The aim of this study was to elucidate if SLC after 24 h storage selects the subpopulation of spermatozoa that better withstands osmotic shock. To test this hypothesis, viability, mitochondrial membrane potential (MMP) and superoxide anion (O(2)(·-)) production of uncentrifuged (UC) and single layer centrifugation (SLC) - selected spermatozoa were analyzed following SLC after storage of the semen. An aliquot of the extended ejaculate (100×10(6) spermatozoa/mL) was centrifuged through a single layer of a silane-coated silica based colloid formulation optimized for equine spermatozoa (Androcoll-E large, SLU, Sweden) and the rest was used as control. UC and SLC-sperm samples were subjected to osmotic challenges (75 and 900 mOsm) with a subsequent return to isosmolarity (300 mOsm) using Biggers-Whitten-Whittingham (BWW) medium. Viability and MMP decreased after the different osmotic stress in UC and SLC spermatozoa, and return to isosmolarity did not reverse these effects. O(2)(·-) production was enhanced after SLC in all osmolarities tested. Interestingly, the percentage of living spermatozoa showing O(2)(·-) production was increased after 900 mOsm stress in UC spermatozoa, this increase being more evident in SLC spermatozoa. Returning spermatozoa to 300 mOsm enhanced this percentage in UC viable cells but not in SLC spermatozoa. The scenario observed for UC spermatozoa shows that O(2)(·-) is produced in response to isolated hyperosmolarities and subsequent osmotic excursions. As the viability, MMP and cell volume remained the same between SLC and UC spermatozoa, we conclude that Androcoll-E large is likely selecting a higher percentage of physiologically O(2)(·-) producing spermatozoa.