The pharmacokinetics of [1,2-14C]ethylene glycol (EG) were evaluated in female Sprague-Dawley rats and CD-1 mice to characterize the plasma disposition after intravenous (IV), peroral (PO), and percutaneous (PC) doses. Rats were given doses of 10 or 1000 mg/kg by each route, and additional PO doses of 400, 600, or 800 mg/kg. Mice were also given IV and PO (bolus gavage) doses of 10 or 1000 mg/kg, and additional PO doses of 100, 200, or 400 mg/kg. PC doses in mice were 100 or 1000 mg/kg, and both species were given a 1000 mg/kg PC dose with a 50% (w/w) aqueous solution (2 ml/kg) to simulate antifreeze exposure. Results from this study have shown that orally-administered EG is very rapidly and almost completely absorbed in both rats and mice, with a bioavailable fraction of 92-100% in rats and similar percentages at the higher doses in mice. In contrast, the absorption of cutaneously applied EG is comparatively slow in both species. A species difference in the overall absorption of PC doses was demonstrated, with higher recoveries of 14C observed after PC doses in mice than for rats and a greater penetration of 14C after applying a 50% aqueous PC dose in mice than in rats, as evidenced by quantifiable plasma 14C concentrations only in mice. The major metabolites in both rats and mice are CO2 and glycolate. Oxidative metabolic pathways are saturated at high PO doses in both species, resulting in a shift from exhaled CO2 as the major excretion route to urinary excretion. The capacity to metabolize more completely EG to CO2 at low doses seems to be greater in the mouse than in the rat, as evidenced by the absence of urinary oxalate from EG-dosed female mice, and saturation of metabolic pathways at a comparatively lower dose in mice than for rats. This evidence suggests that dose-dependent changes in EG excretion in female Sprague-Dawley rats and CD-1 mice probably resulted from capacity-limited effects on EG metabolic pathways for the production of CO2 and a compensatory urine clearance of glycolate. Results from the present study corroborate previous observations in rats for the lower doses, but demonstrate a substantial difference in single-dose pharmacokinetics for IV and PO 1000 mg/kg doses in mice vs. rats. In summary, these data indicate that mice show a nonlinear plasma disposition of total radioactivity (EG and its metabolites) as dose is increased, whereas in rats plasma kinetics were linear over the dose range evaluated, whereas excretion kinetic patterns were nonlinear in both species as dose is increased.