Is the ultimate evolutionary product of disk-disk mergers an elliptical with fine structure? Here again H I observations provided evidence for unexpected merger products. In particular, a number of on-going mergers and merger remnants are found to have large gaseous disks with rotational kinematics. Particularly good examples are Arp 230 (Fig. 2c & d), NGC 520 (Hibbard & van Gorkom 1996), and MCG -5-7-1 (Schiminovich, van Gorkom & van der Hulst in preparation). The very faint loops and streamers imaged around normal disk galaxies (Malin & Haley 1997, Zheng et al. 1999) support the idea that some disk systems may have had a violent origin or experienced a major accretion event.
Finally, there are some systems which simply do not seem to conform to the standard interaction picture. One such example is Mrk 348 (Fig. 3). The main difficulty with the tidal interpretation for this system is that the scale of the H I is tremendous (diameter ~ 280 kpc), and two thirds of the neutral hydrogen (1.4 × 1010 M out of a total of 2.1 × 1010 M) lies outside of the highest contour in Fig. 3, i.e., outside the region containing both the companion and all of the optical light of the disk. It simply does not seem possible that the small companion seen in Fig. 3b could have raised this much material to such large radii. It may be that the progenitor was a very gas-rich low surface brightness galaxy like Malin 1 (Impey & Bothun 1989, Pickering et al. 1997). A more intriguing possibility is that the neutral gas may have condensed out of an extensive halo of ionized gas. In this regard it is interesting to consider the NGC 4532 / DDO 137 system, which has a very irregular distribution of H I lying mostly outside of the optical galaxies. Hoffman et al. (1999) suggest that the H I clumps are simply neutral peaks in sea of mostly ionized hydrogen. The existence of such a sea of baryons may mean that full scale galaxy formation continues to the current epoch.