5alpha-pregnan-3alpha-ol-20-one (3alpha5alpha-THP) in the ventral tegmental area (VTA) mediates lordosis of rodents. If fluoxetine's effects on lordosis are mediated in part by midbrain 3alpha,5alpha-THP, then fluoxetine regimens that decrease and increase lordosis would be expected to respectively lower and elevate midbrain 3alpha, 5alpha-THP levels. Experiment 1: Ovariectomized (ovx) rats received estradiol benzoate (EB; 5 mug, SC) at 0 and 24 h and fluoxetine (20 mg/kg, IP) or vehicle 30 min before sex testing and tissue collection. Other rats received fluoxetine (10 mg/kg, IP) or vehicle for 15 days followed by EB-priming and testing. Systemic acute or chronic fluoxetine significantly decreased lordosis and midbrain 3alpha,5alpha-THP levels compared to vehicle. Experiment 2: Ovx rats with unilateral cannula to the VTA were primed with EB (5 mug; 0, 24 h) and/or progesterone (0 or 100 mug; 44 h, SC). At 47.5 h, fluoxetine (3.6 muM) or vehicle was infused to the VTA. At 48 h, rats were tested. Administering fluoxetine to the VTA significantly increased lordosis and midbrain 3alpha,5alpha-THP levels compared to vehicle infusions. Experiment 3: Ovx EB-primed rats were tested prior to, and 30 min after, treatment with acute fluoxetine (20 mg/kg, IP). Rats were then infused with 3alpha,5alpha-THP (100 mug) or vehicle to the VTA and were retested. 3alpha,5alpha-THP, but not vehicle, to the VTA reversed acute fluoxetine's inhibitory effects on lordosis. Together, these data suggest fluoxetine may alter lordosis in part through actions of 3alpha-THP in the midbrain.