Assessment in contests of male lizards (Anolis carolinensis): how should smaller males respond when size matters?

Within a female-defence polygynous mating system, males of the lizard A. carolinensis use an array of stereotyped signals and aggressive tactics to acquire and defend long-term territories containing multiple, sedentary females. In competition, body size appears to be important because size of free-ranging males correlates positively with volume of male territories and number of patrolled females. Therefore, an ability to assess body size during territorial contests should be an adaptive attribute that would influence the tactics of intermale aggression. To examine this premise, we staged contests between 10 pairs of males matched for size (i.e. symmetric contests) and 10 pairs mismatched for size (i.e. asymmetric contests), while all 20 pairs of males were matched for habitat resources, a mate, and resident status. Overall, we found (1) contest profiles best fit the features of a 'fixed-phase, sequential assessment' model of game theory, (2) body size and mass were highly correlated with contest outcome, (3) none of 12 signal variables predicted contest outcome, and (4) paired males generally matched aggressive tactics and signalling behaviours. We also examined the asymmetric contest profiles for deviations from the profiles of symmetric contests. We tested the proposition that smaller males of size-mismatched contests would assess their disadvantage and choose a bluff strategy to mitigate risky behaviour and avoid fighting. We found, however, that the risk-mitigation hypotheses were unsupported. Smaller males were not playing a bluff strategy, but rather a hawk strategy. They initiated risky tactics by (1) invading the habitats of their larger opponents, an act that invited retaliation, (2) showing no tendency to stay away from larger opponents, (3) maintaining high levels of aggressive signalling as encounters intensified, and (4) engaging larger opponents in physical fighting, despite losing 90% of their encounters. These empirical results support a recent game theoretical construct ('Napoleon complex') that models size-asymmetric contests in which smaller males initiate fights that they usually lose. Our data suggest that, if smaller males of A. carolinensis have breeding territories, then they will engage in costly contests, despite a low probability of defeating larger and equally motivated opponents.