Faster bacterial biomass turnover is expected in water compared to soil, which would result in more rapid community adaption to changing environmental conditions, including temperature. Bacterial community adaptation for growth is therefore predicted to have larger seasonal amplitudes in lakes than in soil. To test this prediction, we compared the seasonal variation in temperature adaptation of bacterial community growth in a soil and lake in Southern Sweden (Tin situ 0-20°C, mean 10°C) during 1.5 years, based on monthly samplings including two winters and summers. An indicator of community adaptation, minimum temperature for growth (Tmin), was calculated from bacterial growth measurements (Leu incorporation) using the Ratkowsky model. The seasonal variation in Tmin (sinusoidal function, R2 = 0.71) was most pronounced for the lake bacterial community, with an amplitude for Tmin of 3.0°C (-4.5 to -10.5°C) compared to 0.6°C (-7 to -8°C) for the soil. Thus, Tmin in water increased by 0.32°C/degree change of Tin situ. Similar differences were also found when comparing four lakes and soils in the winter and summer (amplitudes 2.9°C and 0.9°C for lakes and soils, respectively). Thus, seasonal variation in temperature adaptation has to be taken into account in lakes, while for soils a constant Tmin can be used.