Over the past few years, some evolutionary scenarios have been proposed in order to place the BCDGs and the other types of dwarf galaxies, dEs and dIs, into an unified scheme. Lin & Faber (1983), Davies & Phillipps (1988) have proposed in view of the morphologies and surface brightness distributions that dIs, and dEs might go through a BCD phase. This star formation phase can happen in dE providing that an important gas refueling has occured. The recent work of Sung et al. (1998), based on the comparison of the intrinsic shapes of BCDGs and dEs, dIs, shows that they present morphological similarities. Van Zee et al. (1998) show that HI properties differ largely from one type to another, with some BCDGs showing evidence of ongoing mergers (e.g. II ZW 40). BCDGs have more compact HI clouds, less clumpy than dIs. Finally, they show that the HI within BCDG is rotationally supported, with velocity gradients indicating, however, more complex kinematics, while dEs are most probably not.
The main difficulty of classifying the BCDGs in an evolutionary sequence with both dEs and dIs, and to some extension with the Low Surface Brightness dwarf galaxies (LSB, hereafter), will strongly depend on what component defines the main body of the BCDGs. From dynamical considerations, stars dynamics can be considered completely disconnected from the gas dynamics. Finding a rotating gaseous disk inside a galaxy does not make it a "disk galaxy'' since gaseous compact disks, with some star formation, are also found in giants elliptical galaxies ([Brighenti & Mathews1997]; [Silchenko1997]; [Miller et al.1997]).
Apparently, depending on the tracers used for pinning down the dynamics and evolution of BCDGs, we might end-up with different answers: HI and young stars studies make them similar to low surface brightness HII galaxies, while the morphological studies lead to connect them with dEs or dIs.
Another aspect of the evolution is the environement: dEs and dIs are found mostly in cluster of galaxies, usually associated with a larger elliptical or a spiral. BCDGs and LSBs are preferentially found at the very edge of those clusters or in the fields, isolated.
During our work, we found that the 44 objects, we observed, could be classified into 3 classes: BCDGs whose surface brightness profiles are best described by an r1/4 law, others by an exponential law, and a significant fraction for which it is not clear which of these laws dominates. This classification is somewhat independent of the star formation regions, since we took great care of fitting the profiles away from the light excess due to the starburst. If the light profiles reflect their general dynamics, BCDGs may represent a class of its own, with the LSBs as possible progenitors.
Copyright The European Southern Observatory (ESO)