elmegreen 1993). Flat bars, found primarily in early-type galaxies,
 have a rectangular appearance with flat shoulders in their
photometric profile, are usually associated with grand-design
spiral structure, and lie in the flat part of the rotation curve. They
are also longer relative to their disks and have higher amplitudes.
 Exponential bars, on the other hand, are located in the
rising part of a rotation curve, and are oval, rather than rectangular
 in appearance. Their photometric profile is exponential,
similar to a galactic disk. They are shorter relative to their disks
than flat bars and have a lower amplitude. Exponential bars are
primarily found in late Hubble-type spirals.
Bars in early-type galaxies, i.e., flat bars, are predicted to be
more effective at driving gas inward, because longer bars encounter
 more gas in the disk and because the higher nonaxisymmetry
 leads to a higher inflow rate. N-body (Combes &
Elmegreen 1993) and hydrodynamic models (Regan & Teuben
2004) predict that weaker bars (fatter bars and/or less massive
bars) drive significantly less mass inward. The evidence presented
 here thus supports the predictions of higher inflow rates in
early-type barred spirals.
Another possibility is that the higher CO fluxes in the nuclei of
early-type galaxies may reflect a change in the CO emissivity
due to higher pressures and velocity dispersions in the bulge
region (see the discussion in Regan et al. 2001). The increased
star formation in the nuclei of early-type spirals may also contribute
 to an increase in the pressure. But if there is indeed less
gas at the centers of early types, then the enhanced star formation
implies a higher efficiency and decreased gas consumption times.
Then the presence of a large fraction of early-type bars with high
central concentration of CO emission is surprising. A quantitative
discussion of this awaits better extinction-corrected star formation
rates.
3.3. Secular Evolution From Late- to Early-Type Galaxies?
Several studies have suggested that galaxies undergo secular
evolution via bar-driven inflows. One popular evolutionary
scenario is the change of late-type galaxies into early types via
bar inflows (Friedli & Benz 1995; Zhang 1999). We argue that
such a transformation is unlikely. In order for a late-type galaxy
to build an early-type bulge, it would need to add C241010 to
C241011 MC12 of gas that must then be converted to stars (Binney &
Merrifield 1998). If we take the best-case scenario for such a
transformation, i.e., assume that bars (1) are long lived (10 Gyr),
(2) transport gas inward at sufficiently high rates, (3) have an
adequate supply of molecular gas, (4) convert enough of the
accumulated gas to stars, and (5) scatter the stars in the vertical
dimension to build a bulge, then a late-type galaxy can build an
early-type bulge. Some of these conditions are difficult, if not
impossible, to reconcile with observational data.
While the high bar fraction at z > 0:7 (Sheth et al. 2003)
suggests that bars are likely to be long lived, it is unlikely that the
mass inflow rates or the gas reservoir in late Hubble-type galaxies
 are sufficient to build an early-type bulge. Based on the
differences in fcon-values, we noted that the mass inflow rate in
late Hubble-type galaxies is lower than in early Hubble types.
The inflow rate in the NRO-OVRO sample was estimated by
Sakamoto et al. (1999b) to be C240.1-1 MC12 yrC01. In later type
galaxies the rate would be even lower. We note that at the median
nuclear star formation rate for late-type bars measured by Ho
et al. (1997) data (0.007 MC12 yrC01), it would take 1012 yr to build
even a 1010 MC12 bulge. At the upper end of gas consumption rates
for late-type bars (1 MC12 yrC01), it would be possible to build a
small bulge in 1010 yr; however, the data demonstrate that this is
certainly not typical.
Moreover, late-type bars are small and do not have access to
a vast molecular gas reservoir. In SONG, the average length of
a late-type bars is 3 kpc (the average length of bars in the entire
 sample is 5 kpc; Sheth 2001). To measure the available gas
reservoir, let us assume this length and the average C6tot for
SONG (8 MC12 pcC02). Then the available gas reservoir for a typical
 late-type spiral is only 5 ; 107 MC12. This value may be higher
because the average C6tot is a measurement over the entire disk.
In the bar region, the available gas reservoir may be higher. But
even if we assume a C6 that is 10 times higher, i.e., 80 MC12 pcC02,
the gas reservoir would still be too small. It is possible that
accretion from mergers or gas transport from outer parts of the
galactic disk may increase this number but it is at least 2 orders
of magnitude lower than that required for creating an early-type
bulge in a late-type galaxy. Thus, we believe that it is unlikely
that a late-type galaxy evolves into an early type, consistent
with Kormendy & Kennicutt (2004).
This, however, does not imply that there is no secular evolution
 in barred spiral galaxies. To the contrary, the evidence
presented here for bar-driven gas inflows, combined with the
enhanced star formation activity in barred spirals, indicates that
the central regions of barred galaxies undergo significant changes.
They are probably the most important changes in the evolution of
disks, since z C24 1 when the merging activity began to decline
(Baugh et al. 1996; Ferguson et al. 2000). Another possible secular
 evolutionary sequence could be the dissolution of bars by an
increasing central concentration of mass. Models have been used
to debate the exact impact of increasing central concentration
(e.g., Friedli & Benz 1993; Norman et al. 1996; Bournaud &
Combes 2002; Athanassoula 2003; Shen & Sellwood 2004).
Tentative evidence for such dissolution was presented in Das et al.
(2003), who showed a correlation between the bar ellipticity and
the central mass concentration.
The data presented here lend credence to the formation of
pseudobulges in barred spirals; pseudobulges and secular evolution
 via gas inflows have recently been reviewed by Kormendy
& Kennicutt (2004). In this context, we particularly point out the
significant population of early-type barred spirals (6 of 19) that
have no molecular gas detected in nuclear region and very little
detected in the region within the bar corotation radius. This
suggests that the gas inside the bar corotation radius has already
been driven into the nuclear region, where it has been consumed
by star formation. Among the late-type spirals, there are three
barred galaxies with upper limits to the molecular gas in the
central kiloparsec. One of them has no gas within the bar corotation
 radius. As already discussed, the process of gas accretion
is slower in late types. The star formation rates in late types are
also lower, because late-type spirals have lower C6crit, the critical
 gas surface densities for star formation (e.g., Toomre 1964;
Kennicutt 1989). So it is not unusual to find fewer examples of
late-type barred spirals with no gas in the center and bar regions.
The secular evolutionary sequence of gas inflow and subsequent
star formation simply occurs at a slower pace in these galaxies;
over time, even the late-type spirals will build pseudobulges.
Signatures of poststarburst populations in barred spirals with
low Mnuc and little gas in the bar region would provide further
confirmation of this evolutionary sequence.
Another way to interpret the galaxies without gas in the center
is that a third of the barred galaxies, whether they are early type
or late type, are in a quiescent state, i.e., without measurable gas
in the center or the bar region. This indicates that if the fueling
via bars is periodic, then bars are ''actively'' fueling the center
two-thirds of the time. When better measurements of gas inflow
are obtained, this duty cycle will need to be taken into account to
measure the impact of the gas inflow over the lifetime of the bar
and galaxy.
4. CONCLUSIONS
With a larger and more representative sample of nearby galaxies
 from SONG, we compared the central concentration of
molecular gas in both CO-bright and CO-faint spiral galaxies
spanning a range of Hubble types. In all cases we found clear
evidence of more centrally concentrated molecular gas distributions
 in barred spirals. This is the strongest and most conclusive
 evidence for the bar-driven transport of molecular gas to
the central kiloparsec of galaxies.
We find that barred spirals of late Hubble types are less centrally
 concentrated than early Hubble types. This, coupled with
the enhanced star formation activity observed in early-type bars,
indicates higher mass accretion rates in early Hubble-type spirals.
 The differences are probably related to the longer and
stronger ''flat'' bars that are found preferentially in early Hubbletype
 galaxies. The observation of a significant subset of earlytype
 barred spirals with little or no gas within the bar region is
consistent with higher accretion rates; these galaxies have the
expected conditions of a poststarburst galaxy, where the large
stockpiles of gas driven inward by the bar have already been
converted into stars.
Contrary to previous suggestions, the evidence indicates that
it is highly unlikely for a late-type galaxy to evolve into an early
type via bar-induced gas inflow. Nonetheless, secular evolutionary
 processes are undoubtedly present, and pseudobulges
are inevitable because of the bar-induced gas inflow. Evidence
for pseudobulges is likely to be clearest in early-type galaxies
because of their higher inflow rates and higher star formation
activity.
The authors are indebted to the referee for the detailed and
thorough examination of this manuscript and the many helpful
comments and suggestions that s/he provided. This work would
not have been possible without the rest of the SONG team
members (T. Helfer, T. Wong, L. Blitz, and D. Bock) and the
dedicated observatory staff at Hat Creek and at the Laboratory
for Millimeter-Wave Astronomy at the University of Maryland.
We thank T. Beasley, D. M. Elmegreen, B. G. Elmegreen,
A. Harris, J. Kenney, J. Knapen, N. Scoville, E. Schinnerer,
I. Shlosman, L. E. Strubbe, and T. Treu for invaluable and
insightful discussions.

______________________________________________________________
Received 2004 December 7; accepted 2005 May 18
1 Spitzer Science Center, Mail Stop 220-6, California Institute of Technology,
 Pasadena, CA 91125; kartik@astro.caltech.edu.
2 Department of Astronomy, University of Maryland, College Park, MD
20742-2421.
3 Visiting Astronomer, Kitt Peak National Observatory, National Optical
Astronomy Observatory, which is operated by the Association of Universities
for Research in Astronomy, Inc., under cooperative agreement with the National
Science Foundation (NSF).
4 Space Telescope Science Institute, 3700 San Martin Drive, Baltimore,
MD 21218.
5 Department of Physics, Bucknell University, Lewisburg, PA 17837.
6 The NRO-OVRO survey chose galaxies with integrated intensities of
ICO >10 K km sC01 in any one of the 5000 FCRAO pointings (typically the central
pointing). We use the 10 K km sC01 selection as the division between CO-bright
and CO-faint galaxies for the remainder of this paper.
7 The National Radio Astronomy Observatory is a facility of the NSF, operated
 under cooperative agreement by Associated Universities, Inc.
TABLE 1
BIMA SONG Sample
Galaxy
C11
(J2000.0)
C14
(J2000.0)
D25
(arcmin)
P.A.
(deg)
i
(deg)
VHel
(km sC01)
D
(Mpc) Reference
Type
(RC3)
OTF
Data
Barred
NGC 0925............... 02 27 16.9 +33 34 44.0 10.5 102 55.8 553 9.3 1 SAB(s)d ...
IC 342 ..................... 03 46 48.5 +68 05 46.0 21.4 37 31.0 34 2.1 2 SAB(rs)cd Y
NGC 2403............... 07 36 51.3 +65 36 09.2 21.9 127 55.8 131 2.9 3 SAB(s)cd ...
NGC 2903............... 07 36 51.3 +65 36 09.2 12.6 17 61.4 556 7.3 4 SAB(rs)bc Y
NGC 3184............... 10 18 16.9 +41 25 27.8 7.4 135 21.1 592 8.7 5 SAB(rs)cd ...
NGC 3344............... 10 43 31.1 +24 55 20.0 7.1 24 25.5 586 12.5 6 (R)SAB(r)bc ...
NGC 3351............... 10 43 57.7 +11 42 13.0 7.4 13 47.8 778 10.1 7 SB(r)b Y
NGC 3368............... 10 46 45.7 +11 49 11.8 7.6 5 46.2 897 11.2 8 SAB(rs)ab ...
NGC 3521............... 11 05 48.5 C000 02 09.2 11 163 62.1 805 7.2 9 SAB(rs)bc Y
NGC 3627............... 11 20 15.0 +12 59 28.6 9.1 173 62.8 727 11.1 10 SAB(s)b Y
NGC 3726............... 11 33 21.1 +47 01 44.7 6.2 10 46.2 849 11.7 11 SAB(r)c ...
NGC 3953............... 11 53 49.0 +52 19 35.6 6.9 13 59.9 1054 14.1 12 SB(r)bc ...
NGC 3992............... 11 57 35.9 +53 22 28.3 7.6 68 51.9 1048 14.2 13 SB(rs)bc ...
NGC 4051............... 12 03 09.6 +44 31 52.5 5.3 135 42.2 725 9.4 14 SAB(rs)bc ...
NGC 4258............... 12 18 57.6 +47 18 13.4 18.6 150 67.1 448 7.3 15 SAB(s)bc Y
NGC 4303............... 12 21 54.9 +04 28 24.9 6.5 7 27 1566 15.2 16 SAB(rs)bc Y
NGC 4321............... 12 22 54.8 +15 49 20.6 7.4 30 31.7 1571 16.1 17 SAB(s)bc Y
NGC 4490............... 12 30 36.3 +41 38 37.1 6.3 125 60.7 565 7.7 18 SB(s)d-pec ...
NGC 4535............... 12 34 20.3 +08 11 51.9 7.1 0 44.9 1961 16.0 19 SAB(s)c ...
NGC 4548............... 12 35 26.4 +14 29 46.8 5.4 150 37.4 486 15.9 20 SB(rs)b ...
NGC 4559............... 12 35 57.6 +27 57 35.1 10.7 150 66 815 9.7 21 SAB(rs)cd ...
NGC 4569............... 12 36 49.8 +13 09 46.3 9.6 23 62.8 C0235 16.8 22 SAB(rs)ab Y
NGC 4579............... 12 37 43.5 +11 49 05.1 5.9 95 37.4 1519 16.8 23 SAB(rs)b ...
NGC 4699............... 12 49 02.18 C008 39 51.4 3.8 45 46.2 1427 19.0 24 SAB(rs)b ...
NGC 4725............... 12 50 26.6 +25 30 02.7 10.7 35 44.9 1206 12.6 25 SAB(r)ab-pec ...
NGC 5005............... 12 50 26.6 +25 30 02.7 5.8 65 61.4 946 21.3 26 SAB(rs)bc Y
NGC 5248............... 13 37 32.0 +08 53 06.2 6.2 110 43.6 1153 22.7 27 SAB(rs)bc Y
NGC 5457............... 14 03 12.5 +54 20 55.5 28.8 35 21.1 241 6.5 28 SAB(rs)cd Y
NGC 6946............... 20 34 52.3 +60 09 13.2 11.5 35 31.7 48 6.4 29 SAB(rs)cd Y
Unbarred
NGC 0628............... 01 36 41.7 +15 47 00.5 10.5 25 24.2 657 7.3 30 SA(s)c Y
NGC 1068............... 02 42 40.7 C000 00 47.8 7.1 70 31.7 1136 18.0 31 (R)SA(rs)b Y
NGC 2841............... 09 22 02.6 +50 58 35.3 8.1 147 64.1 638 9.5 32 SA(r)b ...
NGC 2976............... 09 47 15.4 +67 54 59.0 5.9 143 62.8 3 4.3 33 SAc-pec ...
NGC 3031............... 09 55 33.1 +69 03 54.9 26.9 157 58.4 C034 3.3 34 SA(s)ab ...
NGC 3938............... 11 52 49.4 +44 07 14.6 5.4 20 14 809 11.3 35 SA(s)c Y
NGC 4414............... 12 26 27.08 +31 13 24.7 3.6 155 55.8 716 19.1 36 SA(rs)c? Y
NGC 4450............... 12 28 29.6 +17 05 05.8 5.3 175 42.2 1954 16.0 37 SA(s)ab ...
NGC 4736............... 12 50 53.1 +41 07 12.5 11.2 105 35.6 308 6.6 38 (R)SA(r)ab Y
NGC 4826............... 12 56 43.6 +21 40 57.6 10 115 57.5 408 5.0 39 (R)SA(rs)ab Y
NGC 5033............... 13 13 27.5 +36 35 37.1 10.7 170 62.1 875 21.3 40 SA(s)c Y
NGC 5055............... 13 15 49.3 +42 01 45.4 12.6 105 54.9 504 7.2 41 SA(rs)bc Y
NGC 5194............... 13 29 52.6 +47 11 42.9 11.2 163 51.9 463 8.4 42 SA(s)bc-pec Y
NGC 5247............... 13 38 03.0 C017 53 02.5 5.6 20 29.4 1357 15.2 43 SA(s)bc Y
NGC 7331............... 22 37 04.0 +34 24 57.3 10.5 171 69.2 821 15.1 44 SA(s)b Y
Note.--Units of right ascension are hours, minutes, and seconds, and units of declination are degrees, arcminutes, and arcseconds.
a 2MASS Galaxy Centers from the LGA (Jarrett et al. 2003).
b Optical diameter (RC3).
c Position angle (RC3).
d Inclination (RC3).
e Heliocentric velocity (NASA/IPAC Extragalactic Database).
f Adopted distance; reference given in column to the right.
g Indicates whether on-the-fly data were collected as part of SONG.
h Galaxies in common between SONG and NRO-OVRO sample (Sakamoto et al. 1999b).
i From kinematic fitting, Crothwaite et al. (2000).
References.--(1) Sohn & Davidge 1998; (2) Karachentsev & Tikhonov 1993; (3) Metcalfe & Shanks 1991; (4) Planesas et al. 1997; (5) Pompei & Natali 1997;
(6) Corbelli et al. 1989; (7) Graham et al. 1997; (8) Tanvir et al. 1999; (9) Thornley 1996; (10) Saha et al. 1999; (11) van der Kruit 1971; (12) RC3, using H0 = 75;
(13) Gottesman & Hunter 1982; (14) Liszt & Dickey 1995; (15) Herrnstein et al. 1997; (16) Molla et al. 1999; (17) Ferrarese et al. 1996; (18) Tully 1988; (19) Macri
et al. 1999; (20) Graham et al. 1999; (21) Vogler et al. 1996; (22) Barth et al. 1998; (23) Ho 1999a; (24) Bower et al. 1993; (25) Gibson et al. 1999; (26) Barth et al.
1998; (27) Tully 1988; (28) Stetson et al. 1998; (29) Sharina et al. 1997; (30) Sharina et al. 1996; (31) literature values vary from 14 to 22 Mpc; 18 Mpc is assumed
as midpoint value; (32) Block et al. 1999; (33) Karachentsev et al. 1991; (34) Paturel et al. 1998; (35) Jimenez-Vicente et al. 1999; (36) Turner et al. 1998; (37) Tully
1988; (38) Mulder & Driel 1993; (39) Rubin 1994; (40) Ho et al. 1999b; (41) Tully 1988; (42) Feldmeir et al. 1997; (43) Grosbol & Patsis 1998; (44) Hughes etal.1998.
TABLE 2
Derived Properties for Barred SONG Galaxies
Galaxy
(1)
S COnuc COMB
(Jy km sC01)
(2)
S COnuc BIMA
(Jy km sC01)
(3)
rms in BIMA
(Jy km sC01)
(4)
Mnuc
(107 MC12)
(5)
C6nuc
(MC12 pcC02)
(6)
S COtot
(Jy km sC01)
(7)
Mtot
(108 MC12)
(8)
C6tot
(MC12 pcC02)
(9)
C6nuc/C6tot
(10)
Early Types
NGC 2903................... 462 447 13 35.6 454 2740 22.0 3.89 117
NGC 3344 ................. ... 2 8 3.70 47.6 520 12.0 2.30 20.7
NGC 3351................... 207 208 9 31.7 404 700 10.7 2.88 140
NGC 3368................... ... 309 11 58.0 738 610 11.5 2.38 310
NGC 3521 ................. 53 4 15 2.30 30.0 4920 38.2 9.15 3.28
NGC 3627................... 244 233 7 43.0 546 4660 85.8 12.7 43.1
NGC 3953 ................. ... 7 9 5.40 68.0 1790 53.2 8.90 8.04
NGC 3992 ................. ... 6 8 4.80 62.0 ... 7.00 2.00 31.0
NGC 4051................... ... 103 3 13.6 173 740 9.80 5.93 29.2
NGC 4258................... 180 176 13 14.0 179 1240 9.90 0.81 221
NGC 4303................... 135 122 4 42.0 537 2280 78.8 12.2 44.0
NGC 4321................... 140 132 5 51.0 651 3340 129 13.7 47.5
NGC 4548................... ... 26 5 9.80 125 540 20.4 4.20 29.8
NGC 4569................... 141 136 5 57.0 731 1500 63.0 3.66 200
NGC 4579................... ... 43 8 18.2 231 910 38.0 5.88 39.3
NGC 4699 ................. ... 3 6 6.50 83.0 ... ... ... ...
NGC 4725 ................. ... 15 16 7.60 97.0 1950 46.0 3.83 25.3
NGC 5005................... ... 80 2 54.0 691 1260 85.0 8.40 82.3
NGC 5248................... 57 54 2 42.0 515 1190 92.0 6.97 76.0
Late Types
IC 342 ......................... 2379 1713 75 11.3 144 29220 19.0 14.4 10.0
NGC 0925 ................. ... 6 10 2.60 33.0 960 12.0 1.96 16.8
NGC 2403 ................. ... 15 19 0.50 6.10 540 0.70 0.25 24.4
NGC 3184................... ... 60 7 6.80 87.0 1120 13.0 4.60 18.9
NGC 3726................... ... 33 3 6.80 86.0 720 14.7 4.22 20.4
NGC 4490 ................. ... 0 3 0.50 6.80 480 4.00 2.70 2.52
NGC 4535................... ... 110 5 42.0 536 1570 60.0 7.00 76.6
NGC 4559................... ... 21 9 2.96 38.0 ... 4.00 0.50 76.0
NGC 5457................... 123 116 7 7.30 93.0 ... 14.9 4.90 19.0
NGC 6946................... 1691 1747 15 107 1363 12370 76.0 21.1 64.7
a CO flux in central kiloparsec measured from BIMA SONG.
b Total gas mass in the central kiloparsec calculated from the SONG flux. All masses are calculated using M = 1:36 M(H2), M(H2) = 1:1 ; 104D2SCO,whereD
is in Mpc and SCO is in Jy km sC01.WeuseaCOtoH2 conversion factor of 2:8 ; 1020 cmC02 (K km sC01)C01, 7% smaller than Sakamoto et al. (1999b).
c Nuclear gas surface density derived by dividing the mass by C25(500 pc)2.
d Total CO flux from galaxy (disk+nucleus) from the FCRAO survey (Young et al. 1995).
e Total gas mass in the galaxy (disk+nucleus) calculated from the FCRAO flux.
f Total gas surface density in the galaxy (disk+nucleus) calculated from the FCRAO flux.
g Ratio of the nuclear gas surface density to the disk gas surface density, defined as fcon in x 1.
h 2 C27 upper limit.
i Using galaxy-averaged flux from SONG data instead of the FCRAO survey.
TABLE 3
Derived Properties for Unbarred SONG Galaxies
Galaxy
(1)
S COnuc COMB
(2)
S COnuc BIMA
(Jy km sC01)
(3)
rms in BIMA
(4)
Mnuc
(107 MC12)
(5)
C6nuc
(MC12 pcC02)
(6)
S COtot
(Jy km sC01)
(7)
Mtot
(108 MC12)
(8)
C6tot
(MC12 pcC02)
(9)
C6nuc/C6tot
(10)
Early Types
NGC 1068............... 120 116 9 56.0 716 5160 250 23.0 31.1
NGC 2841 ............. ... C011 14 3.80 48.0 1870 25.0 6.40 7.50
NGC 3031............... ... 37 34 1.10 14.0 ... 10.0 1.00 14.0
NGC 4450............... ... 12 5 3.80 49.0 450 17.0 3.60 13.6
NGC 4736............... 239 190 15 12.4 157 2560 17.0 4.60 34.1
NGC 4826............... 760 769 24 28.8 366 2170 8.10 4.90 74.7
NGC 5055............... 302 267 13 20.7 264 5670 44.0 8.00 33.0
NGC 5194............... 103 81 11 8.55 109 9210 97.0 16.5 6.61
NGC 5247............... 29 20 3 6.90 88.0 1030 36.0 7.40 11.9
NGC 7331 ............. 4 C03 8 5.50 69.0 4160 142 8.50 8.12
Late Types
NGC 0628............... 41 26 8 2.10 26.0 2160 17.0 4.40 5.91
NGC 2976............... ... 21 5 0.58 7.40 610 1.70 3.90 1.90
NGC 3938............... 38 30 6 5.70 73.0 1750 33.0 13.5 5.41
NGC 4414............... 10 4 2 2.20 28.0 2740 150 47.0 0.60
NGC 5033............... 35 34 6 23.0 294 1640 111 3.20 91.9
a CO flux in central kiloparsec measured from BIMA SONG.
b Total gas mass in the central kiloparsec calculated from the SONG flux. All masses are calculated using M = 1:36M(H2), M(H2) = 1:1 ; 104D2SCO, where D
is in Mpc and SCO is in Jy km sC01.WeuseaCOtoH2 conversion factor of 2:8 ; 1020 cmC02 (K km sC01)C01, 7% smaller than Sakamoto et al. (1999b).
c Nuclear gas surface density derived by dividing the mass by C25(500 pc)2.
d Total CO flux from galaxy (disk+nucleus) from the FCRAO survey (Young et al. 1995).
e Total gas mass in the galaxy (disk+nucleus) calculated from the FCRAO flux.
f Total gas surface density in the galaxy (disk+nucleus) calculated from the FCRAO flux.
g Ratio of the nuclear gas surface density to the disk gas surface density, defined as fcon in x 1.
h 2 C27 upper limits.
i Using galaxy-averaged flux from SONG data instead of the FCRAO survey.
8 NGC 4699 is not shown in any of the figures because it only has an upper
limit for the single-dish flux. However, it is one of the galaxies with an upper limit
for Mnuc and has no gas in its bar.
Fig. 1.--Plot of molecular gas mass in solar masses in the central kiloparsec
( y-axis) and integrated over the entire galactic disk (x-axis). The diagonal lines
indicate the molecular gas mass fraction in the central kiloparsec. The barred
spirals are represented with circles, and the unbarred spirals with triangles. Solid
symbols are SONG data, whereas open symbols are data from galaxies not in
common with SONG from Sakamoto et al. (1999b). The crosses mark galaxies
that have no gas in the bars (NGB). Galaxies with only upper limits are shown
with a small vertical segment. Larger symbols reflect galaxies that are CO-bright
(ICO > 10 K km sC01 in an FCRAO pointing).
Fig. 2.--Same as Fig. 1, but with the axes now reflecting the molecular gas
surface densities in solar masses per square parsec. The diagonal lines show the
concentration index defined as the ratio of the nuclear molecular gas surface
density to the total molecular gas surface density.
Fig. 3.--Same as Fig. 2, but showing only the CO-faint galaxies. Fig. 4.--Same as Fig. 2, but showing only the early Hubble-type (Sa-Sbc)
galaxies are shown.
Fig. 5.--Same as Fig. 2, but showing only the late Hubble-type (Sc-Sdm)
galaxies.
REFERENCES
Athanassoula, E. 1992a, MNRAS, 259, 328
------. 1992b, MNRAS, 259, 345
------. 2003, MNRAS, 341, 1179
Barth, A. J., Ho, L. C., Filippenko, A. V., & Sargent, W. L. W. 1998, ApJ, 496,
133
Baugh, C. M., Cole, S., & Frenk, C. S. 1996, MNRAS, 282, L27
Binney, J., & Merrifield, M. 1998, Galactic Astronomy (Princeton: Princeton
Univ. Press), chapter 4
Block, D. L., Stockton, A., Elmegreen, B. G., & Willis, J. 1999, ApJ, 522, L25
Bloemen, J. B. G. M., et al. 1986, A&A, 154, 25
Bournaud, F., & Combes, F. 2002, A&A, 392, 83
Bower, G. A., Richstone, D. O., Bothun, G. D., & Heckman, T. M. 1993, ApJ,
402, 76
Combes, F., & Elmegreen, B. G. 1993, A&A, 271, 391
Corbelli, E., Schneider, S. E., & Salpeter, E. E. 1989, AJ, 97, 390
Crothwaite, L., Turner, J. L., & Ho, P. T. P. 2000, AJ, 119, 1720
Das, M., Vogel, S. N., Teuben, P. J., Regan, M. W., Sheth, K., Harris, A. I., &
Jeffreys, W. H. 2003, ApJ, 582, 190
de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H. G., Jr., Buta, R. J., Paturel,
G., & Fouque, P. 1991, Third Reference Catalogue of Bright Galaxies (New
York: Springer) (RC3)
Downes, D., Reynaud, D., Solomon, P. M., & Radford, S. J. E. 1996, ApJ, 461,
186
Elmegreen, B. G., Elmegreen, D. M., & Hirst, A. C. 2004, ApJ, 612, 191
------. 1985, ApJ, 288, 438
Elmegreen, D. M. 1996, IAU Colloq. 91, Barred Galaxies, R. J. Buta, D. A.
Crocker, & B. G. Elmegreen (San Francisco: ASP), 23
Eskridge, P., et al. 2000, AJ, 119, 536
Feldmeier, J. J., Ciardullo, R., & Jacoby, G. H. 1997, ApJ, 479, 231
Ferguson, H. C., Dickinson, M., & Williams, R. 2000, ARA&A, 38, 667
Ferrarese, L., et al. 1996, ApJ, 468, L95
Friedli, D., & Benz, W. 1993, A&A, 268, 65
------. 1995, A&A, 301, 649
Gibson, B. K., et al. 1999, ApJ, 512, 48
Graham, J. A., et al. 1997, ApJ, 477, 535
------. 1999, ApJ, 516, 626
Grosbol, P. J., & Patsis, P. A. 1998, A&A, 336, 840
Gottesman, S. T., & Hunter, J. H., Jr. 1982, ApJ, 260, 65
Hawarden, T. G., Mountain, C. M., Leggett, S. K., & Puxley, P. J. 1986,
MNRAS, 221, P41
Helfer, T. T., Regan, M. W., Thornley, M. D., Sheth, K., Wong, T., Vogel, S. N.,
Bock, D. C.-J., & Blitz, L. 2003, ApJS, 145, 259
Helfer, T. T., Vogel, S. N., Lugten, J. B., & Teuben, P. J. 2002, PASP, 114,
350
Herrnstein, J., Moran, J., Greenhill, L., Inoue, M., Nakai, N., Miyoshi, M., &
Diamond, P. 1997, BAAS, 29, 1252
Ho, L. C. 1999a, ApJ, 516, 672
Ho, L. C., Filippenko, A. V., & Sargent, W. L. W. 1997, ApJ, 487, 591
Ho, L. C., Ptak, A., Terashima, Y., Kunieda, H., Serlemitsos, P. J., Yaqoob, T.,
& Koratkar, A. P. 1999b, ApJ, 525, 168
Huang, J. H., Gu, Q. S., Su, H. J., Hawarden, T. G., Liao, X. H., & Wu, G. X.
1996, A&A, 313, 13
Hughes, S. M. G., et al. 1998, ApJ, 501, 32
Jarrett, T. H., Chester, T., Cutri, R., Schneider, S. E., & Huchra, J. P. 2003, AJ,
125, 525
Jimenez-Vicente, J., Battaner, E., Rozas, M., Castaqeda, H., & Porcel, C. 1999,
A&A, 342, 417
Jogee, S., Scoville, N. Z., & Kenney, J. D. P. 2005, ApJ, 630, 837
Jogee, S., et al. 2004, ApJ, 615, L105
Karachentsev, I. D., & Tikhonov, N. A. B. 1993, A&AS, 100, 227
Karachentsev, I. D., Tikhonov, N. A., Sharina, M. E., Georgiev, Ts. B., &
Bilkina, B. I. 1991, A&AS, 91, 503
Kenney, J. D. P. 1996a, in ASP Conf. Ser. 91, Barred Galaxies, ed. R. Buta, D. A.
Crocker, & B. G. Elmegreen (San Francisco: ASP), 150
Kennicutt, R. C., Jr. 1989, ApJ, 344, 685
------. 1998, ARA&A, 36, 189
Knapen, J. H., Shlosman, I., & Peletier, R. F. 2000, ApJ, 529, 93
Kormendy, J. 1993, IAU Symp. 153, Galactic Bulges, ed. H. Dejonghe & H. J.
Habing (Dordrecht: Kluwer), 209
Kormendy, J., & Kennicutt, R. C., Jr. 2004, ARA&A, 42, 603
Laine, S., Shlosman, I., Knapen, J. H., & Peletier, R. F. 2002, ApJ, 567, 97
Laurikainen, E., Salo, H., & Buta, R. 2004, ApJ, 607, 103
Liszt, H. S., & Dickey, J. M. 1995, AJ, 110, 998
Macri, L. M., et al. 1999, ApJ, 521, 155
Martin, P., & Roy, J. 1994, ApJ, 424, 599
Martinet, L., & Friedli, D. 1997, A&A, 323, 363
Menendez-Delmestre, K., Sheth, K., Scoville, N. Z., Jarrett, T., Schinnerer, E.,
Regan, M. W., & Block, D. 2004, in Penetrating Bars through Masks of
Cosmic Dust: The Hubble Tuning Fork Strikes a New Note, ed. D. Block
et al. (Dordrecht: Kluwer), 787
Metcalfe, N., & Shanks, T. 1991, MNRAS, 250, 438
Molla, M., Hardy, E., & Beauchamp, D. 1999, ApJ, 513, 695
Mulder, P. S., & van Driel W. 1993, A&A, 272, 63
Norman, C. A., Sellwood, J. A., & Hasan, H. 1996, ApJ, 462, 114
Ohta, K., Sasaki, M., & Saito, M. 1986, PASJ, 38, 677
Paturel, G., Lanoix, P., Teerikorpi, P., Theureau, G., Bottinelli, L., Gouguenheim,
L., Renaud, N., & Witasse, O. 1998, A&A, 339, 671
Pfenniger, D., & Norman, C. 1990, ApJ, 363, 391
Planesas, P., Colina, L., & Perez-Olea, D. 1997, A&A, 325, 81
Pompei, E., & Natali, G. 1997, A&AS, 124, 129
Regan, M. W., Helfer, T. T., Thornley, M. D., Sheth, K., Wong, T., Vogel, S. N.,
Blitz, L., & Bock, D. C.-J. 2001, ApJ, 561, 218
Regan, M. W., & Mulchaey, J. 1999, AJ, 117, 2676
Regan, M. W., Sheth, K., & Vogel, S. N. 1999, ApJ, 526, 97
Regan, M. W., & Teuben, P. J. 2003, ApJ, 582, 723
------. 2004, ApJ, 600, 595
Rubin, V. C. 1994, AJ, 107, 173
Sage, L. 1993, A&A, 272, 123
Saha, A., Sandage, A., Tammann, G. A., Labhardt, L., Macchetto, F. D., &
Panagia, N. 1999, ApJ, 522, 802
Sakamoto, K., Okumura, S. K., Ishizuki, S., & Scoville, N. Z. 1999a, ApJS,
124, 403
------. 1999b, ApJ, 525, 691
Sault, R. J., Teuben, P. J., & Wright, M. C. H. 1995, in ASP Conf. Ser. 77,
Astronomical Data Analysis Software and Systems IV, ed. R. A. Shaw, H. E.
Payne, & J. J. E. Hayes (San Francisco: ASP), 433
Sellwood, J. A., & Wilkinson, A. 1993, Rep. Prog. Phys, 56, 173
Sersic, J. L., & Pastoriza, M. 1967, PASP, 79, 152
Sharina, M. E., Karachentsev, I. D., & Tikhonov, N. A. 1997, Astron. Lett., 23, 373
------. 1996, A&AS, 119, 499
Shen, J., & Sellwood, J. A. 2004, ApJ, 604, 614
Sheth, K. 2001, Ph.D. thesis, Univ. Maryland
Sheth, K., Regan, M. W., Scoville, N. Z., & Strubbe, L. E. 2003, ApJ, 592, L13
Sheth, K., Regan, M. W., Vogel, S. N., & Teuben, P. J. 2000, ApJ, 532, 221
Sheth, K., Vogel, S. N., Regan, M. W., Teuben, P. J., Harris, A. I., & Thornley,
M. D. 2002, AJ, 124, 2581
Shlosman, I., Frank, J., & Begelman, M. C. 1989, Nature, 338, 45
Sohn, Y.-J., & Davidge, T. J. 1998, AJ, 115, 130
Stetson, P. B., et al. 1998, ApJ, 508, 491
Tanvir, N. R., Ferguson, H. C., & Shanks, T. 1999, MNRAS, 310, 175
Thornley, M. D. 1996, ApJ, 469, L45
Toomre, A. 1964, ApJ, 139, 1217
Tully, R. B. 1988, Nearby Galaxies Catalog (Cambridge: Cambridge Univ. Press)
Turner, J. L. 1996, in IAU Colloq. 157, Barred Galaxies, ed. R. Buta, D. A.
Crocker, & B. G. Elmegreen (San Francisco: ASP), 146
Turner, J. L., et al. 1998, ApJ, 505, 207
van der Kruit, P. C. 1971, A&A, 15, 110
Vila-Costas, M. B., & Edmunds, M. G. 1992, MNRAS, 259, 121
Vogler, A., Pietsch, W., & Bertoldi, F. 1996, Rontgenstrahlung from the Universe,
 ed. H. U. Zimmermann, J. Trumper, & H. Yorke (MPE Rep. 263;
Garching: MPE), 399
Welch, W. J. 1996, PASP, 108, 93
Young, J. S., & Scoville, N. Z. 1991, ARA&A, 29, 581
Young, J. S., et al. 1995, ApJS, 98, 219
Zaritsky, D., Kennicutt, R. C., Jr., & Huchra, J. P. 1994, ApJ, 420, 87
Zhang, X. 1999, ApJ, 518, 613