2009 NORTHEASTERN NATURALIST 16(3):365–374 Effects of Zebra and Quagga Mussel (Dreissena spp.) Invasion on the Feeding Habits of Sternotherus odoratus (Stinkpot) on Presque Isle, Northwestern Pennsylvania James C. Patterson1,2 and Peter V. Lindeman1,* Abstract - We investigated diet and its relationship to trophic morphology in Sternotherus odoratus (Stinkpot) in northwestern Pennsylvania on Presque Isle, Lake Erie. Three taxa were most prevalent in fecal samples: invasive Eurasian mussels, small snails, and trichopteran larvae. No sexual difference in diet was apparent, although males had relatively wider heads than females. Significant positive correlation of proportion of sample volume composed of mussels with width of the head and alveolar surfaces was accompanied by nonsignificant negative correlation of proportion composed of snails with both variables. The results thus suggest a shift in molluscan prey preference with increasing size of the trophic apparatus. Total consumption of mollusks was high relative to most other reports of Stinkpot diet. In the Laurentian Great Lakes, the Stinkpot is the second turtle species found to prey heavily upon invasive mussels and thereby participate in transferring production from the pelagic zone to the littoral zone. Introduction Trophic morphology of a species is related to the types of food that the animal has the ability to consume. In turtles, characteristics of the skull’s dimensions, shape, and musculature relate to whether the diet is dominated by algae and plant matter, soft-bodied animal prey, an omnivorous mix of animal matter and vegetation, or hard-shelled prey (Berry 1975, Claude et al. 2004, Herrel et al. 2002, Lindeman 2000, Tinkle 1958). The relationship of trophic morphology to diet has been studied in many durophagous species of turtles. Graptemys spp. (map turtles and sawbacks) are extremely sexually dimorphic in both body size and trophic morphology. Females are up to 2.6 times longer than males (Gibbons and Lovich 1990) and display greater trophic specialization for mollusk feeding than do males (Lindeman 2000, Lindeman and Sharkey 2001). In Graptemys versa Stejneger (Texas Map Turtle), both absolute and relative increases in the head and alveolar widths of females are associated with increased sizes of clams that are consumed (Collins and Lindeman 2006). Dreissena polymorpha (Pallas) (Zebra Mussel) was introduced to the Great Lakes of North America in the 1980s from its native habitats in the Caspian Sea of Europe (Hebert et al. 1989, May and Marsden 1992). Recently, the introduction of D. polymorpha and its congener D. bugensis 1Department of Biology and Health Services, Edinboro University of Pennsylvania, Edinboro, PA 16444. 2Current address - Department of Biology, University of Massachusetts Dartmouth, 285 Old Westport Road, North Dartmouth, MA 02747. *Corresponding author - plindeman@edinboro.edu. 366 Northeastern Naturalist Vol. 16, No. 3 Andrusov (Quagga Mussel) has created concern as they have heavily colonized not only hard substrates (Hebert et al. 1991, Mackie 1991), but have also moved onto sedimentary substrates (Haltuch et al. 2000), producing significant ecological change in the Great Lakes. Exotic species may benefit some native species in the environments they invade. For example, Neogobius melanostomus (Pallas) (Round Goby) also arrived in the Great Lakes region from the Caspian Sea and is now a primary food source of Nerodia sipedon insularum (Conant & Clay) (Lake Erie Water Snake). Due to the abundance of gobies, the snakes grow faster and attain larger body sizes (King et al. 2006). On Presque Isle, a Lake Erie sandspit peninsula in Erie County, PA, female Graptemys geographica Lesueur (Common Map Turtle) consume Zebra and Quagga Mussels as a major food source (Lindeman 2006b). Substantial use of Zebra and Quagga Mussels by the same species has also been reported in Lake Opinicon, ON, Canada (Bulté and Blouin-Demers 2008). Sternotherus odoratus (Latreille) (Stinkpot) is a kinosternid turtle species that exhibits little sexual size dimorphism (Gibbons and Lovich 1990). It has been reported to feed on snails and clams, sometimes heavily, but also takes a wide variety of other types of primarily benthic prey, including insects and their larvae, vegetation, seeds, crayfish, amphibians, fish, and carrion (Berry 1975, Ford and Moll 2004, Lagler 1943, Mahmoud 1968). In congeneric species, the diet changes from primarily insects and their larvae to increasing amounts of mollusks as turtles grow larger (Tinkle 1958). As features of the trophic morphology increase in size concomitant with growth in body size, a preference for larger and harder prey may occur, due to increasing ability to consume prey items that are more difficult to crush. No studies to date have addressed how invasion by Zebra and Quagga Mussels might impact the diet of the Stinkpot (Moll and Moll 2004). The purpose of this study is to report on the feeding habits of Stinkpots on Presque Isle, the waters of which have been heavily colonized by Zebra and Quagga Mussels. We hypothesized that Stinkpots would consume large amounts of mussels due to the abundance of the mussels and reports from elsewhere of substantial take of molluscan prey. We further hypothesized that the size of the head and alveolar surfaces would determine the amount of predation on mussels due to the nature of the mussels’ shells and the role these morphological attributes play in allowing the turtles to use them as a major food source. Methods and Materials Data collection was conducted at Presque Isle State Park on the southcentral shoreline of Lake Erie. The peninsula extends out to the northeast for approximately 10 km, almost completely encircling Presque Isle Bay, and is historically considered to have been a floating sandspit that was pushed into its present shape by winds and wave action. The Park has several swimming beaches and bicycle and walking paths. The undeveloped areas of the 2009 J.C. Patterson and P.V. Lindeman 367 peninsula consist of marshes and lagoons located between the southcentral portion and the easternmost tip. The inland portions of the peninsula are primarily sandy soils that are forested with shrubs and cottonwood trees. The areas sampled in the present study were Misery Bay, which is located bayside on the eastern end of the peninsula, and Graveyard Pond directly to the west of Misery Bay. These bodies of water have sandy substrates with heavy vegetative growth in the spring and summer months. Both areas are inhabited by six turtle species: Stinkpot, Common Map Turtle, Chrysemys picta (Schneider) (Painted Turtle), Chelydra serpentina (L.) (Common Snapping Turtle), Apalone spinifera (Lesueur) (Spiny Softshell Turtle), and Emydoidea blandingii (Holbrook) (Blanding’s Turtle). Turtles were captured using fyke nets 15 m in length with a 2.5-cm mesh. Turtles were marked according to the methods of a long-term study employed in the area (Lindeman 2006b). Samples were taken from turtles captured in May−September of 2005 and 2006. Measurements were taken on each specimen using dial calipers to determine straight-line carapace length (SCL) to the nearest mm and head width (HW) and alveolar width (AW) to the nearest 0.1 mm. The turtles were then placed in plastic bins in a few centimeters of water and held overnight in a laboratory to collect feces. All captured turtles were returned to their points of capture within 24–48 hours. Fecal remains were collected by straining with a sieve and preserved in 10% formalin for later analysis under a dissecting microscope. The remains were categorized and then measured by volumetric displacement of water to the nearest 0.1 ml. For prey taxa that displaced less than 0.1 ml, volume was estimated to be either 0.01 or 0.05 ml. For each prey category i, mean percent volume (Vi, averaged across all samples) and percent frequency of occurrence (fi, the percent of samples in which the taxon was present) were used to calculate an index of relative importance (IRIi): IRIi = 100Vifi/ Σ(Vifi), where IRIi values total 100 (Bjorndal et al. 1997). Separate calculations of IRI were conducted for males, females, and all samples. Morphometric data were log10 transformed for analyses. Regression analyses were performed to test for intercorrelation of HW, AW, and SCL. Two separate ANCOVA analyses were performed. The first tested for gender differences in HW with SCL as the covariate. The second tested for gender differences in AW with HW as the covariate. Regression analyses were performed to test for correlation of relative consumption of molluscan taxa with log10 transformed body size and trophic morphology variables, using the arcsine square root of the proportion of each sample composed of mussels or snails. Results There were 39 fecal samples collected for this study. Four were withheld from analyses due to small total sample volume (<0.05 ml), and one 368 Northeastern Naturalist Vol. 16, No. 3 contained only parasitic nematodes and also was not used. Of the remaining 34 samples, 21 were from males and 13 were from females. Measurements of trophic morphology were taken from 28 of the 39 turtles sampled for feces and six additional turtles not detained for fecal samples (Table 1). Zebra and Quagga Mussels, small snails, and caddisfly larvae dominated in fecal samples. Mussels and snails consisted mostly of fragmented shell remains, while caddisfly remains were primarily larval cases. Combined IRI values for these three taxa totaled 88 for males, 91 for females, and 89 for combined samples (Table 2). Mussels were most prevalent, with IRI values of 62 for males, 60 for females, and 62 for combined samples. ANCOVA results demonstrated significantly greater HW in males than in females, but no difference between the sexes in AW (Fig. 1). The covariate SCL significantly influenced HW (F1,30 = 259.6, P < 0.0001) as did sex (F 1,30 = 19.8, P = 0.0001), while the interaction term was not significant (F 1,30 = 2.7, P = 0.11). The covariate HW significantly influenced AW (F 1,30 = 29.7, P < 0.0001), while there was no significant influence of sex (F 1,30 = 3.5, P = 0.07) or the interaction term (F 1,30 = 0.0001, P = 0.99). Transformed values of percent mussels were significantly positively correlated with HW and AW (r = 0.42, P = 0.044 for HW; r = 0.44, P = 0.035 for AW; Fig. 2). Transformed values of percent snails were nonsignificantly Table 1. Summary of morphometric measurements of Stinkpots at Presque Isle State Park, PA. All measurements are in mm. SCL = Straight-line carapace length. Males Females Variable n Mean SE Range n Mean SE Range SCL 26 103.1 2.00 69−117 19 103.0 3.48 74−126 Head width 19 22.60 0.59 14.6−25.7 15 21.23 0.58 16.6−24.8 Alveolar width 19 5.83 0.20 4.4−7.3 15 5.42 0.21 4.2−6.4 Table 2. Dietary data for 34 Stinkpots (21 males and 13 females) from Presque Isle State Park, PA. M = males, F = females, and C = combined. Index of Percent frequency Mean percent volume relative importance M F C M F C M F C Zebra and Quagga Mussels 95 92 94 54 47 51 62 60 62 Snails 71 77 74 24 26 25 21 28 23 Fingernail clams 10 23 15 0.4 1 0.8 0.05 0.45 0.15 Crayfish 10 0 6 0.4 0 0.2 0.05 0 0.02 Trichopterans 71 46 62 5 5 5 5 3 4 Dipterans 5 8 6 0.1 0.07 0.09 0.006 0.007 0.007 Unidentified insects 38 15 29 3 8 5 1 2 2 Fish 5 8 6 0.1 0.2 0.1 0.006 0.02 0.008 Plant leaves 86 54 74 10 4 8 10 3 7 Plant stems 57 23 44 3 0.4 2 2 0.1 1 Filamentous algae 5 0 3 0.06 0 0.03 0.003 0 0.001 Stalked algae 19 8 15 0.4 0.4 0.4 0.09 0.04 0.08 Seeds 5 31 15 0.3 8 3 0.02 3 0.6 2009 J.C. Patterson and P.V. Lindeman 369 negatively correlated with both variables (r = -0.20, P = 0.36 for HW; r = -0.12, P = 0.58 for AW; Fig. 2). Discussion The IRI data show heavy consumption of Zebra and Quagga Mussels, which were present in all but two of the 34 fecal samples and averaged just over half of fecal volume. A shift in primary food source from insects and small snails to mussels occurs with the growth of the turtle and its trophic Figure 1. Correlation of head width (HW) with straight-line carapace length (SCL) and of alveolar width (AW) with HW for the Stinkpot. Males are represented by filled symbols and females by open symbols. The sexual difference in relative HW was significant in ANCOVA with log10 transformation of both variables, while no significant sexual difference in relative AW was found in ANCOVA with log10 transformation of both variables (see text). 370 Northeastern Naturalist Vol. 16, No. 3 apparatus. This trend may be the result of the dietary benefits of eating larger mussels instead of larger numbers of the smaller, thinner-shelled snails. The change in diet was correlated with the increase in head and alveolar widths, which presumably allow greater force for crushing the larger and thicker shells of mussels. The benefits of consuming larger prey may occur in time spent searching for food, as fewer mussels would be needed to sustain the Figure 2. Correlation of transformed proportion of fecal sample composed of molluscan prey taxa with log10- transformed head width (HW) and log10-transformed alveolar width (AW) of the Stinkpot. Regression lines show significant positive relationships involving mussels; relationships involving snails were negative but not signifi- cant (see text). 2009 J.C. Patterson and P.V. Lindeman 371 dietary needs of a turtle. Tinkle (1958) similarly showed increasing consumption of harder prey with increasing body size in three other species of Sternotherus, and Berry (1975) found that in S. minor minor (Agassiz) (Loggerhead Musk Turtle) in Florida, variation in head width was correlated with the dietary abundance of mollusks in five habitats. Three previous studies of Stinkpot diet have reported both percent frequency of occurrence and percent volume (albeit total rather than mean) and so can be used for comparison of IRI values (Table 3). While the four studies show similar variety of prey in the diet, the present study stands out as being least like the others in its high IRI value for mollusks and its low value for insects. An additional study that did not report percent frequency of occurrence found that Stinkpots in one of three Florida populations fed very heavily on mollusks (79% of adult volume and 91% in juveniles), although the primary mollusks taken were tiny snails (Berry 1975). Some bias in fecal samples may occur due to the indigestibility of mollusk shells, but Lagler’s (1943) comparison of material from stomachs and material from colons suggests this bias is not enough to produce the contrast among studies in the IRI values shown in Table 3. Instead, the abundance of a novel food source appears to have caused a shift in the diet of Stinkpot in Lake Erie. The shift toward increased molluscivory is similar to that of species of map turtles whose females have moderately broadened heads and have switched to heavy consumption of invasive bivalves (Lindeman 2000, 2006a, 2006b). In Lake Erie, Stinkpots predominantly feed upon the same major prey taxa as Common Map Turtles, namely Zebra and Quagga Mussels, snails, and caddisfly larvae (Lindeman 2006b). Unlike the case with Common Map Turtles, however, there was little difference in the diets of males and Table 3. Index of relative importance values from literature reports of Stinkpot diet and recalculated from the present study after combining prey taxa for ease of comparison. OklahomaB MissouriC MichiganA (stomachs (stomach flushes PennsylvaniaD Prey taxon Stomachs Colons and colons) and feces) (feces) Mollusks 41 44 25 39 83 Insects 36 32 49 10 8 Crayfish 1 5 3 8 0.01 Amphipods/isopods 0.006 Arachnids 0.002 Amphibians 0.06 Fish 0.3 0.1 0.2 0.009 Fish carrion 15 0.8 1 Plants 3 16 21 18 8 Seeds 8 0.5 Algae 3 2 17 0.1 A Lagler 1943. B Mahmoud 1968. C Ford and Moll 2004. DPresent study. 372 Northeastern Naturalist Vol. 16, No. 3 females, with IRI values nearly matching for all dietary taxa. Ford and Moll (2004) likewise found similar diets in male and female Stinkpots. The lack of dietary differences between the sexes is not surprising, given their similar body size. However, the present study is the first report of a sexual difference in trophic morphology, with males having wider heads than females. The difference does not manifest itself in clear sexual dietary differences (Table 2); however, mussel size was not investigated, and it is possible that males’ larger heads allow them to feed on larger mussels. A similar case occurs in Common Map Turtles in an Ontario population, where larger-headed females are able to consume larger, harder snails (Bulté et al. 2008). Alternative hypotheses for the existence of sexual dimorphism in head width should also be investigated. Male Stinkpots sometimes clasp females in their jaws during courtship (Mahmoud 1967). In many lizards with similar mating behavior, males have larger heads than females (Vincent and Herrel 2007); however, it is unclear whether larger head size is advantageous in male-male combat for mates (a phenomenon unknown in Stinkpots) or for the act of mating (McBrayer and Anderson 2007). Furthermore, the biting hold of the female by the male in Stinkpots is apparently maintained only briefly (Mahmoud 1967), making an adaptive role of increased bite force in males comparatively more difficult to envision. In species of Graptemys, all of which display extreme sexual size dimorphism, sexual differences in head and alveolar width are associated with a difference in diet between the sexes (Lindeman 2000, 2006a; Lindeman and Sharkey 2001). In Graptemys, females of species termed megacephalic have a more exclusively molluscivorous diet than females of species having the mesocephalic and microcephalic morphologies; males exhibit no such trend, due to their smaller body size and general avoidance of mollusks (Lindeman 2000). Although the Stinkpot does not display the same sexual size dimorphism that Graptemys spp. do, they show a shift in diet with growth in head and alveolar width, as exhibited by females of some Graptemys species (Collins and Lindeman 2006; Lindeman 2006a, b). In Poland, Zebra and Quagga Mussels are fed upon by the fish Rutilus rutilus (Linnaeus) (Roach), but only by individuals greater than 160 mm in standard length (and only heavily by individuals greater than 210 mm), apparently because the small mussels that smaller fish could take would entail high handling-time costs and be of low energetic value (Prejs et al. 1990). A similar phenomenon may explain the size-related predation patterns observed for North American turtles reported to feed on Zebra and Quagga Mussels (Bulté et al. 2008, Lindeman 2006b, present study) and warrants future investigation. Bulté and Blouin-Demers (2008) discussed the consumption of Zebra and Quagga Mussels by Common Map Turtles in the light of the bioenergetics of a lake in Ontario. Substantial use of these prey resulted in a shift from participation in a strictly littoral food web to substantial transfer of energy from pelagic trophic pathways (via the phytoplankton on which the mussels 2009 J.C. Patterson and P.V. Lindeman 373 fed) to the littoral zone (where the turtles fed and were active). The same phenomenon appears to occur with Stinkpots. Further studies should be conducted to examine the impact that turtle species such as the Stinkpot and the Common Map Turtle have on the population densities of Zebra and Quagga Mussels in the Great Lakes. Acknowledgments We thank the Regional Science Consortium for use of facilities at the Tom Ridge Environmental Center. The manuscript was read by D. Moll prior to submission. Literature Cited Berry, J.F. 1975. The population effects of ecological sympatry on musk turtles in northern Florida. Copeia 1975:692−701. Bjorndal, K.A., A.B. Bolten, C.J. Lagueux, and D.R. Jackson. 1997. Dietary overlap in three sympatric congeneric freshwater turtles (Pseudemys) in Florida. Chelonian Conservation and Biology 2:430−433. Bulté, G., and G. Blouins-Demers. 2008. Northern Map Turtles (Graptemys geographica) derive energy from the pelagic pathway through predation on Zebra Mussels (Dreissena polymorpha). Freshwater Biology 53:497−508. Bulté, G., D.J. Irschick, and G. Blouins-Demers. 2008. The reproductive role hypothesis explains trophic morphology dimorphism in the northern map turtle. Functional Ecology 22:824–830. Available online at doi 10.1111/j.1365- 2435.2008.01422.x. Claude, J., P.C.H. Pritchard, H. Tong, E. Paradis, and J.-C. Auffray. 2004. Ecological correlates and evolutionary divergence in the skulls of turtles: A geometric morphometric assessment. Systematic Biology 53:933−948. Collins, D., and P.V. Lindeman. 2006. The influence of body size and trophic morphology on the size of molluscan prey of the female Texas Map Turtle (Graptemys versa). Herpetological Review 37:416−418. Ford, D.K., and D. Moll. 2004. Sexual and seasonal variation in foraging patterns in the Stinkpot, Sternotherus odoratus, in southwestern Missouri. Journal of Herpetology 38:296−301. Gibbons, J.W., and J.E. Lovich. 1990. Sexual dimorphism in turtles with emphasis on the Slider Turtle (Trachemys scripta). Herpetological Monographs 4:1−29. Haltuch, M.A., P.A. Berkman, and D.W. Garton. 2000. Geographic information system (GIS) analysis of ecosystem invasion: Exotic mussels in Lake Erie. Limnology and Oceanography 45:1778−1787. Hebert, P. D. N., B.W. Muncaster, and G.L. Mackie. 1989. Ecological and genetic studies on Dreissena polymorpha (Pallas): A new mollusk in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences 46:1587−1591. Hebert, P.D.N., C.C. Wilson, M.H. Murdoch, and R. Lazar. 1991. Demography and ecological impacts of the invading mollusk Dreissena polymorpha. Canadian Journal of Zoology 69:405–409. Herrel, A., J.C. O’Reilly, and A.M. Richmond. 2002. Evolution of bite performance in turtles. Journal of Evolutionary Biology 15:1083−1094. King, R.B., J.M. Ray, and K.M. Stanford. 2006. Gorging on gobies: Beneficial effects of alien prey on a threatened vertebrate. Canadian Journal of Zoology 84:108−115. 374 Northeastern Naturalist Vol. 16, No. 3 Lagler, K.F. 1943. Food habits and economic relations of the turtles of Michigan with special reference to fish management. American Midland Naturalist 29:257−312. Lindeman, P.V. 2000. Evolution of the relative head and alveolar surfaces in Map Turtles (Testudines: Emydidae: Graptemys). Biological Journal of the Linnean Society 69:549−576. Lindeman, P.V. 2006a. Diet of the Texas Map Turtle (Graptemys versa): Relationship to sexually-dimorphic trophic morphology and changes over five decades as influenced by an invasive mollusk. Chelonian Conservation Biology 5:25−31. Lindeman, P.V. 2006b. Zebra and Quagga Mussels (Dreissena spp.) and other prey of a Lake Erie population of Common Map Turtles (Emydidae: Graptemys geographica). Copeia 2006:268−273. Lindeman, P.V., and M.J. Sharkey. 2001. Comparative analyses of functional relationships in the evolution of trophic morphology in the map turtles (Emydidae: Graptemys). Herpetologica 57:313−318. Mackie, G.L. 1991. Biology of the exotic Zebra Mussel, Dreissena polymorpha, in relation to native bivalves and its potential impact in Lake St. Clair. Hydrobiologia 219:251−268. Mahmoud, I.Y. 1967. Courtship behavior and sexual maturity in four species of kinosternid turtles. Copeia 1967:314−319. Mahmoud, I.Y. 1968. Feeding behavior in kinosternid turtles. Herpetologica 24:300−305. May, B., and J.E. Marsden. 1992. Genetic identification and implications of another invasive species of dreissenid mussel in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences 49:1501−1506. McBrayer, L.D., and R.A. Anderson. 2007. Sexual size dimorphisms and bite force in the Northern Alligator Lizard, Elgaria coerulea. Journal of Herpetology 41:554−559. Moll, D.L., and E.O. Moll. 2004. The Ecology, Exploitation, and Conservation of River Turtles. Oxford University Press, Oxford, UK. 393 pp. Prejs, A., K. Lewandowski, and A. Stańczykowska-Piotrowska. 1990. Size-selective predation by Roach (Rutilus rutilus) on Zebra Mussel (Dreissena polymorpha): Field studies. Oecologia 83:378−384. Tinkle, D.W. 1958. The systematics and ecology of the Sternothaerus carinatus complex (Testudinata, Chelydridae). Tulane Studies in Zoology 6:3−56. Vincent, S.E., and A. Herrel. 2007. Functional and ecological correlates of ecologically based dimorphisms in squamate reptiles. Integrative and Comparative Biology 47:172−188.