Here we discuss the sub-samples of each of the Hubble types separately. The light-profiles of all types will be presented in a subsequent paper in conjunction with the ground-based UBVR-profiles, after all the ground-based zero-points have been secured (see Table 2).
4.1. Early-type Galaxies
The regular early-type galaxies (elliptical and S0's) in our sample show in general a significant change in SB from the mid-UV to the red, reflecting an overall lack of a young stellar population (CGCG 097 - 094, NGC 1396, NGC 4478; sections 3.01-3.03). Typically, the µF300W of E/S0's is 3-5 mag arcsec - 2 fainter than µF814W. As was also seen in the far-UV UIT sample of Marcum et al. (2001) and Kuchinski et al. (2000), they are dominated by an older stellar population, and so are generally faint in F300W. Out of 7 early-type galaxies imaged, two have small blue nuclear features (UGC 03426, NGC 3921; Section 3.05 and 3.07). UGC 03426 is a Seyfert 2, and so this feature could be associated with a bi-polar outflow. The other, NGC 3921, is a well studied merger-remnant from the Toomre sequence, and the blue nuclear feature could indicate a location of residual star-formation that is only partially obscured by dust. Three early-type galaxies (NGC 1396, NGC 3516, and UGC 08823; Section 3.02, 3.04, and 3.06) become dominated by point sources in the mid-UV, indicating weak optical-UV AGN, Seyfert, or LINER nuclei. While these are small number statistics, their presence in our current small sample of early-type galaxies is due to our selection of galaxies with high predicted average SB in the mid-UV within their effective (i.e., half-light) area (Section 2.2.3). This resulted in including a number of objects in our sample that are dominated in the U-band and mid-UV by AGN.
If AGN generally reside in bulge-dominated galaxies (e.g., Magorrian et al. 1998), then the red and old stellar population of the underlying early-type galaxies will be generally faint in the UV, but the presence of a (weak) AGN will result in blue (U - B) colors and therefore a high SB the mid-near UV, and so inclusion into UV-selected samples. Similarly, ground-based U-band selected surveys would result in significant numbers of AGN in unrecognized early-type galaxies at moderate redshifts (z 0.3). At redshifts of a few tenths, the ground-based U-band selection would similarly show these objects essentially as point sources, just like some of the nearby early-type galaxies selected in our HST mid-UV sample. However, had these objects been selected at z 0.3 from the ground at much longer wavelengths (i.e., in the I-band), then they would still have shown up as early-type galaxies in images with good ground-based seeing. This leads us to wonder to what extent the (strong) cosmological evolution of AGN selected optically or through their near-UV excess at modest redshifts (Koo & Kron 1988; Boyle & Terlevich 1998; Boyle et al. 2000) could in part be due to a "morphological K-correction" of early-type galaxies with weak AGN. A further discussion of the galaxy light underlying low redshift QSO's is given by Bahcall, Kirhakos & Schneider (1995, 1996, 1997), Kirhakos et al. (1999) and Maoz et al. (1996), who determine the AGN components seen with WFPC2 in early-mid type galaxies.
Two more galaxies, UGC 05101 (Section 3.08) and UGC 08696 (Section 3.09), are tentatively placed in this section of early-type galaxies. These objects are merger remnants and have significant dusty disks in the mid-UV at present, but we believe they will soon evolve into early-type galaxies.