Kormendy & Bender (2012) propose a parallel sequence galaxy classification (Figure 1 here) in which Sph galaxies appear as bulgeless S0s juxtaposed with Im galaxies. In reality, their progenitors can include late-type spirals, but the Fig. 1 tuning fork is designed for simplicity. S0+Sph galaxies are suggested to be star-formation-quenched descendants of S+Im galaxies. It seems essentially guaranteed that all transformation processes discussed in Section 3 are important. Moreover, although there is good agreement between structural parameter correlations for (1) present-day Sphs+S0 disks and (2) present-day Im+S disks, it is of course not guaranteed that the progenitors of Sph and S0 galaxies were exactly like present-day late-type galaxies. The latter are, after all, survivors. The reasons can partly involve stochastic evolution, but they may also involve differences in internal structure and/or environment.
These ideas are slowly gaining acceptance. An observationally biased but still incomplete list of papers includes Grebel et al. (2003); van Zee et al.(2004); Mayer et al. (2006); Boselli et al. (2008a, b); Tolstoy et al. (2009); Rys et al. (2013, 2014); and Janz et al. (2013). KFCB and Kormendy & Bender (2012) address remaining controversies.
From the perspective of this conference, the important - and robust! - conclusion is this:
Dwarf spiral, irregular, and spheroidal galaxies are not new and special kinds of galaxies, as has often been suggested. Rather: There is a complete continuity in structural parameter scaling relations (Figures 2 - 4 and 6), kinematics and dynamics, star formation properties, and metallicity distributions between the disks of giant galaxies and all three subsets of dwarf galaxies. Fainter than MV ~ -18, the properties of galaxies change as their gravitational potential wells get more shallow; this is qualitatively consistent with the increased importance at lower masses of internal and environmental transformation processes. Also, many aspects of galaxy evolution get more stochastic in smaller galaxies. For example, star formation gets more bursty. Many quantitative details remain to be worked out. But this is engineering. A big-picture view of the evolution of dwarf galaxies seems comfortably in place:
All forms of violence - including supernova energy feedback, dynamical harassment, and external ram-pressure stripping - get more important for smaller galaxies. The main result is that smaller galaxies lose more of their baryons or never acquire them. Smaller dwarfs are more dominated by dark matter. Whether or not they retain enough cold gas to feed some star formation is very much a second-order effect. Baryon loss at low halo masses may be so severe that we discover only a small fraction of the smallest, essentially dark halos. This can reconcile the observed scarcity of dwarf galaxies in field environments with our expectations of large numbers of dwarfs that are predicted by the cold dark matter fluctuation spectrum.
It is a pleasure to thank Ralf Bender and Ken Freeman for their hospitality and support during my visits and for always enjoyable collaborations. My attendance at the Seychelles conference was supported by the Max-Planck-Institute for Extraterrestrial Physics and by the Observatory of the Ludwig-Maximilians-University, Munich. This paper was supported by the Curtis T. Vaughan, Jr. Centennial Chair in Astronomy at the University of Texas.