The aerobic capacity model postulates that high basal metabolic rates (BMR) associated with endothermy evolved as a correlated response to the selection on maximum, peak metabolic rate Vo2max. Furthermore, the model assumes that BMR and Vo2max are causally linked, and therefore, evolutionary changes in their levels cannot occur independently. To test this, we compared metabolic and anatomical correlates of selection for high and low body mass-corrected BMR in males of laboratory mice of F18 and F19 selected generations. Divergent selection resulted in between-line difference in BMR equivalent to 2.3 phenotypic standard deviation units. Vo2max elicited by forced swimming in 20 degrees C water was higher in the low BMR than high BMR line and did not differ between the lines when elicited by exposure to heliox at -2.5 degrees C. Moreover, the magnitude of swim- and heliox-induced hypothermia was significantly smaller in low BMR mice, whereas their interscapular brown adipose tissue was larger than in high BMR mice. Our results are therefore at variance with the predictions of aerobic capacity model. The selection also resulted in correlated response in food consumption (C) and masses of metabolically active internal organs: kidneys, liver, small intestine, and heart, which fuel maximum, sustained metabolic rate (SusMR) rather than Vo2max. These correlated responses were strong enough to claim the existence of positive, genetic correlations between BMR and the mass of viscera as well as C. Thus, our findings support the suggestion that BMR evolved as a correlated response to selection for SusMR, not Vo2max. In functional terms BMR should therefore be interpreted as a measure of energetic costs of maintenance of metabolic machinery necessary to sustain high levels of energy assimilation rate.