times the ‘maintenance costs’ of others of the same size and sex? Given the forces of natural selection, such variation in metabolic rate can only persist if it is non-genetic in origin and/or the costs are offset by benefits. In this talk I will use examples from a range of species to discuss the evidence that intraspecific variation in resting or standard metabolic rate has a non-genetic origin, and can be influenced by early nutrition, and to show that metabolic rate is associated with other fitness-related traits. For instance, in salmonid fish a high SMR is linked to greater aggression and dominance, which is likely to confer advantages in some (but not all) environments. Moreover, our recent experiments show that SMR in fish covaries with digestive strategy, so that individuals with a higher SMR pay a greater energetic cost (SDA) but can digest food (and so resume feeding) faster. These results suggest that the fitness consequences of a given metabolic rate will depend on the local circumstances (i.e. different microhabitats may select for different metabolic rates), so leading to the persistence of significant intraspecific variation in this key physiological trait. Email Address for correspondence: n.metcalfe@bio.gla.ac.uk doi:10.1016/j.cbpa.2009.04.276 A9.4 11:00 Wednesday 1st July 2009 Field metabolic rate in seabirds — Influences, limits and con- sequences Jonathan A. Green (University of Liverpool) Seabirds make excellent models for studies of field metabolic rate (FMR). FMR has been evaluated using a variety of methods, in a number of populations and species around the world. However, few studies have looked in depth at what factors might be responsible for the variability we observe between species, populations and indi- viduals. In this presentation I will investigate some of these factors by delving deeper into the behaviour and energetics of seabirds. At the species level, I will discuss how time allocation and the energetic costs of specific activities can explain differences observed between species. Careful allocation of time to energetically costly activities is necessary as animals face limits to the intake and expenditure of energy. Using an example from penguins I will show that in seabirds at least, limits appear to apply to demands on energy expenditure rather than supply. At the population level, differences in the success of populations of gannets can be explained by examining time allocation and calculating the proximities of different populations to energetic limits. Finally, new techniques and methods of analysis can for the first time show how even at the individual level, metabolic performance and reproductive performance are linked. Preliminary analyses from both penguins and gannets suggest that higher performing individuals have lower energetic costs. Email Address for correspondence: jonathan.green@liverpool.ac.uk doi:10.1016/j.cbpa.2009.04.277 A9.5 11:30 Wednesday 1st July 2009 Laboratory model of adaptive radiation: Activity and metabolic rates in bank voles from a multidirectional artificial selection experiment Pawel Koteja (Institute of Environmental Sciences Jagiellonian University Kraków, Poland), Katarzyna Baliga-Klimczyk (Institute of Environmental Sciences Jagiellonian University Kraków, Poland), Katarzyna M. Chrzascik (Institute of Environmental Sciences Jagiel- lonian University Kraków, Poland), Milena Damulewicz (Institute of Environmental Sciences Jagiellonian University Kraków, Poland), Dominika Dragosz-Kluska (Institute of Environmental Sciences Jagiellonian University Kraków, Poland), Justyna Morawska-Ploskon- ka (Institute of Environmental Sciences Jagiellonian University Kraków, Poland) Explaining interspecific variation of metabolic rates is a topical subject in comparative physiology. We asked how the level of metabolism of a wild rodent, the bank vole (Myodes glareolus), would change in response to selection acting in distinct directions: the ability to maintain body mass on a low-quality herbivorous diet (H), high swim- induced aerobic metabolism (A), and intensity of predatory behavior towards crickets (P). Four replicate lines are maintained for each of the directions and an unselected control (C). In generation 7, voles from H lines fed a low-quality diet gained body mass (mean ± SD, 0.35 ± 0.99 g) whereas those from C lines lost mass (-0.65 ± 1.10 g; p <0.0001). The swim-induced metabolic rate was higher in A than in C lines (A: 297± 37; C: 222±31 ml O 2 /h; p <0.0001). The proportion of “predatory” individuals was higher in P (82%) than in C lines (29%; p < 0.001). Continuous records from infra-red motion sensors and visual observa- tions of behaviour (ethograms) showed that voles from A and P lines were more active than those from H and C lines; the pattern of activity differed also between A and P lines. The maximum forced-exercise and cold-induced metabolic rates were higher in A compared to C lines (ml O 2 /h, exercise A: 381±54, C: 312±46; cold A: 311±42, C: 282±39; both p <0.0001). The basal metabolic rate (BMR) was higher in A lines (51.7±9.1 ml O 2 /h) compared to C (47.5±6.7 ml O 2 /h) or H (48.4± 6.4 ml O 2 /h; p =0.03) and was intermediate in P lines (50.3 ±7.4 ml O 2 /h). The results are consistent with a hypothesis linking the evolution of endothermy with a selection for active-predator lifestyle. Email Address for correspondence: pawel.koteja@uj.edu.pl doi:10.1016/j.cbpa.2009.04.278 A9.6 11:50 Wednesday 1st July 2009 Swim more, weigh less: Linking physiological energetic and behavioural ecology in individual European seabass Shaun S. Killen (Université Montpellier II), Stefano Marras (Université Montpellier II), David J. McKenzie (Université Montpellier II) Individual fish exhibit consistent intrinsic differences in standard metabolic rate (SMR), but it is not clear how these differences in physiology may be relevant to their ecology. We investigated relation- ships between SMR, routine metabolic rate (RMR), maximum metabolic rate (MMR), aerobic scope (AS), starvation tolerance as weight loss, and risk-taking behaviour, in 39 juvenile seabass Dicentrarchus labrax. All of these traits showed large inter-individual variation. Standard metabolic rate correlated positively with weight loss during a seven-day period of fasting, although RMR showed a higher correlation, suggesting that individual differences in activity contribute to relative tolerance of starvation. Both MMR and absolute AS showed positive correlations with SMR, indicating that high SMR was associated with increased aerobic capacity. As a consequence, however, MMR also showed a positive correlation with rate of weight loss, suggesting a trade-off between aerobic capacity and starvation tolerance. Behavioural analyses are ongoing, but it appears that individuals show consistent differences in measures of riskiness such as time until first emergence from cover to obtain food after a startling stimulus. The expectation is that these measures of risk-taking tendencies will correlate positively with SMR, Abstracts / Comparative Biochemistry and Physiology, Part A 153 (2009) S145–S150 S146