The objectives of this study were to determine the optimal time of insemination in the pre-ovulatory period (from 32 to 0 h before ovulation) and to evaluate once-daily versus twice-daily inseminations in gilts. In Experiment 1, pre-puberal gilts ( n = 102) were observed for estrus every 8 h and ultrasonography was performed every 8 h from the onset of estrus to confirmation of ovulation. The gilts were inseminated once with 4 × 10 9 spermatozoa at various intervals prior to ovulation. Pregnancy detection was conducted 24 days after AI and gilts were slaughtered 4–6 days later. Corpora lutea and the number of viable embryos were counted and the embryo recovery rate was calculated (based on the percentage of corpora lutea). Inseminations performed <24 h before ovulation resulted in a higher embryo recovery rate ( P = 0.02) and produced 2.1 more embryos ( P = 0.01) than inseminations ≥24 h before ovulation. However, the pregnancy rate was reduced when inseminations were performed >16 h before ovulation ( P = 0.08). In Experiment 2, pre-puberal gilts ( n = 105) were observed for estrus every 12 h and ultrasonography was performed every 12 h from the onset of estrus to confirmation of ovulation. Gilts were inseminated (with 4 × 10 9 spermatozoa) 12 h after the onset of estrus, with inseminations repeated either every 12 h (twice-daily) or 24 h (once-daily) during estrus. The gilts were allowed to farrow. There were no differences (between gilts bred twice-daily versus once-daily) for return to estrus rate ( P = 0.36) and adjusted farrowing rate ( P = 0.19). However, gilts inseminated once-daily had 1.2 piglets less than those inseminated twice-daily ( P = 0.09). In conclusion, gilts should be inseminated up to 16 h before ovulation, as intervals >16 h reduced pregnancy rate and litter size.