The diet and trophic ecology of angler¢sh Lophius piscatorius at the Shetland Islands, UK C.H. Laurenson* P and I.G. Priede O *North Atlantic Fisheries College, Port Arthur, Scalloway, Shetland, UK. O Oceanlab, Newburgh, Aberdeenshire, UK. P Corresponding author, e-mail: Chevonne.Laurenson@nafc.uhi.ac.uk The diet of angler¢sh ( Lophius piscatorius), based on the analysis of stomach contents, was investigated in the Shetland Islands. Samples were obtained during trawling for sandeel (Ammodytidae species) and demersal trawling when mainly white¢sh (gadoids) are targeted. This allowed the diets of samples collected from di¡erent areas and at di¡erent times to be compared. The diets in di¡erent length groups were also compared and niche breadth investigated. A wide range of prey types, mainly ¢sh, were recorded but the main prey was Norway pout ( Trisopterus esmarkii) and lesser sandeel ( Ammodytes marinus). Diet composition varied seasonally. In samples collected during white¢sh trawling the occurrence of empty stomachs decreased from the ¢rst to the third quarters and generally increased with increasing angler¢sh size. Empty stomachs occurred less frequently in samples collected from areas where sandeel ¢shing occurred compared to areas where white¢sh trawling occurred. INTRODUCTION Angler¢sh Lophius piscatorius L. (commonly known as monks or monk¢sh), are distributed throughout the north-east Atlantic and cover a wide bathymetric range, from below the shore to at least 1000 m (Afonso-Dias & Hislop, 1996). The angler¢sh are distinctive in appearance with their dorso-ventrally £attened body and wide upwardly-facing mouth. Feeding occurs when the ¢rst dorsal ray, the illicium, which is situated on the snout, is £icked back and forth to lure prey to within range of an attack. Feeding has also been observed to be opportunistic, occurring when prey stray to within range of a strike (Laurenson et al., 2004). Although angler¢sh are important commercially, parti- cularly in the area around the Shetland Islands, relatively little is known of their biology. The only previous studies to speci¢cally investigate angler¢sh diet in the north-east Atlantic are early work by Fulton (1903) and a study in the Irish Sea by Crozier (1985). More recent studies, where the consumption of cephalopods by ¢sh have been investigated, do however, give mention to angler¢sh. In this paper the diet of angler¢sh caught in waters around the Shetland Islands is analysed. To determine if there were di¡erences in the diet of angler¢sh caught in di¡erent areas, samples were obtained from grounds where sandeel (Ammodytidae) ¢shing occurs and compared to those caught on grounds ¢shed by demersal trawlers targeting white¢sh (mainly gadoids). Seasonal di¡erences were also investigated for samples obtained during white¢sh trawling and the e¡ect of ¢sh length on the diet composi- tion was investigated for each sample group. MATERIALS AND METHODS The study was carried out by collecting stomach samples and analysing them for the remains of prey species. Samples were collected during observer trips on commercial demersal white¢sh trawlers during the second and third quarters of 1999 and the ¢rst and second quar- ters of 2000, providing sample groups Q1w, Q2w and Q3w. During the third quarter of 1999, samples were collected from vessels targeting sandeels (Ammodytidae). Samples from this source (Q3s) were taken in areas spatially separate from grounds sampled during white¢sh trawling. Samples were not taken from ¢sh showing any signs of regurgitation (an estimated 5^10% of ¢sh). Generally, individuals less than 30 cm total length (TL) were taken whole and those greater than 30 cm TL were gutted, the viscera double bagged, individually labelled, and stored frozen. The content of each thawed stomach sample was categorized as follows: Empty (E), Net Feeding only (NF), Prey items only (P) or both Prey items and Net Feeding (P þ NF). Items categorized as ‘net feeding’, were those assumed to have been taken incidentally in the trawl. These items were in a very fresh condition with no visible signs of digestion to the skin or scales. Any items in the buccal cavity or oesophagus were not included in the analysis. To determine the diet, contents were weighed to 0.1g and identi¢ed to the lowest possible taxon. Identi¢cations were undertaken visually or, depending on state of diges- tion, by identifying otoliths using a reference set and an identi¢cation guide (Ha« rko« nen, 1986). Where only premaxillae and vertebrae were recovered, these were identi¢ed as far as possible usingWatt et al. (1997). Levin’s measure of niche breadth ( B) was calculated as: B ¼ 1/ X p 2 j (1) where p j is a fraction of items in the diet that are of prey category j ( p j ¼1). The standardized niche breadth B A is: B A ¼ B  1 n  1 (2) where n is the number of prey categories. J. Mar. Biol. Ass. U.K. (2005), 85, 419^424 Printed in the United Kingdom Journal of the Marine Biological Association of the United Kingdom (2005)