Small-scale environmental variation influences whether coral-dwelling fish promote or impede coral growth.

Mutualistic symbioses are ubiquitous in nature and facilitate high biodiversity and productivity of ecosystems by enhancing the efficiency of energy and nutrient use within ecological communities. For example, small groups of fish that inhabit coral colonies in reef ecosystems potentially enhance coral growth through defense from coral predators, aeration of coral tissue and nutrient provisioning. This study examines whether the prevalence and consequences of fish-coral interactions vary among sites with different environmental conditions in a coral reef lagoon, using the humbug damselfish Dascyllus aruanus and its preferred coral host Pocillopora damicornis as a study system. Using a field experiment, we tested the site-specific effects of D. aruanus on coral growth, and show that the cost-benefit ratio for corals hosting fish varies with local environmental variation. Results of this study also demonstrate that fish prefer to inhabit coral colonies with particular branch-spacing characteristics, and that the local abundance of D. aruanus influences the proportion of coral colonies within a site that are occupied by fish rather than increasing the number of fish per colony. We also show that corals consistently benefit from hosting D. aruanus via defense from predation by corallivorous butterflyfish, regardless of local environmental conditions. These findings highlight the need to consider the potential for multiple scale- and state-dependent interaction effects when examining the ecology of fish-coral associations. We suggest that fluctuating cost-benefit ratios for species interactions may contribute to the maintenance of different colony phenotypes within coral populations.

From cooperation to combat: adverse effect of thermal stress in a symbiotic coral-crustacean community.

Although mutualisms are ubiquitous in nature, our understanding of the potential impacts of climate change on these important ecological interactions is deficient. Here, we report on a thermal stress-related shift from cooperation to antagonism between members of a mutualistic coral-dwelling community. Increased mortality of coral-defending crustacean symbionts Trapezia cymodoce (coral crab) and Alpheus lottini (snapping shrimp) was observed in response to experimentally elevated temperatures and reduced coral-host (Pocillopora damicornis) condition. However, strong differential numerical effects occurred among crustaceans as a function of species and sex, with shrimp (75%), and female crabs (55%), exhibiting the fastest and greatest declines in numbers. These declines were due to forceful eviction from the coral-host by male crabs. Furthermore, surviving female crabs were impacted by a dramatic decline (85%) in egg production, which could have deleterious consequences for population sustainability. Our results suggest that elevated temperature switches the fundamental nature of this interaction from cooperation to competition, leading to asymmetrical effects on species and/or sexes. Our study illustrates the importance of evaluating not only individual responses to climate change, but also potentially fragile interactions within and among susceptible species.

Regulation of protogynous sex change by competition between corticosteroids and androgens: an experimental test using sandperch, Parapercis cylindrica.

Cortisol, the dominant corticosteroid in fish, and 11-ketotestosterone (11KT), the most potent androgen in fish, are both synthesized and (or) deactivated by the same two enzymes, 11beta-hydroxylase and 11beta-hydroxysteroid dehydrogenase. Cortisol is synthesized in response to stress (such as that caused by interaction with a dominant conspecific), whereas 11KT is synthesized during protogynous sex change. It has been hypothesized that corticosteroids (such as cortisol) inhibit 11KT synthesis via substrate competition, thereby providing a mechanism for the regulation of socially mediated, protogynous sex change. We tested this hypothesis by administering cortisol (50 microg g(-1) body weight) to female sandperch (Parapercis cylindrica) under social conditions that were permissive to sex change (i.e. in the absence of suppressive male dominance). Twenty-one days later, mean physiological cortisol concentration in cortisol-treated fish was 4.2-fold greater than that in 'socially stressed' female fish maintained in a semi-natural system. Although the dosage of cortisol was therefore considered to be favorable for engendering competitive inhibition of 11KT synthesis, all cortisol-treated fish changed sex, as did all sham-treated and control fish (n=7 fish per treatment). In addition, there was no effect of cortisol treatment on the rate of sex change or on the pattern of steroidogenesis. Thus, our results refute the hypothesis that protogynous sex change is regulated by substrate competition between corticosteroids and androgens.