Mitochondrial DNA heteroplasmy using the protein-coding cytochrome b (Mtcyb) gene was assessed in laboratory mice (C57BL/6 and BALB/c) exposed to the Chernobyl environment. Subacute to subchronic (30-40 days) exposure resulted in a cumulative radiation dose of 1.2-1.6 Gy ( approximately 0.04 Gy/day). Mice were sampled prior to introduction into the enclosures and again after removal from the enclosures. Nucleotide variation (site heteroplasmy) in 306 pre-exposure Mtcyb gene copies (122400 base pairs) was compared to variation in 354 postexposure gene copies (141600 base pairs). Five mutant copies, each characterized by a single nucleotide substitution, were observed (four in the pre-exposure samples, one in a postexposure sample). The frequencies of mutant gene copies and nucleotide substitutions in pre-exposure and postexposure samples were not significantly different. This suggests that this type of exposure (i.e. low dose rate) does not pose a significant mutation risk to the Mtcyb gene in digit tissue. Furthermore, no significant radiation risk to analogous human tissues may exist when occupational exposures involve low dose rates such as these. Finally, linear, cumulative models of genetic risk currently used to estimate radiation-induced effects are likely to be inappropriate for low-dose-rate exposures and need to be re-evaluated critically.