A central question in evolutionary physiology is how complex physiological adaptations evolve in response to natural selection operating on performance traits hypothetically associated with Darwinian fitness. Selection experiments provide a powerful yet underutilized approach to that question [1,2]. We established an artificial selection experiment with lines of bank voles Myodes glareolus selected for high swim-induced aerobic metabolism (A), the ability to maintain body mass on a low-quality herbivorous diet (H), intensity of predatory behaviour towards crickets (P), and unselected control lines (C) (with four replicate lines in each of the categories [3]). In generation 13, voles from the selected lines achieved 48% higher maximum rate of metabolism, lost 38% less mass in the 3-day test with low-quality diet (at the age of 32-35 days), and attacked crickets 5 times more frequently than voles from control lines. Here we report correlated responses in whole body and internal organs mass, measured in 15-18 voles per line (both sexes, mean age 76 days). The animals were euthanized by an overdose of anaesthetic (isoflurane) and dissected. Organs were weighed to the nearest 0.001g. Although the selection was performed on mass-independent traits (residuals from regression), body mass was higher in H than in C lines (Mixed cross-nested linear model, adjusted least square means±SE; C: 22.5±0.57, A: 22.8±0.58, H: 25.9±0.56, P: 20.8±0.58g; effect of selection p<0.001). Apparently, selection for the capability to maintain energy balance during a period of undernourishment in juveniles resulted in evolution of increased body size in adults. As expected, mass of heart and gasterocniemus leg muscles were increased in A lines (Heart C: 0.117±0.004, A: 0.143±0.004, H: 0.117±0.004, P: 0.127±0.004g, p=0.001; Muscles C: 0.166±0.005, A:0.190±0.005, H:0.159±0.005, P:0.177±0.005, p=0.003). Because our earlier results showed an increased level of basal metabolism and food consumption in A lines [4,5], we also expected an increased mass of organs related to whole-body metabolism, and the results confirmed the expectation (Liver: C:1.20±0.03, A:1.31±0.03, H:1.19±0.03, P:1.27±0.03g; p=0.02; Kidneys: C: 0.29±0.01, A: 0.34±0.01, H: 0.29±0.01, P: 0.32±0.01g, p=0.003; Small intestine: C: 1.07±0.02, A: 1.18±0.02, H: 1.15±0.02, P: 1.16±0.02g, p=0.030; similar pattern was also found for stomach and caecum). As we expected, brain mass increased in P lines, but surprisingly it was also significantly increased in A lines (Total brain mass C: 0.522±0.006, A: 0.557±0.006, H: 0.531±0.006, P: 0.552±0.006g, p=0.006; similar pattern was found for cerebrum and cerebellum analysed separately). The selected lines of voles will provide a unique model to investigate lower-level histological, biochemical and molecular factors underlining evolution of the organismal performance traits.