Studies of locomotor performance have contributed to the elucidation of how suborganismal traits ultimately relate to fitness. In terrestrial populations, exploring the physiological and environmental contributions to whole-animal performance measures has improved our understanding of phenotypic selection. Conversely, very little is known about the links between phenotypic selection and swimming abilities in fish. Most research on swimming performance in fish has focused on morphological, physiological, and biochemical traits contributing to performance or has used swimming performance as a measure of environmental suitability. Few studies have explored how swimming performance is integrated with life-history traits or contributes to Darwinian fitness. In addition, while there are many studies on how the environment influences the swimming performance of fish, few have been done at the individual level. The objective of this study was to broaden our understanding of the relevance of fish swimming performance studies by testing the hypothesis that swimming performance (endurance and sprint) is ontogenetically and temporally stable across fluctuating environmental conditions. We found that individual sprint performances recorded at 12 degrees C were significantly repeatable after a 4-wk acclimation to 22 degrees C, although relative sprint performance of fish that survived 6 mo of natural conditions in a mesocosm was not significantly repeatable. Endurance swimming performance, as measured by critical swimming speed (U(crit)) before and after the 6-mo exposure to simulated natural conditions, was significantly repeatable within survivors. Relative sprint and critical swimming performances were not significantly related to each other. We concluded that within a time frame of up to 6 mo, the swimming performances of individual bass are ontogenetically nearly stable (sprint) to stable (endurance) despite large fluctuations in environmental conditions. Moreover, because they rely on different physiological performance traits, critical swimming and sprinting follow different patterns of change. This observation suggests the absence of a trade-off between these two swimming modes and introduces the possibly of independent selection trajectories.