Southeastern Naturalist 373 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 22001155 SOUTHEASTERN NATURALIST 1V4o(2l.) :1347,3 N–3o9. 62 Trophic Variation in Coastal Plain Stream Predatory Fishes Dean E. Fletcher1,*, Angela H. Lindell1, Garrett K. Stillings1, Gary L. Mills1, Susan A. Blas2, and J Vaun McArthur1 Abstract - Unique morphologies along with associated differences in habitat use and feeding behavior can result in fish at the top of piscine food chains differing in trophic level. Broad size ranges inherent within large species provide opportunity for size-related trophic shifts. Such relationships between size and trophic level can be species specific. Furthermore, individual-based diet variation can bring about differences among similar-sized organisms. A challenge to aquatic ecologists is deciphering these patterns of trophic change both between and within species. Stable isotope analysis has emerged as a powerful tool for evaluating such patterns. Employing stable isotope analyses, we assessed trophic differentiation in 4 large predatory fish species from a coastal-plain stream. We established the trophic base by including 2 herbivorous invertebrates in the analysis and identified a trophic hierarchy among species, with 2 specialized, generally open-water piscivores, Lepisosteus osseus (Longnose Gar) and Micropterus salmoides (Largemouth Bass), occupying the highest trophic position. The largest-bodied and generally benthic-oriented species, Ictalurus punctatus (Channel Catfish), occupied the lowest trophic level among the fishes studied. Trophic position of Largemouth Bass and Longnose Gar increased linearly and gradually with size within the broad size ranges collected. In contrast, Channel Catfish exhibited a more abrupt shift in trophic position with size and much individual variation associated with the shift. Additionally, groups of Longnose Gar had belonged to distinctly different food chains, despite coexisting in a relatively small stream when collected. Differences between the observed patterns and other published accounts indicate further evaluation of trophic patterns of these fishes among habitats is warranted. Introduction Trophic relationships involving predatory fishes have diverse and extensive influences on stream ecosystems at multiple levels of organization. Presence of fish predators has been shown to modify stream-community compositions (Bechara et al. 1992, Jackson et al. 2001, Williams et al. 2003) by reducing both numbers and feeding efficiency of stream herbivores and consequently influencing the processing of stream basal-energy resources such as allochthonous detritus (McIntosh et al. 2005, Ruetz et al. 2002) and attached algae (Bechara et al. 1992, Gelwick 2000, Power 1990, Power et al. 1985). However, herbivore efficiency is not always affected by fish predators (Ruetz et al. 2002). Other effects of predatory stream fishes include habitat shifts of prey species (Englund and Krupa 2000; Power et al. 1985; Schlosser 1987, 1988) and altered prey life histories (Peckarsky et al. 2002). The presence of fish predators in mercury (Hg)-contaminated ponds influenced the Hg Savannah River Ecology Laboratory, University of Georgia, PO Drawer E, Aiken, SC 29802. 2Area Completion Projects, Savannah River Nuclear Solutions, Savannah River Site, Aiken, SC 29808. *Corresponding author - fletcher@srel.uga.edu. Manuscript Editor: Morgan E. Raley Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 374 pool present in the macroinvertebrate community and indirectly the risk to various terrestrial predators of consuming contaminated prey (Henderson et al. 2012). Consequently, understanding the trophic status of large fishes is essential to defining the role the organisms may play in aquatic ecosystems. The trophic level of stream fishes varies not only among species and feeding guilds, but also among individuals within a species. This intraspecific variation is particularly evident in large predatory fishes where broad size ranges of individuals can concurrently reside in a stream. Ontogenetic or size-related trophic shifts can be as important as taxonomic differences (sensu Woodward and Hildrew 2002). Juveniles of piscivorous fish species feed on zooplankton or other small invertebrates and as they grow switch to larger invertebrates or fish larvae before finally shifting to larger fish prey as adults (Keast 1985, Mittelbach and Persson 1998). Within a species or feeding guild, size of prey generally continues to increase with predator body size (Keast 1985, Mittelbach and Persson 1998, Sheldon and Meffe 1993), allowing large predators to consume prey sizes unavailable to relatively smaller predators (Wilson 1975). Thus, the range of prey sizes and number of species consumed often increases with predator body size (Scharf et al. 2000), but prey availability can influence predator diet and consequently trophic level in any particular system (Layman et al. 2005, Wilson 1975). The pace of progression to piscivory may also be heavily influenced by the level of specialization toward piscivory (Keast 1985). Consequently, the size of large fish predators is not always an indicator of trophic status as a result of such morphological or behavioral specializations. Much trophic divergence occurs at the family or generic levels (Sheldon and Meffe 1993) due to morphological distinctions being more pronounced at these levels. Large syntopic predators have partitioned their habitats through evolutionary processes that have altered morphologies, including size and form of their mouths and bodies as well as behaviors (Gatz 1979, Keast and Webb 1966, Wikramanayake 1990, Winemiller 1989). These evolved characteristics interact to influence the potential diets of these predators. Classic diet studies examining fish gut contents provide a snapshot of what a fish has recently eaten. These studies do not make inferences about nutrient assimilation nor are they indicative of what the fish has eaten over a longer time period (Bearhop et al. 2004, Gu et al. 1996, Vander Zanden and Vadeboncoeur 2002). Additionally, diet studies often need to assume trophic interactions among lower trophic levels such as small aquatic invertebrates (Vander Zanden et al. 1997). Stable isotopes of nitrogen (N) and carbon (C) are becoming more commonly incorporated into studies of trophic relationships as tools to understand the importance of different basal resources and the trophic status among and within species. Organismal tissue contains 2 stable isotopes of N: 14N and the heavier isotope 15N. Comparison of the ratios of 15N to 14N (δ15N) is informative of trophic position, because as N is transferred from one trophic level to another, tissue generally becomes more enriched with the heavier isotope by ~3.4‰ (Post 2002). Consequently, a higher δ15N is indicative of a higher trophic level. Two stable isotopes of C are also present in organismal tissue: 12C and the heavier isotope 13C. However, in contrast to N, the ratio of 12C and 13C (δ13C) Southeastern Naturalist 375 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 generally changes less between trophic transfers, with enrichment of the heavier isotope (13C) between trophic levels usually less than 1.6‰ (McCutchan et al. 2003). The δ13C provides information on the source of C acquired by an organism in respect to the basal resources of a system. An advantage of stable isotope analysis (SIA) is that assimilation of these isotopes into tissue results in an analysis of ingested materials integrated over a longer time period than diet studies (Vander Zanden and Vadeboncoeur 2002). Additionally, only assimilated materials are analyzed, and all trophic transfers are assessed—even among invertebrates (Vander Zanden and Vadeboncoeur 2002, Vander Zanden et al. 1997). However it is often difficult to establish variability in stable isotope signature shift between different organisms, spatial and temporal variability in signatures, base autotroph signatures, and the effects of organism physiology and age, sample preparation, and physical factors such as water velocity on signatures (Finlay 2004, Finlay et al. 1999, France 1996, McCutchan et al. 2003, Overman and Parrish 2001, Post 2002). The versatility of SIA of C and N in ecological studies is evident in scientific literature. Trophic arrangement of aquatic communities has been revealed in lentic (Gu et al. 1996, Kupfer et al. 2006) and lotic systems (Jepsen and Winemiller 2002). Moreover size-related trophic shifts within a species have been shown (Gu et al. 1996). Portions of diet items actually assimilated and diet variability has been assessed (Atkinson et al. 2010, Raikow and Hamilton 2001). Distinctions between the assimilation of terrestrial versus aquatic sources of primary production have been made in lotic systems (Finlay 2001, Fuentes Brito et al. 2006, Zah et al. 2001) as has the converse incorporation of C from in-stream primary production into terrestrial food webs (Collier et al. 2002). Within a stream, SIA was used to determine whether assimilated food was derived from local or upstream sources (Finlay et al. 2002). SIA has also revealed the relative contributions of various primaryproduction sources to lentic food webs (Hecky and Hesslein 1995, Herwig et al. 2004, Vander Zanden and Vadeboncoeur 2002). In stream-spawning Salmo trutta L. (Brown Trout), SIA distinguished anadromous from freshwater-resident individuals by differentiating the food webs of each (McCarthy and Waldron 2000). Some types of pollution such as human sewage, barnyard runoff, or synthetic N fertilizers can modify stable isotope signatures (e.g., increase δ15N; Cabana and Rasmussen 1996, Lake et al. 2001, Wayland and Hobson 2001), but SIA of C and N has been used successfully in habitats contaminated by coal-combustion waste (Unrine et al. 2007) such as our study area. Despite the difficulties discussed earlier, SIA provides a powerful tool for use in aquatic ecological studies. Our objectives were to (1) compare the trophic position of 4 species of large predatory fish in a coastal-plain stream, (2) compare the relationship between body size and trophic position among these species, (3) investigate whether specific changes in morphology such as gape or head width influenced the observed relationships between body size and trophic positions, and (4) compare carbon sources among the 4 species of predatory fish and also determine whether they shifted with size or trophic level within a species. Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 376 Field-Site Description Beaver Dam Creek (BDC), a Savannah River tributary, is located on the Savannah River Site (SRS) along the southeast border of South Carolina. The SRS is an 801-km2 National Environmental Research Park operated by the US Department of Energy since 1951. Beaver Dam Creek has been impacted by a coal-fired electric- and steam-generating power plant operated in its headwaters since the early 1950s (Halverson et al. 1997). A more complete study-site description is provided in Fletcher et al. (2014a, b), but summarized here. Upper reaches of BDC were channelized to carry large volumes of Savannah River water from the power house along with effluents from associated ash- and coal-pile runof f basins. Our study sites situated at the edge of the Savannah River floodplain and Aiken Plateau supported species associated with large rivers as well as those typical of smaller upland tributaries. Although many species were rare, we collected 46 species of fish from this area. As part of a larger study, we established 2 sites in the upper 3.2 km of BDC separated by only 430 m, but differing substantially in hydrologic regime. The upstream site A lies on fluvial terraces of the Savannah River and has deeply incised, scoured channel with an inactive floodplain. The downstream site B lies on the Savannah River floodplain with a less-incised channel and an active floodplain that is inundated by both BDC and Savannah River water during floods. Although patterns and processes in the downstream site are not all independent from the upstream site, we sought to make comparisons based on unique hydrology and geomorphology found at each site. Previously reported differences between these sites include stable isotope signatures of invertebrates (Fletcher et al. 2014b). Study Organisms In our study stream, we found 4 species of large fishes: Lepisosteus osseus (L.) (Longnose Gar), Ictalurus punctatus (Rafinesque) (Channel Catfish), Micropterus salmoides (Lacépède) (Largemouth Bass), and Amia calva L. (Bowfin). These native species each exceed 400 mm standard length (SL) as adults, are commonly syntopic in southeastern US coastal plain streams, and vary greatly in head and body forms. Size of these large predators precludes the largest adults of these species from preying upon each other, but it is not visually apparent which species occupy the highest trophic level. Largemouth Bass have a robust fusiform body, a deep emarginate tail, a large terminal mouth, and large eyes relative to their head and body size. These are characteristics that well suit it to be a roving predator (Gatz 1979). The anterior of a Channel Catfish is relatively ventrally flattened and its tail has a deeply forked caudal fin, especially in smaller individuals. A conspicuous feature of Channel Catfish is sensory barbels surrounding its mouth. As apparent by their relatively small eyes and high number of sensory receptors on the barbels and body of Channel Catfish (Caprio et al. 1993), catfishes rely less on vision for capturing prey and more on tactile and chemical responses. Although its subterminal mouth predisposes it to bottom feeding (Gatz 1979, Moyle and Cech Southeastern Naturalist 377 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 1982, Wikramanayake 1990), Channel Catfish will feed at multiple water levels (Eddy and Underhill 1974). Longnose Gar have an elongate, round body, rounded caudal fin, and posteriorly located dorsal fin characteristic of an ambush or stalking predator (Gatz 1979, Keast and Webb 1966). Their head is highly specialized with narrowly elongate, highly toothed jaws well-suited for capturing fish as prey (Netsch 1964, Marcy et al. 2005). Like the Largemouth Bass, the Longnose Gar has large eyes relative to their head size. Bowfin have a long, moderately rounded body anteriorly but more compressed posteriorly with a rounded caudal fin and elongate dorsal fin consistent with a lieand- wait predator (Gatz 1979). Its mottled coloration helps camouflage it in the highly vegetated waters that it typically inhabits (Marcy et al. 2005). Since herbivores integrate sources of primary production feeding a stream (Cabana and Rasmussen 1996,Vander Zanden et al. 1997), we included 2 herbivorous invertebrates to establish isotopic signatures of the food-web base. Corbicula fluminea (Müeller) (Asiatic Clam) filter-feeds particles from the water column and deposit-feeds materials from surface sediments (Hakenkamp et al. 2001). The heptageniid mayfly Maccafertium modestum (Banks) scrapes biofilm from the surface of submerged objects such as wood debris or facultatively collects leaf particles (Kondratieff and Voshell 1980, Merritt et al. 2008). A more complete description of these species is provided in Fletcher et al. (2014b), which compares the trophic levels and carbon sources between these 2 stream invertebrates. Methods Sample collection and tissue handling We collected invertebrates from 14 January through 11 May 2010 using dip nets or seines, and fish from 20 May through 28 July 2010 mostly collected by boat-mounted electrofishing (collected on 16 days) but also with supplemental sampling efforts using seines, minnow traps, hoop nets (trapped on 20 days), and backpack electrofishers (collected on 5 days). All fish were frozen and later thawed and rinsed first in tap water and then in 18-milliohm (MΩ) deionized water prior to processing. We measured SL (mm), total length (TL; mm), and total wet body mass (g), and removed a skinless subsample of muscle from the right anterodorsolateral region of each fish for stable isotope analysis. Tissue samples were lyophilized, ground to a fine powder, and homogenized prior to analysis. We determined sex and maturity by gonadal inspection. Because collection occurred during the late spring through mid-summer, gonads of adult fish were generally not highly regressed. Consequently, the sex was generally apparent; however, we confirmed the sex of the Longnose Gar gonads according to Ferrara and Irwin (2001). We measured gape and head widths on Channel Catfish and Largemouth Bass because these characters can influence size of prey consumed (Gatz 1979, Keast and Webb 1966). Longnose Gar were excluded from the head-feature analyses due to their unique head morphology, and Bowfin were excluded because of small sample size. Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 378 Laboratory handling of M. modestum and Asiatic Clams are detailed in Fletcher et al. (2014b). We analyzed a total of 13 M. modestum composite samples (site A: n = 5; site B: n = 8) and 19 Asiatic Clam composite samples (site A: n = 11; site B: n = 8). We collected and analyzed a total of 5 Bowfin, 51 Channel Catfish, 34 Largemouth Bass, and 19 Longnose Gar (Table 1). Our sampling efforts were focused on many species outside of the scope of this analysis. We attempted to collect a range of sizes of each species, but some species were more difficult to collect adequate sample sizes of than others. We released all Longnose Gar and Channel Catfish after an adequate sample was attained, but needed to keep all captured Bowfin and Largemouth Bass for analysis. Stable isotope analyses Stable isotope analyses of C and N were performed on a Finnigan Delta Plus mass spectrometer (Thermo-Finnigan, Bremen, Germany) in the Stable Isotope and Soil Biology Laboratory, Odum School of Ecology, University of Georgia. Isotope ratios were expressed in the delta (δ) format: δ 13C or δ15N (units of ‰) = ([Rsample - Rstandard] /Rstandard) x 1000, where R is the 13C:12C ratio or 15N:14N ratio (Atkinson et al. 2010). A bovine standard was referenced against an international standard (Atkinson et al. 2010), and precision averaged ≤0.1%. Statistical analyses After assessing normality of our data, we improved the distribution of δ15N, δ13C, SL, gape width, and wet body mass by loge transformation prior to analyses. Additionally, we arcsine square-root transformed relative gape width (Zar 1984). Using 2-sample t-tests, we compared the δ15N of Channel Catfish, Largemouth Bass, and Longnose Gar between sexes and between adults versus juveniles. Similar comparisons of δ13C for these species between sexes, size groups, and between adults versus juveniles were made. We were unable to determine the sex of 1 juvenile Largemouth Bass, so we excluded it from analyses of sex. Using least squares linear regression, we individually evaluated the relationships of δ15N and gape width with total wet body mass and SL, as well as the relationship between δ15N and gape width. We graphically assessed homoscedasticity of the regression residuals. The graphs illustrating linear regression lines include the 95% confidence intervals. We compared the δ15N and δ13C between sites and among species using analysis of variance (ANOVA) models that included site, species, and the associated interaction term and used Tukey pairwise comparisons to further evaluate differences among species. The ANOVA models were repeated including only the 4 fish species to confirm the reliance of the significant interaction term on spatial differences in the less-mobile invertebrates. Spatial variation of the invertebrates has been previously addressed (Fletcher et al. 2014b). Southeastern Naturalist 379 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 Results Intraspecific trophic variation We collected broad size ranges of Channel Catfish and Largemouth Bass that spanned from small juveniles to large adults (Table 1). However, we did not collect any Longnose Gar smaller than 444 mm SL or Bowfin smaller than 367 mm SL. Trophic position (δ15N) did not differ between sexes in Channel Catfish, Largemouth Bass, or Longnose Gar (Table 2), and sample size precluded Table 1. Species, sizes, and sex of fish included in the analyses. SL = standard length, F = female, M = male, and U = unknown sex, A = adult, J = juvenile. Species n SL (mm) Whole wet weight (g) Maturity (n) Bowfin (A. calva) Females 1 506 2012 1 A, 0 J Males 4 423 (367–465) 1200 (765–1551) 4 A, 0 J Channel Catfish (I. punctatus) Females 28 410 (240–563) 1515 (224–3497) 23 A, 5 J Males 23 382 (99–700) 1623 (16–6870) 9 A, 14 J Largemouth Bass (M. salmoides) Females 21 214 (97–480) 385 (16–2250) 9 A, 12 J Males 12 179 (97–317) 187 (16–736) 4 A, 8 J Unknown sex 1 Longnose Gar (L. osseus) Females 7 801 (612–962) 2255 (825–3913) 7 A, 0 J Males 12 527 (444–645) 529 (261–1084) 12 A, 0 J Table 2. Comparison of the average δ15N and δ13C between males and females, and adults and juveniles for each of our study species. Mean ± standard deviation with the limits in parentheses. Associated P values for 2-sample t-tests are provided. Males Females P Adults Juveniles P δ15N (‰) Longnose Gar 13.9 ±0.8 14.6 ± 1.4 0.30 14.2 ± 1.0 - - (12.7–15.1) (11.8–15.7) (11.8–15.7) - Largemouth Bass 13.1 ±0.7 13.6 ± 1.2 0.15 14.1 ± 1.3 13.0 ±0.6 0.01 (12.0–14.1) (12.0–17.6) (12.6–17.6) (12.0–14.2) Channel Catfish 11.4 ±0.8 10.9 ± 1.1 0.63 11.1 ± 1.1 11.2 ±0.7 0.44 (10.1–12.9) (8.8–13.7) (8.8–13.7) (10.1–12.8) Bowfin 13.0 ±1.5 11.4 - 12.7 ± 1.5 - - (11.1–14.7) - (11.1–14.7) - δ13 C (‰) Longnose Gar -23.9 ± 3.2 -22.7 ± 3.2 0.39 -23.4 ± 3.2 - - (-26.6 to -15.9) (-25.8 to -17.9) (-26.6 to -15.9) - Largemouth Bass -25.6 ± 0.5 -25.7 ± 0.8 0.61 -25.7 ± 1.0 -25.7 ±0.4 0.81 (-26.6 to -24.8) (-28.4 to -24.6) (-28.4 to -24.6) (-26.9 to -25.2) Channel Catfish -26.1 ± 1.1 -26.0 ± 1.1 0.92 -26.1 ± 1.3 -26.0 ±0.5 0.38 (-29.0 to -23.3) (-27.8 to -23.6) (-29.0 to -23.3) (-26.9 to -25.2) Bowfin -27.3 ± 1.2 -28.3 - -27.5 ± 1.1 - - (-28.8 to -26.0) - (-28.8 to -26.0) - Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 380 comparison within Bowfin. However, with the wide range of fish sizes within each species, size-related trophic shifts were noted. The δ15N of Longnose Gar spanned a narrow range: 3.9‰ (11.8–15.7, SD = 1.0). A weak linear relationship occurred between δ15N and both body mass (R2 = 0.17, β = 0.03, P = 0.08; Fig. 1a) and SL (R2 = 0.18, β = 0.12; P = 0.07) of that species. The δ15N of Largemouth Bass spanned the broadest range of the 4 species: 5.6‰ (12.0–17.6, SD = 1.1‰). However, 32 of 34 Largemouth Bass ranged only 2.2‰ (12.0–14.2). The δ15N was higher in adult than juvenile Largemouth Bass (Table 2), but this could be driven by body size. Across all sizes of Largemouth Bass, the relationship between body size and δ15N was a simple linear relationship indicating a gradual increase of trophic position with increasing body mass (R2 = 0.31, β = 0.02, P < 0.01; Fig. 1b) and SL (R2 = 0.30, β = 0.08, P < 0.01). Additionally gape width increased with body mass (R2 = 0.99, β = 0.33, P < 0.01; Fig. 2a) and SL (R2 = 0.99, β = 1.06, P < 0.01). Consequently the Largemouth Bass δ15N and gape width were also linearly related (R2 = 0.29, β = 0.07, P < 0.01; Fig. 2c). Although allometric growth in Largemouth Bass does result in relative gape width (as a proportion of SL) being greater (P < 0.01) in adults (mean = 9.4 %, SD = 0.40) than juveniles Figure 1. Relationship of δ15N and whole wet body mass for (a) Longnose Gar, (b) Largemouth Bass, (c) Channel Catfish with gape width of <34 mm, and (d) Channel Catfish with gape width of >34 mm. Regression lines and 95% confidence interv als are illustrated. Southeastern Naturalist 381 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 (mean = 8.3 %, SD = 0.55), the difference is not great. Across all sizes of Largemouth Bass, gape width ranged 38.4 mm (7.5–45.9), but relative gape width was only 2.6% SL (7.5–10.1). For both Largemouth Bass and Longnose Gar there was 1 fish that was aberrant from the rest. These outliers were particularly evident in the linear-regression residuals. Since we were unable to rerun tissue samples from these 2 fish, it is not known if the observed values are handling mistakes or curious observations. For Largemouth Bass, removal of the putative outlier increases the amount of variation explained by linear regression of δ15N and whole body mass (R2 = 0.41, β = 0.02, P < 0.01). Similarly for Longnose Gar, exclusion of the single outlier results in a more highly significant linear relationship between δ15N and whole body mass (R2 = 0.35, β = 0.04, P = 0.01). The δ15N of Bowfin spanned 3.6‰ (11.1–14.7, SD = 1.5), but collection of only 5 individuals precluded examination of the relationship between body mass and δ15N for the Bowfin. Figure 2. Relationship of gape width and whole wet body mass for (a) Largemouth Bass and (b) Channel Catfish. Relationship δ15N and gape width for (c) Largemouth Bass and (d) Channel Catfish (open circles = Channel Catfish with gape width of <34 mm; closed circles = Channel Catfish with gape width of >34 mm). Regression lines and 95% confidence intervals are illustrated. Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 382 The δ15N of Channel Catfish ranged 4.8‰ (8.8–13.7, SD = 1.0). Across all sizes of individuals, the δ15N of Channel Catfish was not linearly related to body mass (R2 < 0.01, β = 0.00, P = 0.64) or SL (R2 = 0.01, β = 0.02, P = 0.61). The δ15N also did not differ between adults and juveniles (Table 2). Across all sizes of Channel Catfish, gape width ranged 87.2 mm (11.2–98.4), which corresponds to a range in relative gape width (% SL) of 7.2 (8.2–15.5). Gape width linearly increased with body mass (R2 = 0.93, β = 0.40, P < 0.01; Fig. 2b) and SL (R2 = 0.93, β = 1.24, P < 0.01). However, the δ15N was also not linearly related to gape width (R2 = 0.03, β = 0.03, P = 0.25). Further investigation revealed a more complex relationship. The relationship between gape width and δ15N changed at a gape width of around 34 mm. When less than 34 mm, gape width ranged 20.3 mm and relative gape width ranged 4.1% SL. When greater than 34 mm, gape width ranged 63.4 mm and relative gape width ranged 6.1% SL. When gape width was less than 34 mm, the δ15N decreased with gape width (R2 = 0.32, β = -0.10, P < 0.05; Fig. 2d) and body mass (R2 = 0.38, β = -0.03, P < 0.03; Fig. 1c). In contrast, δ15N increased with gape width (R2 = 0.28, β = 0.17, P < 0.01; Fig. 2d) and body mass (R2 = 0.24, β = 0.08, P < 0.01; Fig. 1d) when gape width exceeded 34 mm. Community trophic hierarchies The δ15N differed among species, with the 2 herbivorous invertebrates occupying the lowest trophic position, Channel Catfish intermediate, and Bowfin, Largemouth Bass and Longnose Gar the highest (ANOVA: R2 = 0.90, species P < 0.01, site P = 0.01, species*site P <0.01, pairwise comparisons P < 0.05; Fig. 3a). On average, Channel Catfish is at a lower trophic position than the other fishes. However, with the wide range of δ15N and associated size-related shifts occurring in Channel Catfish, some larger individuals begin to overlap with that of smaller individuals of both Largemouth Bass and Longnose Gar (Figs. 3, 4). The latter 2 species broadly overlap in trophic position. The position of Bowfin is less certain due to relatively high variability among the small sample size. Pairwise comparisons indicated the significant site and species*site interaction term was driven by spatial variability in the δ15N of M. modestum (pairwise comparison P < 0.01). In contrast, the δ15N of the other species did not differ between sites (pairwise comparisons P > 0.90). Correspondingly, the same ANOVA model including only the fish indicated differences between species, but not sites (R2 = 0.63, species P < 0.01, site P = 0.99, species*site P = 0.40). Interestingly, with all fish species included, δ15N was negatively correlated to body mass (rranks = -0.203, P < 0.05) and not correlated to body length (rranks = -0.082, P > 0.10). However, Figure 4 clearly shows these relationships were driven by the Channel Catfish occupying a lower trophic level despite its heavy body. Carbon sources The δ13C of Largemouth Bass, Bowfin, and Channel Catfish spanned 3.8‰ (-28.4 to -24.6, SD = 0.7), 2.9‰ (-28.8 to -26.0, SD = 1.1), and 5.7‰ (-29.0 to -23.3, SD = 1.1), respectively. Whether comparing size groups established from size frequencies or adults versus juveniles (Channel Catfish: 16–1612 g and 1936–6870 g, Southeastern Naturalist 383 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 Figure 3. Mean δ15N and δ13C of C. fluminea (Asiatic Clam), M. modestum, A. calva (Bowfin), I. punctatus (Channel Catfish), M. salmoides (Largemouth Bass), and L. osseus (Longnose Gar) represented by black circles. Error bars represent + 1 standard error. Gray lines represent measures of individuals. Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 384 respectively; Largemouth Bass: 14–180 g and 264–2250 g, respectively; Longnose Gar: 261–1145 g and 1644–3913 g, respectively), δ13C was not dependent upon size (all P > 0.35; Table 2). Similarly, the δ13C also did not differ between sexes of Channel Catfish, Largemouth Bass, or Longnose Gar (all P > 0.38; Table 2). Spanning a range of 10.66‰, the δ13C of Longnose Gar was substantially more variable than the other 3 species. Additionally, 2 groups could be identified in the Longnose Gar population, with the δ13C of 1 group averaging -19.29‰ (n = 6, SD = 2.16, range = -21.72 to -15.93) and the other averaging -25.34‰ (n = 13, SD = 0.66, range = -26.59 to -24.47). We were unable to contribute this δ13C distribution to size, sex, site, or trophic position. Carbon sources based on δ13C statistically differed between groups of species. The δ13C was lowest and similar among Asiatic Clam, M. modestum, and Bowfin, intermediate and similar in Channel Catfish and Largemouth Bass, and on average highest in Longnose Gar (R2 = 0.71, species P < 0.01, site P < 0.01, species*site P < 0.01; pairwise comparisons P < 0.05; Fig. 3b). However, the average falls in between the 2 groups of Longnose Gar. Consequently, only Longnose Gar individuals of the more enriched δ13C group do not overlap with any of the other species. The mean δ13C of Bowfin appears only marginally lower than Channel Catfish (pairwise comparison P = 0.06). Again, the significant site and interaction term was driven by spatial variability in the invertebrates, and fish did not significantly differ between sites (pairwise comparison P > 0.77). The same ANOVA model including only the fish also indicated differences between species, but not sites (R2 = 0.35, species P < 0.01, site P = 0.23, species*site P = 0.32). Figure 4. Species comparison of the relationship of δ 15N and body mass. Southeastern Naturalist 385 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 Discussion Trophic position calculations A mean trophic fractionation or enrichment rate of 15N by 3.4‰ has been reported from diverse habitats including aquatic and terrestrial systems, laboratory and field studies, and organisms ranging from copepods to polar bears (Post 2002). However, Post (2002) elaborated that an enrichment range from about 2 to 5‰ may occur for any given trophic transfer. Similarly, a 3.3‰ enrichment of 15N between trophic levels was calculated in a hypereutrophic subtropical lake (Gu et al. 1996) and 2.8‰ in tropical streams (Jepsen and Winemiller 2002). Using laboratory studies and literature, McCutchan et al. (2003) reported an average enrichment of 2.3‰, but indicated enrichment varies among organisms and stressed the uncertainty inherent with this variability. Because of spatial, temporal, and taxonomic variability, as well as anthropogenic influences, establishing an isotopic signature representative of basal primary producers can also be notoriously difficult (Anderson and Cabana 2007, Cabana and Rasmussen 1996, Jardine et al. 2006, Post 2002). Long-lived herbivores such as freshwater mussels have been used as integrators of primary production to convert the δ 15N to a trophic position using the equation Trophic position = ([Predator δ15N–Herbivore δ15N]/3.4) + C, where C equals the position of the organisms used to calibrate trophic position (Cabana and Rasmussen 1996, Overman and Parrish 2001, Post et al. 2000, Vander Zanden 1997). In a broad comparison of 5 invertebrate functional-feeding groups in streams, scrapers have also been recommended for use in establishing a trophic baseline (Anderson and Cabana 2007). Freshwater mussels are very rare in the BDC system, so we used the average δ15N of our 2 herbivores, the scraper M. modestum and deposit/filter feeding Asiatic Clam (trophic position = 2 ), to estimate trophic position of the fishes. This may not be as optimal as larger and longer-lived mussels, but because filter feeding, deposit feeding, and scraping biofilm are included in these 2 species, many of the primary stream-energy sources are represented. We calculated average trophic positions of 3.04, 3.50, 3.72, and 3.94 for Channel Catfish, Bowfin, Largemouth Bass, and Longnose Gar, respectively. However future work should substantiate the enrichment rate in these organisms, as we acknowledge the uncertainties inherent in using stable isotope signatures. Because of their widespread distributions in coastal-plain streams, further evaluation of using M. modestum and Asiatic Clam to establish a baseline for SIA food web studies are warranted. Interspecific trophic variation Based on review of diet-survey literature, Roach et al. (2009) classified Longnose Gar, Largemouth Bass, and Bowfin from the upper Mississippi River in Minnesota as piscivores and Channel Catfish as an omnivore/detritivore. Subsequent employment of SIA in the same study indicated a trophic hierarchy with Longnose Gar > Largemouth Bass > Bowfin > Channel Catfish based on average trophic positions of 4.25 > 3.98 > 3.78 > 3.62.. A similar hierarchy occurred among Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 386 these same 4 species in BDC, despite all 4 species being at lower trophic positions of 3.94, 3.72, 3.50, and 3.04, respectively. Since longer food chains lead to higher trophic positions and food-chain length may increase with ecosystem size (Post et al. 2000), it is not surprising that fish inhabiting a small stream such as BDC were at lower trophic positions than the same species from a large system like the main stem and backwaters of the Mississippi River. A similar pattern to BDC was also observed in both forested and open canopy habitats within a Texas reservoir with Lepisosteus oculatus Winchell (Spotted Gar), Largemouth Bass, and Channel Catfish ranked 4.5 > 4.1 > 3.3 in forested areas and 4.4 > 4.2 > 3.5 in open water, respectively (Chumchal and Hambright 2009). Similarly, trophic ranks were Lepisosteus platyrhincus DeKay (Florida Gar) > Largemouth Bass > Bowfin in near shore habitats of a Florida lake, but were Bowfin > Florida Gar > Largemouth Bass in marsh habitats of the lake inhabited by smaller Largemouth Bass and more Bowfin (Fry et al. 1999). Other SIA studies of freshwater ecosystems have placed gar species at the top of food webs dominated by fishes. Florida Gar held the highest trophic position in another Florida lake (Gu et al. 1996). Lepisosteus platostomus Rafinesque (Shortnose Gar) had the most enriched δ15N of fishes in backwater lakes of the Mississippi and Illinois Rivers, IL (Herwig et al. 2004, 2007). Even though the Florida Gar and Shortnose Gar reach a smaller maximum size and have less-elongate jaws compared to the Longnose Gar, the specialized morphology suited for piscivory often places gars at or near the top of piscine food chains. This ranking is consistent with lower trophic specialization at the species level (Keast 1985, Sheldon and Mef fe 1993). Largemouth Bass also held the second highest trophic level in a Florida lake (Gu et al. 1996). Literature and our study support Channel Catfish often inhabiting a lower trophic position than equal-sized Bowfin, Largemouth Bass, and Gar. The negative correlation between body mass and δ15N when all 4 BDC species were combined resulted from the heaviest-bodied fish, Channel Catfish, occupying the lowest trophic level. A general average trophic hierarchy of Longnose Gar (or other gar species) > Largemouth Bass > Bowfin > Channel Catfish is consistent with our data and published literature, acknowledging lack of statistical significance in some of our comparisons and lack of statistical testing in some other studies. However, location-specific habitat or community composition has the potential to favor individual species. Interspecific carbon source variation δ 13C statistically differed among species, but generally followed the same pattern as trophic level. The difference among species generally falls within the range that would occur with an enrichment rate of around 1.6‰ per trophic transfer as reported by McCutchan et al. (2003). Consequently basal C resources appeared to broadly overlap among the 4 predatory fishes. This overlap of basal carbon sources by these predators that integrate lower trophic levels into their diet was not surprising in a relatively small stream. However, a portion of the Longnose Gars with entirely unique carbon signatures was surprising and will be described in detail below. Southeastern Naturalist 387 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 Intraspecific trophic variation Specific patterns of trophic change are less established than the above-described trophic hierarchies. However, general relationships between body size and diet or trophic level have been widely studied. For example, size of eaten prey often increases with predator size (Mittelbach and Persson 1998, Sheldon and Meffe 1993). Such relationships between predator and prey size can arise from gape limitation for predators that swallow their prey whole, thereby necessitating growth to a point where large prey can be consumed (Winemiller 1989). Correspondingly, transitions from eating invertebrates to a diet of fish can be size related and the portion of fish found in a facultative piscivore’s diet increases with predator body size (Winemiller 1989). Two patterns in the relationship between trophic level and body size were apparent in our study. Trophic position increased linearly within the collected size ranges of both Longnose Gar and Largemouth Bass. Any abrupt change from lower to observed trophic positions occurred at sizes smaller than those in our samples. The smallest Longnose Gar and Largemouth Bass were 444 and 97 mm SL, respectively. These patterns were also characterized by relatively low trophic variability within size ranges (Fig. 1). In contrast, Channel Catfish exhibited a more abrupt trophic shift. Morphological specialization for piscivory can reduce the size and age at which a species becomes piscivorous (Keast 1985). Moreover such specialization can reduce a species’ ability to eat other prey types (Winemiller 1989), which will further reduce trophic variability. Longnose Gars are a particularly specialized piscivore with long, beak-like jaws for capturing prey by biting. Prey is swallowed whole after capture. Diet studies have indicated that Longnose Gar can become largely piscivorous at less than 100 mm TL, and adults have a diet of almost exclusively fish (Lagler and Hubbs 1940, Mittelbach and Persson 1998, Netsch 1964). Fast growth can also enable species to become piscivorous at younger ages (Mittelbach and Persson 1998). In addition to specialization, Longnose Gar is a fast-growing species that can reach lengths of 400 mm TL in 1 year (Netsch and Witt 1962). Similarly, captive Longnose Gar grew to 335 mm TL in 183 days in Florida and to an average of 290 mm TL in 267–325 days in Missouri (Carlander 1969). Our smallest Longnose Gar was over 500 mm TL; thus all specimens captured likely had transitioned to piscivory. Despite ranging from 444 to 962 mm SL, with the 1 outlier excluded, our Longnose Gar spanned a range of only 0.87 trophic positions (3.51–4.38), with low variability among similar-sized individuals. We suspect the observed variability in trophic level in Longnose Gar that did occur stems from preying on fish of different trophic positions. Largemouth Bass are a large-mouthed suction feeder well suited for piscivory (Keast and Webb 1966, Mittelbach and Persson 1998). Based on diet analysis and SIA, Largemouth Bass young of the year (YOY) became piscivorous at 85 mm TL in a small Michigan lake (Post 2003). Another study in Michigan lakes found 60% of the YOY bass to be piscivorous at sizes over 40–50 mm SL (Olson 1996). In a survey of diet studies, Mittelbach and Persson (1998) also indicated that Largemouth Bass may begin to become piscivorous at a size less than 100 mm TL. Our smallest Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 388 collected individuals exceeded these sizes. Positive relationships among gape size, body size, and δ15N were observed in Largemouth Bass. Largemouth Bass trophic level (range = 1.67 [3.28–4.95]) varied across a broader span than Longnose Gar, but the increase was still relatively gradual. Thus again, any abrupt trophic shift occurred at a size below our smallest individuals of 97 mm SL (16 g). Our largest Largemouth Bass weighed 2.25 kg, twice that of our next largest bass. This largest individual occupied a trophic position of 4.49 compared to 3.75 for the next largest individual and the species average of 3.72. The trophic position of this single large individual fell near the regression line between trophic position and body size, suggesting a continuous and gradual increase in trophic level with size (Fig. 1b). Habitat-related trophic shifts were noted in Largemouth Bass in a Florida lake, with trophic position gradually increasing with body size (Fry et al. 1999). The δ15N of Largemouth Bass also appeared to increase gradually with body size at locations in the Colorado River system, CO (Martinez et al. 2001). In Rhode Island ponds and small lakes, the δ15N of predators such as Largemouth Bass were less variable than those of benthic-feeding fishes (Lake et al. 2001). Although relatively uncommon in a hypereutrophic Florida lake, the δ15N of Largemouth Bass appeared to increase more abruptly with size (Gu et al. 1996) in contrast to the gradual increase we found in BDC. Consequently, variability among habitats and locations in the relationship between trophic level and size should be the focus of further investigation. Largemouth Bass can have a diverse diet including aquatic insects, crustaceans, and fish (Marcy et al. 2005, Wiltz 1993). However in BDC, the broad overlap of the Largemouth Bass δ15N with the extremely specialized Longnose Gar also suggests piscivory in Largemouth Bass within most individuals. In contrast, Channel Catfish exhibited more substantial trophic shifts in large individuals. The shift began at gape widths exceeding 34 mm, with substantially more individual variation in trophic level at the beginning of the shift (i.e., individuals with gape widths just greater than 34 mm). Abrupt trophic changes can occur as fish reach size thresholds that allow changes in feeding habits. For example, Cheilodactylus spectabilis Hutton (Red Moki) off the coast of New Zealand exhibited an abrupt diet shift thought to arise from an increased feeding suctorial force resulting from larger body size (McCormick 1998). A similar abrupt trophic shift is suggested by the relationship of size and trophic position in BDC Channel Catfish. Allometric growth of body dimensions such as head and gape width resulted in trophic level increasing with size in larger individuals. Subsequently, the largest Channel Catfish adults began to overlap in trophic level with smaller Largemouth Bass. In Lake Texoma, size-related diet shifts occurred in Channel Catfish, with piscivory increasing in larger individuals (Edds et al. 2002). In that population, the diet shift occurred at about 300 mm SL. Though trophic level increased with body size, an abrupt trophic shift was not apparent in Channel Catfish in a Florida lake (Fry et al. 1999); however, all Channel Catfish in that study were less than 800 g. Channel Catfish from BDC with a gape greater than 34 mm ranged from 719 to 6870 g. We observed a trophic shift at a 34-mm gape width in our study stream, but acknowledge that prey availability can influence such shifts in a location-specific Southeastern Naturalist 389 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 manner (Winemiller 1989). Channel Catfish in the Savannah River had a diverse diet of fish, aquatic insects, and crustaceans as observed through gut analyses (Wiltz 1993). Bowfin are voracious opportunistic predators with a diet of primarily fish and crayfish (Lagler and Applegate 1942, Lagler and Hubbs 1940, Reighard 1903). Bowfin trophic position increased with size in a Florida lake (Fry et al. 1999). The δ15N of Bowfin from BDC ranged an equivalent of 1 trophic level (3.0–4.1), but small sample size precluded thorough analyses of intraspecific variation. As expected, neither the δ15N nor δ13C of any of the large, mobile predators differed between sites. As described above, Longnose Gar (Johnson and Noltie 1996, McGrath et al. 2012, Snedden et al. 1999), Largemouth Bass (Paller et al. 2005), and Channel Catfish (Humphries 1965, Wendel and Kelsch 1999) regularly move further than the 430 m separating these sites. Intraspecific carbon source variation Movement of animals between food webs or ecosystems can result in individuals possessing unique stable isotope signatures that appear as outliers in data (Jardine et al. 2006). Accordingly, SIA has been employed to identify migrants into a river from foreign food webs (Martinez et al. 2001, McCarthy and Waldron 2000). Longnose Gar were unique among the 4 predators in having 2 distinct groups of individuals based on carbon resources as indicated by δ13C. Interestingly, the δ13C of the more 13C-enriched group of Longnose Gar does not overlap with any of the other 5 species in this analysis. Moreover, they do not overlap that of 8 genera of odonate (Dragonfly) nymphs or 3 species of Ameiurus (Bullhead Catfishes) analyzed from this system during the same time period (D.E. Fletcher, unpubl. data). Additionally, we could not find evidence of size, sex, reproductive condition, or site contributing to this difference. Either the 2 unique groups of Longnose Gar were part of 2 distinct food chains within BDC, which seems unlikely particularly given the apparent uniqueness of the δ13C, or a portion of the population had recently immigrated into that part of BDC. Longnose Gar are known to be residents of mid-size streams such as BDC (Marcy et al. 2005), but are also known to migrate from larger rivers into tributary streams during spawning runs in the spring (Johnson and Noltie 1996, McGrath et al. 2012). Documented migration distances of Longnose Gar have ranged from 10 to over 50 km (Johnson and Noltie 1996, McGrath et al. 2012), which exceed the total channel length of BDC and supports the notion that migration is contributing to our observed δ13C patterns. Following a spawning migration, Longnose Gar may remain in the stream as long as 3 months before dispersing. Our collection times would have fallen within the period when migrants may reside in the stream. Additionally, inundation of the floodplain of large river systems can stimulate movements of several kilometers (Snedden et al. 1999), and systems with large floodplains can result in seasonal increases in niche breadth of riverine species (Jepsen and Winemiller 2002). The Savannah River floodplain had been inundated the winter and early spring before our fish collections. The importance of identifying these distinct groups of Longnose Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 390 Gar in BDC was illustrated by Fletcher et al. (2014a), who found the group with the unique δ13C had accumulated an order of magnitude higher concentrations of arsenic in addition to other higher contaminant concentrations. Channel Catfish (Humphries 1965) and Largemouth Bass (Paller et al. 2005) are also often seasonally migratory, but we did not observe distinct groupings based on 13C in these fishes. Either the Channel Catfish and Largemouth Bass collected in BDC all had similar movement patterns or the populations shared the same carbon source. Conclusions Though SIA signatures are known to exhibit extensive spatial variation (Guzzo et al. 2011), a pattern we have observed for less-mobile organisms in the BDC system (Fletcher et al. 2014b), the signatures of these highly mobile fishes did not differ between sites. Overall, stable isotope analyses identified a trophic hierarchy among species, with 2 specialized piscivores—Longnose Gar and Largemouth Bass—occupying the highest trophic position. The largest bodied and more benthic-oriented species, Channel Catfish, occupied the lowest trophic level. The trophic level of Bowfin remains less clear due to our small sample size. Largemouth Bass exhibited greater trophic variation than Longnose Gar, but trophic level in both species increased in a gradual linear pattern across the broad size ranges analyzed. Even though our Largemouth Bass ranged over a 160-fold difference in body mass, trophic level increased gradually with body size as did Longnose Gar, which exhibited a 15-fold body mass range. If any abrupt trophic shifts occurred in these 2 species, the shifts occurred at sizes smaller than we analyzed. Trophic variation among similar-sized individuals was relatively low, but appeared to be highest in the smallest individuals. In contrast, Channel Catfish showed a more abrupt trophic shift associated with greater individual variation, particularly in the size ranges near where the shift started. Trophic position increased with gape width in large individuals. The importance of such trophic shifts is illustrated in Channel Catfish, for which mercury concentrations of human health concern were found in the muscle only of individuals that had begun the trophic shift (Fletcher et al. 2014a). The δ13C analysis identified 2 groups of Longnose Gar, suggesting membership in different food chains. Stable isotope analyses hold much potential in studying spatial foraging patterns (Jepsen and Winemiller 2002) and the role of fish migrations in connecting geographically separate food webs in coastal-plain streams. Future studies should investigate the underlying causes of the observed variation among and within these species. As many streams in the Southeast’s upper coastal plain have been extensively impacted by anthropogenic stressors (EPA Report 2013), understanding trophic relationships is an essential component to be incorporated into assessments related to stream disturbance, preservation, restoration, or recovery in the region. Acknowledgments Funding was provided by the Area Completion Projects-SRNS. This work was also supported by the Department of Energy under Award Number DE-FC09-07SR22506 to the Southeastern Naturalist 391 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 University of Georgia Research Foundation. We thank Gary Meffe for insightful comments that improved this manuscript and David Kling, Cynthia Tant, Beryl Walker, and Chandler Tuckfield for field and lab assistance, and Tom Maddox for SIA. Disclaimer: This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. Literature Cited Anderson, C., and G. Cabana. 2007. Estimating the trophic position of aquatic consumers in river food webs using stable nitrogen isotopes. Journal of the North American Benthological Society. 26:273–285. Atkinson, C.L., S.P Opsahl, A.P Covich, S.W. Golladay, and L.M. Conner. 2010. Stable isotopic signatures, tissue stoichiometry, and nutrient cycling (C and N) of native and invasive freshwater bivalves. Journal of the North American Benthological Society 29:496–505. Bearhop, S., C.E. Adams, S. Waldron, R.A. Fuller, and H. MacLeod. 2004. Determining trophic niche width: A novel approach using stable isotope analysis. Journal of Animal Ecology 73:1007–1012. Bechara, J.A., G. Moreau, and D. Planas. 1992. Top-down effects of Brook Trout (Salvelinus fontinalis) in a boreal forest stream. Canadian Journal of Fisheries and Aquatic Sciences 49:2093–2103. Cabana, G., and J.B. Rasmussen. 1996. Comparison of aquatic food chains using nitrogen isotopes. Proceedings of the National Academy of Sciences of the USA 93:10844–10847. Caprio, J.G. Brand, J.H. Teeter, T. Valentincic, D.L. Kalinoski, J. Kohbara, T. Kumazawa, and S. Wegert. 1993. The taste system of the Channel Catfish: From biophysics to behavior. Trends in Neurosciences 16:192–197. Carlander, K.D. 1969. Handbook of Freshwater Fishery Biology, Vol. 1. The Iowa State University Press, Ames, IA. 752 pp. Chumchal, M.M., and K.D. Hambright. 2009. Ecological factors regulating mercury contamination of fish from Caddo Lake, Texas, USA. Environmental Toxicology and Chemistry 28:962–972. Collier, K.J., S. Bury, and M. Gibbs. 2002. A stable isotope study of linkages between stream and terrestrial food webs through spider predation. Freshwater Biology 47:1651–1659. Edds, D.R., W.J. Matthews, and F.P. Gelwick. 2002. Resource use by large catfishes in a reservoir: Is there evidence for interactive segregation and innate differences? Journal of Fish Biology 60:739–750. Eddy, S., and J.C. Underhill. 1974. Northern Fishes. University of Minnesota Press, Minneapolis, MN. 414 pp. Englund, G. and J.J. Krupa. 2000. Habitat use by crayfish in stream pools: Influence of predators, depth, and body size. Freshwater Biology 43:75–83. Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 392 Environmental Protection Agency (EPA). 2013. National rivers and streams assessment 2008–2009: A collaborative survey. Draft. EPA/841/D-13/001. Washington, DC. Ferrara, A.M., and E.R. Irwin. 2001. A standardized procedure for internal sex identification in Lepisosteidae. North American Journal of Fisheries Management 21:956–961. Finlay, J.C. 2001. Stable-carbon-isotope ratios of river biota: Implications for energy flow in lotic food webs. Ecology 82:1052–1064. Finlay, J.C. 2004. Patterns and controls of lotic algal stable carbon isotope ratios. Limnology and Oceanography 49:850–861. Finlay, J.C., M.E. Power, and G. Cabana. 1999. Effects of water velocity on algal carbon isotope ratios: Implications for river food-web studies. Limnology and Oceanography 44:1198–1203. Finlay, J.C., S. Khandwala, and M.E. Power. 2002. Spatial scales of carbon flow in a river food web. Ecology 83:1845–1859. Fletcher, D.E., A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur. 2014a. Variation in trace-element accumulation in predatory fishes from a stream contaminated by coal-combustion waste. Archives of Environmental Contamination and Toxicology 66:341–360. Fletcher, D.E., A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur. 2014b. Spatial and taxonomic variation in trace-element bioaccumulation in two herbivores from a coal combustion-waste–contaminated stream. Ecotoxicology and Environmental Safety 101:196–204. France, R.L. 1996. Carbon-13 conundrums: Limitations and cautions in the use of stable isotope analysis in stream ecotonal research. Canadian Journal of Fisheries and Aquatic Sciences 53:1916–1919. Fry, B., P.L. Mumford, F. Tam, D.D. Fox, G.L. Warren, K.E. Havens, and A.D. Steinman. 1999. Trophic position and individual feeding histories of fish from Lake Okeechobee, Florida. Canadian Journal of Fisheries and Aquatic Sciences 56:590–600. Fuentes Brito, E., T.P. Moulton, M.L. De Souza, and S.E. Bunn. 2006. Stable isotope analysis indicates microalgae as the predominant food source of fauna in a coastal forest stream, southeast Brazil. Austral Ecology 31:623–633. Gatz, Jr., A.J. 1979. Ecological morphology of freshwater stream fishes. Tulane Studies in Zoology and Botany 21:91–124. Gelwick, F.P. 2000. Grazer identity changes the spatial distribution of cascading trophic effects in stream pools. Oecologia 125:573–583. Gu, B., C.L. Schelske, and M.V. Hoyer. 1996. Stable isotopes of carbon and nitrogen as indicators of diet and trophic structure of the fish community in a shallow hypereutrophic lake. Journal of Fish Biology 49:1233–1243. Guzzo, M.M., G.D. Haffner, S. Sorge, S.A. Rush, and A.T. Fisk. 2011. Spatial and temporal variabilities of 13C and 15N within lower trophic levels of a large lake: Implications for estimating trophic relationships of consumers. Hydrobiologia 675:41–53. Hakenkamp, C.C., S.G. Ribblett, M.A. Palmer, C.M. Swan, J.W. Reid, and M.R. Goodison. 2001. The impact of an introduced bivalve (Corbicula fluminea) on the benthos of a sandy stream. Freshwater Biology 46:491–501. Halverson, N.V., L.D. Wike, K.K. Patterson, J.A. Bowers, A.L. Bryan, K.F. Chen, C.L. Cummins, B.R. de Carmen, K.L. Dixon, D.L. Dunn, G.P. Friday, J.E. Irwin, R.K. Kolka, H.E. Mackey, Jr., J.J. Mayer, E.A. Nelson, M.H. Paller, V.A. Rogers, W.L. Specht, H.M. Westbury, and E.W. Wilde. 1997. SRS environmental information document— SRS ecology. Chapter 5. Streams, reservoirs, and the Savannah River, Section 5.2-Beaver Dam Creek drainage description and surface hydrology. WSRC-TR -97- 0223. Westinghouse Savannah River Company, Savannah River Site, Aiken, SC. Pp. 5-67–5-119. Southeastern Naturalist 393 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 Hecky, R.E., and R.H. Hesslein. 1995. Contributions of benthic algae to lake food webs as revealed by stable isotope analysis. Journal of the North American Benthological Society 14:631–653. Henderson, B.L., M.M. Chumchal, R.W. Drenner, Y. Deng, P. Diaz, and W.H. Nowlin. 2012. Effects of fish on mercury contamination of macroinvertebrate communities of grassland ponds. Environmental Toxicology and Chemistry 31:870–876. Herwig, B.R., D.A. Soluk, J.M. Dettmers, and D.H. Wahl. 2004. Trophic structure and energy flow in backwater lakes of two large floodplain rivers assessed using stable isotopes. Canadian Journal of Fisheries and Aquatic Sciences 61:2–22. Herwig, B.R., D.H. Wahl, J.M. Dettmers, and D.A. Soluk. 2007. Spatial and temporal patterns in the food-web structure of a large floodplain river assessed using stable isotopes. Canadian Journal of Fisheries and Aquatic Sciences 64:495–508. Humphries, R.L. 1965. A study of the movements of the Channel Catfish, Ictalurus lacustris punctatus, in the Savannah River and one of its tributaries within the AEC Savannah River Operations Area. Institute of Radiation Ecology, University of Georgia, Athens, GA. TID-21791. 53 p.p. Jackson, D.A., P.R. Peres-Neto, and J.D. Olden. 2001. What controls who is where in freshwater fish communities: The roles of biotic, abiotic, and spatial factors. Canadian Journal of Fisheries and Aquatic Sciences 58:157–170. Jardine, T.D., K. Kidd, and A. Fisk. 2006. Applications, considerations, and sources of uncertainty when using stable isotope analysis in ecotoxicology. Environmental Science and Technology 40:7501–7511. Jepsen, D.B., and K.O. Winemiller. 2002. Structure of tropical river food webs revealed by stable isotope ratios. Oikos 96:46–55. Johnson, B.L., and D.B. Noltie. 1996. Migratory dynamics of stream-spawning Longnose Gar (Lepisosteus osseus). Ecology of Freshwater Fish 5:97–107. Keast, A. 1985. The piscivore feeding guild of fishes in small freshwater ecosystems. Environmental Biology of Fishes 12:119–129. Keast, A., and D.Webb. 1966. Mouth and body form relative to feeding ecology in the fish fauna of a small lake, Lake Opinicon, Ontario. Journal of the Fisheries Research Board of Canada 23:1845–1874. Kondratieff, B.C., and J.R. Voshell, Jr. 1980. Life history and ecology of Stenonema modestum (Banks) (Ephemeroptera: Heptageniidae) in Virginia, USA. Aquatic Insects 2:177–189. Kupfer, A., R. Langel, S. Scheu, W. Himstedt, and M. Maraun. 2006. Trophic ecology of a tropical aquatic and terrestrial food web: Insights from stable isotopes (15N). Journal of Tropical Ecology 22:469–476. Lagler, K.F., and V.C. Applegate. 1942. Further studies of the food of the Bowfin (Amia calva) in southern Michigan, with notes on the inadvisability of using trapped fish in food analysis. Copeia 1942:190–191. Lagler, K.F., and F.V. Hubbs. 1940. Food of the Longnosed Gar (Lepisosteus osseus oxyurus) and the Bowfin (Amia calva) in southern Michigan. Copeia 1940:239–241. Lake. J.L., R.A. McKinney, F.A. Osterman, R.J. Pruell, J. Kiddon, S.A. Ryba, and A.D. Libby. 2001. Stable nitrogen isotopes as indicators of anthropogenic activities in small freshwater systems. Canadian Journal of Fisheries and Aquatic Sciences 58:870–878. Layman, C.A., K.O. Winemiller, D.A. Arrington, and D.B. Jepsen. 2005. Body size and trophic position in a diverse tropical food web. Ecology 86:253 0–2535. Marcy, B.C., D.E. Fletcher, F.D. Martin, M.H. Paller, and M.J.M. Reichert. 2005. Fishes of the Middle Savannah River Basin: With Emphasis on the Savannah River Site. The University of Georgia Press, Athens, GA. 462 pp. Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 394 Martinez, P.J., B.M. Johnson, and J.D. Hobgood. 2001. Stable isotope signatures of native and nonnative fishes in upper Colorado River backwaters and ponds. Southwestern Naturalist 46:311–322. McCarthy, I.D., and S. Waldron. 2000. Identifying migratory Salmo trutta using carbon and nitrogen stable isotope ratios. Rapid Communications in Mass Spectrometry 14:1325–1331. McCormick, M.I. 1998. Ontogeny of diet shifts by a microcarnivorous fish, Cheilodactylus spectabilis: Relationship between feeding mechanics, microhabitat selection, and growth. Marine Biology 132:9–20. McCutchan, J.H., Jr., W.M. Lewis, Jr., C. Kendall, and C.C. McGrath. 2003. Variation in trophic shift for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos 102:378–390. McGrath, P.E., E.J. Hilton, and J.A. Musick. 2012. Seasonal distributions and movements of Longnose Gar (Lepisosteus osseus) within the York River System, Virginia. Southeastern Naturalist 11:375–386. McIntosh, A.R., H.S. Greig, S.A. McMurtrie, P. Nystrӧm, and M.J. Winterbourn. 2005. Top-down and bottom-up influences on populations of a stream detritivore. Freshwater Biology 1206–1218. Merritt, R.W., K.W. Cummins, and M.B. Berg (Eds.). 2008. An Introduction to the Aquatic Insects of North America. 4th Edition. Kendall/ Hunt Publishing Co., Dubuque, IA. 1214 pp. Mittelbach, G.G., and L. Persson. 1998. The ontogeny of piscivory and its ecological consequences. Canadian Journal of Fisheries and Aquatic Sciences 55:1454–1465. Moyle, P.B. and J.J. Cech, Jr. 1982. Fishes: An Introduction to Ichthyology. Prentice-Hall, Englewood Cliffs, NJ. 744 pp. Netsch, N.F. 1964. Food and feeding habits of Longnose Gar in central Missouri. Proceedings of the Annual Conference Southeastern Association of Game and Fish Commission. 18:506–511. Netsch, N.F. and A. Witt, Jr. 1962. Contributions to the life history of the Longnose Gar (Lepisosteus osseus) in Missouri. Transactions of the American Fisheries Society 91:251–262. Olson, M.H. 1996. Ontogenetic niche shifts in Largemouth Bass: Variability and consequences for first-year growth. Ecology 77:179–190. Overman, N.C. and D.L. Parrish. 2001. Stable isotope composition of Walleye: 15N accumulation with age and area-specific differences in δ13C. Canadian Journal of Fisheries and Aquatic Sciences 58:1253–1260. Paller, M.H., D.E. Fletcher, T. Jones, S.A. Dyer, J.J. Isely, and J.W. Littrell. 2005. Potential of Largemouth Bass as vectors of 137Cs dispersal. Journal of Environmental Radioactivity 80:27–43. Peckarsky, B.L., A.R. McIntosh, B.W. Taylor, and J. Dahl. 2002. Predator chemicals induce changes in mayfly life history traits: A whole-stream manipulation. Ecology 83:612–618. Post, D.M. 2002. Using stable isotopes to estimate trophic position: Models, methods, and assumptions. Ecology 83:703–718. Post, D.M. 2003. Individual variation in the timing of ontogenetic niche shifts in Largemouth Bass. Ecology 84:1298–1310. Post, D.M., M.L. Pace, and N.G. Hairston, Jr. 2000. Ecosystem size determines food-chain length in lakes. Nature 405:1047–1049. Power, M.E. 1990. Effects of fish in river food webs. Science 250:81 1–814. Power, M.E., W.J. Matthews, and A.J. Stewart. 1985. Grazing minnows, piscivorous bass, and stream algae: Dynamics of a strong interaction. Ecology 66: 1448–1456. Southeastern Naturalist 395 D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 Raikow, D.F. and S.K. Hamilton. 2001. Bivalve diets in a midwestern US stream: A stable isotope enrichment study. Limnology and Oceanography 46:514–522. Reighard, J. 1903. The natural history of Amia calva Linneaeus. Pp. 57–109, In G.H. Parker, (Ed.). Mark Anniversary Volume. Henry Holt and Company, New York, NY. 513 pp. Roach, K.A., J.H. Thorp, and M.D. Delong. 2009. Influence of lateral gradients of hydrologic connectivity on trophic positions of fishes in the Upper Mississippi River. Freshwater Biology 54:607–620. Ruetz, C.R., III, R.M. Newman, and B. Vondracek. 2002. Top-down control in a detritusbased food web: Fish, shredders, and leaf breakdown. Oecologia 132:307–315. Scharf, F.S., F. Juanes, and R.A. Rountree. 2000. Predator size–prey size relationships of marine fish predators: Interspecific variation and effects of ontogeny and body size on trophic-niche breadth. Marine Ecology Progress Series 208:229–2 48. Schlosser, I.J. 1987. The role of predation in age- and size-related habitat use by stream fishes. Ecology 68:651–659. Schlosser, I.J. 1988. Predation risk and habitat selection by two size classes of a stream cyprinid: Experimental test of a hypothesis. Oikos 52:36–40. Sheldon, A.L., and G.K. Meffe. 1993. Multivariate analysis of feeding relationships of fishes in blackwater streams. Environmental Biology of Fishes 37 :161–171. Snedden, G.A., W.E. Kelso, and D.A. Rutherford. 1999. Diel and seasonal patterns of Spotted Gar movement and habitat use in the Lower Atchafalaya River Basin, Louisiana. Transactions of the American Fisheries Society 128:144–154. Unrine, J.M., W.A. Hopkins, C.S. Romanek, and B.P. Jackson. 2007. Bioaccumulation of trace elements in omnivorous amphibian larvae: Implications for amphibian health and contaminant transport. Environmental Pollution 149:182–192. Vander Zanden, M.J., and Y. Vadeboncoeur. 2002. Fishes as integrators of benthic and pelagic food webs in lakes. Ecology 83:2152–2161. Vander Zanden, M.J., G. Cabana, and J.B. Rasmussen. 1997. Comparing trophic position of freshwater fish calculated using stable nitrogen isotope ratios (15N) and literature dietary data. Canadian Journal of Fisheries and Aquatic Sciences 54:1142–1158. Wayland, M., and K.A. Hobson. 2001. Stable carbon, nitrogen, and sulfur isotope ratios in riparian food webs on rivers receiving sewage and pulp-mill effluents. Canadian Journal of Zoology 79:5–15. Wendel, J.L., and S.W. Kelsch. 1999. Summer range and movement of Channel Catfish in the Red River of the North. Pp. 203–214, In E.R. Irwin, W.A. Hubert, C.F.Rabeni, H.L. Schramm, Jr., and T. Coon (Eds.). Catfish 2000: Proceedings of the International Ictalurid Symposium. American Fisheries Society, Symposium 24, Bethesda, MD. 516 pp. Wikramanayake, E.D. 1990. Ecomorphology and biogeography of a tropical stream fish assemblage: Evolution of assemblage structure. Ecology 71:1756–1764. Williams, L.R., C.M. Taylor, and M.L. Warren. 2003. Influence of fish predation on assemblage structure of macroinvertebrates in an intermittent stream. Transactions of the American Fisheries Society 132:120–130. Wilson, D.S. 1975. The adequacy of body size as a niche difference. The American Naturalist 109:769–784. Wiltz, J.W. 1993. A survey of the fishes of Beaverdam Creek and the adjacent Savannah River, Burke County, Georgia. Georgia Power Company. Atlanta, GA. 126 pp. Winemiller, K.O. 1989. Ontogenetic diet shifts and resource partitioning among piscivorous fishes in the Venezuelan Ilanos. Environmental Biology of Fishes 26:177–199. Southeastern Naturalist D.E. Fletcher, A.H. Lindell, G.K. Stillings, G.L. Mills, S.A. Blas, and J V. McArthur 2015 Vol. 14, No. 2 396 Woodward, G., and A.G. Hildrew. 2002. Body-size determinants of niche overlap and intraguild predation within a complex foodweb. Journal of Animal Ecology 71:1063–1074. Zah, R., P. Burgherr, S.M. Bernasconi, and U. Uehlinger. 2001. Stable isotope analysis of macroinvertebrates and their food sources in a glacier stream. Freshwater Biology 46:871–882. Zar, J.H. 1984. Biostatistical Analysis, Second Edition. Prentice-Hall, Inc, Englewood Cliffs, NJ. 718 pp.