This study was conducted to investigate whether the dose of estradiol (E) administered acutely, or chronic delivery of one dose of E impacts acquisition and subsequent cocaine self-administration in ovariectomized (OVX) female rats. Five groups of female rats were compared: OVX females treated with 0, 1, 2, or 5 µg 17 β-E, 30 min prior to the self-administration session, and OVX rats that received a 1.5 mg E pellet (designed to chronically release 25 µg E/d X 60 d) implanted 1 week before cocaine self-administration initiation. Rats were tested in 1 hr sessions on a FR1 schedule with the dose of cocaine increasing every week (testing occurred 5 day/wk; doses: 0.2, 0.3, 0.4, 0.5 and 0.75 mg/kg/ infusion). We report that OVX rats treated with 2 µg E acquired self-administration more rapidly than all of the other groups, and animals that received 1 or 2 µg E self-administered significantly more cocaine compared to OVX+vehicle at 0.3 and 0.4 mg/kg/infusion. In contrast, OVX rats given 5 µg E acutely, or chronic E via slow-release pellets did not take more cocaine than the OVX+vehicle group at any time point. Physiological serum concentrations of E were seen with 1 or 2 µg E, but 5 µg E and the E pellet produced supra-physiological concentrations. These results suggest an inverted U-shaped dose-response curve for the effect of E on acquisition of cocaine self-administration.