Vol. 464: 1–15, 2012 MARINE ECOLOGY PROGRESS SERIES Published September 19 doi: 10.3354/meps09964 Mar Ecol Prog Ser OPENPEN FEATURE ARTICLE ACCESSCCESS Host preference and habitat segregation among Red Sea anemonefish: effects of sea anemone traits and fish life stages Lindsay K. Huebner1,*, Brianna Dailey1, Benjamin M. Titus1, Maroof Khalaf2, Nanette E. Chadwick1 1Department of Biological Sciences, Auburn University, Auburn, Alabama 36849, USA 2Department of Marine Biology, The University of Jordan-Aqaba, Aqaba 77110, Jordan ABSTRACT: Competition drives habitat segregation between adults and juveniles in many types of organ- isms, but little is known about this process in anemonefish that compete for host sea anemones which differ in habitat quality. We performed field and laboratory experiments to determine causes of habitat segregation in 2-band anemonefish Amphi prion bi - cinctus on coral reefs in the northern Red Sea, where juvenile fish mainly occupy leathery sea ane mones Heteractis crispa, and breeding adults almost exclu- sively inhabit bulb-tentacle sea anemones Entac maea quadricolor. E. quadricolor were usually larger than H. crispa, and expanded more in response to fish pres- ence. Adult fish visually concealed a larger proportion of their body surface area among the relatively thick tentacles of E. quadricolor than among the thinner The tentacles of the sea anemone Entacmaea quadricolor tentacles of H. crispa, while juveniles were concealed offer optimal cover for anemonefish Amphiprion bicinctus. equally well in both hosts. During field experiments, Photo: L. K. Huebner vacated E. quadricolor were colonized rapidly by fish, whereas H. crispa were not. In laboratory choice ex- periments, fish at all post-settlement life stages pre- INTRODUCTION ferred E. quadricolor, and large individuals monopo- lized this host and relegated subordinates to H. crispa. Habitat segregation occurs between juvenile and We conclude that competitive exclusion drives habitat segregation among life stages of this anemonefish and adult life stages in many organisms, due in part to that host anemone traits underlie this process. The different habitat requirements at each life stage. In non-preferred host H. crispa may func tion as a refuge marine fishes, nursery habitats such as seagrass for juvenile fish while they wait for space to become meadows often provide special conditions that sup- available in the preferred host E. quadricolor, where port the growth and survival of juveniles, which then they are able to attain sexual maturity. migrate to adult habitats for breeding (Beck et al. 2001). Habitat segregation among life stages also KEY WORDS: Intraspecific competition · Symbiosis · may result from active exclusion of juveniles from Amphiprion bicinctus · Entacmaea quadricolor · optimal habitat by adults. The processes of habitat Heteractis crispa · Coral reef preference, competition, and predation all affect Resale or republication not permitted without patterns of habitat use among coral reef fishes, with written consent of the publisher *Email: [email protected] © Inter-Research 2012 · www.int-res.com 2 Mar Ecol Prog Ser 464: 1–15, 2012 important implications for their population dynamics tribute nutrients to both the anemones and their (Holbrook & Schmitt 2002, Dirnwöber & Herler 2007, endosymbiotic microalgae (Cleveland et al. 2011), Ben-Tzvi et al. 2009, Bonin et al. 2009). Many reef and aerate host anemone tissues (Szczebak 2011). fishes behave aggressively toward conspecifics, with Anemones thus experience faster growth, higher sur- larger individuals maintaining dominance over smaller vival, and enhanced asexual reproduction when ones (Forrester et al. 2006), and denying access to hosting anemonefish (Holbrook & Schmitt 2005). benthic habitats for juveniles as they recruit to the Additionally, anemones expand more fully when reef from the plankton (Bay et al. 2001, Ben-Tzvi et occupied by anemonefish than when alone, and, if al. 2009). Among reef fishes that form obligate sym- their fish are removed, they may contract com- bioses with benthic cnidarians, the cnidarian hosts pletely—despite minimized photosynthesis by their are a limiting resource for the population growth of microalgae in such a state—until the anemonefish their fish associates (Allen 1972, Holbrook et al. have returned (Porat & Chadwick-Furman 2004). In 2000). Coral characteristics such as body size, mor- contrast, the only documented benefit provided by phology, health, and surrounding habitat all can sea anemones to anemonefishes is shelter from pre- affect selection and preference for them as hosts by dation, to both the fish and their egg masses (Fautin coral-dwelling gobies and damselfishes (Holbrook et 1991). Sea anemones vary in their traits that relate to al. 2000, Schiemer et al. 2009). Especially among sheltering fish, for example in terms of their chemical damselfishes (Family Pomacentridae), patterns of defenses (Elliott et al. 1994), tentacle morphology coral habitat use are frequently altered by aggres- (Fautin & Allen 1997), types of microhabitats occu- sion among individuals for coral occupancy (Robert- pied (Chadwick & Arvedlund 2005), and in the provi- son 1996, Bay et al. 2001, Ben-Tzvi et al. 2009, Bonin sion of nearby substratum for the incubation of fish et al. 2009). egg clutches (Allen 1972). Anemonefishes potentially Anemonefishes comprise 28 species of damsel - compete for preferred host species based on these fishes that engage in obligate shelter symbioses with factors, causing their distributional patterns to differ 10 species of sea anemones on Indo-Pacific coral from those expected based on host and fish abun- reefs (Elliott et al. 1999). They are protandrous and dances alone (Srinivasan et al. 1999). No studies to form social groups with rigid size-based hierarchies date have quantified variation among anemone in each host anemone: the terminal female is largest species in their morphological or behavioral traits in size and dominant, her mate is second in size and that affect their abilities to shelter anemonefishes, or rank, and any other individuals are smaller sub- determined the resulting impacts on anemonefish adults or juveniles (e.g. Fautin 1991, Buston 2004). population dynamics. Depending on host size and anemonefish species, In the Gulf of Aqaba, northern Red Sea, endemic members of the breeding pair may not tolerate the 2-band anemonefish Amphiprion bicinctus mainly presence of other group members, and will regularly occupy bulb-tentacle sea anemones Entacmaea evict juveniles that attempt to immigrate or recruit quadricolor and leathery sea anemones Heteractis from the planktonic larval phase into their anemones crispa (Chadwick & Arvedlund 2005). H. crispa are (Buston 2003a). Because anemonefishes are more solitary with long, thin tentacles, and mainly host diverse than are host anemones, some anemone juveniles of A. bicinctus (Fishelson 1970, Chadwick species host >1 anemonefish species, but many ane - & Arvedlund 2005), while E. quadricolor possess monefish also may associate with >1 host species thicker tentacles that may be long and digitiform or (Fautin 1991). However, some hosts appear to be short and bulbous (Dunn 1981). Polyps of E. quadri- more desirable than others, resulting in potential color replicate clonally to form aggregations in some monopolization by competitively dominant individu- parts of the Indo-Pacific (Fautin & Allen 1997); how- als (Fautin 1986). Thus, breeding pairs of adult fish ever, only solitary individuals occur in the northern may defend certain types of anemones, while con- Red Sea, where they primarily host adult breeding specific juvenile fish are relegated to other types of pairs of fish, as well as occasional juveniles (Fishel- host anemones (Fautin 1991, Chadwick & Arvedlund son 1970, Chadwick & Arvedlund 2005). Thus, habi- 2005). However, the contributing factors and demo- tat partitioning occurs among life stages of this graphic impacts of this habitat segregation among anemonefish, in which individuals recruit to H. crispa anemonefish life stages are not well understood. but do not breed there. Additionally, these 2 host Anemonefishes provide an array of benefits to sea anemone species both usually occur in habitats anemones: they protect them from predation by but- which offer adjacent hard substrate for anemonefish terflyfishes (Porat & Chadwick-Furman 2004), con- to lay their egg masses: H. crispa often occur along Huebner et al.: Anemonefish host preference and habitat segregation 3 the reef–sand interface and E. quadricolor occur dead reef substratum (after Porat & Chadwick- along the reef−sand interface or are completely sur- Furman 2004). During all 3 census years, body size rounded by reef rock (Chadwick & Arvedlund 2005, (tentacle crown long and short diameter, to calculate authors’ pers. obs.). Fish preferences for these hosts tentacle crown surface area, TCSA; after Hattori likely occur in large part due to host behavior and 2002) and the number and TL of resident fish were morphology, rather than host microhabitat use per se. recorded for all anemones at the site. Some ane - Here we examine 2 major hypotheses concerning mones occurred near (<0.5 m distant) other ane - anemonefish habitat segregation in this system: mones, and their associated fish moved between, and (1) the 2 host anemone species differ in their morpho - sheltered in, both hosts. To simplify the assessment logical and behavioral traits related to sheltering of host use patterns,