Administration of naltrexone or the long-acting morphine antagonist chlornaltrexamine before infertile mating had no effect on the length of the resulting pseudopregnancy in mice. Naltrexone in doses of 10 to 200 mg/kg s.c. given on Days 2 or 3 of pregnancy showed no consistent effects on the maintenance of pregnancy. Multiple doses or intracerebroventricular administration of naltrexone also had no effect. Chronic infusion of naltrexone, provided by mini-osmotic pumps, from Day 1 of pregnancy had no effect on the incidence of pregnancy or the number of embryos implanted. These results suggest that endogenous opioids do not play a critical role in this prolactin-dependent physiological process.