The pulmonary toxicity of airborne lunar dust was assessed in rats exposed by nose-only inhalation to 0, 2.1, 6.8, 20.8 and 60.6 mg/m3 of respirable size lunar dust. Rats were exposed for 6 h/d, 5 d/week, for 4 weeks (120 h). Biomarkers of toxicity were assessed in bronchial alveolar lavage fluid (BALF) collected at 1 d, 1 week, 4 weeks or 13 weeks post-exposure for a total of 76 endpoints. Benchmark dose (BMD) analysis was conducted on endpoints that appeared to be sensitive to dose. The number of endpoints that met criteria for modeling was 30. This number was composed of 13 endpoints that produced data suitable for parametric analysis and 17 that produced non-normal data. Mean BMD values determined from models generated from non-normal data were lower but not significantly different from the mean BMD of models derived from normally distributed data. Thus BMDs ranged from a minimum of 10.4 (using the average BMD from all 30 modeled endpoints) to a maximum of 16.6 (using the average BMD from the most restricted set of models). This range of BMDs yields safe exposure estimate (SEE) values of 0.6 and 0.9 mg/m3, respectively, when BMDs are extrapolated to humans, using a species factor of 3 and extrapolated from a 1-month exposure to an anticipated 6-month lunar surface exposure. This estimate is very similar to a no-observable-adverse-effect-level (NOAEL) determined from the same studies (0.4 mg/m3) and a SEE derived from a study of rats that were intratracheally instilled with lunar dusts (0.5-1.0 mg/m3).