Nocturnal Ecophysiology of the Anemonefish-Sea Anemone Mutualism: Patterns of Dark Oxygen Consumption and Symbiont Behavior at Night by Joseph Thomas Szczebak A thesis submited to the Graduate Faculty of Auburn University in partial fulfilment of the requirements for the Degree of Master of Science Auburn, Alabama December 12, 2011 Keywords: mutualism, ecophysiology, anemonefish, sea anemone, coral reef, oxygen consumption, nocturnal behavior Copyright 2011 by Joseph Thomas Szczebak Approved by Nanete E. Chadwick, Chair, Asociate Profesor of Biological Sciences Raymond P. Henry, Profesor of Biological Sciences Carol Johnston, Profesor of Fisheries and Alied Aquacultures ii Abstract The mutualism betwen anemonefishes and giant sea anemones is one of the most wel known interactions on coral reefs. While the symbiotic benefits provided to each partner have been researched for over 100 years, litle is known about the mutualism at night. Further, the ecophysiological mechanisms that underpin the ecological benefits of the mutualism remain greatly unexplored. Here, I conducted foundational research on the metabolic and behavioral interactions of the anemonefish-sea anemone mutualism at night. Physical contact betwen anemonefish and sea anemones elevates the net dark oxygen (O 2 ) consumption of the partners. Further, anemonefish engage in more flow- modulating activities when sea anemones are present than when anemonefish are alone. Lastly, sea anemone O 2 consumption increases with water flow. I conclude that anemonefish behavior at night modulates sea anemone O 2 consumption by forced convection of ambient water flow. iii Acknowledgements I am indebted to my advisor, Dr. Nanete Chadwick. While my project, like most, had its share of complications, you made sure I had everything I needed to solve the isue(s) at hand. Further, your confidence in my ideas and abilities encouraged me push my limits and think outside the box. I also thank my commite members, Dr. Ray Henry and Dr. Carol Johnston. Your regular advice and fedback throughout the design, research, and writing proceses of this thesis are greatly appreciated. I am grateful to the scientists and staf of the Marine Science Station in Aqaba, Jordan. My time in the Red Sea was personaly and profesionaly inspiring, and it was a trip I wil never forget (funded by a NSF IRES Grant, awarded to Dr. Nanete Chadwick). I also would like to thank the Chad Lab crew: Kathy Morrow, Lindsay Huebner, Ben Titus, Ashley Isbel, Stanton Belford, and Joe Krieger. A solid crew of felow grad students is an indispensable resource, and I appreciate al the advice and good times. A special thanks goes to Mom, Dad, Jenn, Dan, Tim, and Anna. Your love and support have never been in short supply, and you were a key ingredient in everything I achieved here at Auburn. Finaly, I would like to thank Maria Piraino, my only true constant for the last 2.5 yr. Without your love, encouragement, and spunk, I can?t even imagine how crazy I would be. P.S. Thanks to my pup, Penny, for geting me out of the lab. iv Table of Contents Abstract .............................................................................................................................. ii Acknowledgements ........................................................................................................... iii List of Tables ...................................................................................................................... v List of Figures .................................................................................................................... vi Chapter I: The benefits and ecophysiology of mutualistic symbioses on coral reefs .........1 Benefits of coral reef mutualisms ............................................................................1 Ecophysiological adaptations to environmental stres ............................................5 Chapter II: Anemonefish behavior at night modulates sea anemone oxygen consumption .............................................................................................................9 Summary .................................................................................................................9 Introduction ...........................................................................................................10 Materials and Methods. .........................................................................................13 Results ...................................................................................................................19 Discussion .............................................................................................................21 References .........................................................................................................................40 v List of Tables Table 1. Flow-through respirometry treatments used to ases the efects of symbiotic interactions on dark oxygen consumption (VO 2 ) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor) ........................................29 Table 2. Statistical summary (repeated-measures ANOVA) of efects of respirometry treatment on dark oxygen consumption (mean VO 2 ?1 s.e.m.) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor) ...................32 Table 3. Statistical summary (repeated-measures ANOVA) of percent time and bout frequency (bouts 5 min -1 ) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus) when alone (Fish) and when with host sea anemone (Entacmaea quadricolor, Fish+Anem) ..................................................................35 Table 4. Statistical summary (Friedman?s Chi Square Test) of efects of respirometry treatments (Table 1) on the percent time and bout frequency (bouts 5 min -1 ) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus) ......36 Table 5. Statistical summary (repeated-measures ANOVA) of efects of time (20:00- 6:00) on percent time and bout frequency (bouts 5 min -1 ) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus) in the presence of host sea anemone (Entacmaea quadricolor) .....................................................39 vi List of Figures Fig. 1. Flow-through respirometry setup used to measure dark oxygen consumption (!mol O 2 hr -1 ) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor) .......................................................................................28 Fig. 2. Representative plot of an oxygen meter reading during a flow-through respirometry experiment (unit treatment) on an anemonefish (Amphiprion bicinctus) and sea anemone (Entacmaea quadricolor) at Auburn University .......30 Fig. 3. Dark oxygen consumption (mean ?1 s.e.m.) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor) across respirometry treatments (Table 1) ...............................................................................................31 Fig. 4. Efects of rate of water flow on the dark oxygen consumption mean?1 s.e.m.) of the sea anemone Entacmaea quadricolor in flow-through respirometry ..........3 Fig. 5. Percent time (mean?1 s.e.m.) that anemonefish (Amphiprion bicinctus) engaged in five types of nocturnal behavior in experimental aquaria and in respirometry chambers ...........................................................................................34 Fig. 6. Bout frequencies (mean?1 s.e.m.) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus) in experimental aquaria and in respirometry chambers ...........................................................................................37 Fig. 7. Nocturnal and diurnal stroke frequencies (mean?1 s.e.m.) of dorsal and caudal fins of anemonefish (Amphiprion bicinctus) .........................................................38 1 CHAPTER I The benefits and ecophysiology of mutualistic symbioses on coral reefs BENEFITS OF CORAL REEF MUTUALISMS The high biodiversity on coral reefs is unparaleled in other marine habitats, and globaly is second only to that of tropical rainforests (Reaka-Kudla, 1997). Symbiotic interactions on coral reefs, such as the complex mutualistic asociations betwen sedentary cnidarian hosts and their symbiotic guests, contribute to the biodiversity acknowledged in coral reef systems (Davies, 1992). While the ecological importance of mutualisms is recognized, the extent of the benefits and the underlying biological proceses remain greatly unexplored. Most research on mutualisms betwen cnidarian hosts (i.e., corals and sea anemones) and symbiotic guests on coral reefs has focused on the direct benefits provided to the participants. One of the most visible benefits is mutual protection aforded through habitat and/or physical defense. Reef cnidarians are vulnerable to predation as a result of their sedentary nature (e.g., Porat and Chadwick-Furman, 2004). However, many corals and sea anemones function as structural habitat for obligate and/or facultative fish and crustacean guests that protect their hosts from predation. Pociloporid corals on Western Pacific coral reefs host obligate guard crabs (Trapezia and Tetralia sp.) that efectively deter predation on the coral host by the crown-of-thorns sea star (Acanthaster planci) (Pratchet et al., 2000). The Trapezia crabs drive the low frequency of pociloporid corals in the diet of A. planci (Pratchet, 2001). Similarly, on Indo-Pacific coral reefs, anemonefishes (family Pomacentridae) chase away butterflyfishes (family Chaetodontidae) that prey upon host sea anemone tentacles (Fautin, 1991; Fautin and 2 Alen, 1997; Porat and Chadwick-Furman, 2004), and the stinging nematocysts of host sea anemones protect anemonefishes from piscivores (Mariscal, 1970b). Although not as apparent as the protection of symbiotic partners, mutualisms also contribute to the nutrient dynamics on coral reefs. Coral reefs are considered to be productive oases amid oligotrophic marine deserts (Hoegh-Guldberg, 1999). The succes of coral reefs within highly unproductive tropical waters is largely atributed to the eficient use and recycling of esential nutrients (i.e., organic and inorganic nitrogen and carbon compounds) (Davies, 1992). Corals and sea anemones metabolize organic carbon (Muscatine, 1990; Biel et al., 2007) and nitrogen (Wang and Douglas, 1998; Wang and Douglas, 1999) produced by their endosymbiotic dinoflagelate microalgae, which then utilize the inorganic carbon and nitrogen from the metabolic byproducts of the cnidarian hosts (Furla et al., 2005; Venn et al., 2008). By recycling the metabolic wastes of their symbiotic partner, the union betwen coral reef cnidarians and microalgae facilitates coral reef formation over a broader range of environmental conditions than possible in the absence of the mutualism (Bruno et al., 2003). Ectosymbiotic guests of coral reef cnidarians (e.g., anemonefishes) also contribute to the tight nutrient cycling of mutualisms. Laboratory studies on anemonefish (Amphiprion bicninctus) and sea anemones (Entacmaea quadricolor) by Roopin et al. (2008) strongly indicate that sea anemones uptake amonia excreted by anemonefish, leading to increased nutritional benefits for the host (Roopin and Chadwick, 2009). Godinot and Chadwick (2009) additionaly suggested that anemonefish (A. bicinctus) supply phosphate to sea anemones (E. quadricolor), although sea anemone demand exceds anemonefish supply. Using 13 C- and 15 N-labeled isotopes, Cleveland et al. (2010) 3 confirmed that sea anemones (Heteractis crispa) uptake nitrogen and carbon compounds from anemonefishes (A. clarki, A. perideraion), and documented the subsequent uptake of carbon and nitrogen by microalgae within sea anemone tisue. Amonia recycling has also been suggested betwen anemoneshrimps (Periclimenes yucatanicus) and sea anemones (Condylactis gigantea) (Spotte, 1996). Nutritive benefits may be universal among intimate coral reef mutualisms; however, more research on the nutrient dynamics of similar symbiotic asociations on coral reefs is needed. The defensive and nutritive benefits of mutualisms betwen reef cnidarians and their symbionts drive a wide range of indirect ecological benefits. For example, the temperate coral Oculina arbuscula is an inferior competitor to the brown seaweds (e.g., Sargassum, Padina, and Dictyota) that dominate wel-lit reef structure off North Carolina, USA (Stachowicz and Hay, 1999). O. arbuscula host facultative omnivorous crabs (Mithrax forceps) that consume the algal competitors of the coral host. Consequently, colonies of O. arbuscula that hosted M. forceps grow 10x faster than colonies without M. forceps, and M. forceps grow faster on live corals than dead corals (Stachowicz and Hay, 1999). Similarly, in the Red Sea, stony corals (Stylophora pistilata) that host obligate damselfish (Dascylus marginatus) experience faster long term (>7 mo) growth and reproductive output than S. pistilata without damselfish guests (Liberman et al., 1995). The ecological benefits provided by coral reef mutualisms can extend beyond the symbiotic participants to other organisms. For example, increased nutrition to photosynthetic dinoflagelates within sea anemone tisues, such as the nitrogen provided by anemonefishes (Roopin et al., 2008; Roopin and Chadwick, 2009; Cleveland et al., 2010), increases rates of microalgal photosynthesis and mitotic division (Fit and Cook, 4 2001). The microalgae can then supply more organic compounds to sea anemones and increase the growth and reproduction of their host (Hoegh-Guldberg and Smith, 1989; Fit and Cook, 2001). Further, anemonefish residency enhances sea anemone asexual reproduction (Holbrook and Schmit, 2005), growth, survivorship (Porat and Chadwick- Furman, 2004; Holbrook and Schmit, 2005), dinoflagelate abundance, and tisue regeneration (Porat and Chadwick-Furman, 2004) . The microalgae- and anemonefish- induced nutrient enhancements increase the ecological performance of sea anemones which then cover a greater net area on coral reefs (Schmit and Holbrook, 2003). Of the coast of Moorea in French Polynesia, the increased area of sea anemone cover produced by the tripartite mutualism of sea anemones (Heteractis magnifica), microalgae (Symbiodinium sp.), and anemonefish (A. chrysopterus) enhances biodiversity on coral reefs, by providing more habitat for three-spot damselfish (Dascylus trimaculatus), which are outcompeted by anemonefish when sea anemones cover les net area (Schmit and Holbrook, 2003). Mutualisms on coral reefs further benefit the reef community by connecting pelagic and litoral food webs. Planktivorous fishes, many of which are facultative or obligate guests of reef cnidarians, transfer nutrients from the open water to the benthos via nitrogen- and phosphorous-rich feces (Pinnegar and Polunin, 2005; Holbrook et al., 2008) and excretions (Porat and Chadwick-Furman, 2005; Roopin et al., 2008; Godinot and Chadwick, 2009; Roopin and Chadwick, 2009; Cleveland et al., 2010). By serving as net importers of nutrients to coral reef organisms, the ectosymbiotic guests of many marine mutualisms contribute to the productivity and biodiversity asociated with coral reefs. 5 Mutual protection and nutrient recycling by symbiotic partners serve as esential mechanisms leading to enhanced habitat for the partners themselves, as wel as other organisms on coral reefs. Currently, we have only scratched the surface of the complexity and importance of facilitative interactions, such as mutualistic symbioses (Bruno et al., 2003). Continued research on the benefits of mutualisms, as wel as the underlying biological mechanisms, wil lead to a more complete understanding of the ecology of tropical coral reefs. ECOPHYSIOLOGICAL ADAPTATIONS TO ENVIRONMENTAL STRESS The sedentary nature of the organisms involved in many marine symbioses (e.g., mutualisms betwen cnidarian hosts and obligate guests) may make them more susceptible than free-ranging organisms to environmental stresors on coral reefs. For example, the oxygen concentration ([O 2 ]) surrounding coral reefs is highly variable, especialy during low tide and at night. The [O 2 ] of the ambient water around coral reefs off Heron Island on the Great Barrier Reef decreases substantialy when the lagoons are cut off from the ocean by the reef crest during low tide (Kinsey and Kinsey, 1967). Additionaly, phase shifts from coral- to algal-dominated reefs reduce the ambient [O 2 ] above the reef structure, most likely by increasing labile disolved organic mater and subsequent microbial metabolism (Niggl et al., 2010; Wild et al., 2010). On a microhabitat scale, the [O 2 ] of the water among coral branches becomes severely hypoxic at night, as documented on reefs off Lizard Island at the Great Barrier Reef (Nilson et al., 2004) and in the northern Red Sea at Eilat (Goldshmid et al., 2004). Lastly, reef cnidarians experience diel cycles of O 2 availability within their tisues, which become hyperoxic during the daytime due to microalgal photosynthesis, and hypoxic at night due 6 to the respiration of both microalgae and host (Shashar et al., 1993; Richier et al., 2003). Even when the [O 2 ] is normoxic (70-100% saturation), local flow variability can restrict the availability of O 2 and other substances to many sedentary coral reef organisms. Water flow is one of the most important abiotic factors afecting the growth and survivorship of sesile marine invertebrates (Sebens et al., 2003), which generaly lack the ability to self-regulate the mas transfer of disolved particles across their tisues (Shick, 1990). The flow-induced reduction or elimination of the difusive boundary layer surrounding sedentary organisms notably enhances gas exchange (Paterson and Sebens, 1989; Paterson et al., 1991; Bruno and Edmunds, 1998; Sebens et al., 2003; Fineli et al., 2006; Schutter et al., 2010), nutrient uptake (Stambler et al., 1991; Atkinson and Bilger, 1992; Leser et al., 1994; Thomas and Atkinson, 1997), prey capture (Helmuth and Sebens, 1993; Sebens, 1997; Sebens et al., 1998), and debris removal (Nugues and Roberts, 2003; Box and Mumby, 2007). The waters above coral reefs are generaly asociated with wave-induced oscilatory flow (Helmuth and Sebens, 1993), which reduces the thicknes of the difusive boundary layer that limits gas and nutrient transfer in sedentary invertebrates (Reidenbach et al., 2006). However, coral reefs are topographicaly complex and water flow at the surface of the reef structure is likely restricted. For example reef crests and flats experience high flow regimes from waves and tides, while protected lagoons, fore reefs, and reefs at greater depths are sheltered from wave action and experience weaker water flow regimes (Sebens, 1997; Sebens et al., 1997). Moreover, within a microhabitat, water flow can be reduced or diverted by dense colonies of structuraly complex organisms (e.g., reef-building corals) (Sebens, 1997). Similarly, the flow regime around crevice dwelers (e.g., sea anemones) may be obstructed by protrusions of the reef structure. 7 The variable [O 2 ] and water flow encountered by sedentary invertebrates on coral reefs threatens the residency of their symbiotic guests. However, symbiotic guests employ unique ecophysiological adaptations to maintain the benefits of the mutualisms. At Lizard Island on the Great Barrier Reef, coral-dweling fishes (Gobiidae, Scorpaenidae) that take refuge among coral branches at night exhibit substantial hypoxia tolerance and air breathing abilities (Nilson et al., 2004; Nilson and Ostlund-Nilson, 2004; Nilson et al., 2007b). Further, the relative expresion of these adaptations is correlated to species-specific habitat preferences, such that fishes among the branches of coral colonies near the water surface (i.e., more likely to become air-exposed) posses greater air breathing abilities than fishes among coral colonies at greater depths (Nilson et al., 2007b). Similarly, low rates of water flow on coral reefs at Eilat in the northern Red Sea contribute to hypoxic conditions among coral branches at night (Goldshmid et al., 2004). Three species of damselfishes that reside among the corals actively aerate their hosts by beating their fins at stroke frequencies 2x faster than during diurnal swiming. This modulation of the hydrodynamic conditions among coral branches efectively restores [O 2 ] to that of the ambient water (Goldshmid et al., 2004). Cnidarian hosts and their endosymbiotic dinoflagelates have evolved physiological adaptations to oxidative stres within host tisue. The hyperoxic environment within cnidarian tisues during the day can increase the abundance of harmful O 2 derivatives (i.e., reactive oxygen species, ROS) that cause severe celular damage (Li and Jackson, 2002; Lushchak and Bagnyukova, 2006). To protect against oxidative stres, corals and sea anemones posses antioxidant systems, such as the superoxide dismutase enzyme (SOD), which reduces ROS to les harmful derivatives (Li and Jackson, 2002). Richier et al. (2005) reported that symbiotic cels of the sea anemone 8 Anemonia viridis have higher SOD activity and diversity than aposymbiotic cels. Furthermore, within symbiotic sea anemone tisue (Anthopleura elegantisima), SOD activity positively correlates with chlorophyll concentrations (proxy for endosymbiotic microalgae) (Dykens and Shick, 1984). Coral reef mutualisms employ a myriad of ecophysiological adaptations to buffer the symbiotic participants against environmental stresors, such as O 2 variability and water flow paterns. The evolution of esential physiological and behavioral connections betwen mutualistic partners underpins the nutritive and ecological advantages asociated with these facilitative interactions. Further investigation of the ecophysiology and behavior of mutualistic interactions wil clarify how coral reef organisms adapt to changes in their naturaly variable environment, as wel as to the human- and climate- induced changes currently afecting the world?s coral reefs. The goal of this thesis is to explore potential nocturnal benefits and underlying ecophysiological components of the anemonefish-sea anemone mutualism. As a model, I used the two-band anemonefish (Amphiprion bicinctus) and its preferred host in the Red Sea, the bulb-tentacle sea anemone (Entacmaea quadricolor) (Chadwick and Arvedlund, 2005). While previous investigations demonstrate that nocturnal physiology and asociated behaviors are esential components to the maintenance of similar symbiotic asociations (Goldshmid et al., 2004; Nilson et al., 2004), litle is known about the nocturnal interactions betwen anemonefishes and sea anemones. Further, the Red Sea anemonefish-sea anemone mutualism is a practical model to addres the metabolic and behavioral interactions of mutualisms on coral reefs at night because of the intimate nature of the partners at night, in which anemonefish spend the entire night nestled among the sea anemone tentacles (Alen, 1974; Fautin and Alen, 1997). 9 CHAPTER II Anemonefish behavior at night modulates sea anemone oxygen consumption SUMMARY Mutualisms betwen sessile cnidarian hosts (i.e., corals and sea anemones) and their fish guests involve complex metabolic and behavioral components, especialy at night when the fishes rest in asociation with the host. While some aspects of the mutualism betwen anemonefishes and giant sea anemones have been wel examined, the benefits derived by the partners at night, and the underlying biological proceses involved, remain greatly unexplored. The present study investigated the metabolic and behavioral components of the anemonefish-sea anemone mutualism at night, using two- band anemonefish (Amphiprion bicinctus) and bulb-tentacle sea anemones (Entacmaea quadricolor). The net dark oxygen consumption (VO 2 , !mol O 2 hr -1 ) of pairs (fish+anemone) were measured separately, together as a unit, and together but separated by a mesh screen that prevented physical contact. The net VO 2 of the symbionts when incubated together was 1.4x higher than that of the partners in isolation of each other, or separated by a mesh barrier. The symbiotic asociation betwen anemonefish and sea anemones elevates the VO 2 of at least one partner, and physical contact betwen partners is needed to induce the metabolic elevation. The VO 2 of isolated sea anemones increased with water flow until 2 cm s -1 , after which VO 2 remained constant up to 8 cm s -1 . Using infrared video, I observed the nocturnal behavior of anemonefish in the absence and 10 presence of sea anemone hosts to categorize the behavioral repertoire of anemonefish at night and to discern the efect of sea anemone residency on anemonefish behavior. The percent time and bout frequency of several types of anemonefish behavior (i.e., fanning, wedging, switching) increased significantly when the host sea anemone was present. Based on the enhancement of flow-modulating behaviors by anemonefish when they occurred with sea anemones, and the increase of sea anemone VO 2 with flow, I conclude that anemonefish behavior at night likely oxygenates host anemones and augments metabolism in both partners. INTRODUCTION The symbiosis betwen anemonefishes and giant sea anemones on Indo-Pacific coral reefs is one of the most conspicuous and endeared mutualisms in marine environments, and the benefits provided to the participants have been heavily researched. A fundamental benefit to this symbiotic asociation is mutual protection against predation; anemonefishes chase away butterflyfishes (Chaetodontidae) that prey on sea anemone tentacles (Fautin, 1991; Fautin and Alen, 1997; Porat and Chadwick-Furman, 2004), and the nematocysts within sea anemone tisue ward off piscivorous predators of anemonefishes (Mariscal, 1970b). Moreover, the stinging tentacles of sea anemones provide a protective veil behind which anemonefishes lay their benthic egg clutches (Moyer, 1976; Moyer and Stene, 1979; Fautin, 1991; Arvedlund et al., 2000). In addition to protection from predators, anemonefish residence provides nutritional benefits that enhance host sea anemone growth, reproduction, and survivorship (Porat and Chadwick-Furman, 2004; Holbrook and Schmit, 2005; Porat and 11 Chadwick-Furman, 2005). Through excretion, anemonefishes provide inorganic nitrogen, phosphate, and carbon to the endosymbiotic dinoflagelates within sea anemone tisue (Porat and Chadwick-Furman, 2005; Roopin et al., 2008; Godinot and Chadwick, 2009; Roopin and Chadwick, 2009; Cleveland et al., 2010). Nutrient fertilization of sea anemones by resident anemonefishes and their microalgae contributes to increased rates of sea anemone regeneration and proliferation, which in turn enhances the net area of sea anemone habitat for anemonefishes and other facultative fish residents that are outcompeted by anemonefishes when sea anemones are smaler and les abundant (Schmit and Holbrook, 2003). Despite over 100 yr of research on the symbiotic interactions betwen sea anemones and their anemonefishes, the net benefits and underlying biological proceses of the mutualism at night remain greatly unexplored. This is particularly surprising given the intimate nature of the mutualism at night. During the day, anemonefishes swim meters above host sea anemones to forage for zooplankton in the water column, but spend the entire night among sea anemone tentacles for rest and protection (Alen, 1974; Fautin and Alen, 1997). Further, intimate symbioses betwen coral reef cnidarians (i.e., corals and sea anemones) and fishes are common in marine systems, and the unique metabolic and behavioral adaptations of these symbionts at night are esential to maintain symbiotic benefits during variable nocturnal conditions. At Lizard Island on the Great Barrier Reef, the oxygen availability among living coral branches can drop severely at night (Nilson et al., 2004), and coral-dweling fishes that take refuge among coral branches exhibit substantial hypoxia tolerance and air breathing abilities (Nilson et al., 2004; Nilson and 12 Ostlund-Nilson, 2004; Nilson et al., 2007b). Moreover, in the northern Red Sea, hypoxic conditions at night are compounded by reduced water flow over reef structure (Goldshmid et al., 2004), and damselfishes that reside among coral branches engage in sleep-swiming behaviors to increase oxygen availability among coral branches (Goldshmid et al., 2004). Sea anemones, like many sesile marine invertebrates on coral reefs, are largely unable to self-modulate the bulk flow of sea water across their tisues, and thus rely on ambient water flow for the mas transfer of esential gases and nutrients (Sebens, 1987). Currently, the potential for anemonefishes to modulate the flow regime surrounding host sea anemones has only been hypothesized (Mariscal, 1970b; Alen, 1974; Fautin, 1991; Porat and Chadwick-Furman, 2004, 2005). Further, qualitative observations of in situ nocturnal behavior suggest that anemonefishes are generaly inactive at night (Alen, 1974). Regardles, the variable oxygen concentrations (Kinsey and Kinsey, 1967; Niggl et al., 2010; Wild et al., 2010) and flow rates (Sebens and Done, 1992; Helmuth and Sebens, 1993; Johnson and Sebens, 1993) of coral reefs demonstrate the potential importance of nocturnal interactions betwen sea anemones and anemonefishes. Like similar symbiotic networks on coral reefs, the metabolic and behavioral asociation of sea anemones and anemonefish at night may be an esential aspect of this mutualism. In the present study, I explored the nocturnal interactions betwen the bulb- tentacle sea anemone and its obligate fish guest in the northern Red Sea, the two-band anemonefish. I measured the efects of the symbiotic asociation on the metabolism (oxygen consumption) of the symbionts at night, and the efect of water motion on the gas exchange of sea anemones. Further, I conducted nocturnal surveilance on 13 anemonefish to characterize fish activity at night and determine the efect of fish behaviors on the nocturnal metabolism of the mutualism. MATERIALS AND METHODS Animal collection and maintenance At the Marine Science Station (MS) at Aqaba, Jordan, two-band anemonefish (Amphiprion bicinctus, R?ppel 1828), 7-11 cm fork length [FL], and bulb-tentacle sea anemones (Entacmaea quadricolor, R?ppel and Leuckart 1828), 11-16 cm tentacle crown diameter [TCD]), were obtained in June 2010 from shalow coral reefs adjacent to the MS (29?27.250" N, 34?58.359" E). Al animals were distributed haphazardly among flow-through aquaria (80 L) circulating seawater pumped from the Red Sea. Aquaria received a 12:12 light:dark photoperiod using halogen lighting (Aqua-Medic, Fort Collins, CO, USA). Fish were fed Formula One Marine Flake (Ocean Nutrition, San Diego, CA, USA) daily, and anemones were hand fed frozen fish (Atherinomorus spp.) wekly. At Auburn University (AU) in Alabama, USA, A. bicinctus (10-12 cm FL) were obtained from Oceans, Reefs & Aquariums (Florida, USA) in 2006, and E. quadricolor (10-18 cm TCD) were obtained from SunPet, Inc. (GA, USA) and from the New England Aquarium (MA, USA) during August 2010-August 2011. At AU, animals were distributed haphazardly among 150 L glas aquaria (two fish, 1 anemone tank -1 ) circulating artificial seawater and receiving a 12:12 light:dark photoperiod using high- output fluorescent lighting (Sunlight Supply, Inc., Pompano, FL, USA). Fish were fed a mixed diet of Formula One Marine Pelet (Ocean Nutrition, San Diego, CA, USA) and 14 frozen foods (Mysid Shrimp and Emerald Entr?e; San Fransisco Bay Brand, Inc., Newark, CA, USA) daily, and anemones were hand-fed raw shrimp wekly. All animals appeared to be in good physiological condition prior to and during experimental use. Nocturnal oxygen consumption patterns The metabolic efects of the symbiotic asociation betwen fish and anemones were asesed at both the MS and AU using flow-through respirometry (Fig. 1). Animals were placed in custom-made acrylic chambers connected to a recirculating seawater reservoir (150 L, flow=1.0?0.1 cm s -1 ). Two Clark-type oxygen (O 2 ) electrodes (Strathkelvin Instruments, Ltd., North Lanarkshire, Motherwel, Scotland) were atached to the inflow and outflow ports of the chamber. To ensure a uniform O 2 concentration ([O 2 ]) within the chamber, a stir bar covered by a smal mesh cage in the bottom of the chamber gently mixed the seawater. To atain a standard metabolic state (post-absorptive and quiescent), animals were starved for !24 hr prior to experimental use, and al experiments were conducted under dark conditions by draping darkroom curtains over the chamber. Dark conditions also simulated nighttime, when anemonefishes reside among sea anemone tentacles for rest and protection (Alen, 1974). Animals were alowed to aclimate within the chamber until standard metabolism was reached (3-6 hr), at which point 20 min of dark O 2 consumption (VO 2 , ?mol O 2 hr -1 ) was measured (1 measurement 6 s -1 ) using a two-channel O 2 meter (Strathkelvin Instruments, Ltd.). The time at which animals were reduced to standard metabolism was determined visualy, based on stable VO 2 on the O 2 meter display (Fig. 2). 15 Dark VO 2 of symbiotic pairs (1 fish+1 anemone) was measured in three experimental treatments in random order (Table 1). The control treatment measured the dark VO 2 of the fish and anemone separately, and the two rates were added together for a single net dark VO 2 . The unit treatment measured the net dark VO 2 of the fish and anemone within the chamber as a pair. Lastly, the mesh treatment measured the net dark VO 2 of the fish and anemone within the same chamber, but physicaly separated by a flow-permeable screen (1 mm mesh) that permited visual and chemical interaction but prevented any physical contact betwen partners. A subsample of pairs (N=6) was subjected to the mesh treatment twice; once with the anemone in the bottom of the chamber such that incoming seawater pased by the anemone before reaching the fish (mesh 1 ), and again with the fish in the bottom of the chamber such that incoming seawater first pased by the fish before reaching the anemone (mesh 2 ). The mean VO 2 diference betwen mesh 1 and mesh 2 was compared to ases the possibility that (a) basal nitrogen excretion by fish (Roopin et al., 2008) efects anemone VO 2 (mesh 2 >mesh 1 ), or (b) fish detection of anemone scent efects fish VO 2 (mesh 1 >mesh 2 ). Dark VO 2 of animals at the MS was measured within 5-10 d of collection. Due to time and collection limitations, VO 2 of MSS animals was measured betwen only the control and unit treatments, and a single anemone was used with al fish examined (N=6). Dark VO 2 of animals at AU was measured after !2 yr (fish) or 4-5 wk (anemones) of maintenance in laboratory aquaria. Dark VO 2 of AU animals was measured across al three treatments, and each pair (N=12) consisted of a unique fish and anemone. 16 Effects of water motion on sea anemone respiration at night The efects of water motion on dark VO 2 of anemones (N=8) were asesed at AU using flow-through respirometry (Fig 1). Anemones were transferred to the respirometry chamber and aclimated for 3-6 hr, as described above. In contrast to the low uniform water flow rate (1.0?0.1 cm s -1 ) used in the above experiments, this experiment exposed anemones to increasing levels of water motion by varying the speed of the caged stir bar within the chamber. Water flow levels generated by the stir bar were estimated with a Flo-Mate 2000 portable flow meter (Marsh-McBirney, Inc., Frederick, MD, USA). Anemones were exposed to flow speeds (N=9, 0.5?8.0 cm s -1 ) commonly encountered by A. bicinctus and E. quadricolor at coral reef field sites in the northern Red Sea (Goldshmid et al., 2004). At each flow speed, 10 min of VO 2 was measured (1 measurement 6 s -1 ). Maximum O 2 consumption (VO 2 max ) and maximum change in O 2 consumption across flow regime (VO 2 dif ) were calculated. Anemonefish nocturnal behavior Observations in experimental aquaria Potential efects of fish behavior on net dark VO 2 of the symbiotic partners were investigated at AU by observing the nocturnal activity of the fish. Six symbiotic pairs (fish and anemone) were selected randomly from laboratory stock (N=13). In random order, each fish was observed under two treatments: anemone absent and present. For each treatment, the individual(s) was transferred to experimental aquaria (150L) and alowed to aclimate for 24 hr. Eight 20-min video segments (1 hr apart) were recorded throughout the night (20:00-6:00) per fish per treatment using a Sony DCR-SR68 IR 17 video camera (Sony, San Diego, CA, USA) and IRLamp6 LED infrared lamps (Wildlife Engineering, Carlisle, PA, USA). From each 20 min segment, a random 5 min subsegment (8 subsegments x 5 min each = 40 min total per fish per treatment) was observed to (a) categorize the nocturnal behavioral repertoire of the fish, and (b) determine efects of anemone presence on fish behavior at night. The percent time and bout frequency of five distinct fish behaviors were measured: fanning, wedging, switching, swiming, and no motion. When fanning, fish are motionles, aside from continuous pectoral fin strokes. When wedging (1-2 s), fish use using both caudal and pectoral fins to forcefully wiggle deeper into the anemone tentacle crown (or into the bottom substrate, in the absence of an anemone). When switching (1-2 s), fish change orientation (usualy 180?) while wedging. During swiming events (2-8 s), fish hover in one spot or slowly move around the aquarium, usualy to forage. Lastly, during periods of no motion, fish lay completely stil on the substrate or among anemone tentacles. Fish fin stroke frequencies (pectoral and caudal, reported as strokes per 5 s) were compared betwen the two treatments using five random 5 s segments (25 s per fish per treatment) selected from 24:00-3:00. Nocturnal fin stroke frequencies were compared to diurnal fin stroke frequencies, using five random 5 s video segments collected for the same fish from 12:00-15:00. Observations in respirometry chambers To ases variation in fish behavior when in experimental aquaria (above) versus in the smaler respirometry chambers, IR video of a random subsample of the fish and 18 anemones (N=6) during the four respirometry treatments (control, unit, mesh 1 , mesh 2 ) was recorded during the 20 min of VO 2 used for respirometry analysis. From each 20 min segment, one random 5 min subsegment was observed (fish -1 treatment -1 ) for behavioral analysis. Similar to the behavioral analysis in the experimental aquaria, the percent time and bout frequency of the five distinct fish behaviors were analyzed. Data analysis Statistical analyses were conducted using SAS 9.2 (Cary, NC, USA). Diferences in fish and anemone VO 2 among treatments, and the efects of water motion (semi-ln transformed) on anemone VO 2, were examined using one-way repeated-measures analysis of variance (rmANOVA). The VO 2 of mesh 1 and mesh 2 were compared using a paired t-test. Efects of treatment (i.e., anemone absent, present) and time of night on fish behavior within the experimental aquaria (percent time and bout frequency) and the efect of anemone presence on fish fin stroke frequency (pectoral and caudal) were analyzed with one- or two-way rmANOVA. The efects of treatment on fish behavior within the respirometry chamber were analyzed with the nonparametric Friedman?s Chi Square test. When appropriate, post hoc multiple pairwise comparisons were analyzed using Tukey?s studentized range (HSD) tests. For rmANOVA models, where the asumption of sphericity was not met, Greenhouse-Geiser approximations were used. The significance level for al analyses was set at P<0.05. Al reported values are means?1 s.e.m. 19 RESULTS Nocturnal oxygen consumption patterns The net dark oxygen consumption (VO 2 ) of fish and anemones at the MS was significantly higher (1.4x) when measured together within the same chamber (unit) than the summed VO 2 of both partners in isolation (control) (Fig. 3A, Table 2A). Similarly, at AU, the net VO 2 of anemonefish and anemone partners together (unit) was significantly higher (1.4x) than both the summed VO 2 of the partners in isolation (control) and the VO 2 of the partners separated by a mesh barrier (mesh) (Fig. 3B, Table 2B). The positions of the anemonefish and anemone in the chamber during the mesh treatments (mesh 1 , mesh 2 ) had no significant efect on the net VO 2 of the symbionts at night (Table 2C). Mean VO 2 dif betwen the unit treatment and control or mesh treatments was 70.96?10.03 or 83.34?14.81 !mol O 2 hr -1 , respectively. Effects of water motion and sea anemone oxygen consumption at night Anemone VO 2 increased significantly when exposed to water flow rates betwen 0.5-2.0 cm s -1 , then reached an asymptote at 2.0 cm s -1 (rmANOVA, F=41.32, P<0.0001) (Fig. 4). VO 2 max for anemones was 111.53?13.32 !mol O 2 hr -1 . Maximum VO 2 dif across flow regimes (mean diference betwen VO 2 at 0.5 and 3.0 cm s -1 ) was 23.69?2.64 !mol O 2 hr -1 . Anemonefish nocturnal behavior In the experimental aquaria, regardles of treatment (i.e., anemone absent or present), fish spent !98% of the night in a single location. When the anemone was 20 present, it always served as the singular location. In the absence of the anemone, the fish rested against a rock or the aquarium wal. In the respirometry chambers, fish spent !80% of the night in a single location. When the anemone was acesible (unit), it always served as the singular location. Percent time spent performing each behavior Within the experimental aquarium, anemone presence had no significant efect on the percent time fish spent fanning, swiming, and not moving. However, time spent wedging and switching increased 11x and 47x, respectively, when the anemone was present (Fig. 5A, Table 3A). Similarly, within the respirometry chambers, treatment (control, unit, mesh 1 , mesh 2 ) had no significant efect on the percent time fish spent fanning, swiming, or not moving. When the anemone was acesible (unit treatment), fish spent a significantly higher percent time wedging and switching (20x and 2.5x, respectively) than during treatments when the anemone was absent or inacesible (control, mesh 1 , mesh 2 ) (Fig. 5B, Table 4A). Frequencies of behaviors In the experimental aquarium, bouts of fanning, wedging, and switching were more frequent when the anemone was present (1.6x, 20x, and 36x, respectively), while the frequency of swiming and periods of no motion were unafected (Fig. 6A, Table 3B). In the respirometry chambers, fish fanned, wedged, and switched more frequently (3x, 6x, and 30x) when the anemone was acesible (unit) than when inacesible (control, mesh 1 , mesh 2 ) (Fig. 6B, Table 4B). Further, during the mesh 1 treatment (fish 21 downstream of anemone), fish engaged in switching more frequently than during the control and mesh 2 treatments. Within the experimental aquarium, anemone presence did not afect the pectoral fin stroke frequency of fish (rmANOVA, F=0.32, P=0.32) (Fig. 7). Further, both nocturnal fin stroke frequencies (i.e., anemone absent, present) were significantly lower than diurnal pectoral stroke frequencies (rmANOVA, F=3.32, P=0.12). Caudal fin stroke frequency, however, was significantly higher when the anemone was present (rmANOVA, F=14.29, P=0.013) and was comparable to the diurnal caudal fin stroke frequency (rmANOVA, F=0.01, P=0.95). Effect of time of night on anemonefish behavior Time of night had no significant efect on the percent time fish engaged in any of the five behaviors exhibited in the experimental aquarium, regardles of treatment (i.e., anemone absent, present) (Table 5A). However, fish engaged in wedging at a substantialy higher frequency during the first time segment (20:00?20:20) than during the remaining seven time segments (20:20?6:00) (rmANOVA, F=4.12, P=0.002) (Table 5B). DISCUSSION The symbiotic asociation betwen anemonefish (A. bicninctus) and sea anemones (E. quadricolor) increases the net dark O 2 consumption (VO 2 ) of one or both partners at night. Additionaly, physical contact betwen the partners is needed to produce metabolic elevation. The relative importance of each partner?s contribution to this metabolic 22 elevation could not be discerned because of the technical constraints of measuring the VO 2 of multiple species incubated together. However, it is likely that fish behavior elevates anemone VO 2 when the partners are together, because (a) artificialy increasing water motion elevates anemone VO 2 , and (b) anemone presence increases the expresion of flow-modulating behaviors by the fish. It is unlikely that anemones are wholly responsible for the elevated VO 2 when the partners are together. The mean VO 2 dif betwen control and unit treatments (70.96?10.03 ?mol O 2 hr -1 ), when added to the mean VO 2 of isolated sea anemones (control treatment,108.28?9.44 ?mol O 2 hr -1 ), is substantialy higher than the mean VO 2 max of anemones during the flow experiments (114.21?13.89 ?mol O 2 hr -1 ). Thus, the metabolic elevation observed during the unit treatment is too large to be achieved by the anemone alone, and increased fish metabolism is also required to explain the total increase in VO 2 when the partners are together. Contrary to previous studies that report that anemonefishes, like most pomacentrids, remain generaly inactive at night (Alen, 1974), we demonstrate that anemonefish spend the majority of the night in some form of localized motion. Further, some anemonefish behaviors appear to be tailored specificaly toward interactions with sea anemone hosts (i.e., wedging and switching). Within both the experimental aquaria and the respirometry chamber, anemonefish spent significantly more time wedging and switching when resting among sea anemone tentacles than when alone. Increased instances of wedging and switching by anemonefish potentialy (a) elevated the energy expenditure of the anemonefish through increased activity and (b) increased sea anemone gas exchange through enhanced ambient water flow. Together, these efects provide a 23 likely explanation for the increased net dark VO 2 of the anemonefish and sea anemone partners during the unit treatment. Bouts of wedging and switching, though brief in duration, involve rapid caudal and pectoral fin movement, and forcefully propel the fish deeper into the anemone tentacle crown. These vigorous behaviors rely heavily upon the caudal fin and posterior musculature, and likely require more energy than the other localized movements. For example, bouts of fanning involve no fish movement aside from alternating pectoral fin strokes, and swiming primarily involves simultaneous pectoral rowing. Further, fish change behaviors at a substantialy higher rate when anemones are present than when not. High levels of activity at night have been documented for other marine fishes that rely on symbiosis with sedentary invertebrates on coral reefs. Slep-swiming behaviors have been documented in three species of damselfishes (Dascylus marginatus, D. aruanus, Chromis viridis) that shelter among the branches of stony corals at night (Goldshmid et al., 2004). During slep swiming, the damselfishes move among coral branches at fin strokes frequencies 2x higher than during diurnal activities. Enhanced activity by anemonefish among sea anemone tentacles also afects the cnidarian host. The wedging and switching behaviors of anemonefish clearly enhance the hydrodynamic conditions surrounding their host, as evidenced by the neutraly buoyant tentacle crown of sea anemones, which moves pasively with ambient water flow. The rapid and forceful bouts of wedging and switching by the anemonefish likely disrupt the difusive boundary layer surrounding sea anemone tisue, and enhance gas exchange with ambient water. Ultimately, anemonefish facilitate regular bouts of tentacle movement 24 over most of the sea anemone tentacle crown that is noticeably greater than tentacle movement induced by ambient water flow alone. Anemonefish-induced water flow among sea anemone tentacles could have pronounced efects on the physiology and biology of the host. For example, in the present study, sea anemone dark VO 2 increased with water flow, suggesting sea anemones are potentialy flow-limited. Similarly, Paterson and Sebens (1989) demonstrated that reduced difusive boundary layer thicknes increased gas exchange in the temperate sea anemone Metridium senile. Oxygenation of host sea anemone through anemonefish behavior could be especialy important in the light of recent research that implicates O 2 limitation as a major abiotic selection presure on coral reefs (Goldshmid et al., 2004; Nilson et al., 2007a; Niggl et al., 2010; Wild et al., 2010). Similar to the benefits provided to sea anemones by increased water flow, anemonefish residency catalyzes sea anemone growth, reproduction, and survivorship (Porat and Chadwick-Furman, 2004; Holbrook and Schmit, 2005; Porat and Chadwick- Furman, 2005). Some benefits are directly atributed to the nutrient contributions from anemonefishes to their host sea anemones and endosymbiotic dinoflagelates, in the form of phosphorous (Godinot and Chadwick, 2009), nitrogen (Porat and Chadwick-Furman, 2005; Roopin et al., 2008; Roopin and Chadwick, 2009; Cleveland et al., 2010), and carbon (Cleveland et al., 2010). The results of the present study suggest that anemonefishes likely aid in the eficient uptake of nutrients and disolved gases, via forced convection of ambient seawater. Beyond the nocturnal paterns presented here, anemonefish-induced flow modulation also could have important diurnal efects. While anemonefishes generaly 25 spend most of the daylight hours in the water column above their host sea anemones (Alen, 1974; Fautin and Alen, 1997), periodic ?bathing? forays by anemonefishes back to their sea anemones (Alen, 1974) can flush the difusive boundary layer surrounding the host and increase primary production within sea anemone tisue. Positive efects of enhanced water flow on intracelular primary production have been documented for other reef cnidarians, including sea anemones (Paterson and Sebens, 1989) and reef-building corals (Paterson and Sebens, 1989; Paterson et al., 1991; Bruno and Edmunds, 1998; Sebens et al., 2003; Fineli et al., 2006). Moreover, anemonefish-induced flow modulation may clear sediments and algae from sea anemone tentacles (Nugues and Roberts, 2003; Box and Mumby, 2007), and flush metabolic wastes, such as harmful O 2 species, that can acrue within sea anemone tisues (Lushchak and Bagnyukova, 2006). The enhanced removal of free O 2 radicals by anemonefish could buffer sea anemones against bleaching, or expedite recovery after a bleaching event (Nakamura and van Woesik, 2001; Nakamura et al., 2003). Physical interaction betwen anemonefishes and sea anemones appears to be required for increased sea anemone gas exchange; however, chemical compounds released by sea anemones may play a role in initiating the anemonefish behaviors responsible for the elevation in net VO 2 of the partners when together. For example, anemonefish engaged in switching behaviors significantly more frequently when positioned downstream of sea anemones (mesh 1 treatment) than upstream (mesh 2 treatment). Further, an indication of the same patern was observed for the percent time anemonefish spent fanning, but the diference was not significant. Sea anemone chemical compounds directly influence the recruitment and recognition behaviors of 26 anemonefishes toward host sea anemones (Murata et al., 1986; Arvedlund et al., 1999); however, the extent to which sea anemone chemical cues influence anemonefish behavior at night has yet to be discerned. While it is clear that anemonefish behavior at night modulates the hydrodynamic conditions surrounding host sea anemones, it is unclear if flow modulation is the intended purpose of these behaviors. In both French Polynesia and the Red Sea, the number and size of resident anemonefishes correlate positively with sea anemone body size (Holbrook and Schmit, 2005; Porat and Chadwick-Furman, 2005). It is possible that anemonefish wedging and switching stimulate sea anemones to alter their morphology (e.g., expand). If anemonefish behavior promotes sea anemone expansion, this proces may contribute to increased sea anemone VO 2 by exposing more surface area for gas exchange. Alternatively, anemonefishes kept in captivity without sea anemone hosts occasionaly ?bathe? in airstream bubbles and stringy algal tufts (Mariscal, 1970a). As such, the tactile stimulation that anemonefishes receive from sea anemone tentacles may be beneficial to anemonefish wel being (Mariscal, 1970a). More research is needed to clarify the factors that enhance the expresion of certain behaviors (i.e., wedging and switching) when anemonefishes reside among sea anemone tentacles. My findings demonstrate that the asociation betwen anemonefishes and sea anemones can elevate the VO 2 of the symbionts at night. Also, I observed that anemonefish activity is afected by sea anemone presence, and certain behaviors (i.e., wedging and switching) modulate water flow among sea anemone tentacles and appear to increase sea anemone gas exchange. It is important to note that wild individuals of A. bicinctus and E. quadricolor in the Red Sea can be much larger than the specimens used 27 in the present study (Chadwick and Arvedlund, 2005). Further, in the Red Sea, E. quadricolor usualy host multiple anemonefish, including mated pairs of A. bicinctus plus 0-3 juveniles (Chadwick and Arvedlund, 2005). The efects of anemonefish size, quantity, and social hierarchy on their nocturnal behavior in the wild are currently not wel understood. Regardles, these results provide foundational evidence of anemonefish- induced flow modulation of sea anemone hosts, a previously debated benefit of this mutualism. Further, this study documents the metabolic consequences to both partners of anemonefish behavior at night, and thus joins a growing body of knowledge indicating the importance of ecophysiological underpinnings to the ecological advantages asociated with symbiotic asociations on coral reefs. 28 Fig. 1. Flow-through respirometry setup used to measure dark oxygen consumption (!mol O 2 hr -1 ) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor). Arrows indicate water flow direction (1.0?0.1 cm s -1 ). Numbers 1 and 2 depict inflow and outflow oxygen electrodes, respectively. Caged-enclosed stir bar (A) and magnetic stir plate (B) are displayed. 29 Table 1. Flow-through respirometry treatments used to ases the efects of symbiotic interactions on dark oxygen consumption (VO 2 ) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor). Se Fig. 1 for details of chamber setup. (1) and (2) depict the position of the anemonefish and sea anemone during mesh 1 and mesh 2 treatments, respectively. Treatment Description Chamber setup Control Anemonefish and sea anemone VO 2 measured seperately, then summed for a single VO 2 Unit Anemonefish and sea anemone VO 2 measured together Mesh Anemonefish and sea anemone VO 2 measured together, but separated by a mesh barrier that prevented physical contact + (2) (1) 30 Fig. 2. Representative plot of an oxygen meter reading during a flow-through respirometry experiment (unit treatment) on an anemonefish (Amphiprion bicinctus) and sea anemone (Entacmaea quadricolor) at Auburn University. Plot depicts the oxygen concentrations of seawater pasing electrode 1 (imediately before entering the respirometry chamber) and electrode 2 (imediately after exiting the chamber). Leters refer to the time at which experimental animals were added to the chamber (a), and the time at which standard metabolic rate was achieved (b). Oxygen consumption rate of the experimental animals was derived from the mean diference betwen the two electrode readings for 20 min after (b). 31 Fig. 3. Dark oxygen consumption (mean?1 s.e.m.) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor) across respirometry treatments (Table 1) at the Marine Science Station in Aqaba, Jordan (A) and at Auburn University in Alabama, USA (B). Asterisks depict significant diference. 32 Table 2. Statistical summary (repeated-measures ANOVA) of efects of respirometry treatment on oxygen consumption (mean VO 2 ?1 s.e.m.) of anemonefish (Amphiprion bicinctus) and sea anemones (Entacmaea quadricolor) at the Marine Science Station in Aqaba, Jordan (A) and at Auburn University in Alabama, USA, (B,C). Significant results are in bold type. Treatment VO 2 df MS F P A Control 163.62 ? 4.78 1 10168.578 16.26 0.01 Unit 221.27 ? 12.24 B Control 213.84 ? 14.92 2 24266.53 19.22 0.0001 Unit 283.10 ? 14.08 Mesh* 203.14 ? 10.71 C Mesh 1 213.84 ? 14.92 1 351.84 0.29 0.6 Mesh 2 283.10 ? 14.08 * * No diference in VO 2 acros mesh treatments (C), so only mesh 1 is presented. 33 Fig. 4. Efects of rate of water flow on the dark oxygen consumption (mean?1 s.e.m.) of the sea anemone Entacmaea quadricolor in flow-through respirometry. Mean VO 2 max =114.40?14.47 !mol O 2 hr -1 . Asterisks depict significant diference. 34 Fig. 5. Percent time (mean?1 s.e.m.) that anemonefish (Amphiprion bicinctus) engaged in five types of nocturnal behavior in experimental aquaria (A) and in respirometry chambers (B). F=fanning, W=wedging, S=switching, Sw=swiming, and N=no motion. C and D depict magnified views of the percent time anemonefish engaged in wedging and switching behaviors, which were significantly diferent across treatments in experimental aquaria and respirometry chambers. In experimental aquaria, anemonefish were observed alone (Fish) and with host sea anemone (Entacmaea quadricolor, Fish+Anem). In the respirometry chambers, anemonefish were observed in each of four experimental treatments (Table 1). Asterisks depict significant diference within each behavior type. 35 Table 3. Statistical summary (repeated-measures ANOVA) of percent time (A) and bout frequency (bouts 5 min -1 , B) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus) when alone (Fish) and when with host sea anemone (Entacmaea quadricolor, Fish+Anem). Significant results are in bold type. Behavior Treatment Repeated-measures ANOVA Fish Fish+Anem df MS F P A Faning 66.38 ? 13.08 83.79 ? 3.85 1 0.808 1.13 0.337 Wedging 0.28 ? 0.18 3.26 ? 0.54 1 0.399 54.41 0.001 Switching 0.03 ? 0.02 1.42 ? 0.41 1 0.154 13.57 0.014 Swiming 0.29 ? 0.16 0.49 ? 0.28 1 0.003 1.02 0.359 No Motion 33.01 ? 13.15 11.03 ? 4.15 1 1.669 2.33 0.188 B Faning 7.67 ? 0.75 12.80 ? 1.70 1 640.667 24.09 0.004 Wedging 0.44 ? 0.12 9.02 ? 1.88 1 1768.167 66.62 0.001 Switching 0.06 ? 0.04 2.19 ? 0.79 1 108.375 20.84 0.006 Swiming 0.17 ? 0.09 0.31 ? 0.09 1 0.510 1.32 0.302 No Motion 6.79 ? 0.78 5.38 ? 0.55 1 48.167 0.94 0.376 36 Table 4. Statistical summary (Friedman?s Chi Square Test) of efects of respirometry treatments (Table 1) on the percent time (A) and bout frequency (bouts 5 min -1 , B) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus). Significant results are in bold type. Behavior Treatment Friedman's Chi Square Test Control Unit Mesh 1 Mesh 2 df Q P A Fanning 59.56 ? 17.55 79.00 ? 9.10 79.06 ?15.87 70.33 ? 15.40 3 2.39 0.4955 Wedging 0.33 ? 0.33 8.39 ? 1.57 0.00 ? 0.00 0.00 ? 0.00 3 18.8 5 0.0008* Switching 0.00 ? 0.00 2.61 ? 0.71 1.06 ? 0.44 0.00 ? 0.00 3 14.3 6 0.0025* Swiming 5.78 ? 3.45 0.00 ? 0.00 19.89 ? 16.10 17.44 ? 16.53 3 5.03 0.1697 No Motion 34.33 ? 15.44 10.00 ? 8.63 0.00 ? 0.00 12.22 ? 7.01 3 4.36 0.2254 B Faning 4.67 ? 1.65 17.67 ? 3.85 4.50 ? 1.31 3.33 ? 1.15 3 10.1 6 0.0173* Wedging 0.50 ? 0.50 15.17 ? 6.86 0.00 ? 0.00 0.00 ? 0.00 3 12.7 5 0.0052* Switching 0.00 ? 0.00 6.00 ? 1.39 3.00 ? 1.27 0.00 ? 0.00 3 14.3 6 0.0025** Swiming 1.00 ? 0.45 0.00 ? 0.00 1.00 ? 0.37 0.33 ? 0.21 3 7.69 0.0529 No Motion 5.00 ? 2.42 3.67 ? 2.89 0.00 ? 0.00 2.50 ? 1.15 3 4.36 0.2254 * Unit treatment was significantly higher than control, mesh 1 , and mesh 2 treatments. 37 Fig. 6. Bout frequencies (mean?1 s.e.m.) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus) in experimental aquaria (A) and in respirometry chambers (B). F=fanning, W=wedging, S=switching, Sw=swiming, and N=no motion. In experimental aquaria, anemonefish were observed alone (Fish) and with host sea anemone (Entacmaea quadricolor, Fish+Anem). In the respirometry chambers, anemonefish were observed in each of four experimental treatments (Table 1). Asterisks depict significant diference within each behavior type. 38 Fig. 7. Nocturnal and diurnal stroke frequencies (mean?1 s.e.m.) of dorsal and caudal fins of anemonefish (Amphiprion bicinctus). During the night, fin stroke frequencies were measured when each anemonefish was alone (F) and with host sea anemone (Entacmaea quadricolor, F+A). Asterisks depict significant diference within fin type. 39 Table 5. Statistical summary (repeated-measures ANOVA) of efects of time (20:00- 6:00) on percent time (A) and bout frequency (bouts 5 min -1 , B) for five types of nocturnal behavior by anemonefish (Amphiprion bicinctus) in the presence of host sea anemone (Entacmaea quadricolor). The asumption of sphericity was not met and Greenhouse and Geiser (G-G) approximations were used. Significant results are in bold type. Behavior Source G-G epsilon G-G df1 G-G df2 G-G adjusted P A Faning time 0.354 2.475 12.376 0.589 trt X time 0.400 2.799 13.997 0.447 Wedging time 0.473 3.309 16.545 0.199 trt X time 0.336 2.351 11.753 0.192 Switching time 0.403 2.823 14.116 0.087 trt X time 0.411 2.879 14.396 0.052 Swiming time 0.300 2.099 10.497 0.459 trt X time 0.296 2.073 10.367 0.176 No Motion time 0.351 2.458 12.289 0.505 trt X time 0.370 2.588 12.940 0.628 B Faning time 0.416 2.909 14.546 0.061 trt X time 0.321 2.249 11.246 0.118 Wedging time 0.287 2.012 10.059 0.0338* trt X time 0.269 1.882 9.408 0.046 Switching time 0.253 1.768 8.838 0.076 trt X time 0.265 1.854 9.268 0.070 Swiming time 0.313 2.193 10.966 0.470 trt X time 0.338 2.367 11.837 0.212 No Motion time 0.429 3.005 15.026 0.350 trt X time 0.406 2.843 14.217 0.504 * Wedging ocured at a significantly higher frequency during the first time segment (20:0- 20:20) than the seven remaining segments. 40 REFERENCES Alen, G. R. (1974). Anemonefishes. Neptune City: T.F.H. Publications. Arvedlund, M., Bundgaard, I. and Nielsen, L. (2000). Host imprinting in anemonefishes (Pisces: Pomacentridae): Does it dictate spawning site preferences? Environmental Biology of Fishes 58, 203-213. Arvedlund, M., McCormick, M., Fautin, D. and Bildsoe, M. (1999). Host recognition and possible imprinting in the anemonefish Amphiprion melanopus (Pisces: Pomacentridae). Marine Ecology Progres Series 188, 207-218. Biel, K. Y., Gates, R. D. and Muscatine, L. (2007). Efects of free amino acids on the photosynthetic carbon metabolism of symbiotic dinoflagelates. Russian Journal of Plant Physiology 54, 171-183. Box, S. J. and Mumby, P. J. (2007). Efect of macroalgal competition on growth and survival of juvenile Caribbean corals. Marine Ecology Progres Series 342, 139- 149. Bruno, J. and Edmunds, P. (1998). Metabolic consequences of phenotypic plasticity in the coral Madracis mirabilis (Duchasaing and Michelotti): the efect of morphology and water flow on aggregate respiration. Journal of Experimental Marine Biology and Ecology 229, 187-195. Bruno, J., Stachowicz, J. and Bertnes, M. (2003). Inclusion of facilitation into ecological theory. TRENDS in Ecology and Evolution 18, 119-125. Chadwick, N. E. and Arvedlund, M. (2005). Abundance of giant sea anemones and paterns of asociation with anemonefish in the northern Red Sea. Journal of the Marine Biology Asociation of the U.K. 85, 1287-1292. Cleveland, A., Verde, E. and Lee, R. (2010). Nutritional exchange in a tropical tripartite symbiosis: direct evidence for the transfer of nutrients from anemonefish to host anemone and zooxanthelae. Marine Biology 158, 589-602. 41 Davies, S. P. (1992). Endosymbiosis in marine cnidarians. In Plant-Animal Interactions in the Marine Benthos, (ed. D. M. Johns S. J. Hawkins and J. H. Price), pp. 511- 540. New York: Oxford University Pres. Dykens, J. and Shick, M. (1984). Photobiology of the symbiotic sea anemone, Anthopleura elegantisima: Defenses against photodynamic efects, and seasonal photoaclimation. The Biological Bulletin 167, 683-697. Fautin, D. G. (1991). The anemonefish symbiosis: what is known and what is not. Symbiosis 10, 23-46. Fautin, D. G. and Alen, G. R. (1997). Anemonefishes and their host sea anemones. Perth: Western Australian Museum. Fineli, C. M., Helmuth, B. S. T., Pentcheff, N. D. and Wethey, D. S. (2006). Water flow influences oxygen transport and photosynthetic eficiency in corals. Coral Reefs 25, 47-57. Fit, W. K. and Cook, C. B. (2001). The efects of feding or addition of disolved inorganic nutrients in maintaining the symbiosis betwen dinoflagelates and a tropical marine cnidarian. Marine Biology 139, 507-517. Furla, P., Alemand, D., Shick, M., Ferier-Pages, C., Richier, S., Plantivaux, P. and Tambutte, S. (2005). The symbiotic anthozoan: A physiological chimera betwen alga and animal. Integrative and Comparative Biology 45, 595-604. Godinot, C. and Chadwick, N. E. (2009). Phosphate excretion by anemonefish and uptake by giant sea anemones: demand outstrips supply. Bulletin of Marine Science 85, 1-9. Goldshmid, R., Holzman, R., Weihs, D. and Genin, A. (2004). Aeration of corals by sleep-swiming fish. Limnology and Oceanography 49, 1832-1839. Helmuth, B. and Sebens, K. P. (1993). The influence of colony morphology and orientation to flow on particle capture by the scleractinian coral Agaricia agaricites (Linnaeus). Journal of Experimental Marine Biology and Ecology 165, 251-278. 42 Hoegh-Guldberg, O. (1999). Climate change, coral bleaching and the future of the world's coral reefs. Marine and Freshwater Research 50, 839-866. Hoegh-Guldberg, O. and Smith, G. J. (1989). The efect of sudden changes in temperature, light and salinity on the population density and export of zooxanthelae from the reef corals Stylophora pistilata Esper and Seriatopora hystrix Dana. Journal of Experimental Marine Biology and Ecology 129, 279- 303. Holbrook, S. J. and Schmitt, R. J. (2005). Growth, reproduction and survival of a tropical sea anemone (Actiniaria): benefits of hosting anemonefish. Coral Refs 24, 67-73. Holbrook, S. J., Brooks, A. J., Schmitt, R. J. and Stewart, H. L. (2008). Efects of sheltering fish on growth of their host corals. Marine Biology 155, 521-530. Johnson, A. S. and Sebens, K. P. (1993). Consequences of a flatened morphology: efects of flow on feding rates of the scleractinian coral Meandrina meandrites. Marine Ecology Progres Series 99, 99-114. Kinsey, D. W. and Kinsey, B. E. (1967). Diurnal changes in oxygen content of the water over the coral reef platform at Heron I. Australian Journal of Marine and Freshwater Research 18, 23-34. Li, C. and Jackson, R. M. (2002). Reactive species mechanisms of celular hypoxia- reoxygenation injury. American Journal of Physiology: Cel Physiology 282, C227-C241. Liberman, T., Genin, A. and Loya, Y. (1995). Efects on growth and reproduction of the coral Stylophora pistilata by the mutualistic damselfish Dascylus marginatus. Marine Biology 121, 741-746. Lushchak, V. and Bagnyukova, T. (2006). Efects of diferent environmental oxygen levels on free radical proceses in fish. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 144, 283-289. Mariscal, R. N. (1970a). The nature of the symbiosis betwen Indo-Pacific anemone fishes and sea anemones. Marine Biology 6, 58-65. 43 Mariscal, R. N. (1970b). An experimental analysis of the protection of Amphiprion xanthurus Cuvier & Valenciennes and some other anemone fishes from sea anemones. Journal of Experimental Marine Biology and Ecology 4, 134-149. Moyer, J. (1976). Reproductive behavior of the anemonefish Amphiprion clarki at Miyake-Jima, Japan. Japanese Journal of Ichthyology 23, 23-32. Moyer, J. and Steene, R. (1979). Nesting behavior of the anemonefish Amphiprion polymnus. Japanese Journal of Ichthyology 26, 209-214. Murata, M., Miyagawa-Kohshima, K., Nakanishi, K. and Naya, Y. (1986). Characterization of compounds that induce symbiosis betwen sea anemone and anemonefish. Science 234, 585-587. Muscatine, L. (1990). The role of symbiotic algae in carbon and energy flux in reef corals. In Coral Refs, (ed. Z. Dubinsky), pp. 75-88. Amsterdam: Elsevier. Nakamura, T. and van Woesik, R. (2001). Water-flow rates and pasive difusion partialy explain diferential survival of corals during the 1998 bleaching event. Marine Ecology Progres Series 212, 301-304. Nakamura, T., Yamasaki, H. and van Woesik, R. (2003). Water flow facilitates recovery from bleaching in the coral Stylophora pistilata. Marine Ecology Progres Series 256, 287-291. Niggl, W., Haas, A. F. and Wild, C. (2010). Benthic community composition afects O 2 availability and variability in a Northern Red Sea fringing reef. Hydrobiologia 644, 401-405. Nilson, G. E. and Ostlund-Nilson, S. (2004). Hypoxia in paradise: widespread hypoxia tolerance in coral reef fishes. Procedings of the Royal Society B: Biological Sciences 271, S30-S33. Nilson, G. E., Hobbs, J.-P. A. and Ostlund-Nilson, S. (2007a). Tribute to P. L. Lutz: respiratory ecophysiology of coral-reef teleosts. Journal of Experimental Biology 210, 1673-1686. 44 Nilson, G. E., Hobbs, J.-P., Munday, P. L. and Ostlund-Nilson, S. (2004). Coward or braveheart: extreme habitat fidelity through hypoxia tolerance in a coral- dweling goby. Journal of Experimental Biology 207, 33-39. Nilson, G. E., Hobbs, J.-P. A., Ostlund-Nilson, S. and Munday, P. L. (2007b). Hypoxia tolerance and air-breathing ability correlate with habitat preference in coral-dweling fishes. Coral Refs 26, 241-248. Nugues, M. M. and Roberts, C. M. (2003). Coral mortality and interaction with algae in relation to sedimentation. Coral Refs 22, 507-516. Patterson, M. R. and Sebens, K. P. (1989). Forced convection modualtes gas exchange in cnidarians. Procedings of the Natural Academy of Science USA 86, 8833- 8836. Patterson, M. R., Sebens, K. P. and Olson, R. R. (1991). In situ measurements of flow efects on primary production and dark respiration in reef corals. Limnology and Oceanography 36, 936-948. Pinegar, J. K. and Polunin, N. V. C. (2005). Planktivorous damselfish support significant nitrogen and phosphorus fluxes to Mediterranean reefs. Marine Biology 148, 1089-1099. Porat, D. and Chadwick-Furman, N. E. (2004). Efects of anemonefish on giant sea anemones: expansion behavior, growth, and survival. Hydrobiologia 530/531, 513-520. Porat, D. and Chadwick-Furman, N. E. (2005). Efects of anemonefish on giant sea anemones: Amonium uptake, zooxanthela content and tisue regeneration. Marine and Freshwater Behaviour and Physiology 38, 43-51. Pratchett, M. (2001). Influence of coral symbionts on feding preferences of crown-of- thorns starfish Acanthaster planci in the western Pacific. Marine Ecology Progres Series 214, 111-119. Pratchett, M., Vytopil, E. and Parks, P. (2000). Coral crabs influence the feding paterns of crown-of-thorns starfish. Coral Refs 19, 36-36. 45 Reaka-Kudla, M. L. (1997). The Global Biodiversity of Coral Reefs: A Comparison with Rainforests. In Biodiversity II: Understanding and Protecting Our Biological Resources, (ed. M. L. Reaka-Kudla D. E. Wilson and E. O. Wilson), pp. 83-108. Washington: Joseph Henry Pres. Reidenbach, M. A., Koseff, J. R., Monismith, S. G., Steinbuck, J. V. and Genin, A. (2006). The efects of waves and morphology on mas transfer within branched coral reefs. Limnology and Oceanography 51, 1134-1141. Richier, S., Furla, P., Plantivaux, A., Merle, P.-L. and Alemand, D. (2005). Symbiosis-induced adaptation to oxidative stres. Journal of Experimental Biology 208, 277-285. Richier, S., Merle, P.-L., Furla, P., Pigozzi, D., Sola, F. and Alemand, D. (2003). Characterization of superoxide dismutases in anoxia- and hyperoxia-tolerant symbiotic cnidarians. Biochimica et Biophysica Acta 1621, 84-91. Roopin, M. and Chadwick, N. E. (2009). Benefits to host sea anemones from amonia contributions of resident anemonefish. Journal of Experimental Marine Biology and Ecology 370, 27-34. Roopin, M., Henry, R. P. and Chadwick, N. E. (2008). Nutrient transfer in a marine mutualism: paterns of amonia excretion by anemonefish and uptake by giant sea anemones. Marine Biology 154, 547-556. Schmitt, R. J. and Holbrook, S. J. (2003). Mutualism can mediate competition and promote coexistence. Ecology Leters 6, 898-902. Sebens, K. P. (1987). Coelenterata. In Animal Energetics, vol. 1 (ed. T. J. Pandian and F. J. Vernberg), pp. 55-108. San Diego: Academic Pres, Inc. Sebens, K. P. (1997). Adaptive responses to water flow: morphology, energetics, and distribution of reef corals. In Procedings of the 8th International Coral Ref Symposium, vol. 2, pp. 1053-1058. Sebens, K. P. and Done, T. J. (1992). Water flow, growth form and distribution of scleractinian corals: Davies Reef (GBR), Australia. In Procedings of the 7th International Coral Ref Symposium, vol. 1, pp. 557-568. 46 Sebens, K. P., Witting, J. and Helmuth, B. (1997). Efects of water flow and branch spacing on particle capture by the reef coral Madracis mirabilis (Duchasaing and Michelotti). Journal of Experimental Marine Biology and Ecology 211, 1-28. Sebens, K. P., Helmuth, B., Carrington, E. and Agius, B. (2003). Efects of water flow on growth and energetics of the scleractinian coral Agaricia tenuifolia in Belize. Coral Refs 22, 35-47. Shashar, N., Cohen, Y. and Loya, Y. (1993). Extreme diel fluctuations of oxygen in difusive boundary layers surrounding stony corals. Biology Bulletin 185, 455- 461. Shick, J. M. (1990). Difusion limitation and hyperoxic enhancement of oxygen consumption in zooxanthelate sea anemones, zoanthids, and corals. The Biological Bulletin 179, 148-158. Spotte, S. (1996). Supply of regenerated nitrogen to sea anemones by their symbiotic shrimp. Journal of Experimental Marine Biology and Ecology 198, 27-36. Stachowicz, J. J. and Hay, M. E. (1999). Mutualism and coral persistence: the role of herbivore resistance to algal chemical defense. Ecology 80, 2085-2101. Ven, A., Loram, J. and Douglas, A. (2008). Photosynthetic symbioses in animals. Journal of Experimental Botany 59, 1069-1080. Wang, J. and Douglas, A. (1998). Nitrogen recycling or nitrogen conservation in an alga-invertebrate symbiosis? Journal of Experimental Biology 201, 2445-2453. Wang, J. T. and Douglas, A. E. (1999). Esential amino acid synthesis and nitrogen recycling in an alga-invertebrate symbiosis. Marine Biology 135, 219-222. Wild, C., Niggl, W., Naumann, M. and Haas, A. (2010). Organic mater release by Red Sea coral reef organisms?potential efects on microbial activity and in situ O 2 availability. Marine Ecology Progres Series 411, 61-71.