Metabolic costs of ghting are driven by contest performance in male convict cichlid sh Donald L. Copeland a , Bryan Levay b , Boopathy Sivaraman c , Caili Beebe-Fugloni b , Ryan L. Earley c, * a Program in Integrated Bioscience, Department of Biology, University of Akron, Akron, OH, U.S.A. b Department of Biology, California State University Fresno, Fresno, CA, U.S.A. c Department of Biological Sciences, University of Alabama, Science & Engineering Complex, Tuscaloosa, AL, U.S.A. article info Article history: Received 31 October 2010 Initial acceptance 23 November 2010 Final acceptance 19 April 2011 Available online 31 May 2011 MS. number: A10-00752R Keywords: Amatitlania nigrofasciata assessment body size cichlid cortisol ghting lactate One of the most important decisions that an animal faces is whether to persist or surrender in an aggressive contest. It pays to win these encounters, but it might be just as economical to evaluate ghting costs and adjust behavioural decisions accordingly. Relatively few studies have explored the physio- logical basis of contest costs and fewer have determined the internal signals that animals might attend to during combat. We explored relationships between physiological parameters, including energy substrates, anaerobic metabolites and steroid hormones, and both contest performance and contest success in male convict cichlids, Amatitlania nigrofasciata. Size-matched pairs were allowed to ght until one member of the pair submitted, and tissues were subsequently harvested for analysis. There were no signicant status differences (winner, loser, control) in mean postght physiological measures. Domi- nance reinforcement during a short postsettlement period buffered the winnersstress response. In both winners and losers, muscle lactate concentrations correlated positively with contest intensity, while only losers showed a signicant increase in plasma glucose concentrations as contests intensied. Larger cichlids used more costly ghting tactics than smaller conspecics, and size asymmetries affected contest outcomes between larger animals signicantly more than they did those between smaller animals. These ndings suggest that ghting is metabolically costly and that the ramications of these costs may differ in a size-dependent fashion. We use a recent literature on the scaling of anaerobic tolerance to evaluate the possibility that larger cichlids may be more tolerant of lactate accumulation, and thus able to both engage in more intense ghting tactics and persist longer in escalated ghts than marginally smaller opponents. Contest intensity varied independently of mass asymmetries but varied positively with the absolute size of both the winner and loser, suggesting that convict cichlids use self-assessment strategies during contests with similarly sized opponents. Ó 2011 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. The factors that inuence ght outcomes among animals are complex and multifaceted. Classic views centre on ghting ability, termed resource-holding potential (RHP), and its role in moder- ating animal contest dynamics (Parker 1974; Briffa & Sneddon 2010). A major component of RHP is the animals body size, and there is generally a positive correlation between relative body size/ mass and contest success (Arnott & Elwood 2009a). The specic mechanisms through which size may confer greater RHP (e.g. strength, hormone titres, or mass-specic physiological tolerances) have not been thoroughly investigated in most animals. However, it is reasonable to suspect that, during ghts, animals use specic mechanisms to assess their own RHP, that of their opponent, or both. Selection should favour quick resolution of aggressive inter- actions between highly asymmetrical pairs, particularly for inter- actions that are energetically costly and/or tend to result in injury. Therefore, animals must decide whether to engage or escalate in contests based on, for example, visual or chemical cues indicative of opponent ghting ability, or internal cues reecting accumulation of energetic costs or injury (Briffa & Elwood 2001; Arnott & Elwood 2009a). Animals may decide whether and for how long they will ght by using self-assessment and/or mutual assessment (Hsu et al. 2008; Arnott & Elwood 2009a). Mutual assessment (MA) posits that contestants can gauge each others ghting ability and thus forgo futile escalation once they realize their relative RHP (Enquist & Leimar 1983). Models of self- assessment (SA) state that contestants assess only their own ability to perform in a contest and while a positive relationship should exist between the losers RHP and contest duration, no * Correspondence: R. L. Earley, Department of Biological Sciences, University of Alabama, Science & Engineering Complex, Box 870344, 300 Hackberry Lane, Tuscaloosa, AL 35487, U.S.A. E-mail address: rlearley@bama.ua.edu (R. L. Earley). Contents lists available at ScienceDirect Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav 0003-3472/$38.00 Ó 2011 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2011.05.001 Animal Behaviour 82 (2011) 271e280