Optimal foraging patterns reflect integration of costs, gains, and risks associated with food availability, capturing and handling time, exposure to mortality sources, density of competing conspecifics, and time away from other tasks. Adaptive balance between costs and gains of foraging often involves an inverse relationship between food quality and risk. Cannibalism can add an extreme endpoint to that relationship, with its peculiar tension of prey that may be of especially high nutritional quality but are also especially risky because they are potential predators. Discovery of perfect intact skeletons of the large, common Caribbean starfish, Oreaster reticulatus, as well as living starfish with patches of exposed skeleton on their aboral surfaces, inspired experiments and observations aimed at uncovering the circumstances leading to both successful and unsuccessful cannibalism attempts. Every starfish in 7344 m2 of seagrass partially bordered by shallow coral reef was marked, measured, and its location and feeding activities recorded on each of 68 days to identify relationships between distance traveled, location with respect to the reef and other starfish, food quality, and cannibalism. To reveal and confirm what situations result in cannibalism, behavior with conspecifics was quantified under a variety of circumstances by experimental manipulations in the field.
Results/Conclusions
Meal quality varies widely, is inversely related to availability, can be predicted by spatial position, and can influence distances moved by starfish in 24 hr. Population density is correlated with increased cannibalism attempts, a result found in other cannibalism studies. Experimental manipulations of starfish to generate cannibalism opportunities revealed that in most cases the intended meal starfish can escape, exhibiting a stereotyped behavioral sequence that suggests this may be a common challenge for Oreaster. As found in previous cannibalism studies, size disparity between interacting starfish plays an important role in determining whether or not cannibalism attempts are successful. However Oreaster do not conform with predictions of foraging theory and data from previous studies that agree on an association of lower food quality with higher rates of cannibalism. Availability of high quality food increased cannibalism attempts and success. Intended victims enthralled by high quality meals were more likely to lose their lives to a cannibal star, and when they did escape, the cannibal star often took over their meal. While Oreaster behavior fails to match specific predictions from theory, their foraging is optimal in the context of natural history idiosyncrasies of this system, highlighting the value of understanding system-specific nuances.