Epidemiologic evidence indicates that maternal smoking increases the risk of preterm birth. While a number of plausible mechanisms for early delivery have been offered, the role of gestational hormones in this smoke-induced outcome is uncertain. Thus, a toxicologic study was performed to examine the effects and underlying hormonal mechanisms of mainstream cigarette smoke (MCS) exposure on gestational duration. Pregnant B6C3F1 mice were exposed by inhalation to MCS for 5 days/week (4 hrs/day) from Gestational Day (GD) 4 to parturition. Smoke-induced effects on gestational length, interpubic ligament length, maternal hormone secretion patterns (estradiol-17beta, progesterone, prolactin, and relaxin), body weight gain, postimplantation loss, litter size, and offspring sex ratio were examined. Dams exposed to MCS at a concentration equivalent to smoking less than one pack of cigarettes/day (carbon monoxide = 25 parts per million, total suspended particulates = 16 mg/m3) demonstrated a significant (P < 0.05) shortening of gestational duration (compared with pregnant, air-exposed mice). In addition, MCS-exposed mice sacrificed on GD 18 had significantly (P < 0.05) increased interpubic ligament length, elevated serum estrogen levels, and a reduced progesterone to estradiol-17beta ratio (compared with air-exposed controls); levels of progesterone and prolactin were only modestly decreased and increased, respectively, in the MCS-exposed mice. Smoke exposure had no significant effects on maternal relaxin levels, body weight gain, postimplantation loss, litter size, or sex ratio. Results of this study demonstrate that inhalation exposure of pregnant mice to a low dose of MCS shortens gestation and alters hormone secretory patterns, which are important for maintaining pregnancy and inducing parturition. These findings support the view that pregnant women who smoke (even modestly) may be at increased risk for preterm birth, and that early delivery may be related (at least partly) to MCS-induced.