The prevalence of weapon damage: a proportional meta-analysis

Sexually selected weapons are tools that are used in physical fights over mating opportunities. Individuals can face dire fitness consequences if they break their weapon and can no longer fight effectively; however, the costs and consequences of weapon damage have largely been ignored. In this study, the literature was systematically searched to examine the prevalence of weapon damage, and a phylogenetically controlled proportional meta-analysis was conducted to explore factors influencing the patterns of weapon damage across the animal kingdom. A total of 36 suitable studies were identified across 74 species and 13 orders for a total of 82 effect sizes. We found that weapon damage is common across taxa, with an average rate of damage of 9.4% across all observations. We found strong evidence that regenerative ability influences the patterns of weapon damage, with species capable of weapon regeneration exhibiting high prevalence of weapon damage. Herein, we discuss additional factors that may affect the likelihood, costs and consequences of weapon damage, including weapon size, weapon allometry, fighting behaviour and when and to what degree a species can regenerate its weapon. We argue that the risk of weapon damage is an important but underappreciated cost that could affect strategic fighting decisions, providing new insights into the role of weapon damage on the evolution of weapon morphology and contest behaviour within and among species.