1.3. Where and why are the primeval galaxies hiding?

To resolve the issue of where and why are the long sought primeval galaxies hiding, direct observational evidence is critical. The difficulty is in finding large numbers of faint, and presumably distant, galaxies at or near their time of formation. Many theories of galaxy formation predict that there should exist a widespread population of primeval galaxies at a cosmic epoch somewhere before z 2. These primeval galaxies, undergoing an intense initial burst of star formation, should be detectable in several emission lines (e.g., Ly, Ha, [O-III], etc.). Exhaustive ground-based Ly searches have been made over the last 15 years for young galaxies with strong continuous star-formation at high redshift (e.g., TDT95), but have not yielded a large number of candidates for several possible reasons. First, even small amounts of dust, if properly distributed, can effectively quench Ly and UV continuum emission (CK92, K96). Second, these objects may primarily occur at much higher redshifts (z 5-10) than have thus far been explored. Third, they might exist in protoclusters or larger-scale structures at high redshifts (although the clustering amplitude is predicted to be lower at high redshifts by CDM theories), causing most narrow-band searches to look in between any such structures (see Section 3.4). And/or fourth, in a hierarchical formation scenario - in which galaxies were assembled from pieces over a long time interval - the long-sought Ly emitting primeval galaxies may be out there, but in a large number of small pieces (P96b; see also Section 3), which would be largely beyond the ground-based flux detection limits. Finding these elusive objects, or a similar population of very compact star-forming systems, may help us to improve our understanding of how exactly galaxies formed.