Dihaloalkanes are of toxicological interest because of their high-volume use in industry and their abilities to cause tumors in rodents, particularly dichloromethane and 1,2-dichloroethane. The brominated analogues are not used as extensively but are known to produce more toxicity in some systems. Rats and mice were treated i.p. with (14)C-dichloromethane, -dibromomethane, -1,2-dichloroethane, or -1,2-dibromoethane [5 mg (kg body weight)(-1)], and livers and kidneys were collected to rapidly isolate DNA. The DNA was digested using a procedure designed to minimize processing time, because some of the potential dihalomethane-derived DNA-glutathione (GSH) adducts are known to be unstable, and the HPLC fractions corresponding to major adduct standards were separated and analyzed for (14)C using accelerator mass spectrometry. The level of liver or kidney S-[2-(N(7)-guanyl)ethyl]GSH in rats treated with 1,2-dibromoethane was approximately 1 adduct/10(5) DNA bases; in male or female mice, the level was approximately one-half of this. The levels of 1,2-dichloroethane adducts were 10-50-fold lower. None of four known (in vitro) GSH-DNA adducts was detected at a level of >2/10(8) DNA bases from dibromomethane or dichloromethane. These results provide parameters for risk assessment of these compounds: DNA binding occurs with 1,2-dichloroethane but is considerably less than from 1,2-dibromoethane in vivo, and low exposure to dihalomethanes does not produce appreciable DNA adduct levels in rat or mouse liver and kidney of the doses used. The results may be used to address issues in human risk assessment.