Prey selection by flounder, Platichthys flesus, in the Douro estuary, Portugal By C. Vinagre, H. Cabral and M. J. Costa Instituto de Oceanografia, Faculdade de Cie ˆncias, Universidade de Lisboa, Campo Grande, Lisboa, Portugal Summary Prey selection by the flounder, Platichthys flesus (Linnaeus, 1758), in an estuarine nursery was investigated and the major factors influencing food choice by this species were assessed. Diet breadth was narrow, reflecting the low prey diversity observed in the benthos. A gradual ontogenetic shift from small prey such as amphipods to larger prey like polychaetes and bivalves was observed. Amphipods had positive electivity values in the upper estuary and negative values in the lower estuary.Polychaetes showed the inverse pattern. Bivalve elec- tivity values were always positive. Differential selectivity throughout the estuary was mainly related to spatial segrega- tion of flounders according to size, with the smaller individuals concentrating in the upper estuary and larger individuals concentrating in the lower estuary. Amphipods such as Corophium spp. play a crucial role in the flounder diet because of their small size, low mobility and diel activity pattern. As prey, the polychaete value increases throughout flounder ontogeny since the flounder size range is compatible with the larger mouth gapes and detection ability of larger fish. Bivalve electivity values for flounder are mainly related to high calorific values. The absence of Crangon crangon (Linnaeus, 1758) in the diet may be due to low water temperature since the cost–benefit involved in the capture of highly mobile prey is too high at low temperatures. It was concluded that flounder must use several sensory features to detect and capture prey in turbid estuarine waters and that field studies provide impor- tant background information on the actual predator prefer- ences under natural conditions. Introduction Predator–prey interactions are one of the most important forces structuring biological communities. Predators must weigh costs against benefits to maximize energy intake, leading to different food selection criteria used by foraging animals. Cost–benefit studies have shown that foraging decisions incur costs in metabolic energy and time that must be outweighed by energy intake provided by prey (Rovero et al., 2000; Smalleg- ange and van der Meer, 2003). Most studies on prey selection by fish focus on visual selection in open water environments (e.g. Zander and Berg, 1984; Saunders and Montgomery, 1985). Therefore, many of the factors found to affect selection are related to visual detection, such as prey movement (e.g. Holmes and Gibson, 1986; Nelson and Coull, 1989), size (e.g. Mauck and Coble, 1971; Main, 1985; Ryer, 1988), pigmentation and visual crypticity (e.g. Minello and Zimmerman, 1984; Main, 1985). However, few studies deal with selection in estuarine waters, which are characterized by a high degree of turbidity and low visibility (Brewer and Warburton, 1992), and fewer on selectivity by fish that prey at night or dusk. There is, however, information on several features of prey organisms that potentially influence their selection. For instance, gustatory cues are used by Cyprinus carpio (Linnaeus, 1758) to determine shell thickness and consequently select gastropods (Stein et al., 1975). Burying ability prevents some prey species from being selected (Minello and Zimmerman, 1984). Siphon tips of bivalves are often reported as preferred by several species due to their high energy content (e.g. Cyrus and Blaber, 1983; Tomiyama et al., 2005). The European flounder, Platichthys flesus (Linnaeus, 1758), is a macrobenthic predator common in shallow waters from the North Sea to the Mediterranean (Quero et al., 1986). Juveniles of this species spend their first 2 years in estuarine and coastal nurseries. Little is known about the sensory mechanisms used by juveniles of this species to detect prey. Flounder are usually classified as visual predators (Moore and Moore, 1976; Holmes and Gibson, 1983), yet juveniles forage in turbid estuarine waters mainly at dawn, dusk and night (Bregnballe, 1961; Muus, 1967; Pihl, 1982). This study describes food selection by P. flesus in the Douro estuary (Portugal) and investigates the major factors influenc- ing its foraging behaviour. Materials and methods Study area The Douro estuary, located in northern Portugal, covers an area of 7.8 km 2 (Anonymous, 1986). Its land limit is set by the Crestuma-Lever dam, located 21.8 km upstream from the river mouth. The average annual river flow amounts to 480 m 3 s )1 , and the estuary is dominated by winter floods (average high winter flow varies between 917 and 2160 m 3 s )1 ) and summer droughts (average low summer flow varies between 25 and 178 m 3 s )1 ) (Bordalo, 1991). The tidal regime is semidiurnal with a range of approximately 4 m. Average estuary width is 400 m, while the maximum width is 1.1 km near the mouth (Fig. 1). Sampling procedures Fish samples were collected bimonthly from November 2000 to March 2002, in 12 sites within the estuary (Fig. 1) using a 12 m otter-trawl with 18-mm mesh size (stretched mesh). Trawls were towed at dusk on spring tides at a constant speed of 2.5 knots and lasting for 15 min. To ensure trawl contact with the bottom to maintain a high catching efficiency, the ground rope of the trawl was equipped with a heavy metal chain. Temperature was measured using a probe (Fig. 2). J. Appl. Ichthyol. 24 (2008), 238–243 Ó 2008 The Authors Journal compilation Ó 2008 Blackwell Verlag, Berlin ISSN 0175–8659 Received: January 31, 2007 Accepted: September 15, 2007 doi: 10.1111/j.1439-0426.2008.01055.x U.S. Copyright Clearance Centre Code Statement: 0175–8659/2008/2403–0238$15.00/0 www.blackwell-synergy.com