4.3. Embedded Disks - II. Flat Pseudobulges

That some pseudobulges are essentially as flat as disks is inferred when we observe spiral structure (Figure 5), but it is observed directly in surface photometry of highly inclined galaxies (see Kormendy 1993 and Kormendy & Kennicutt 2004 for reviews). Figures 7 - 9 show examples.

Figure 7. NGC 4371 pseudobulge - top image: 130" × 80" WFPC2 F606W image from the HST archive; middle and bottom: 406" × 249" CFHT 12K R-band images from Kormendy et al. (2004) at different logarithmic intensity stretches. North is up and east is at left. The plots show surface photometry, including brightness cuts along the major and bar axes (see the text) shifted to the V-band INT zeropoint derived using aperture photometry from Poulain (1988).

The SB0 galaxy NGC 4371 contains one of the most rotation-dominated "bulges" in Figure 6. This result (Kormendy 1982b, 1993) already implies that NGC 4371 contains a pseudobulge.

Photometry strengthens the evidence that the "bulge" is disky. Kormendy (1979) concluded that "the spheroid is distorted into a secondary bar; i. e., is prolate". Wozniak et al. (1995) saw this, too, but noted that it could be a projection effect due to high inclination. Based on unsharp-masked images, Erwin & Sparke (1999) conclude that NGC 4371 contains a smooth nuclear ring and identify this - in effect, if not in name - as a pseudobulge.

Figure 7 shows our photometry. The ellipticities and position angles PA are based on ellipse fits to the isophotes. However, the isophotes are far from elliptical at some radii, so the bottom two panels show brightness cuts in 25° wedges along the major and bar axes. The bar is obvious as a shelf in surface brightness and as corresponding features in the and PA profiles. Interior to the outer exponential disk shown by the major-axis cut is a steep central rise in surface brightness that would conventionally be identified as the bulge. However, its properties are distinctly not bulge-like. It contains a shelf in surface brightness with radius r 10"; this is shown in more detail in Figure 8. The outer rim of a shelf looks like a ring when an image is divided by a smoothed version of itself. The shelf has the brightness profile of a lens (cf. the prototype in NGC 1553: Freeman 1975; Kormendy 1984). We interpret it as a nuclear lens. The important point is this: The nuclear lens has essentially the same apparent flattening and position angle as the outermost disk. We cannot tell from Figure 7 whether it really is a disk or whether it is thicker than a disk and therefore prolate (a nuclear bar). Rapid rotation (Figure 6) makes the disk interpretation more likely. In either case, the nuclear lens is diagnostic of a pseudobulge (see Kormendy & Kennicutt 2004 for further discussion). Photometric criteria (Figures 7 and 8) and dynamical criteria (Figure 6) for identifying pseudobulges agree very well in NGC 4371.

Figure 8. NGC 4371 nuclear lens. North is up and east is at left. The HST PC F606W image at the top is 28."7 × 17."6. The intensity stretch is logarithmic. The bottom panel shows isophotes chosen to distinguish the high-ellipticity lens from the rounder center. The contour levels are 19.6, 19.1, 18.6, 18.0, 17.6, 17.3, 16.8, and 16.3 V mag arcsec-2. Compare the lens in NGC 1553 (Freeman 1975; Kormendy 1984).

The same is true in NGC 3945 (Figure 9). As in NGC 4371, the bar of this SB0 galaxy is oriented almost along the apparent minor axis. Therefore, when the "bulge" at r 10" looks essentially as flat as the outermost disk, there is ambiguity about whether the inner structure is flat and circular or axially thick and a nuclear bar. It was interpreted as a nuclear bar in Kormendy (1979) and in Wozniak et al. (1995) and is illustrated as such in Kormendy & Kennicutt (2004). In contrast, Erwin & Sparke (1999) interpret it as "probably intrinsically round and flat - an inner disk". Erwin et al. (2003) reach the same conclusion in a detailed photometric study. Based on and PA profiles, they also identify an "inner bar" with radius 2". All of these features are well confirmed by our photometry (Figure 9). The main and "inner" bars are clear in the and PA profiles. The main bar also makes an obvious shelf in the bar-axis brightness cut, while the nuclear bar is so subtle that we regard it as an interpretation rather than a certainty. Our photometry is consistent with the interpretation either that the shelf at r 10" in the major-axis cut is a nuclear bar (in which case the galaxy has three nested bars) or that this is a nuclear lens which is nearly circular and very flat. For the purposes of this paper, we do not have to decide between these alternatives. Either one is characteristic of a disk. Consistent with the conclusions of all of the above papers, either interpretation implies that the central rise in surface brightness above the galaxy's primary lens and outer ring is caused by a pseudobulge. This may have been added to a pre-existing classical bulge, but if so, the classical bulge dominates the light only in the central 1."5 (Figure 9; Erwin et al. 2003).

Figure 9. NGC 3945 pseudobulge - top image: 29" × 29" PC F450W image from the HST archive; middle and bottom: 6.'1 × 6.'1 WIYN B-band images from Buta et al. (2004) at different logarithmic intensity stretches. North is up and east is at left. The plots show ellipticity and position angle profiles from ellipse fits to the isophotes and major- and bar-axis cuts in 28° wedges. The McDonald Observatory 0.8 m telescope V-band images were zeropointed using aperture photometry from Burstein et al. (1987) and from Angione (1988). The other profiles are shifted to this zeropoint. Outer rings are usually elongated perpendicular to bars, so the apparent ellipticity of flat, circular isophotes is likely to be the one observed at the largest radii.

Present-day gas inflow and star-formation rates imply that dissipative secular evolution should be most important in Sbc - Sc galaxies (Kormendy & Kennicutt 2004). Classical bulges are the rule in Sas, but it is remarkable how easily one can find S0s with pseudobulges. We interpret this result as additional evidence for van den Bergh`s (1976) "parallel sequence" classification, which recognizes that some S0s have smaller (pseudo)bulge-to-disk luminosity ratios than do Sa galaxies. The hint is that the secular evolution happened long ago, when the galaxies were gas-rich and before they were converted to S0s.