Depredación

Oecologia (1998) 113:231±238

Ó Springer-Verlag 1998

Margaret W. Miller á Mark E. Hay

Effects of ®sh predation and seaweed competition on the survival and growth of corals

Received: 31 May 1997 / Accepted: 2 September 1997

Abstract On Caribbean coral reefs, high rates of grazing by herbivorous ®shes are thought to bene®t corals because ®shes consume competing seaweeds. We conducted®eld experiments in the Florida Keys, USA, to examine the e€ects of grazing ®shes on coral/seaweed competition. Initially, fragments of Porites divaracata from an inshore habitat were transplanted into full-cage, half-cage, and no-cage treatments on a fore-reef. Within 48 h, 56% of the unprotected corals in half-cage and nocage treatments (62 of 111) were completely consumed. Stoplight parrot®sh(Sparisoma viride) were the major coral predators, with redband parrot®sh (S. aurofrenatum) also commonly attacking this coral. Next, we transplanted fragments of P. porites collected from the fore-reef habitat where our caging experiments were being conducted into the three cage treatments, half in the presence of transplanted seaweeds, and half onto initially clean substrates. The corals wereallowed to grow in these conditions, with concurrent development of competing seaweeds, for 14 weeks. Although seaweed cover and biomass were both signi®cantly greater in the full-cage treatment, coral growth did not di€er signi®cantly between cage treatments even though corals placed with pre-planted seaweeds grew signi®cantly less than corals placed on initially clean substrate. This surprisingresult occurred because parrot®shes not only grazed algae from accessible treatments, but also fed directly on our coral transplants. Parrot®sh feeding scars

were signi®cantly more abundant on P. porites from the half and no-cage treatments than on corals in the full cages. On this Florida reef, direct ®sh predation on some coral species (P. divaracata) can exclude them from forereef areas, as haspreviously been shown for certain seaweeds and sponges. For other corals that live on the fore-reef (P. porites), the bene®ts of ®shes removing seaweeds can be counterbalanced by the detrimental e€ects of ®shes directly consuming corals. Key words Coral á Corallivory á Florida á Predation á Seaweed competition

Introduction
Invertebrate corallivores are widely recognized as having importante€ects on reef development and reef coral populations in both the Caribbean (e.g., the gastropod Coralliophila: Ott and Lewis 1972; Brawley and Adey 1982; Knowlton et al. 1990) and Paci®c (e.g., Acanthaster and other echinoderms: Endean 1973; Glynn et al. 1979; Glynn and Krupp 1986). In contrast, much of the literature on corallivory by reef ®shes refers to specialized Paci®c butter¯y®shes (Reese1977; Neudecker 1979; Cox 1986). Most research in the Caribbean and temperate Atlantic has focused on how grazing ®shes indirectly bene®t corals by consuming seaweeds and thus reducing competitive pressure (Birkeland 1977; Lewis 1986; Steneck 1988; Hughes 1994; Miller and Hay 1996). These studies suggest that a signi®cant reduction of ®sh grazing can result in the overgrowth of juvenile and adultcorals by other benthic life forms, especially ¯eshy macroalgae. Observations by Birkeland (1977) in the tropical eastern Paci®c indicate that herbivorous ®shes actively avoid consuming juvenile corals greater than a few millimeters in size. A similar result was reported by Brock (1979) for Hawaiian parrot®sh. In his laboratory study, ®eld densities of parrot®sh were determined to be ``optimal'' inpromoting coral recruitment and benthic community development.

M.W. Miller (8) National Marine Fisheries Service, Southeast Fisheries Science Center, 75 Vivginia Beach Dr. Miami, FL 33149, USA, e-mail: margaret.w.miller@noaa.gov; Fax: (305) 361-4562 M.E. Hay University of North Carolina at Chapel Hill, Institute of Marine Sciences, 3431 Arendell St., Morehead City, NC 28557, USA

232...