Progesterone and luteal blood flow effects of an i.u. 2-h infusion of 0.25 mg/h of prostaglandin F 2a (PGF) that simulated a natural pulse of 13,14-dihydro-15-keto-PGF (PGFM) were compared to the effects of a single bolus i.u. injection of PGF (4 mg) that induced complete luteolysis in heifers. Blood sampling and an estimate of the percentage of luteal area with colour-Doppler signals of blood flow were performed every 2 min for 20 min and less frequently thereafter for 6 h. After the beginning of PGF infusion or a bolus injection, progesterone increased to a peak at 14 and 10 min respectively, and was accompanied by an increase in blood flow in the bolus group but not in the infusion group. Progesterone then decreased for 1 or 2 h and was accompanied by a continued elevation in blood flow in the PGF bolus group and by a slight increase in the PGF infusion group. Progesterone then rebounded in both groups, but the rebound was greater in the infusion group. Blood flow decreased during the descending arm of the progesterone rebound. Cortisol and prolactin began to increase 6 min after the bolus PGF injection but did not increase during or after PGF infusion. The increases in cortisol, prolactin and blood flow after a PGF bolus treatment but not during a simulated PGFM pulse indicated that the bolus treatment was pharmacologic, and its use may lead to faulty conclusions on the nature of physiologic luteolysis. The comparisons between progesterone and blood flow are novel.