Our purpose was to determine whether estradiol cypionate (ECP) could be substituted for the second GnRH injection of the standard Ovsynch protocol (injection of GnRH given 7 d before and 48 h after PGF(2alpha), with timed AI [TAI] 12 to 20 h after the second GnRH injection). Lactating dairy cows ranging from 61 to 82 d in milk at TAI were studied in 14 replicates. Main effects were hormone (ECP vs. GnRH) to induce ovulation and exposure to progesterone (P4) or not during the week preceding PGF(2alpha)-induced luteolysis. Four treatments were: 1) 100 microg of GnRH at 48 h after PGF(2alpha) (Ovsynch; n = 27); 2) same as Ovsynch, plus a P4-releasing intravaginal insert (CIDR) placed for 7 d beginning at the first GnRH injection (Ovsynch + CIDR, n = 20); 3) same as Ovsynch, but substituting 1 mg of ECP for GnRH, and injecting ECP at 24 h after PGF(2alpha) (Heatsynch; n = 33); or 4) Heatsynch + CIDR (n = 26). The largest follicle was identified by ultrasonography 24 h after PGF(2alpha) and was monitored every 6 h until ovulation. Incidence of estrus was less after GnRH (54%) than after ECP (87%), but more GnRH-treated cows had LH surges detected (95 vs. 65%) and ovulated (100 vs. 86%). Duration of LH surges, but not peak concentrations, was less after GnRH than after ECP (6.1 +/- 0.7 vs. 12.2 +/- 0.9 h). Pre-treatment with P4 reduced the incidence of LH surges but had no effects on incidence of estrus or ovulation. Intervals to the LH surge and ovulation were less after GnRH than after ECP, but intervals between onset of the LH surge and ovulation did not differ (26 +/- 2 vs. 30 +/- 3 h). We concluded that substituting ECP for GnRH resulted in more cows in estrus and slightly fewer ovulating.