A collection of anaerobically induced gene fusions.were isolated, and representative isolates were characterized with respect to their regulatory properties, phenotypes, and approximate map locations. Four fusion strains that had defects in the anaerobic metabolism of asparagine or aspartate were found. These fusions were all repressed by alternate electron acceptors, ammonia, and glucose but were induced by other sugars. Several other fusion strains which demonstrated no observable phenotype showed diverse regulatory responses. The anaerobically induced fusions were scattered around the Escherichia coli chromosome more or less at random, suggesting that aUl the isolates examined were in separate genes.The natural habitat ofEscherichia coli is the large intestine of humans and other animals, from which E. coli emerged to colonize the scientific laboratory about half a century ago. Although the exact nature of the nutrients used by E. coli in the Wild remains obscure (20), it is clear that growth normally occurs in the absence of oxygen and that energy must be supplied either by fermentation or by the anaerobic respiration of alternative electron acceptors such as nitrate (15), fumarate (13, 17), trimethylamine oxide (TMAO) (23,32), or dimethyl sulfoxide (DMSO) (2). Work on E. coli and Salmonella typhimurium has established a large group of genes which are.involved in anaerobic respiration (13,15,31). These genes are repressed by oxygen (26, 31) and require the presence of the appropriate electron acceptor as an inducer (30). In addition, the product of thefnr gene (oxrA in S. typhimurium) appears to be an activator protein, necessary for transcription of many anaerobically regulated genes, including those involved in anaerobic respiration (7,27,28) and several others such as pepT of S. typhimurium (16,31 We previously demonstrated by using gene fusions that E. coli has approximately 50 anaerobically regulated genes, mnost of which are unmapped and biochemically uncharacterized (10). Our first collection of anaerobic gene fusions was made with the original Mu dl phage of Casadaban (3) and proved to be too unstable for convenient further characterization. We therefore made a second collection of anaerobically induced gene fusions by using the modified phage, Mu