THE SEARCH FOR PRIMEVAL GALAXIES

Isolated examples of young star-forming galaxies have been discovered, but the predicted widespread population of primeval galaxies (estimated to have surface densities in the range 10(4-5) deg-2) has so far escaped detection. The search for this extensive (and, so far, elusive) population concerns the formation of spheroidal systems (bulges and halos), since there is considerable evidence that disk formation was more quiescent. The review begins with a summary of the expected model independent properties of primeval galaxies. Particular attention is paid to the redshift of galaxy formation; empirical evidence is presented that supports galaxy formation at z < 5; however, there exist persuasive arguments that at least some galaxy formation must have taken place at higher redshifts. A major part of this review concerns primeval-galaxy search strategies. Most (although not all) constraints on the properties of primeval galaxies come from searches for Lyman-alpha-emitting objects. Flux limits for these searches have now reached levels at which surveys should, according to simple models, have detected a total of 10(1-3) objects, yet no really good candidates have been found. This discrepancy between models and observations can possibly be explained by a combination of (i) dust absorption, and (ii) finite angular extent, which degrades flux limits. Technological advances in submillimeter instrumentation should result in the detection of the far-IR flux of primeval galaxies in the next few years.