In vitro fertilization in horses has had limited and inconsistent success for oocytes matured in vivo or in vitro. However, oocytes transferred to the oviducts of inseminated recipient mares are consistently fertilized and result in normal embryo development. Potentially, equine sperm are unable to penetrate the zona pellucida (ZP) in vitro, which could be a result of failure of capacitation or hyperactivation. We hypothesised that ZP of equine oocytes impairs fertilization in vitro and bypassing the ZP would be beneficial for fertilization rates. Oocytes were retrieved using transvaginal follicle aspirations from oestrous mares 22 to 24 h after administration of a gonadotropin-releasing hormone analog (deslorelin, 1.1 mg). After culture for 20 h in TCM-199 with Earle's salts and 10% FCS, oocytes were assigned to 1 of 3 groups: 1) IVF, equine (n = 9) and bovine (n = 80) cumulus–oocyte complexes co-incubated with sperm; 2) PVI, sperm injections into the perivitelline space of equine (n = 13) and bovine (n = 11) oocytes; and 3) ICSI, intracytoplasmic sperm injections of equine oocytes (n = 13). Fresh sperm from 2 stallions was used. For IVF, sperm were treated with 20 μM dilauroyl phosphatidylcholine (PC12) to induce capacitation but not the acrosome reaction; sperm for PVI and ICSI were treated with 40 μM PC12 to capacitate and acrosome-reacted sperm. For IVF, each equine oocyte was co-incubated with 8 to 11 bovine oocytes and 250 000 sperm in fertilization media (De La Torre-Sanchez et al. 2006 Reprod. Fertil. Dev. 18, 597–607) for 18 h. For PVI, 10 to 15 sperm were injected into the perivitelline space of each equine and bovine oocyte and for ICSI, 1 sperm was injected into the cytoplasm of each equine oocyte. Oocytes were then placed in embryo culture medium (DMEM-F12 and 10% FCS) for 24 h before examination for cleavage. Cleavage rates for each group were compared using chi-square analysis. After IVF or PVI, no equine oocytes cleaved; in comparison, 11 of 13 (85%) cleaved in the ICSI group. However, of the bovine oocytes, 10 of 80 (12%) cleaved after IVF and 1 of 11 (9%) cleaved after PVI with equine sperm. An additional 6 PVI injections with equine (n = 6) oocytes and sperm were stained with Hoechst 33324 to determine sperm localization within the oocytes. All injected sperm remained in the perivitelline space and did not appear to fuse with the oolemma or enter the ooplasm. The results of this study suggest that the equine sperm prepared with PC12 were unable to penetrate the oolemma. In our experiment, sperm treated with PC12 resulted in high fertilization rates when placed directly into the ooplasm during ICSI; however, the sperm were unable to penetrate the ZP or oolemma. Additional research is needed to determine the cause of penetration failure and whether this is due to inadequate sperm treatments or a missing oviducal component in our in vitro systems. Funding for this project was provided by the benefactors for the Preservation of Equine Genetics and the Cecil and Irene Hylton Foundation.