The goal of our research is to produce an ordered set of cosmid clones for each of several species of halobacteria for use in physical and genetic mapping. These maps will answer questions about genome evolution and about gene organization and regulation in this archaebacterial lineage. Progress in cloning and mapping the genome of Halobacterium volcanii DS2 (synonym Haloferax volcanii DS2) is reported. Overlapping cosmids are recognized by a strategy which makes use of the distinctive restriction fragments around relatively rare restriction sites. Each site recognized by the infrequently cutting restriction enzymes is a landmark from which to identify different regions of the genome. The main advantage of this strategy is that only a small overlap (10–20%) between cosmid clones is required, resulting in a correspondingly small number of cosmid clones to be analyzed. The certainty of overlap is high, and computation is simple. The final 5–10% of each genome is cloned, linked, and identified by chromosome walking methods. Hybridization of cloned homologous or heterologous genes and of stable RNAs to the minimal cosmid set localizes these genes on the physical map. Additional genes have been and will be cloned by complementation of auxotrophic mutants, or as determinants of resistance to antibiotics.Key words: landmark strategy, genome mapping, archaebacteria, cosmid.