In: Greenwood, J.G. & Hall, N.J., eds (1998) Proceedings of the Australian Coral Reef Society 75th Anniversary Conference, Heron Island October 1997. School of Marine Science, The University of Queensland, Brisbane. pp. 15-??. Grazing Pressure of Roving and Territorial Fishes on Patch Reef in One Tree Lagoon David J. Booth Department of Environmental Sciences, University of Technology, Sydney, Westbourne Street, Gore Hill NSW 2065 ABSTRACT Grazing by fishes has a strong influence on coral reef community structure and production. Temporal and spatial patterns in fish grazing pressure may affect algal production, coral growth and fish recruitment. In this study, I monitored grazing rates of roving fishes (mainly small scarids) and territorial pomacentrids over the period June 1993 to July 1996 on reeftops in One Tree Island lagoon. Pomacentrus wardi territories covered between 13 and 35% of the top of each reef in winter, and between 11 and 28% in summer, within which roving grazers were effectively excluded. Feeding rates averaged 1300 bites per m 2 per day in winter and 4400 bites per m 2 per day in summer, although breeding males did not defend territories in November- December. Outside territories, groups of between 2 and 20 small grazing scarids (mainly Scarus spp.) were the principal grazing fishes, with larger scarids, acanthurids and siganids making minor contributions to the grazing pressure. Feeding rates of small scarids varied from 1400 bites per m 2 per day in winter to 3800 bites per m 2 per day in summer. Coupled with estimates of algal intake per bite for pomacentrids and small scarids, these results suggest that grazing pressure is higher in summer, and higher in P. wardi territories. INTRODUCTION Coral reefs harbour significant biodiversity and are characterised by high rates of primary production (Connell, 1978). The herbivorous fishes of coral reefs consist of grazing and browsing species (Ogden & Lobel, 1978; Russ, 1985), and account for 10 to 25% of the total fish diversity on coral reefs (Talbot, 1965; Randall, 1967; Goldman & Talbot, 1976). There are many potential grazers of algae on coral reefs. These include roving herbivorous fishes, such as scarids and acanthurids, site-attached blennies and damselfishes, plus a variety of invertebrates (Klumpp & Polunin, 1989). Carpenter (1986) recorded more than 60 species of herbivores (excluding micrograzers) on a Caribbean reef. Benthic community structure can be influenced by fish grazing (Hatcher, 1981). Intense grazing by fish has been shown to directly affect algal diversity and biomass. Although coral reefs have high rates of primary production, this is not reflected in high algal standing crop, because grazing may remove as much as 6% of the standing crop per day (Bakus, 1969; Hatcher & Larkum, 1983). Caging experiments designed to exclude herbivorous invertebrates and fish have shown large increases in algal biomass inside cages (e.g. Bakus, 1967; Ogden & Lobel, 1978), and these algae may hinder coral growth and development (e.g. Potts, 1977). Intense grazing may also keep epilithic algal communities at an early successional stage (e.g. Vine, 1974; Hixon & Brostoff, 1996). Within damselfish territories, grazing pressure from other species is substantially reduced (Booth, unpubl. data), and epilithic algae within territories may be subjected to "intermediate disturbance" (Hixon & Brostoff, 1981). The activities of herbivorous fishes represent a significant source of remineralisation of bioavailable nutrients, and may enhance the recruitment of crustose-coralline algae and reef- building corals (Steven, 1994). Grazers rework sediment, reduce grain size (Hiatt & Strasburg,