Field evidence from underground storage tank sites where leaded gasoline leaked indicates the lead scavengers 1,2-dibromoethane (ethylene dibromide, or EDB) and 1,2-dichloroethane (1,2-DCA) may be present in groundwater at levels that pose unacceptable risk. These compounds are seldom tested for at UST sites. Although dehalogenation of EDB and 1,2-DCA is well established, the effect of fuel hydrocarbons on their biodegradability under anaerobic conditions is poorly understood. Microcosms (2 L glass bottles) were prepared with soil and groundwater from a UST site in Clemson, South Carolina, using samples collected from the source (containing residual fuel) and less contaminated downgradient areas. Anaerobic biodegradation of EDB occurred in microcosms simulating natural attenuation, but was more extensive and predictable in treatments biostimulated with lactate. In the downgradient biostimulated microcosms, EDB decreased below its maximum contaminant level (MCL) (0.05 microg/L) at a first order rate of 9.4 +/- 0.2 yr(-1). The pathway for EDB dehalogenation proceeded mainly by dihaloelimination to ethene in the source microcosms, while sequential hydrogenolysis to bromoethane and ethane was predominant in the downgradient treatments. Biodegradation of EDB in the source microcosms was confirmed by carbon specific isotope analysis, with a delta13C enrichment factor of -5.6 per thousand. The highest levels of EDB removal occurred in microcosms that produced the highest amounts of methane. Extensive biodegradation of benzene, ethylbenzene, toluene and ortho-xylene was also observed in the source and downgradient area microcosms. In contrast, biodegradation of 1,2-DCA proceeded at a considerably slower rate than EDB, with no response to lactate additions. The slower biodegradation rates for 1,2-DCA agree with field observations and indicate that even if EDB is removed to below its MCL, 1,2-DCA may persist.