Estuaries and Coasts Predator-prey interactions of the polyclad, Euplana gracilis, and the amphipod, Apocorophium lacustre, in the Chesapeake Bay --Manuscript Draft-- Manuscript Number: ESCO-D-16-00077R2 Full Title: Predator-prey interactions of the polyclad, Euplana gracilis, and the amphipod, Apocorophium lacustre, in the Chesapeake Bay Article Type: Original Article Keywords: Chesapeake Bay; Euplana gracilis; Apocorophium lacustre; predator-prey interactions Corresponding Author: Dean Janiak Smithsonian Environmental Research Center Edgewater, MD UNITED STATES Corresponding Author Secondary Information: Corresponding Author's Institution: Smithsonian Environmental Research Center Corresponding Author's Secondary Institution: First Author: Dean Janiak First Author Secondary Information: Order of Authors: Dean Janiak Julia N Adams Benjamin Rubinoff Richard W Osman Order of Authors Secondary Information: Funding Information: Abstract: Predation within the marine environment has been well-studied and shown to be of major importance in shaping patterns of biodiversity. Typically larger predators, such as fishes, are examined because of the ease of manipulation and strong detectable results whereas effects of smaller micro-predators are much more difficult to evaluate. Here, we examined the distribution and prey consumption of the polyclad flatworm, Euplana gracilis, in the Chesapeake Bay. Euplana gracilis is a common, micro- predator but no data exist on its ecological function. Flatworms were found to actively prey upon a single species, the tube-building amphipod Apocorophium lacustre, in lab trials when tested against several other commonly encountered species. To examine natural population densities of flatworms, large-scale field sampling was conducted via benthic grabs and E. gracilis abundances were found to be significantly correlated with A. lacustre particularly in areas close to the shoreline. Some predator-prey interactions were examined including timed observations of consumption, predator size, and tube protection. Flatworm body size was found to correlate strongly with number of prey consumed over time. Tubes constructed by amphipods were examined as a means of refuge when in the presence of E. gracilis but provided very little protection as flatworms could easily penetrate tubes in search of prey. Our results are the first to show predation of an estuarine/marine polyclad flatworm on amphipods as well as provide some insight into the dynamics of this previously unknown predator-prey relationship. Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation 8/3/2016 To the Editor: Please find the attached final version of our manuscript. All edits were corrected and contribution numbers added in the Acknowledgements section. We would like to thank you as well as the reviewers for the edits/comments that helped make the manuscript better. Sincerely, Dean Janiak Cover Letter Specific Edits/comments: -“et al.” throughout the text should not be in italics, remove italics Throughout the text, all italics have been removed. -where sentences start with E. gracilis (e.g., line 65), please write out the genus name The full genus was written out for all sentences that started with a species name. -remove title on Fig. 3 Title for Figure 3 was removed. -Fig. 3 add per grab on axis labels (e.g., E. gracilis (average no. ind. per grab or similar) and also use the same format on axes of Fig 4 for consistency. Both Figure 3 and Figure 4 axis labels changed. -Font size of numbers on y-axes differ for the two species in Fig 3. Also make sure all fonts (numeric and text) font on all figures 3-5 is consistent Font size on y-axis was corrected. Font size was checked and fixed when needed in Fig 3-5. -There seems to be an error in the first sentence of the figure legend 3: “from based on the different shoreline types??” Sentence was corrected, “based on” was deleted. -incorrect long dash (sorry this was my copy paste error) please use “–“ throughout text for long dash and in the citations The corrected long dash “–“ was replaced throughout the text and references. – Response to Reviewer Comments 1 Title: 1 Predator-prey interactions of the polyclad, Euplana gracilis, and the amphipod, Apocorophium lacustre, in the 2 Chesapeake Bay 3 4 Authors: 5 Dean S. Janiak1,2, Julia N. Adams1,3, Benjamin Rubinoff1,4, and Richard W. Osman1 6 7 Address: 8 1Smithsonian Environmental Research Center, 647 Contees Wharf Rd., Edgewater, MD. 21037, USA 9 2Smithsonian Marine Station, 701 Seaway Dr., Ft. Pierce, FL. 34982, USA 10 3Smithsonian Tropical Research Institute, PO Box 0843–03092, Panama, Republic of Panama 11 4The Ohio State University, Columbus, OH. 43210, USA 12 13 Corresponding Author: 14 janiakd@si.edu 15 16 Key words: 17 Chesapeake Bay, Euplana gracilis, Apocorophium lacustre, predator-prey interactions, 18 19 Abstract 20 Predation within the marine environment has been well-studied and shown to be of major importance in 21 shaping patterns of biodiversity. Typically larger predators, such as fishes, are examined because of the ease of 22 manipulation and strong detectable results whereas effects of smaller micro-predators are much more difficult to 23 evaluate. Here, we examined the distribution and prey consumption of the polyclad flatworm, Euplana gracilis, in 24 the Chesapeake Bay. Euplana gracilis is a common, micro-predator but no data exist on its ecological function. 25 Flatworms were found to actively prey upon a single species, the tube-building amphipod Apocorophium lacustre, in 26 lab trials when tested against several other commonly encountered species. To examine natural population densities 27 of flatworms, large-scale field sampling was conducted via benthic grabs and E. gracilis abundances were found to 28 Manuscript Click here to view linked References 2 be significantly correlated with A. lacustre particularly in areas close to the shoreline. Some predator-prey 29 interactions were examined including timed observations of consumption, predator size, and tube protection. 30 Flatworm body size was found to correlate strongly with number of prey consumed over time. Tubes constructed by 31 amphipods were examined as a means of refuge when in the presence of E. gracilis but provided very little 32 protection as flatworms could easily penetrate tubes in search of prey. Our results are the first to show predation of 33 an estuarine/marine polyclad flatworm on amphipods as well as provide some insight into the dynamics of this 34 previously unknown predator-prey relationship. 35 36 Introduction 37 Within the marine environment, the role of predation has been well-studied and shown to be an important 38 factor in shaping patterns of biodiversity. Predators can have direct or indirect effects on multiple trophic levels 39 within a community, causing alterations in the interactions among prey and their competitors for resources (Paine, 40 1980; Duffy, 2002; Bruno and O’Conner, 2005; Clemente et al., 2010; Vieira et al., 2012). Most studies have 41 focused on the effects of larger predators on populations or communities (Lubchenco and Menge, 1978; Myers and 42 Worm, 2003; Thrush et al., 2006; McCauley et al., 2012) whereas few studies have documented the effects of micro-43 predators on their prey (e.g. Newell et al., 2007). What little work has been done shows the importance of these 44 species as predators in different marine habitats (Ambrose, 1991; Osman and Whitlatch, 1992; Newell et al., 2000; 45 Osman and Whitlatch, 2004; Lavender et al., 2014). One typically overlooked though ubiquitous group of small 46 marine predators are the polyclad flatworms. 47 The Polycladida are a diverse order within the phylum Platyhelminthes (Class Rhabditophora) consisting of 48 almost entirely marine, non-parasitic forms. They are globally distributed with an extensive dietary breadth, and 49 found in most habitats as well as in close association with a variety of invertebrates (e.g. crustaceans and 50 echinoderms) (Newman and Cannon, 2003). The majority of species within this group primarily consume sessile 51 prey including bivalves, barnacles, corals, and ascidians (see review by Galleni et al., 1980; Newman et al., 2000; 52 Rawlinson and Stella, 2012). Other than feeding selectivity, little is known about the ecological role polyclad 53 flatworms have in the marine environment. What has been identified comes from a group of studies that has 54 focused primarily on members of the Stylochidae (Galleni et al., 1980; Chintala and Kennedy, 1993; Merory and 55 Newman, 2005; Lee et al., 2006). Members of this family are typically recognized as pests on a variety of 56 3 commercial aquaculture species including clams, mussels, and oysters (Littlewood and Marsbe, 1990; Newman et 57 al., 1993; Jennings and Newman, 1996; O’Connor and Newman, 2003). 58 The trophic structure within the Chesapeake Bay is fairly well-characterized (Baird and Ulanowicz, 1989; 59 Krause et al., 2003) and an important and well-studied relationship includes the predator-prey interaction between 60 the eastern oyster, Crassostrea virginica (Gmelin, 1791) and the polyclad flatworm, Stylochus ellipictus (Girard, 61 1850) (Landers and Rhodes, 1970; White and Wilson, 1996; Newell et al., 2000). Oysters provide economic and 62 ecological services vital to the bay and therefore, have received much attention over the years (Newell, 1988; 63 Ulanowicz and Tuttle, 1992; Rodney and Paynter, 2006). Within the bay, a second flatworm, Euplana gracilis 64 Girard, 1853, is found in high densities (personal observation) though little is known of its ecological role. Euplana 65 gracilis is a common inhabitant in the Chesapeake Bay as well as most of the eastern Atlantic coastline from Maine 66 to the Gulf of Mexico (Hyman, 1940). Despite the fairly large distributional range of E. gracilis, there exists little 67 knowledge on the species. The aim of the current study was to examine the local distribution and predatory impacts 68 of E. gracilis in the upper Chesapeake Bay. To do this, field collections were made and a series of laboratory 69 experiments were conducted to examine the predatory role of E. gracilis. Experiments were designed to test prey 70 selectivity as well as examine specific interactions between E. gracilis and its prey. Specifically we asked: 1) what 71 is the local distribution of E. gracilis within a representative area of the upper Chesapeake Bay, 2) what is the prey 72 selectivity for E. gracilis among commonly encountered benthic species, and 3) what are some specific predator-73 prey interactions between E. gracilis and its prey. 74 75 Materials and Methods 76 Study site and field collections 77 All sampling and species collections were done in the Rhode River (38° 53.03’ N, 76° 32.4’ W), a 78 subestuary in the northwestern portion of the Chesapeake Bay in Maryland, USA (Figure 1). The river covers an 79 area of approximately 4 km2, is shallow (2 – 4 m depth), and mesohaline having salinity ranges from 0 to 20 with 80 highest salinities occurring during the drier parts of the year. The mean tidal amplitude is roughly 35 cm though can 81 be influenced by local wind patterns. For laboratory studies, all species were collected from wooden pilings and 82 docks at the Smithsonian Environmental Research Center located on the western shore of the Rhode River. 83 Structures were scraped with a paint scrapper and all mobile and sessile animals were brought back to the lab for 84 4 identification and sorting. Animals were retained in the lab with fresh river water changed often under ambient 85 temperature. Artificial habitats were used to collect live specimens due to practicality and a diverse suite of species 86 found throughout the area. Collections of flatworms and potential prey were made as needed throughout the 87 duration of the project. Once an animal was used in a trial or spent > 3 days in the lab, it was returned to the field 88 and new collections were made. Field collections detailed below were done in natural habitats throughout the 89 entirety of the river. 90 Predation on local species 91 To examine prey selectivity of Euplana gracilis, experiments were conducted with potential prey species 92 commonly encountered and found in close proximity to flatworms in the Rhode River. The prey species used in 93 each of the trials were 1) the tube-building amphipod Apocorophium lacustre (Vanhöffen, 1911), 2) the free-94 swimming amphipod Gammarus mucronatus Say, 1818, 3) the barnacle Amphibalanus improvisus (Darwin, 1854), 95 4) the tube-building spionid polychaete Polydora cornuta Bosc, 1802, 5) the nudibranch Cratena pilata (Gould, 96 1870), 6) the ctenostome bryozoan Victorella pavida Saville-Kent, 1870, 7) the nereid polychaete Alitta succinea 97 (Leuckart, 1847), and 8) Tanypus sp. larvae (Insecta: Chironomidae). All experiments were run as paired trials, 10 98 replicates with and 10 replicates without a single individual exposed to a flatworm. For the colonial species V. 99 pavida, clumps consisting of 6 actively feeding zooids were used for each replicate. All experiments were done in 100 square 250 mL containers with newly-collected river water and allowed to run for 24 h. All species were collected 101 within 48 h of the start and held in separate containers without food. All trials were monitored to note any particular 102 interactions that occurred. After the allotted time, all prey species were counted as either dead or alive. A Fisher’s 103 exact test was used to compare the survivorship (nominal variables “dead” or “alive”) of potential prey with the null 104 hypothesis that the proportion of prey alive is the same when exposed or not exposed to a potential predator. 105 Distribution and abundance 106 As part of a separate monitoring project to examine the distribution and abundance of infaunal 107 communities, 151 benthic grab samples were collected throughout the entirety of the Rhode River (Figure 1) in June 108 2014. Approximately half of the samples were collected at nearshore sites (0 – 3 m from shoreline) and the other 109 half collected at offshore sites (> 3 m from shoreline). Environmental data collected for each site included depth, 110 temperature, dissolved oxygen, salinity, and sediment type (visually assessed as fine sand, coarse sand, mud, and 111 mix). Furthermore, samples collected at nearshore sites were classified by their shoreline type (forest, marsh, beach, 112 5 bulkhead, riprap, and offshore). Samples were collected using a Petite Ponar benthic grab (WILDCO®). This 113 particular grab can sample a variety of benthic substrate types and samples approximately an area of 15.2 cm2 (2.4 L 114 of sediment). Samples were sieved at 500 µm, fixed in 10 % formalin for one week, and transferred to 70 % ethanol 115 for sorting and enumeration. For the purposes of this study, only E. gracilis and A. lacustre abundances were used 116 in the data analysis. Data were analyzed using a distance-based analysis of a linear model (DistLM; PRIMER v7) 117 (Clarke and Gorley, 2015) to examine species abundance in relation to environmental factors. Abundance data were 118 square-root transformed and a resemblance matrix was constructed using Bray-Curtis similarities. Environmental 119 data was normalized and a resemblance matrix was constructed using Euclidean distances. Selection of 120 environmental factors was step-wise and AICc was used as the selection criterion to choose the best-fit 121 environmental parameters explaining species distribution. A Two-Way ANOVA was run on abundances of both A. 122 lacustre and E. gracilis using the fixed, categorical factors shoreline type and sediment type. The interaction 123 between the two factors was not included because sediment type within and between shoreline type strongly varied. 124 In both ANOVAs, Student-Newman-Keuls pairwise comparisons test within factors was used. Abundances of 125 flatworms and amphipods were also compared using linear regression to test for any pattern in natural abundances 126 between the two species. 127 Timed Observations 128 A series of trials were run to examine the length of time it took for E. gracilis to attack and consume A. 129 lacustre as well as if that same individual would consume a second amphipod after 24 h. Twenty-seven adult 130 flatworms of a similar size were randomly chosen and placed in separate containers (250 mL) along with a single 131 adult amphipod in each. All flatworms and amphipods were collected within 48 h and starved during that time. 132 Amphipods that were chosen were all of a similar size. The time it took for an initial attack was recorded as well as 133 how long it took for amphipods to become immobilized and fully consumed. After the first amphipod was 134 consumed, a second was added and the amount of time it took the same flatworm to prey upon the second amphipod 135 was recorded. 136 Size versus consumption 137 A series of trials were run to examine if the size of individual E. gracilis was a significant factor in the 138 number of amphipods consumed over a given time period. Flatworms (n = 45) of various sizes were randomly 139 chosen and each was placed in a drop of water on top of a ruler and photographed at least 3 times when fully 140 6 extended. The area (mm2) of each flatworm was measured using ImageJ (Abramoff et al., 2004), and the average 141 area from the separate photographs was used to indicate each individual’s size. This was necessary as flatworms 142 were quite active and a single measurement could be misleading. Sizes of flatworms used were found to range from 143 0.8 mm2 to 9 mm2. After photographs were taken, each flatworm was put into a separate square 500 mL container 144 with new river water. In each container, 4 randomly chosen, adult amphipods of a similar size were added. The 145 prey density was kept constant at 4 and dead amphipods were noted every few hours and replaced with lives ones 146 over the duration of the experiment. The experiment was allowed to run for 120 h (5 days) and at the end of each 147 day, water in each container was carefully siphoned out and exchanged with new river water. After the allotted 148 time, all amphipods left were marked as either dead or alive. A control with 45 amphipods in a separate container 149 without flatworms was run simultaneously to monitor the health of amphipods over the duration of the experiment. 150 The number of amphipods consumed was compared against the size of each individual E. gracilis (mm2) via linear 151 regression. 152 Amphipod tubes and protection 153 Three separate trials were run in which amphipods were allowed to build tubes prior to predator exposure. 154 Both trials 1 and 2 contained 5 replicates and trial 3 contained 12 replicates. Trial 3 had increased replication to 155 ensure results were consistent. Each pair of replicates consisted of a single adult amphipod all of a similar size, 156 either with a tube or without a tube exposed to a single flatworm. In each trial, half of the amphipods were placed in 157 separate square 250 mL containers with defaunated sediment and new river water and allowed to build a tube 158 whereas the other half were placed in containers without sediment. After amphipods in all replicates had built tubes 159 (approximately 24 h), a single flatworm was randomly picked and added to each of the containers and the 160 experiment was allowed to run for 120 min. Once time had expired, all amphipods were counted as dead or alive. 161 Fisher’s exact test was used to compare the survivorship of amphipods either with or without tubes using the 162 nominal variables “dead” or “alive”. 163 164 Results 165 Predation on local species 166 Out of all potential prey species tested, only Apocorphium lacustre was significantly consumed by Euplana 167 gracilis (Table 1). Predation on A. lacustre was rapid and in all trials happened within 30 min. There was a 90 % 168 7 and 80 % survival rate of Gammarus mucronatus and Amphibalanus improvisus and based on personal observations, 169 mortality resulted from damage during collection rather than predation. One Polydora cornuta was found dead and 170 after the experiment ended, a flatworm was found on the worm but it was unclear whether or not the worm was 171 being preyed upon by the flatworm. The last 4 species, Cratena pilata, Victorella pavida, Alitta succinea, and 172 Tanypus sp., were all alive by the end of the trials. In all trials with flatworms absent, all species were accounted for 173 and alive. 174 Distribution and abundance 175 A total of 151 benthic ponar grabs were taken throughout the majority of the Rhode River (Figure 1). 176 Apocorophium lacustre was found in 93 samples (62 % of total) and E. gracilis was found in 51 samples (34 % of 177 total). Based on the DistLM analysis, the environmental parameters temperature (p = 0.001), depth (p = 0.008), and 178 salinity (p = 0.085) were the best predictors of abundances for the two species (Figure 2) though only explained 20 179 % of the variation in the data (r2 = 0.202). Both A. lacustre and E. gracilis were found to be more abundant at 180 nearshore sites (collectively for all shoreline types, 78.36 ± 18.41 S.E.) as compared to offshore sites (10.04 ± 3.78 181 S.E.). In separate Two-Way ANOVAs, there was a significant effect of shoreline type found for both E. gracilis (p 182 = 0.028) and A. lacustre (p < 0.001) however no effect was found for substrate type. In both cases, the shoreline 183 type forest had the highest abundances (Figure 3). There was a strong positive relationship (r2 = 0.59, p < 0.001) 184 between the abundance of A. lacustre and the abundance of E. gracilis (Figure 4). Out of all samples containing A. 185 lacustre, 55 % had flatworms present and there was roughly a 10:1 ratio of amphipods to flatworms. Of those 186 without flatworms, amphipod densities were quite low (1 – 10 per benthic grab) and no samples were collected that 187 contained only flatworms with no amphipods. 188 Timed Observations 189 In all trials, predation occurred at a rapid rate and all initial A. lacustre were consumed by E. gracilis. The 190 initial attack took on average 14 min (± 18 S.D.) from when the two species were added together. When an 191 amphipod was added, the flatworm would increase its speed of movement in search of the prey and once 192 encountered, would swiftly attack the ventral portion between two pereopods, injecting its pharynx into the tissue. 193 The flatworm then moved to the dorsal side of the amphipod and in many cases with the pharynx removed, while the 194 amphipod was still mobile. After approximately 3 min (± 1 S.D.), the amphipod was fully immobilized and the 195 flatworm moved back to the ventral side and began to actively digest the internal tissues. Flatworms fed on average 196 8 65 min (± 28 S.D.) prior to abandoning the carcass and after, went into a short “resting” phase whereby movement 197 was limited. In 85 % of trials, flatworms had already consumed the second amphipod within 24 h. 198 Size versus consumption 199 There was a significant positive linear relationship between the size of E. gracilis and the number of 200 amphipods consumed in 120 h (r2 = 0.49, p < 0.001, Figure 5). Flatworms in the smaller size classes exhibited 201 increased handling time and some variation in prey consumption but still consumed at least 3 amphipods in the 202 allotted amount of time. Observations suggest that all amphipod mortality was due to predation by flatworms. In 203 the control container, with no flatworms added, all amphipods were accounted for and alive. 204 Amphipod tubes and protection 205 In all 3 trials, there was no significant difference between amphipods with tubes present and those without 206 tubes (Fishers exact test, p > 0.05). In the first and second trial, 4 of 5 amphipods having tubes were consumed 207 whereas all 5 were consumed without tubes and in the third trial, 10 of 12 amphipods with tubes were consumed 208 whereas all 12 were consumed without tubes. During the trials with tubes present, prey attack and consumption 209 occurred both in and out of tubes. Flatworms were attracted to tubes that contained amphipods and either entered 210 the tube or attacked the amphipod from the outside. 211 212 Discussion 213 The flatworm, Euplana gracilis, was found to consume a single species when tested against several 214 common species in the Rhode River. Flatworms readily consumed the corophioid, Apocorophium lacustre, but prior 215 to our study, consumption of amphipods in an estuarine or marine setting has only been reported once as anecdotal 216 observations (Jennings, 1957). Our study is the first to present and quantify any ecological data for E. gracilis, 217 including its natural distribution and predator-prey interactions. Polyclad flatworms are generally highly selective in 218 their prey choice (Galleni et al., 1980 and references within) and it is therefore not surprising that E. gracilis was 219 found to consume one species throughout the study. Both E. gracilis and A. lacustre are found throughout the 220 eastern Atlantic coastline though A. lacustre is restricted to brackish waters (Bousfield, 1973) whereas E. gracilis 221 has been found in salinities ranging from 0 – 37 (personal observation). We included in our laboratory trials a free-222 swimming amphipod (Gammarus mucronatus) though this species was found to easily avoid any encounters when 223 flatworms were in search of food. It is possible that flatworms could be more general in prey choice though the 224 9 escape behavior of non-tube building amphipods could separate them as prey. In contrast, A. lacustre tended to be a 225 very poor swimmer and was easily captured. It is feasible that E. gracilis can consume other species, possibly other 226 tube-building amphipods, within its range though this needs further study. 227 An extensive survey of the benthic habitat within the Rhode River showed a significant relationship 228 between E. gracilis and A. lacustre. Flatworms were only found in samples that contained A. lacustre and increases 229 in flatworms were concomitant with prey density. Both flatworm and amphipod densities were highest in nearshore 230 habitats as compared to offshore though it is unknown why A. lacustre was more abundant along the coast as the 231 majority of the river is soft, unsorted mud. Krause et al. (2003) developed an empirical food web model for the 232 Chesapeake Bay which included several fish species (i.e. spot, catfish, and hogchoker) as the main predators of A. 233 lacustre so increased predation may occur away from the shoreline. Sites with a forested shoreline also had the 234 highest abundances though this type of shoreline made up the majority of the sampling sites and therefore could be 235 an artifact of site selection. 236 Our study investigated some specific predator-prey interactions between E. gracilis and A. lacustre. Timed 237 observations of feeding were relatively consistent with predation occurring within 15 min of adding prey and took 238 roughly 65 min for consumption. Predator size was also a significant factor in consumption rates and as size 239 increased, there was a steady positive increase in the number of amphipods a single individual could consume. 240 Interestingly, the smallest flatworms used (0.8 – 2 mm2) were found to easily capture and consume an average-sized 241 amphipod, though with increased handling time. 242 Corophioid amphipods are a common tube-building group found throughout much of the world. Tubes are 243 thought to have several uses including feeding, acting as a storage deposit for food, and facilitating mating 244 efficiency by limiting search time (Shillaker and Moore, 1987; Borowsky, 1991; Dixon and Moore, 1997). The 245 majority of studies have examined tubes in relation to feeding and it is generally thought that Corophium spp. are 246 primarily filter feeders (Foster-Smith and Shillaker, 1977; Gerdol and Hughes, 1994), and use their tubes to filter 247 water through to capture particles (Dixon and Moore, 1997). The tubes themselves have never been tested as a 248 means of refuge from predators. Our results suggest that tubes provided very little protection from flatworms. 249 Collectively in all 3 trials, approximately 80 % of amphipods were consumed with tubes as compared to 100 % 250 without tubes. Although with or without tubes was not found to be statistically different, some amphipods did 251 survive and therefore tubes could be somewhat useful. However, increasing the allotted time during the experiment 252 10 could have increased the consumption rate to 100 %. Flatworms were attracted to occupied tubes and would 253 actively examine the tube either attempting to enter or attacking the amphipod from the outside. In rare cases, 254 amphipods would leave tubes in an attempt to crawl away though with little success. Throughout the trials, E. 255 gracilis was also found to deposit eggs on the inside surface of tubes. The successful development and release of 256 larvae was not followed but other species of polyclad flatworms are known to deposit eggs within shells of their 257 prey after consumption (Hurley, 1976; Galleni et al., 1980; Lee et al., 2006). 258 Studies on polyclad flatworms have generally shown that they can have a strong negative effect on 259 populations, particularly those with long generation times such as corals (Rawlinson et al., 2011; Rawlinson and 260 Stella, 2012), barnacles (Hurley, 1975; Branscomb, 1976) and bivalves (Pearse and Wharton, 1938; Loosanoff, 261 1956; Littlewood and Marshe, 1990). Corophioids have short generation times (1 – 4 cohorts per year), with direct 262 development leading to high local abundances (Fish and Mills 1979; Moore 1981; Peer et al., 1986; Cunha et al., 263 2000; Pérez et al., 2007). Several studies have examined the population dynamics of Corophium volutator (Pallas, 264 1766) and have shown negative effects of larger predators including shorebirds (Hicklin and Smith, 1984) and fish 265 (McCurdy et al., 2005). Our results demonstrate that E. gracilis is a predator on A. lacustre, but it is unclear 266 whether or not there is any top-down control on populations given their high recruitment throughout the year. Data 267 from field collections did show that flatworms were positively correlated with amphipod abundances and only found 268 in samples that contained amphipods, however, this correlation does suggest that flatworms do not limit populations 269 of amphipods. 270 The ecological role of micro-predators within the marine environment is poorly understood particularly 271 because of the challenges in constructing manipulative experiments. Despite this, these predators are typically 272 thought of as abundant and important components within the habitats that they are found. Traditionally, micro-273 predators have been associated with newly settled or juvenile prey altering the composition of communities over 274 succession (e.g. Osman et al., 1992). Euplana gracilis preys upon adult corophioids and is an example of a micro-275 predator that could potentially have a large effect on amphipod populations given their rapid rates of consumption as 276 well as observed field densities found throughout the study site. Despite the fact that this species is fairly ubiquitous 277 throughout the eastern Atlantic coastline, this is the first study to acknowledge its ecological importance and 278 therefore, this is significant for future considerations of the trophic structure within the Chesapeake Bay as well as 279 within the distributional range of E. gracilis. 280 11 281 Acknowledgements 282 Partial funding was provided to DSJ through the Smithsonian’s Tennenbaum Marine Observatiories 283 Network (TMON) and the Seward Johnson Trust for Oceanography. Funding was awarded to JNA and BR through 284 an REU-NSF grant through the Smithsonian Environmental Research Center. We thank Midge Kramer for 285 assistance in field collections. This is contribution number # 10 from TMON as well as # 1038 from the 286 Smithsonian Marine Station at Ft. Pierce. 287 288 Literature Cited 289 290 Abramoff, M.G., P.J. Magalhaes, S.J. Ram. 2004. Image processing with ImageJ. Biophotonics International. 11: 291 36–42. 292 293 Ambrose, W.G. 1991. Are infaunal predators important in structuring marine soft-bottom communities? American 294 Zoologist 31: 849–860. 295 296 Baird, D., and R.E. Ulanowicz. 1989. The seasonal dynamics of the Chesapeake Bay ecosystem. Ecological 297 Monographs 59: 329–364. 298 299 Borowsky, B. 1991. Patterns of reproduction of some amphipod crustaceans and insights into the nature of their 300 stimuli. Pp 33–49 in Crustacean Sexual Biology, R.T. Bauer and J.W. Martin, eds. Columbia University Press, New 301 York. 302 303 Bousfield, E.L. 1973. Shallow-water Gammaridean Amphipoda of New England. Cornell University Press, London. 304 305 Branscomb, E.S. 1976. Proximate causes of mortality determining the distribution and abundance of the barnacle 306 Balanus improvisus Darwin in Chesapeake Bay. Chesapeake Science 17: 281–288. 307 308 12 Bruno, J.F., and M.I. O’Conner. 2005. Cascading effects of predator diversity and omnivory in a marine food web. 309 Ecology Letters 8: 1048–1056. 310 311 Chintala, M.M., and V.S. Kennedy. 1993. Reproduction of Stylochus ellipticus (Playthelminthes:Polycladida) in 312 response to temperature, food, and presence or absence of a partner. Biological Bulletin 185: 373–387. 313 314 Clarke, K.R., and R.N. Gorley. 2015. PRIMER v7: User Manual/Tutorial. Plymouth: PRIMER-E. 315 316 Clemente, S., J.C. Hernández, A. Rodríguez, and A. 2010. Identifying keystone predators and the importance of 317 preserving functional diversity in sublittoral rocky-bottom areas. Marine Ecology Progress Series 413: 55–67. 318 319 Cunha, M.R., J.C. Sorbe, and M.H. Moreira. 2000. The amphipod Corpohium multisetosum (Corophiidae) in Ria de 320 Aveiro (NW Portugal). I. Life history and aspects of reproductive biology. Marine Biology 137: 637–650 321 322 Dixon, M.T., and P.G. Moore. 1997. A comparative study on the tubes and feeding behavior of eight corophioid 323 Amphipoda and their bearing on phylogenetic relationships within the Corophioidea. Philosophical Transactions of 324 the Royal Society B: Biological Sciences 352: 93–112. 325 326 Duffy, J.E. 2002. Biodiversity and ecosystem function: the consumer connection. Oikos 99: 201–219. 327 328 Fish, J.D., and A. Mills. 1979. The reproductive biology of Corophium volutator and C. arenarium 329 (Crustacea:Amphipoda) Journal of the Marine Biological Association of the United Kingdom 59: 355–368. 330 331 Foster-Smith, R.L., and R.O. Shillaker. 1977. Tube irrigation by Lembos websteri Bate and Corophium bonnelli 332 Milne Edwards (Crustacea:Amphipoda). Journal of Experimental Marine Biology and Ecology 26: 289–296. 333 334 Galleni, L., P. Tongiorgi, E. Ferrero, and U. Salghetti. 1980. Stylochus mediterraneus (Turbellaria:Polycladida), 335 predator on the mussel Mytilus galloprovincialis. Marine Biology 55: 317–326. 336 13 337 Gerdol, V., and R.G. Hughes. 1994. Feeding behavior and diet of Corophium volutator in an estuary in southeastern 338 England. Marine Ecology Progress Series 114: 103–108. 339 340 Hicklin, P.W., and P.C. Smith. 1984. Selection of foraging sites and invertebrate prey by migrant semipalmated 341 sandpipers, Calidris pusilla (Pallas), in Minas Bassin, Bay of Fundy. Canadian Journal of Zoology 62: 2201–2210. 342 343 Hurley, A.C. 1975. The establishment of populations of Balanus pacificus Pilsbury (Cirripedia) and their 344 elimination by predatory Turbellaria. Journal of Animal Ecology 44: 521–532. 345 346 Hurley, A.C. 1976. The polyclad flatworms Stylochus tripartitus Hymen as a barnacle predator. Crustaceana 36: 347 110–111. 348 349 Hyman, L.H. 1940. The polyclad flatworms of the Atlantic coast of the United States and Canada. Proceedings of 350 the United States National Museum 89: 449–495. 351 352 Jennings, J.B. 1957. Studies on feeding, digestion, and food storage in free-living flatworms 353 (Platyhelminthes:Turbellaria). Biological Bulletin 112: 63–80. 354 355 Jennings, K.A., and L.J. Newman. 1996. Four new Stylochid flatworms (Platyhelminthes:Polycladida) associated 356 with commercial oysters from Moreton Bay, southeast Queensland, Australia. Raffles Bulletin of Zoology 44: 493–357 508. 358 359 Krause, A.E., K.A. Frank, D.M. Mason, R.E. Ulanowicz, and W.W. Taylor. 2003. Compartments revealed in food-360 web structure. Nature 426: 282–285. 361 362 Landers, W.S., and E.W. Rhodes. 1970. Some factors influencing predation by the flatworm, Stylochus ellipticus 363 (Girard), on oysters. Chesapeake Science 11: 55–60. 364 14 365 Lavender, J.T., K.A. Dafforn, and E.L. Johnston. 2014. Meso-predators: A confounding variable in consumer 366 exclusion studies. Journal of Experimental Marine Biology and Ecology 456: 26–33. 367 368 Lee, K., A.M. Beal, E.L. Johnston. 2006. A new predatory flatworm (Platyhelminthes, Polycladida) from Botany 369 Bay, New South Wales, Australia. Journal of Natural History 39: 3987–3995. 370 371 Littlewood, D.T.J., and L.A. Marshe. 1990. Predation on cultivated oysters, Crassostrea rhizophorae (Guilding), by 372 the polyclad turbellarian flatworm, Stylochus (Stylochus) frontalis Verrill. Aquaculture 88: 145–150 373 374 Loosanoff, V.L. 1956. Two obscure oyster enemies in New England waters. Science 123: 1119–1120. 375 376 Lubchenco, J., and B.A. Menge. 1978. Community development and persistence in a low rocky intertidal zone. 377 Ecological Monographs 48: 67–94. 378 379 McCauley, D.L., H.S. Young, R.B. Dunbar, J.A. Estes, B.X. Semmens, and F. Micheli. 2012. Assessing the effects 380 of large mobile predators on ecosystem connectivity. Ecological Applications 22: 1711–1717. 381 382 McCurdy, D.G., M.R. Forbes, S.P. Logan, D. Lancaster, and S.I. Mautner. 2005. Foraging and impacts by benthic 383 fish on the intertidal amphipod Corophium volutator. Journal of Crustacean Biology 25: 558–564. 384 385 Merory, M., and L.J. Newman. 2005. A new stylochid flatworm (Platyhelminthes, Polycladida) from Victoria, 386 Australia and observations on its biology. Journal of Natural History 39: 2581–2589. 387 388 Moore, P.G. 1981. The life histories of the amphipods Lembos websteri and Corophium bonnellii Milne Edwards in 389 kelp holdfasts. Journal of Experimental Marine Biology and Ecology 49: 1–50. 390 391 Myers, R.A., and B. Worm. 2003 Rapid worldwide depletion of predatory fish communities. Nature 423: 280–283. 392 15 393 Newell, R.I.E. 1988. Ecological changes in the Chesapeake Bay: Are they the result of overharvesting the American 394 Oyster, Crassostrea virginica? Pp. 536–546 in Understanding the Estuary. Advances in Chesapeake Bay research. 395 CRC Publication 129. Chesapeake Bay Consortium, Solomons, MD. 396 397 Newell, R.I.E., G.S. Alspach, V.S. Kennedy, and D. Jacobs. 2000. Mortality of newly metamorphosed eastern 398 oysters (Crassostrea virginica) in mesohaline Chesapeake Bay. Marine Biology 136: 665–676. 399 400 Newell, R.I.E., V.S. Kennedy, and K.S. Shaw. 2007. Comparative vulnerability to predators, and induced defense 401 responses, of eastern oysters Crassostrea virginica and non-native Crassostrea ariakensis oysters in Chesapeake 402 Bay. Marine Biology 152: 449–460. 403 404 Newman, L.J., and L.R.G. Cannon. 2003. Marine flatworms: the world of polyclads. CSIRO publishing. 405 406 Newman, L.J., L.R.G. Cannon, and H. Govan. 1993. Stylochus (Imogene) matatasi n. sp. (Platyhelminthes, 407 Polycladida): pest of cultured giant clams and pearl oysters from Solomon Islands. Hydrobiologia 257: 185–189. 408 409 Newman, L.J., J.L. Norenburg, and S. Reed. 2000. Taxonomic and biological observations on the tiger flatworm, 410 Maritigrella crozieri (Hyman, 1939), a new combination (Playhelminthes, Polycladida, Euryleptidae) from Florida 411 Waters. Journal of. Natural History 34: 799–808. 412 413 O’Connor, W.A., and L.J. Newman. 2003. Predation of cultured mussels, Mytilus galloprovincialis, by stylochid 414 flatworms, Imogine mcgrathi, from Twofold Bay, New South Wales, Australia. Asian Fisheries Science 16: 133–415 140. 416 417 Osman, R.W., R.B. Whitlatch, and R.J. Malatesta. 1992. Potential role of micro-predators in determining 418 recruitment into a marine community. Marine Ecology Progress Series 83: 35–43. 419 420 16 Osman R.W., and R.B. Whitlatch. 2004. The control of the development of a marine benthic community by 421 predation on recruits. Journal of Experimental Marine Biology and Ecology 311: 117–145. 422 423 Paine, R.T. 1980. Food Webs: Linkage, Interaction Strength and Community Infrastructure. J. Animal Ecology 49: 424 666–685. 425 426 Pearse, A.S., and G.W. Wharton. 1938. The oyster “leech,” Stylochus inimicus Palombi, associated with oysters on 427 the coasts of Florida. Ecological Monographs 8: 605–656. 428 429 Peer, D.L., L.E. Linkletter, and P.W. Hicklin. 1986. Life history and reproductive biology of Corophium volutator 430 (Crustacea:Amphipoda) and the influence of shorebird predation on population structure in Chignecto Bay, Bay of 431 Fundy, Canada. Netherlands Journal of Sea Research 20: 359–373. 432 433 Pérez, V., M.A. Marquiegui, and M.J. Belzunce. 2007. Life history and production of Corophium urdaibaiense 434 (Crustacea:Amphipoda) in the Urdaibai estuary (NE Spain). Marine Biology 151: 1163–1174. 435 436 Rawlinson, K.A., J.A. Gillis, R.E. Billings, and E.H. Borneman. 2011. Taxonomy and life history of the Acropora-437 eating flatworm Amakusaplana acroporae nov. sp. (Polycladida:Prosthiostomidae). Coral Reefs 30: 693–705. 438 439 Rawlinson, K.A., and J.S. Stella. 2012. Discovery of the corallivorous polyclad flatworm, Amakusaplana acroporae, 440 on the Great Barrier Reef, Australia – the first report from the wild. PloS One 7: e42240. 441 442 Rodney, W.S., and K.T. Paynter. 2006. Comparisons of macrofaunal assemblages on restored and non-restored 443 oyster reefs in mesohaline regions of Chesapeake Bay in Maryland. Journal of Experimental Marine Biology and 444 Ecology 335: 39–51. 445 446 Shillaker, R.O., and P.G. Moore. 1987. The feeding habits of the amphipods Lembos websteri Bate and Corophium 447 bonellii Milne Edwards. Journal of Experimental Marine Biology and Ecology 110: 93–112. 448 17 449 Thrush, S.F., J.E. Hewitt, M. Gibbs, C. Lundquist, and A. Norkko. 2006. Functional role of large organisms in 450 intertidal communities: community effects and ecosystem function. Ecosystems 9: 1029–1040. 451 452 Ulanowicz, R.E., and J.H. Tuttle. 1992. The trophic consequence of oyster stock rehabilitation in Chesapeake Bay. 453 Estuaries 15: 298–306. 454 455 Vieira, E.A., L.F.L. Duarte, and G.M. Dias. 2012. How the timing of predation affects composition and diversity of 456 species in a marine sessile community? Journal of Experimental Marine Biology and Ecology 412: 126–133. 457 458 White, M.E. and E.A. Wilson. 1996. Predators, pests, and competitors. Pp. 559–580 in The eastern oyster, 459 Crassostrea virginica, V.S. Kennedy, R.I.E Newell, and A.F. Eble, eds. Maryland Sea Grant Publication 99, College 460 Park, Maryland. 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 18 Figure 1. Map of the Rhode River. Star indicates the location of the Smithsonian Environmental Research Center 489 where live collections were made for experimental trials. Grey shading indicates water and the river empties to the 490 south-east into the mainstem of the Chesapeake Bay. Black circles indicate field sampling sites used for benthic 491 ponar grabs (n = 151). 492 493 Figure 2. Distance-based linear model (DistLM) plot based on step-wise selection of environmental parameters 494 fitted to abundances for E. gracilis and A. lacustre taken from benthic grabs. Vectors indicate the direction of effect 495 for environmental parameters in the plot. Split-plot bubbles represent the number of individuals of each of the two 496 species found at each site. 497 498 499 Figure 3. Average A. lacustre and E. gracilis abundances (± S.E.) per benthic grab (approximate area = 15.2 cm2) 500 from the different shoreline types. Offshore indicates all samples that were taken roughly 3m or more from the 501 shoreline. Note the difference in scale on the left and right y-axis. 502 503 Figure 4. Linear regression for the relationship between amphipod (x-axis) and flatworm (y-axis) abundances per 504 benthic grab (n = 151) collected during field sampling. 505 506 Figure 5. Linear regression for the relationship of E. gracilis (n = 45) size to the amount of A. lacustre consumed 507 over 120 h. 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 19 545 546 547 548 Fig. 1 Figure 1 Fig 2. Figure 2 Fig 3. Shoreline Type Forest Bulkhead Offshore Marsh Riprap Beach A . la c u st r e ( a v e ra g e n o . i n d . p e r g r a b ) 0 50 100 150 200 E . g ra ci li s (a v e ra g e n o . in d . p e r g ra b ) 0 10 20 30 40 50 A. lacustre E. gracilis Figure 3 Fig. 4 A. lacustre (no. ind. per grab) 0 200 400 600 E . g ra ci li s (n o . in d . p e r g ra b ) 0 20 40 60 r 2 =0.59 p < 0.001 Figure 4 Fig. 5 E. gracilis Size (mm2) 0 2 4 6 8 10 A . la cu st r e C o n s u m e d ( 1 2 0 h ) 0 2 4 6 8 10 12 r2 = 0.49 p < 0.001 Figure 5 Table 1 Common species tested as potential prey of E. gracilis. Percent Alive Fisher's exact test Apocorophium lacustre 0 p < 0.001 Gammarus mucronatus 90 NS Amphibalanus improvisus 80 NS Polydora cornuta 90 NS Cratena pilata 100 NS Victorella pavida 80 NS Alitta succinea 100 NS Tanypus sp. larvae 100 NS Fisher’s exact test on survivorship (“dead” or “alive”, n = 10). NS = not significant. Table 1