The goal of this investigation was to determine the percentage of male and female bovine embryos produced after oocyte fertilization with Percoll density gradient centrifugation or with self-migration (swim-up) selected semen. In Experiment 1, sperm selection was performed by 90-45% discontinuous Percoll density gradient centrifugation (T1) and swim-up (T2). In Experiment 2, in addition to the discontinuous gradient, a 67.5% continuous gradient, and 5 or 10 min centrifugation were used. Four treatment groups were defined (TI = continuous, 5 min, TII = discontinuous, 5 min, TIII = continuous, 10 min and TIV = discontinuous, 10 min). Embryos were sexed using PCR. In Experiment 1, Percoll density gradient centrifugation (n = 185) resulted in 48.6% (n = 90) male and 51.4% (n = 95) female embryos, and swim-up (n = 142) in 58.4% (n = 83) male and 41.6% (n = 59) female embryos. In Experiment 2, the percentages of male and female embryos obtained in TI (n = 93), TII (n = 70), TIII (n = 82) and TIV (n = 82) were 49.5% (n = 46) and 50.5% (n = 47), 57.1% (n = 40) and 42.9% (n = 30), 36.6% (n = 30) and 63.4% (n = 52) and 48.8% (n = 40) and 51.2% (n = 42), respectively. Swim-up (T2) and continuous Percoll gradient centrifugation for 10 min (TIII) showed a deviation toward males (P = 0.044) and females (P = 0.015), respectively, when compared to the 50% expected percentage of each gender. There was no difference in male and female percentages in the other treatment groups from Experiments 1 and 2, when compared to the 50% expected percentage of each gender neither between treatments.