Within an in vitro production system, bulls differ with respect to their semen potential in generating embryos when the variables of maternal effects are minimized (Marquant-le-Guienne and Humblot 1992 Ann. Zootech. 41, 361-370). We have tested the hypothesis that even with this variation among bulls, there is also an intra-bull variation among frozen sexed semen batches (Sexing Technologies, Navasota, TX, USA; personal communication) when used with IVF. In an embryo commercial production system, 5058 viable oocytes obtained by ovum pick-up with ultrasound from 193 Nelore cows (Bos indicus) over a 12-month period were matured in vitro for 24 h in TCM-199 (Gibco, Life Technologies, Carlsbad, CA, USA) containing 0.5 μg mL^-1 FSH (Bioniche Animal Health, Belleville, Ontario, Canada), 50 μg mL^-1 LH (Intervet, Boxmeer, the Netherlands) and 10% fetal bovine serum (Gibco). Mature oocytes were inseminated in vitro for 18 h in IVF-Talp medium (BSA-FAF 6 mg mL^-1; 10 ng of heparin, Sigma, France), using 3 different batches (I, II, III) of frozen-thawed sexed semen from 4 bulls (A, B, C, D), separated with a Percoll gradient (45:90; Sigma, St-Quentin Fallavier, France). Putative zygotes surrounded in cumulus cells were transferred in CR2aa medium droplets (Rosenkrans and First 1994 J. Anim. Sci. 72, 434-437) with 3 mg mL^-1 BSA-FV (Sigma) for 163 h in a humidified incubator at 39°C, with an atmosphere of 5% CO² in air. Total number of oocytes, total number of blastocysts, and embryonic developmental rates for each bull and respective batch are reported in Table 1. The chi-square test was measured with a significance level of P < 0.05 and showed that there is difference between the batches used with respect to developmental rates of blastocysts. Therefore, there is intra-bull variation in the ability to produce in vitro embryos according to the batch of frozen sexed semen.