VARIATION IN THE PRESENCE AND CAUSE OF DENSITY-DEPENDENT MORTALITY IN THREE SPECIES OF REEF FISHES

Abstract

Determining the mechanisms by which natural populations are regulated is a key issue in ecology. Identifying the biological causes of density dependence has, however, proved difficult in many systems. In this study we tested whether adults of three species of reef fish (all gobies) suffered density-dependent mortality, and whether the density-dependent component of mortality was caused by predation. We used field experiments to test for density dependence in each prey species, manipulating the presence of predators and prey density in a factorial design. Prey were stocked on replicate patches of reef constructed of natural materials, with each reef receiving a different density of gobies. Predatory fishes were excluded from half of the reefs using a combination of removals and exclusion cages. Survival of the first species, Lythrypnus dalli, was high and density-independent on reefs free of predators, but declined rapidly with increasing density on reefs to which predators had access. Density dependence in L. dalli was thus a result of mortality inflicted by predatory fishes. In the second species, Coryphopterus nicholsii, predators caused a large reduction in the survival in one experiment but had a negligible effect in a second experiment. More importantly, though, survival of C. nicholsii was always independent of its density, regardless of predator presence. In the final species, Coryphopterus glaucofraenum, two separate experiments showed that natural changes in adult abundance (experiment 1) and survival of stocked adults (experiment 2) were density-dependent regardless of predator presence. Both experiments thus indicated that the density-dependent component of loss in C. glaucofraenum was caused, at least in part, by an interaction other than predation. The presence, intensity, and biological cause of density-dependent survival were thus strikingly different for each of these three fishes, despite the fact that they are taxonomically closely related and ecologically similar. These findings suggest a need for further studies aimed at predicting under what circumstances different mechanisms of population regulation will operate.