2.6.3. Dwarf Galaxies and Low Surface Brightness Galaxies
The fainter end of the galaxy luminosity function has been a matter of debate for some time, but recent work with larger volume surveys (Lin et al. 1996; Zucca et al. 1997) indicates the luminosity function is close to flat for at least 5 mag down from L* for the local field galaxies. There is some evidence for a rise of the fainter tail right after the shoulder of L* (Ellis et al. 1996), but this is visible only in a high z sample selected in the blue band, for which small galaxies show extraordinary star formation activity without contributing much to the baryon budget. This means the contribution of dwarf galaxies to the baryon budget is small, and the estimate from the local luminosity density likely is more secure. A rise in the luminosity function somewhat below L* for cluster members also has been reported. (For the most recent literature see Phillipps et al. 1997). The contribution to the mass density from these cluster dwarf galaxies cannot be dominant unless this sharp rise continues to very low mass, which is not likely since it would violate measurements of cluster surface brightnesses.
The review by Bothun, Impey, & McGaugh (1997) indicates the importance of galaxies with surface brightnesses lower than the more readily detectable ``normal'' galaxies. It is not clear however that low surface brightness (LSB) galaxies make a significant contribution to the baryon budget. LSB galaxies contribute to the budget by their stars and star remnants, neutral gas, and plasma. The 21-cm observations by Briggs (1997) indicate there is not a significant contribution to the mean density of atomic hydrogen from gas-rich LSB galaxies in the field at distances 10h-1 Mpc. A direct constraint on diffuse plasma around LSB galaxies is much more difficult, but since LSB galaxies avoid the voids defined by the more visible galaxies this is the problem of counting the plasma concentrated around galaxies outside the great clusters, as summarized in lines 7a, 7b, and 7'. As we have noted, diffuse baryons initially associated with LSB galaxies that end up as cluster members would be counted in the X-ray measurements of (HII)cl (eq. ) and in the scaled value for the plasma in the field (eq.  and line 7' in Table 3). The mean luminosity density in the LSB sample considered by Sprayberry et al. (1997) is 15% of our adopted value (eq. ). The mean luminosity density from all stars not shrouded by dust, including those in systems with luminosities or surface brightnesses below detection thresholds, is constrained by measurements of the extragalactic contribution to the sky surface brightness. Absolute measurements in progress (Bernstein 1997) are capable of reaching ~ 3 (eq. ), but differential measurements of background fluctuations already constrain plausible new populations to contribute much less than (Dalcanton et al. 1997, Vogeley 1997). We cannot exclude the possibility that there is a significant mass in brown dwarfs or baryonic MACHOs in LSB galaxies, but we can note that since many of these galaxies seem to be in early states of evolution they would not seem to have had much opportunity to have sequestered mass in dark stars, and that the integrated baryon content in present-day LSB galaxies likely is counted in the measures of diffuse gas at redshift z ~ 3, as discussed next.