Smaller fish body sizes are often believed to be a universal response to warming. Many experiments show that fish reared in warmer conditions grow faster as juveniles but reach smaller adult size. Yet, despite decades of research it is still not clear why warmed fish are smaller and how commonly this applies to natural populations. In this talk I will present our research on temperature and mean fish species body sizes through space and time, explore the widespread assumption that warming leads to faster metabolic rates and present examples of temperature impacts on fish growth in inter-generational field and laboratory experiments. I conclude that fish body sizes vary extensively across its distribution range and that describing a species with one general attribute, such as maximum body size, is likely to be misleading.