* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Effects of Food Type, Habitat, and Fish Predation on the Relative
Survey
Document related concepts
Unified neutral theory of biodiversity wikipedia , lookup
Biodiversity action plan wikipedia , lookup
Overexploitation wikipedia , lookup
Reconciliation ecology wikipedia , lookup
Habitat destruction wikipedia , lookup
Biological Dynamics of Forest Fragments Project wikipedia , lookup
Mission blue butterfly habitat conservation wikipedia , lookup
Theoretical ecology wikipedia , lookup
Myxobolus cerebralis wikipedia , lookup
Occupancy–abundance relationship wikipedia , lookup
Transcript
J. Great Lakes Res. 30(1):100–113 Internat. Assoc. Great Lakes Res., 2004 Effects of Food Type, Habitat, and Fish Predation on the Relative Abundance of Two Amphipod Species, Gammarus fasciatus and Echinogammarus ischnus María J. González* and Greta A. Burkart1 Department of Zoology Miami University Oxford, Ohio 45056-1400 ABSTRACT. We evaluated the abundance patterns of Gammarus fasciatus and Echinogammarus ischnus in dreissenid and macrophyte areas in Hatchery Bay, Lake Erie before (1997) and after round goby (2001, 2002) invaded the area. Total amphipod abundance was higher before round goby invasion in both habitats. In mussel beds, E. ischnus abundance was similar or significantly higher than G. fasciatus. In macrophytes, G. fasciatus was significantly more abundant than E. ischnus. In laboratory experiments, we compared amphipod survivorship and growth when fed mussel feces and pseudofeces (F+P) or macrophytes with epiphytes (M+E). Gammarus fasciatus survivorship and growth were higher when fed F+P than M+E. Echinogammarus ischnus showed similar survivorship under both diets, but significantly higher growth when fed M+E than F+P. Therefore inter-habitat differences in food resources cannot explain the abundance patterns observed in the lake. We also estimated the relative vulnerability of G. fasciatus and E. ischnus to yellow perch (Perca flavescens) and round goby (Neogobius melanostomus) predation in laboratory feeding trials using mussel colonies or macrophyte beds as substrate. Both fish strongly preferred E. ischnus in macrophytes, but consumed relatively more G. fasciatus than E. ischnus in dreissenid habitats. Our results suggest that dreissenid establishment may have facilitated the invasion of E. ischnus. However, habitat-specific differences in vulnerability to fish predation may mediate the coexistence of G. fasciatus and E. ischnus by minimizing expansion of E. ischnus to macrophyte areas. Our results also suggest that round goby invasion can alter amphipod abundance patterns in Lake Erie. INDEX WORDS: tence. Amphipods, Dreissena spp., macrophytes, round goby, yellow perch, species coexis- INTRODUCTION The success of biological invasions depends upon physical and biotic conditions in the target community (Elton 1958, Simberloff 1986). However, once an exotic species is established, it can modify abiotic and biotic conditions and facilitate the colonization success of future invaders (Aplet 1990). Introduced species can increase resource availability (Vitousek et al. 1987, Vitousek and Walker 1989), which can facilitate invasibility by other exotic species (Vitousek 1990, Aplet 1990). Furthermore, if invaders can increase resource heterogeneity they may increase the number of species that can coexist (MacArthur et al. 1972, Huston 1994). The success of an invader, however, may be constrained by predation risk (Robinson and Wellborn 1988). Predators can often decrease success of invaders by selectively consuming an exotic prey (Orians 1986). Conversely, predators can also enhance invasibility by selectively consuming a competitively superior native prey, therefore reducing interspecific competition and allowing coexistence of competitors (Lodge 1993). Habitat heterogeneity can be an important factor mediating predator-prey interactions among native and exotic species (Crowder and Cooper 1982). Differences in refuge availability among habitats can mediate interspecific competition and lead to differences in the distribution of potential competitor species (Sih 1987, Duffy and Hay 1991). Predator-prey interactions also are dependent on differences in prey vulnerability to predation, which *Corresponding author. E-mail: [email protected] address: Utah State University, Department of Fisheries and Wildlife, Logan UT 84322. 1Current 100 Amphipod Abundance Patterns in Macrophyte and Dreissenid Habitats are determined by the size, coloration and behavior of prey in relation to the habitat (see Heck and Crowder 1991 for a review). The Great Lakes have been drastically affected by exotic species. The remarkable abundance of the exotic zebra (Dreissena polymorpha) and quagga (Dreissena bugensis) mussels have greatly modified littoral habitats and energy flow in the Great Lakes. Habitat complexity is greater in Dreissena clusters compared to bare rocks (González and Downing 1999). Dreissenids filter seston from the water column and subsequently deposit feces and pseudofeces (seston that is filtered, coated with mucous, then ejected without being ingested) at the lake bottom. Their high filtering capacity has led to increased water clarity (Leach 1992, Budd et al. 2001) and increased light penetration has led to an increase in near-shore macrophytes abundance (Stuckey and Moore 1995, Skubinna et al. 1995) and periphyton (Lowe and Pillsbury 1995). Thus, the dreissenid invasion has diversified littoral habitats due to increased habitat complexity in mussel colonies and increased macrophytes abundance. Mussel colonies and macrophytes may provide macroinvertebrates with different food resources and refuge from predation. The effects of dreissenid invasion on macroinvertebrates on rocky substrates have been extensively studied in the last decade, while the role of increased macrophytes resulting from the dreissenid invasion is not well understood. Benthic invertebrate abundance on rocky substrates increased dramatically following the dreissenid invasion (Dermott et al. 1992, Stewart and Haynes 1994). In particular Gammarus fasciatus, a widely distributed amphipod species in North America (Clemens 1950), became one of the most abundant macroinvertebrates associated with dreissenid colonies (Dermott et al. 1992, Stewart and Haynes 1994, Ricciardi et al. 1997). Higher amphipod abundance on mussels is related to increased habitat complexity (Ricciardi et al. 1997, Stewart et al. 1998, González and Downing 1999). Macrophytes also increased the littoral habitat complexity and are utilized by benthic invertebrates (Gilinsky 1984). Information about amphipod populations in macrophytes in the Great Lakes is scarce, although amphipods can utilize such habitat (Clemens 1950, Dermott et al. 1998, Kashian and Burton 2000). Recently two species from the Ponto-Caspian region have invaded the Great Lakes: the amphipod, Echinogammarus ischnus (Witt et al. 1997) and a fish, the round goby, Neogobius melanostomus 101 (Jude et al. 1992). Echinogammarus ischnus, commonly associated with dreissenid colonies (Köhn and Waterstraat 1990), invaded Lake Erie in 1994 (Witt et al. 1997). Dermott et al. (1998) documented a higher abundance of E. ischnus than G. fasciatus on mussel colonies in several areas in the Great Lakes and suggested that E. ischnus had displaced G. fasciatus from mussel colonies. Echinogammarus ischnus, however, was scarce in macrophyte beds in Lake St. Clair and Lake Ontario (Witt et al. 1997, Dermott et al. 1998). In Lake Michigan, E. ischnus was abundant in sites with dreissenid colonies but was not present in sites with submerged vegetation (Nalepa et al. 2001). Round goby, a benthivorous fish species associated with riprap, vegetated areas and dreissenid colonies (Jude and DeBoe 1996), invaded Lake Erie in 1996 (Charlebois et al. 1997). Round gobies can feed on soft benthic invertebrates, including amphipods, but also prey on dreissenids (Charlebois et al. 1997, Ghedotti et al. 1995, Jude et al. 1995). These two exotic species are likely to interact with G. fasciatus and native fish species such as yellow perch (Perca flavescens). Yellow perch is an important component of the commercial and sport fisheries in the Great Lakes, especially Lake Erie and Lake Michigan (Francis et al. 1995). Perch undergo three ontogenetic niche shifts (Wu and Culver 1991). Larval perch are pelagic and feed predominantly on zooplankton. Juveniles (30–200 mm) shift to become more littoral in their habitat and feed mainly macroinvertebrates. Perch spend several years at this feeding stage until they are large enough to become piscivorous (Hayes and Taylor 1990). In Lake Erie and Lake Michigan, benthic invertebrates such as amphipods are a major component of juvenile yellow perch diet, particularly after Dreissena spp. invasion (Robillard et al. 1999, Tyson and Knight, 2001). Yellow perch also consume Dreissena spp. (Morrison et al. 1997). Food resources vary between mussel colonies and macrophytes beds. Biodeposition of organic matter, mainly in the form of feces and pseudofeces, is higher in the presence than in the absence of mussels (Klerks et al. 1996, Stewart et al. 1998). Feces and pseudofeces are consumed by amphipods (Fisher et al. 1992), although the implications for amphipod fitness are not well understood. Amphipods can also consume macrophytes and associated epiphytes (Clemens 1950, Newman 1991, Delong et al. 1993). Therefore, dreissenid establishment may facilitate the invasion of E. ischnus by 102 González and Burkart providing increased food resources. In addition differences in food resources between dreissenid and macrophyte dominated habitats may determine spatial distributions of E. ischnus and G. fasciatus. Differences between the type of refuge from fish predation provided by mussel and macrophytes may also mediate the distribution of native and exotic amphipods, although the value of the refuge provided by each habitat may depend on predator species. Similar benthic invertebrate biomass was observed in experimental ponds colonized by zebra mussels, with and without yellow perch (Thayer et al. 1997). Kuhns and Berg (1999) reported higher total invertebrate densities in colonization cages with zebra mussel colonies when round gobies were excluded. Therefore, yellow perch and round goby have the potential to affect amphipod abundance patterns. Here we document differences in the distribution of G. fasciatus and E. ischnus in mussel colonies and macrophytes before and after the invasion of round goby. We also analyze how habitat differences in food resources (dreissenid egestion products vs. macrophytes and epiphytes) and predation by native yellow perch and the exotic round goby may influence abundance patterns of these amphipod species. METHODS Amphipod Abundance Patterns We estimated the abundance of G. fasciatus and E. ischnus in mussel colonies and macrophyte beds during late summer in 1996, 2001, and 2002. In 1996, sample sites were chosen at random within habitats dominated by dreissenids (100–80% dreissenid; 0 –20% macrophytes) or macrophytes (100–80% macrophytes; 0–20% dreissenids) in Hatchery Bay (Peach Point and Gibraltar Island). We collected three samples from each habitat type (Dreissena spp. and macrophytes). In each habitat, a bucket (30-cm diameter) with a 90-µm sieve attached to the bottom was placed upside-down and buried 2 cm into the substrate. A thin metal sheet, attached to the bucket using bungee cords, was slid between the bucket and the substrate, and the bucket was carefully lifted from the sediment and placed into a pillowcase. Pillowcases were transported to the laboratory inside coolers in less than 2 hours. Substrates were agitated in a bucket with lake water for at least 25 min to remove amphipods. Amphipods samples were preserved with a 10% formalin solution and identified to species using a dissecting microscope. In 2001 and 2002 (after round goby establishment), amphipod densities were considerably lower than in 1996 and we modified our sampling methodology. A 1 m2 quadrat was randomly placed on the substrate and four samples from each quadrat were taken using the bucket sieve method described above. Samples from each quadrat were pooled. In 2001, we only collected samples from six randomly placed quadrats in areas dominated by dreissenids (100% dreissenids). In 2002, we collected samples from 14 quadrats chosen at random along two 50-m transects (Peach Point and near Perry’s Monument, seven quadrats in each transect). However one quadrat sample from Perry’s Monument was not included in the analysis because no amphipods were present in the sample. The habitat composition (% dreissenids and % macrophytes) in the quadrats was more variable in 2002, so samples were categorized into three groups according to mussel cover (100–70%, n = 3; <70–30%, n = 4; and < 30–0%, n = 6). Macrophytes occupied substrate without dreissenids, except in one quadrat, where the macrophyte cover was 30% and the remaining 70% was bare rock. In both surveys, amphipods were removed from the substrate similarly to 1996, but amphipods were identified and counted immediately without preservation. The percentage of juvenile amphipods that could not be identified was high only in 2001, so we did not include these individuals in our statistical analysis. In 1996 and 2002, we tested for differences in amphipod abundance among habitats using fixed-effects multivariate analysis of variance (MANOVA: treating the abundance of the two amphipod species as dependent variables, SAS Institute 1990). Since significant treatment effects were detected in the MANOVA, we performed interhabitat comparisons (mussel vs. macrophytes) for each species and inter-species comparisons (G. fasciatus vs. E. ischnus) within each habitat using ttests. In 2001, we performed inter-specific comparisons in dreissenid dominated habitat using a Hotelling t-test. All data were log-transformed to meet assumptions of variance homogeneity (Sokal and Rohlf 1988). Effect of Diet on Amphipod Survivorship and Growth We determined the effect of diet on survivorship and growth of G. fasciatus and E. ischnus by rear- Amphipod Abundance Patterns in Macrophyte and Dreissenid Habitats ing each species on two different foods for eight weeks. Individuals of both amphipod species were collected from both mussel colonies and macrophytes in Hatchery Bay during late June, 1997. Amphipods were transported to the laboratory, identified as G. fasciatus or E. ischnus, and placed into two separated aerated tanks containing both mussel egestion products and macrophytes. Therefore we used amphipods from both habitats in the experiment. After 24 hours, ovigerous females of each species were placed in individual cups of lake water. As the F 1 generation hatched, the mother was removed and the new generation was acclimated and fed 5-cm-long macrophyte pieces (Vallisneria americana) for 5 days. We used macrophytes instead of mussel egestion products as a food source during acclimation because in preliminary experiments recently hatched amphipods incubated clung to macrophytes. This facilitated finding and transferring amphipods with minimum handling because there was no detritus accumulation in the bottom of the cups in comparison to cups with egestion products. After 5 days of acclimation amphipods were raised on one of two diets: mussel feces and pseudofeces (F+P) or macrophytes (V. americana) with epiphytes (M+E). Vallisneria americana was used in the M+E treatment because it was the dominant macrophyte in Hatchery Bay (Stuckey and Moore 1995). To ensure that no other food source was present, we filtered lake water used in the experiment through a 1-µm glassfiber filter before adding it to each cup. We also raised amphipods in filtered lake water (no food treatment) to assess the effect of food deprivation on amphipod survivorship and growth. In preliminary observations using several amphipods per cup dead organisms were consumed by live amphipods within a few hours. Therefore we used one amphipod per cup for our experiments. Each amphipod was placed in a 300-mL plastic cup containing 250 mL of filtered lake water and a 1 × 1-cm piece of plastic netting as a substrate. Cups were randomly arranged in six rows, with 27 cups per row. We added 2 mL of a concentrated mixture of feces and pseudofeces or a 15-mm macrophyte segment to each replicate. Feces and pseudofeces for the experiment were collected in a funnel trap placed underneath suspended mussels in a flow-through tank containing Lake Erie water. Fresh macrophytes were collected from the Hatchery Bay on each feeding day. Macrophytes were rinsed gently with filtered lake water to remove at- 103 tached macroinvertebrates without disturbing epiphytes. Food and water were changed weekly. Food was always present in the cups, indicating that amphipods did not starve. Once a week, dissolved oxygen was measured in a randomly chosen sample of each treatment. Dissolved oxygen never dropped below 6.6 mg/L during the entire experiment. Temperature during the experiment was 22.5–23.5°C Each treatment for G. fasciatus had 33 replicates. Replicate numbers for E. ischnus treatments were lower because we obtained fewer offspring from ovigerous E. ischnus than from G. fasciatus. The F+P, M+E and no food treatments for E. ischnus had 24, 19, and 20 replicates respectively. Number of amphipods in E. ischnus treatments was chosen based on preliminary data that suggested amphipod growth was more variable in F+P treatments than in M+E and no food treatments. Amphipod survivorship was measured weekly. Following the recommendations of Pyke and Thompson (1986), survivorship was analyzed using a non-parametric test for estimating survival distributions when not all individuals under treatment conditions are followed until death (LIFETEST procedure, SAS Institute 1990). We performed intraspecific comparisons among food treatments (M+E vs. F+P) and interspecific comparisons (G. fasciatus vs. E. ischnus) within each food treatment. To quantify growth, we measured head length weekly by gently pouring an individual onto a 100 µm sieve and measuring from the rostrum tip to the posterior margin of the head using a microscope with an ocular micrometer. Following measurement, the amphipod was placed in a clean cup containing filtered lake water and fresh food. The measurement of head length instead of body length minimized the time that amphipods were under the microscope. Previous studies have shown a positive correlation between amphipod head length and total body length (Cooper 1965, Delong et al. 1993, Morrison et al. 1997). To verify this relationship we collected amphipods from Lake Erie and regressed total body length (TL) against head length (HL) for both amphipods. The results of the statistical analysis were similar using either head length or body length estimates. Here we present the results from the analysis using head length. We calculated weekly growth (head length(week n) – head length(week n–1)) for each treatment using amphipods that survived to the end of each week. We tested for intraspecific differences in growth among treatments using a repeated measures ANOVA (SAS Institute 1990). We tested for differences only 104 González and Burkart between the F+P and M+E because after the first week there were not enough survivors in the nofood treatment to satisfy ANOVA assumptions. We evaluated interspecific differences in mean growth using a two-way ANOVA to test for diet (F+P vs. M+E) and species (G. fasciatus vs. E. ischnus) effects. Because we detected a strong species*diet interaction (see results) we compared treatment means using a Tukey-Kramer test (Sokal and Rohlf 1988). All data were log-transformed to meet assumptions of variance homogeneity (Sokal and Rohlf 1988). Fish Predation Experiments We conducted laboratory experiments to estimate the relative vulnerability of G. fasciatus and E. ischnus to predation by yellow perch and round goby in two simulated habitats, mussel colonies and macrophyte beds. Experiments were conducted in 80-L aerated aquaria containing either cobbles with dreissenid colonies or macrophytes in a 0.0625 m 2 area. Macrophyte habitats consisted of 17 shoots of V. americana (shoot density, 272 shoots/m2) cut to the height of the aquarium (30 cm) and weighted down by metal nuts. Mussel habitats consisted of three pieces of cobble (cobble density, 48/m2) colonized by mussels (mean # mussels/m2 = 1,203 ± 44SD). Mean # of mussels/tank was never significantly different among tanks (p > 0.9). Shoot and mussel densities used in the trials were within the range observed in the lake during early August (215–312 shoots/m2, 906–2,172 mussels/m2). Aquarium walls were covered with black plastic material to minimize disturbance of fish and to ensure that fish were not influenced by activity in neighboring tanks. Laboratory temperature during the fish experiments was 23–24°C . Yellow perch and round goby were collected in Hatchery Bay by seining or cast netting. Fish were acclimated in the laboratory for two weeks during which they were fed both amphipod species. Fish were starved for 48 hours prior to an experiment. In all experiments we removed the fish after 24 hours, and the remaining amphipods were counted by species. Fish were measured and weighed at the end of each experiment. Yellow Perch Experiments We estimated yellow perch predation rates in August 1997 and June 2001. In both experiments we added 100 G. fasciatus and 100 E. ischnus to each aquarium and allowed them to acclimate for 12 hours. We then added three fish to each aquarium. In 1997 there were three replicates per habitat, while in 2001 there were four replicates per habitat. A two-way ANOVA (date and habitat effects) indicated significant differences in total fish length and biomass between dates. Because consumption rates can depend on fish size (Kitchell et al. 1977), we analyzed each experiment independently. Round Goby Experiments We estimated round goby predation rates in August 2000 and 2001. We had difficulty collecting E. ischnus during summer 2000, therefore we only added 65 G. fasciatus and 65 E. ischnus to each aquarium. We added only two fish per tank due to the lower amphipod number. Fish mortality in two tanks (one dreissenid and one macrophyte habitat) reduced the number of replicates in both treatments to only two. In 2001 there were four replicates per habitat. Similar to the yellow perch experiments, we added 100 G. fasciatus and 100 E. ischnus and three round gobies to each tank. A two-way ANOVA detected significant differences in fish biomass/tank between dates. Due to the differences in fish biomass and amphipod density we decided to analyze each experiment independently. In all experiments, fish consumption rates (total number of amphipods consumed/g fish biomass/ day) and the ratio of the number of G. fasciatus and E. ischnus consumed (G:E) were analyzed using ttests (SAS Institute 1990) to assess differences in fish predation between mussel and macrophyte habitats. Data were log-transformed to meet assumptions of variance homogeneity (Sokal and Rohlf 1988). We also tested if the Gammarus: Echinogammarus ratio consumed in each habitat was different from one using a two-tailed t-test. Ratios > 1 indicated a higher vulnerability of G. fasciatus to fish predation, while ratios < 1 indicated a higher vulnerability of E. ischnus to fish predation. RESULTS Amphipod Abundance Patterns In 1996, we detected significant differences in amphipod abundance patterns between habitats (Wilk’s lambda value = 0.148, F 1,4 = 22.94, p = 0.009). Inter-habitat comparisons showed that G. fasciatus abundance was significantly higher in macrophytes than in dreissenid colonies (Fig. 1a, Amphipod Abundance Patterns in Macrophyte and Dreissenid Habitats 105 abundance tended to be slightly higher than G. fasciatus, although this trend was not statistically significant (t0.05,4 = 1.73, p = 0.15). In 2001 and 2002, total amphipod abundance was ten-fold lower than in 1996. Echinogammarus ischnus abundance was significantly higher than G. fasciatus abundance in dreissenid colonies in 2001 (Fig. 1b; Hotelling-Law value = 172.5, F2,3 = 258.8, p = 0.0004). In 2002, we also detected significant differences in amphipod abundance patterns between habitats (Wilk’s lambda value = 0.153, F 2,10 = 27.6, p < 0.001). Inter-habitat abundance patterns differed from those found in 1996. Gammarus fasciatus abundance was similar among habitats (F 2,10 = 4.74, p = 0.09), while E. ischnus abundance was significantly higher in dreissenid dominated habitat (100 –70%) than in habitats with < 70% dreissenid cover (Fig. 1.c, F 2,10 = 77.75, p < 0.001). As in 1996, inter-species comparisons showed that G. fasciatus was significantly more abundant than E. ischnus in habitats with a < 70% dreissenid cover (F1,6 = 8.36, p = 0.03 in < 70–30% and F1,10 = 230, p < 0.0001 in 0 – < 30%). Echinogmmarus ischnus abundance also tended to be higher in dreissenid dominated habitat, but the trend was not statistically significant (F1,4 = 6.07, p = 0.06). FIG. 1. Mean abundance (± 1 SE) of G. fasciatus and E. ischnus in dreissenid colonies and macrophyte beds in (a) 1996, (b) 2001, and (c) 2002. In 2001 samples were only collected in dreissenid colonies. Error bars that are not apparent are smaller than the symbols. t0.05,4 = 4.92, p = 0.008), while E. ischnus abundance was similar in both habitats (Fig. 1a, t0.05,4 = 0.12, p = 0.70). Inter-species comparisons revealed that G. fasciatus was nearly 2.5 times more abundant than E. ischnus (t 0.05,4 = 3.94, p = 0.02) in macrophyte beds. In dreissenid colonies, E. ischnus Effect of Diet on Amphipod Survivorship and Growth Survivorship of both amphipod species was significantly lower in the no food treatment than in the food treatments (G. fasciatus: X2(1, 0.05) = 19.7 for F+P and 34.2 for M+E; E. ischnus: X2(1,0.05) = 23.3 for F+P and 19.02 for M+E; p = 0.0001; Fig. 2). Survivorship of G. fasciatus fed F+P was significantly higher than those fed M+E (X2(1, 0.05) = 5.5; p = 0.02; Fig. 2a), but E. ischnus showed similar survivorship when fed either F+P or M+E (X2(1, 0.05) = 0.18; p = 0.6; Fig. 2b) Interspecific comparisons showed that E. ischnus survivorship was significantly higher than G. fasciatus survivorship when fed M+E (X 2 (1, 0.05) = 5.80; p = 0.02; Fig. 2). However, there were no significant differences between the two species when raised on F+P (X2 (1, 0.05) = 0.09; p = 0.8; Fig. 2). We found a strong correlation between total body length and head length for G. fasciatus (TL (mm) = 10.768 (HL) – 1.460; r2 = 0.91 n = 64) and E. ischnus (TL (mm) = 9.554 (HL) - 0.871; r2 = 0.86 n = 123). Both amphipod species showed an increase in mean body length over the eight-week period (Fig. 106 González and Burkart FIG. 2. Weekly survivorship of (a) G. fasciatus and (b) E. ischnus in the feces and pseudofeces (F+P), macrophytes (M+E) and no food (NF) treatments over an 8-week period. 3). Intraspecific comparisons detected significant date effects for both species (F8,296 = 115.2 for G. Gammarus and F8,296 = 58.56 for E. ischnus; p < 0.0001) and a significant diet effect only for G. fasciatus (F1,395 = 27.2, p < 0.001). However the date *diet interactions were significant for both species (F8,395 = 2.64, p < 0.02 for G. fasciatus and F8,296 = 5.46, p < 0.0001 for E. ischnus), mainly because differences in growth between treatments varied with time. In 5 out of 8 weeks, G. fasciatus grew more when fed F+P than when fed M+E, while in the other 3 weeks growth was similar on both diets (Fig. 3a). Echinogammarus ischnus grew more when fed M+E than when fed F+E only during the last 2 weeks of the experiments (Fig. 3b). Interspecific comparisons of mean growth at the end of the experiment showed significant species effects (F1,65 = 171.6, p < 0.0001), but a species* diet interaction (F1,65 = 25.5, p < 0.0001). Gammarus fasciatus grew more when fed F+P than when fed M+E, whereas E. ischnus showed the opposite pattern (Fig. 4). FIG. 3. Weekly mean body length (± 1 SE) of (a) G. fasciatus and (b) E. ischnus in the feces and pseudofeces (F+P), macrophytes (M+E), and no food (NF) treatments over an 8-week period. Error bars that are not apparent are smaller than the symbols. FIG. 4. Mean growth (± 1 SE) of (a) G. fasciatus and (b) E. ischnus fed feces and pseudofeces (F+P) or macrophytes (M+E). Letters indicate that growth was different at a 0.05 significance level (Tukey-Kramer). Amphipod Abundance Patterns in Macrophyte and Dreissenid Habitats Fish Predation Experiments Yellow Perch Experiments Yellow perch consumption rate was higher in the 1997 than in the 2001 experiments, probably because yellow perch individuals were smaller in 1997 than in 2001. We observed significant differences in total fish length (6.6 cm ± 0.1 TL in 1997 and 14.0 cm ± 0.2 in 2001; F 1,10 = 329.7, p < 0.0001) and biomass (14 g ± 0.4 in 1997 and 87g ± 4.2 in 2001; F 1,10 = 856.7, p < 0.0001) between dates. However no significant habitat effects or date*habitat interactions were observed in total fish length and fish biomass/tank. In both experiments, yellow perch consumption rates were similar between habitat treatments (Fig. 5a and b). However, G:E ratios were significantly lower in the macrophyte than in the mussel treatments (t0.05,4 = 8.63, p = 0.04 in 1997; t0.05,6 = 2.94, p = 0.03 in 2001; Fig. 5c and d). G:E ratios were 107 significantly less than 1 in macrophyte treatments indicating a higher vulnerability of E. ischnus to yellow perch predation in this habitat (t 0.05,2 = 18.58, p = 0.001 in 1997; t0.05,3 = 4.57, p = 0.01 in 2001). In mussel treatments the G:E ratios were not significantly different than 1, indicating similar vulnerability of G. fasciatus and E. ischnus to yellow perch predation. Round Goby Experiments No significant differences in size (9.7 cm ± 0.2 TL in 2000 and 9.8 cm ± 0.2 in 2001) were detected between dates. However, the fish biomass/tank was significantly lower in 2000 that in 2001 (27.1 g ± 0.2 in 2000 and 38.6 g ± 0.5 in 2001, F1,8 = 58.1, p < 0.001). No significant habitat effects or date*habitat interactions were observed in total fish length and fish biomass/tank. Round goby consumption rates were significantly FIG. 5. (a,b) Mean consumption rates (amphipod/g/d ± 1 SE) and (c, d) mean consumed G. fasciatus:E. ischnus ratios in mussel and macrophyte habitats after three yellow perch were allowed to feed for 24 hours. Error bars that are not apparent are smaller than the symbols. 108 González and Burkart FIG. 6. (a, b) Mean consumption rates (amphipod/g/d ± 1 SE) and (c, d) mean consumed G. fasciatus:E. ischnus ratios in mussel and macrophyte habitats after 2 (2000) –3 (2001) round gobies were allowed to feed for 24 hours. Error bars that are not apparent are smaller than the symbols. higher in macrophyte treatments than in mussel treatments in both experiments (t0.05,2 = 7.18, p = 0.02 in 2000; t0.05,6 = 4.34, p = 0.005 in 2001, Fig. 6a and b). The G:E ratio was significantly lower in the macrophyte treatment than in the mussel treatment only in 2001 (t0.05,6 = 2.87, p = 0.02, Fig. 6c and d). In macrophyte treatments the G:E ratio was significantly less than 1 only in 2001 (t0.05,3 = 7.47, p = 0.002), indicating a higher vulnerability of E. ischnus to round goby predation. In mussel treatments, the G:E ratios were not significantly different from 1 indicating similar vulnerability of G. fasciatus and E. ischnus to round goby predation in this habitat. DISCUSSION Our survey results showed inter-habitat differences in the abundance of both amphipod species. Gammarus fasciatus abundance was consistently higher in macrophytes than in dreissenid colonies, while E. ischnus abundance tended to be higher in dreissenid colonies. Previous studies have also reported higher abundance of E. ischnus in mussel colonies than in vegetation dominated areas (Witt et al. 1997, Dermott et al. 1998, Nalepa et al. 2001). Our results support previous work that suggests that the exotic E. ischnus may be displacing G. fasciatus on rocky substrate areas in the Great LakesSt. Lawrence system (Dermott et al. 1998). An earlier study suggested that G. fasciatus was equally abundant in macrophyte beds and rocky substrates (Clemens 1950) in Hatchery Bay before the invasion of Dreissena sp. and E. ischnus. Prior to the invasion of E. ischnus, G. fasciatus was the dominant amphipod in mussel colonies, in terms of numbers and biomass, in Lake Erie including Hatchery Bay (Dermott et al. 1992, Dermott et al. 1998). However, the higher abundance of G. fasciatus in macrophyte beds suggests that the presence Amphipod Abundance Patterns in Macrophyte and Dreissenid Habitats of macrophytes may allow G. fasciatus to coexist with the exotic species. The drastic decrease in amphipods between 1996 and 2000-2001 cannot be attributed to differences in methodology because a similar decrease was observed in the average number of amphipods colonizing experimental cages placed in dreissenid colonies in 2001 (92 ± 15 (SE) individuals/cage; González unpublished) compared to 1995–1996 (1630 ± 275 (SE) individuals/cage; González and Downing 1999). The drastic decrease in the total number of amphipods in both habitats may be related to the invasion of round goby in the study area during 1998. Our experimental cages in those experiments excluded predators. However, regular observations by SCUBA divers observed no round gobies located around the cages in 1995 and 1996, while the average number of round gobies located around the cages was 4.45 + 0.9(SE)/cage in 2001 (González unpublished). Kuhns and Berg (1999) reported lower benthic invertebrate abundance in colonization cages where round gobies was allowed to feed compared to cages excluding round gobies. Food (Klerks et al. 1996, Stewart et al. 1998) and shelter (Ricciardi et al. 1997, González and Downing 1999) for benthic invertebrates have increased since the invasion of Dreissena sp. Although many researchers have speculated about the potential importance of feces and pseudofeces in the diet of benthic invertebrates (Klerks et al. 1996, Griffiths 1992, Fisher et al. 1992), ours is the first study to document the effects of this food on amphipod survivorship and growth. The higher survivorship and growth of G. fasciatus when fed dreissenid egestion products provide evidence that the drastic increase of this amphipod after the Dreissena sp. invasion was related to increased in food availability in addition to increased habitat complexity. Our survivorship and growth experiments, however, suggest that differences in the relative abundance of G. fasciatus and E. ischnus cannot be explained by differences in food quality associated with each habitat. Based on our intraspecific and interspecific comparisons we would expect the relative abundance of G. fasciatus to be higher in mussel colonies than in macrophyte beds, because growth and survivorship of G. fasciatus was significantly greater when fed F+P than when fed M+E. Although survivorship of both amphipods was similar under F+P diet, G. fasciatus net growth was higher than E. ischnus. Results for E. ischnus are more complex. Based on the intraspecific comparisons we would expect similar E. ischnus abun- 109 dance in both habitats. The growth and survivorship of E. ischnus was similar when fed F+P or M+E. Our interspecific comparison, however, suggest that E. ischnus exhibited greater survivorship and growth than G. fasciatus under M+E diet. Predation risk is an important factor constraining the spatial distribution of marine and freshwater organisms (Werner et al. 1983, Duffy and Hay 1991, González and Tessier 1997). Our results suggest that differential vulnerability to fish predation between habitats may explain differences in relative abundance of G. fasciatus and E. ischnus in mussel colonies and macrophyte beds. The lower relative abundance of E. ischnus in macrophytes may be caused by higher vulnerability of this amphipod to fish predation in this habitat. The G:E ratios were significantly lower in macrophyte than mussel habitats in three out of four experiments (Figs. 5 and 6) indicating fish preference for E. ischnus in this habitat. Also G:E ratios were significantly < 1 in the same three experiments indicating higher vulnerability of E. ischnus to fish predation than G. fasciatus. In the mussel habitat fish tended to consume more G. fasciatus than E. ischnus, but G:E ratios were never significantly > 1 indicating similar vulnerability of the two amphipod species to fish predation in this habitat. The lack of statistically significant results in the 2000 round goby experiment was probably related to low statistical power due to the loss of some replicates. However it is important emphasize that G:E ratio tend to be lower in macrophyte than mussel habitats, as in the other three experiments. The similar trends in consumption patterns between yellow perch and round goby suggest that predator preferences were related to prey-specific characteristics mediated by habitat. Differences in coloration, size, and behavior between G. fasciatus and E. ischnus may lead to variability in their susceptibility to fish predators between habitats. In macrophytes the light greenish-gray color and clinging behavior (“C-shaped posture”) of G. fasciatus (Clemens 1950) may be better camouflage, whereas the relatively dark reddish color and more active behavior of E. ischnus (Burkart 1999) may make it more visible to predators in macrophyte beds. Active swimming behavior has been documented in E. ischnus (reviewed by Nalepa et al. 2001). Behavioral differences (movement) among Crangonyx pseudogracilis, G. duebebeni, and G. pulex was an important factor determining the feeding preference of brown trout on these three amphipod species (MacNeil et al. 1999). Crangonyx 110 González and Burkart pseudogracilis was preferentially consumed by brown trout. The most common escape response exhibited by C. pseudogracilis was to swim away, while both Gammarus species usually adopted a “C-shaped posture” that make handling by trout difficult. The lower overall amphipod consumption in mussel colonies in both round goby experiments may be related to mussel consumption by round goby. Stomach content analyses have showed that round gobies > 7 cm TL rely more on dreissenids, compared to round gobies < 7 cm TL (Jude et al. 1995, Ray and Corkum 1997, French and Jude 2001). The mean size of round goby in our experiments was 9.8 cm and we observed fragments of mussel shell in several tanks after the feeding trials. The similarity in prey preference exhibited by both fish species, and the drastic decrease in benthic invertebrates after round goby invasion, suggest potential competition between round goby and yellow perch. Under intense competition for resources during the benthivorous feeding stage, round goby can create a bottleneck in yellow perch growth and delay the ontogenetic shift of perch to piscivory (Olson 1996, Heath and Roff 1996). Hence yellow perch may be vulnerable to predation by piscivores for a longer period of time, and high predator-induced mortality on young yellow perch may cause poor recruitment (Hartman and Magraf 1993). However, competition may depend on fish size. In our experiments small yellow perch (6.6 cm TL) show overall higher consumption rates than round gobies (9.8 cm TL), while larger yellow perch (14 cm TL) consumption was considerably lower than round goby. Studies investigating competition between yellow perch and round goby should consider size structure of populations. Intraguild predation has been identified as an important factor determining the establishment of exotic amphipods in streams in Ireland and England (Dick et al. 1993, 1999). Intraguild predation between G. fasciatus and E. ischnus has recently been evaluated under laboratory conditions (Kozac 2001). Gammarus fasciatus prey more often on E. ischnus than vice versa, regardless of whether amphipods are fed M+E or F+P. Based on these findings we would expect G. fasciatus to be the dominant amphipod in both habitats. Fish predation may mediate the importance of intraguild predation on E. ischnus populations differently in mussel and macrophyte habitats. In mussel colonies higher fish predation on G. fasciatus may release E. ischnus from intraguild predation and competition by G. fasciatus. In macrophyte beds a strong vulnerability to fish predation and intraguild predation may limit the abundance of E. ischnus. Future studies should focus on how fish and intraguild predation mediate amphipod population distributions in mussel colonies and macrophyte beds. In summary, we found differences in the relative abundance of G. fasciatus and E. ischnus between mussel and macrophyte beds. Amphipod species were differentially affected by food resources and fish predation pressure associated with mussel colonies and macrophyte beds. However, habitatspecific differences in amphipod relative abundance seem to be related to differential vulnerability to fish predation between amphipod species rather than differences in food availability between habitats. Our results suggest that diversification of the benthic habitat caused by dreissenid invasion (establisment of mussel colonies and proliferation of macrophytes) as well as habitat-specific differences in vulnerability to fish predation seem to mediate the coexistence of G. fasciatus and E. ischnus in Lake Erie by allowing the displacement of G. fasciatus by E. ischnus in mussel colonies and minimizing the expansion of E. ischnus into macrophyte areas. This study also suggests that the establishment of round goby may expand the overall predation pressure on amphipod populations in macrophytes and mussel colonies causing substantial changes in benthic invertebrates communities in the Great Lakes. Furthermore, due to the dietary overlap between round goby and yellow perch, the drastic decreases in benthic invertebrate abundance observed in this study after the round goby invasion may cause a bottleneck effect on growth rate of benthic yellow perch. ACKNOWLEDGMENTS We thank J. Hageman, M. Thomas, and the staff of F. T. Stone Laboratory for their assistance with several aspects of the project. We also thank J. Bebout, B. Burkart, E. Croissant, A. Downing, L. Jeter, L. Knoll, K. Kozac, D. Lavoie, Z. Sutphin, M. Thomas, and D. Vallance for their field and laboratory assistance. Comments by D. Berg, J. Stoeckel, M. J. Vanni, and two anonymous reviewers greatly improved the manuscript. This research was supported by grants from the National Sea Grant College Program, National Oceanic and Atmospheric Administration, in the U. S. Department of Commerce (R/ER-23-40), Ohio Sea Grant (R/NIS-7), and Ohio Board of Regents /Research Challenge Amphipod Abundance Patterns in Macrophyte and Dreissenid Habitats Grant to M.J.G. and Ohio Sea Grant Seed Funding to M.J.G. and G.B. REFERENCES Aplet, G.H. 1990. Alteration of earthworm community biomass by the alien Myrica faya in Hawai’i. Oecologia 82:414–416. Budd, J.W., Drummer, T.D., Nalepa, T.F., and Fahnenstiel, G.L. 2001. Remote sensing of biotic effects: Zebra mussels (Dreissena polymorpha) influence on water clarity in Saginaw Bay, Lake Huron. Limnol. Oceanogr. 46:213–223 Burkart, G.A. 1999. Habitat use and nutrient release by native and exotic amphipods in Western Lake Erie: Assessing the influence of the exotic zebra mussel (Dreissena polymorpha). Master’s thesis. Wright State University, Dayton, Ohio, USA. Charlebois, P.M., Marsden, J.E., Goettel, R.G., Wolfe, R.K., Jude, D.J., and Rudnika, S. 1997. The round goby, Neogobius melanostomus (Pallas), a review of European and North American literature. IllinoisIndiana Sea Grant Program and Illinois Natural History Survey. INHS Special Publication No. 20. Clemens, H.P. 1950. Life cycle and ecology of Gammarus fasciatus Say. Franz Theodore Stone Inst. Hydrol. Contrib 12. Cooper, W.E. 1965. Dynamics and production of a natural population of a fresh-water amphipod, Hyalella azteca. Ecol. Monogr. 35:377–394. Crowder, L.B., and Cooper, W.E. 1982. Habitat structural complexity and the interaction between bluegills and their prey. Ecology 63:1802–1813. Delong, M.D., Summers, R.B., and Thorp, J.H. 1993. Influence of food type on the growth of a riverine amphipod, Gammarus fasciatus. Can. J. Fish. Aquat. Sci. 50:1891–1896. Dermott, R., Mitchell, J., Murray, I., and Fear, E. 1992. Biomass and production of zebra mussel (Dreissena polymorpha) in shallow waters of Northeastern Lake Erie. In Zebra mussels: biology, impacts, and control, eds. T. Nalepa and D. Schloesser, pp. 399–414. Chelsea, MI: Lewis Publishers. ———, Witt, J., Um, Y.M., and González, M.J. 1998. Distribution of the Ponto-Caspian amphipod Echinogammarus ischnus in the Great Lakes and displacement of native Gammarus fasciatus. J. Great Lakes Res. 24:442–452. Dick, J.T.A., Montgomery, W.I., and Elwood, R.W. 1993. Replacement of the indigenous amphipod Gammarus duebeni celticus by the introduced G. pulex: differential cannibalism and mutual predation. J. Animal Ecol. 62:79–88. ——— , Montgomery, W.I., and Elwood, R.W. 1999. Intraguild predation may explain an amphipod replacement, evidence from laboratory populations. J. Zool. Lond. 249:463–468. 111 Duffy, J.E., and Hay, M.E. 1991. Food and shelter as determinants of food choice by an herbivorous marine amphipod. Ecology 72:1286–1298. Elton, C.S. 1958. The Ecology of Invasions by Animals and Plants. Chapman and Hall. Fisher, S.W., Gossiaux, D.C., Bruner, K.A., and Landrum P.F. 1992. Investigations of the toxicokinetics of hydrophobic contaminants in the zebra mussel (Dreissena polymorpha). In Zebra mussels: biology, impacts, and control, eds. T. Nalepa and D. Schloesser, pp. 465–490. Chelsea, MI: Lewis Publishers. Francis, J.T., Robillard, S.R., and Marsden, J.E. 1995. Yellow perch management in Lake Michigan: A multi-jurisdictional challenge. Fisheries 21:18–20. French, J.R.P., and Jude, D.J. 2001. Diets and diet overlap of nonindigenous gobies and small benthic native fishes co-inhabiting the St. Clair River, Michigan. J. Great Lakes Res. 27:300–311. Ghedotti, M.J., Smihula, J.C., and Smith, G.R. 1995. Zebra mussel predation by round gobies in the laboratory. J. Great Lakes Res. 21:665–669. Gilinsky, E. 1984. The role of predation and spatial heterogeneity in determining benthic community structure. Ecology 65:455–468. González, M.J., and Downing, A. 1999. Mechanisms underlying amphipod responses to zebra mussel invasion and implications for fish-amphipod interactions. Can. J. Fish. Aquat. Sci. 50:1891–1896. ———, and Tessier, A.J. 1997. Habitat segregation and interactive effects of multiple predators on a prey assemblage. Freshwat. Biol. 37:179–191. Griffiths R.W. 1992. Effects of zebra mussel (Dreissena polymorpha) on benthic fauna of Lake St. Clair. In Zebra mussels: biology, impacts, & control, eds. T. Nalepa and D. Schloesser, pp. 415–438. Chelsea, MI: Lewis Publishers. Hartman, K.J., and Magraf, F.J. 1993. Evidence of predatory control of yellow perch (Perca flavescens) recruitment in Lake Erie, U.S.A. J. Fish Biol. 43:109–119. Hayes, D.B., and Taylor, W.W. 1990. Reproductive strategy in yellow perch (Perca flavescens): Effects of diet ontogeny, mortality, and survival costs. Can. J. Fish. Aquat. Sci. 47:921–927. Heath, D.D., and Roff, D.A. 1996. The role of trophic bottlenecks in stunting: a field test of an allocation model of growth and reproduction in yellow perch, Perca flavescens. Env. Biol. Fish. 45:53–63. Heck, K.L., and Crowder, L.B. 1991. Habitat structure and predator-prey interactions in vegetated aquatic systems. In Habitat structure: the physical arrangement of objectives in space, eds. S.S. Bell, E.D. McCoy, and H. R. Mushinski. London England: Chapman and Hall. Huston, M.A. 1994. Biological Diversity: The coexis- 112 González and Burkart tence of species on changing landscapes. Cambridge University Press. Jude, D.J., and DeBoe, S.F. 1996. Possible impact of gobies and other introduced species on habitat restoration efforts. Can. J. Fish. Aquat. Sci. 53:136–141. ———, Reider, R.H., and Smith, G.R. 1992. Establishment of Gobiidae in the Great Lakes Basin. Can. J. Fish. Aquat. Sci., 49:416–421. ———, Janssen, J., and Crawford, G. 1995. Ecology, distribution, and impact of the newly introduced round and tubenose gobies on the biota of the St. Clair and Detroit Rivers. In The Lake Huron ecosystem: ecology, fisheries and management. Ed. M. Munawar, T. Edsall, and J. Leach, pp. 447–460. Ecovision World Monogr. Ser. S.P.B. Amsterdam, The Netherlands: Academic Publishing. Kashian, D.R., and Burton, T.M. 2000. A comparison of macroinvertebrates of two Great Lakes coastal wetlands: testing potential metrics for an index of ecological integrity. J. Great Lakes Res. 26:460–481. Kitchell, J.F., Stewart, D.J., and Weininger, D. 1977. Applications of a bioenergetics model to yellow perch (Perca flavenscens) and walleye (Stizostedion vitreum vitreum). J. Fish. Res. Board Can. 34:1922–1935. Klerks, P.L., Fraleigh, P.C., and Lawniczak, J.E. 1996. Effect of zebra mussels (Dreissena polymorpha) on seston levels and sediment deposition in Western Lake Erie. Can. J. Fish. Aquat. Sci. 53:2284–2291. Köhn, J., and Waterstraat, A. 1990. The amphipod fauna of Lake Kummerow (Mecklenburg, German Democratic Republic) with reference to Echinogammarus ischnus Stebbing, 1899. Crustaceana 58:74–82. Kozac, K. 2001. Biotic interactions among round goby, yellow perch and amphipods within zebra mussel colonies and macrophyte beds. Master’s thesis. Wright State University, Dayton, Ohio, USA. Kuhns, L.A., and Berg, M.B. 1999. Benthic invertebrate community responses to round goby (Neogobius melanostomus) and zebra mussel (Dreissena polymorpha) invasion in southern Lake Michigan. J. Great Lakes Res. 25:910–917. Leach, J.H. 1992. Impacts of the zebra mussel (Dreissena polymorpha) on water clarity and fish spawning reefs in Western Lake Erie. In Zebra mussels: biology, impacts, and control, eds. T. Nalepa and D. Schloesser, pp. 381–398. Chelsea, MI: Lewis Publishers. Lodge, D.M. 1993. Biological invasions: lessons for ecology. Trends in Ecology and Evolution 8:133–137 Lowe, R.L., and Pillsbury, R.W. 1995. Shifts in benthic algal community structure and function following the appearance of zebra mussels (Dreissena polymorpha) in Saginaw Bay, Lake Huron. J. Great Lakes Res. 21:558–566. MacArthur, R.H., Diamond, J.M., and Karr, J.R. 1972. Density compensation in island faunas. Ecology 53: 330–342. MacNeil, C., Elwood, R.W., and Dick, J.T.A. 1999. Predator-prey interactions between brown trout Salmo trutta and native introduced amphipods; their implications for fish diets. Ecography 22:415–423s. Morrison, T.W., Lynch, W.E. Jr., and Dabrowski, K. 1997. Predation on zebra mussels by freshwater drum and yellow perch in Western Lake Erie. J. Great Lakes Res. 23:177–189. Nalepa, T.F., Schloesser, D.W., Pothoven, S.A., Hondorp, D.W., Fanslow, D.L., Tuchman, M.L., and Fleischer, G.W. 2001. First finding of the amphipod Echinogammarus ischnus and the mussel Dreissena bugensis in Lake Michigan. J. Great Lakes Res. 27:384–391. Newman, R.M. 1991. Herbivory and detritivory on freshwater macrophytes by invertebrates: a review. J. North Am. Benth. Soc. 10:89–114. Olson, M.H. 1996. Ontogenetic niche shifts in largemouth bass: Variability and consequences for firstyear growth. 1996. Ecology 77:179–190. Orians, G.H. 1986. Site characteristics favoring invasions. In Ecology of biological invasions of North America and Hawaii, eds. H.A. Mooney and J. Drake, pp. 133–148. New York, New York: Springer-Verlag. Pyke, D.A., and Thompson, J.N. 1986. Statistical analysis of survival and removal rate experiments. Ecology 67:240–245. Ray, W.J., and Corkum, L.D. 1997. Predation of zebra mussels by round gobies, Neogobius melanostomus. Environ. Biol. Fishes 50:267–273. Ricciardi, A., Whoriskeyk, F.G., and Rasmussen, J.B. 1997. The role of zebra mussel (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrata. Can. J. Fish. Aquat. Sci. 53: 2596–2608. Robillard, S.R., Weis, A., and Dettmers, J.M. 1999. Yellow perch population assessment in southwestern Lake Michigan, including evaluation of sampling techniques and the identification factors that determine yellow perch year-class strength. Annual report to Illinois Department of Natural Resources. Illinois Natural History Survey Technical Report 99/5. Robinson, J.V., and Welborn, G.A. 1988. Ecological resistance to the invasion of a freshwater clam, Corbicula fluminea: fish predation effects. Oecologia 77: 445–452. Sih, A. 1987. Predator and prey lifestyles an evolutionary and ecological overview. In Predation: Direct and Indirect Impacts on Aquatic communities, eds. W.C. Kerfoot and A. Sih, pp. 203–224. University Press of New England. Simberloff, D. 1986. Introduced insects: a biogeographic and systematic perspective. In Ecology of biological invasions of North America and Hawaii, eds. H.A. Mooney and J.A. Drake, pp. 3–26. New York, New York: Springer-Verlag. Amphipod Abundance Patterns in Macrophyte and Dreissenid Habitats Skubinna, J.P., Coon, T.G., and Batterson, T.R. 1995. Increased abundance and depth of submersed macrophytes in response to decreased turbidity in Saginaw Bay, Lake Huron. J. Great Lakes Res. 21:476–488. Sokal, R. R. and Rohlf, F.R. 1988. Biometry. New York: W. H. Freeman and Co. Statistical Analysis System Institute. 1990. SAS/STAT user’s guide, version 6. Fourth edition. Cary, North Carolina: SAS Institute. Stewart, T.W., and Haynes, J.M. 1994. Benthic macroinvertebrate communities of southwestern Lake Ontario following invasion of Dreissena. J. Great Lakes Res. 20:479–493. ———, Miner, J.G., and Lowe, R.L. 1998. Quantifying mechanisms for zebra mussel effects on benthic macroinvertebrates: organic matter production and shelter-generated habitat. J. North Am. Benth. Soc. 17:81–94. Stuckey, R.L., and Moore, D.L. 1995. Return and increase in abundance of aquatic flowering plants in Put-In-Bay Harbor, Lake Erie, Ohio. Ohio J. Sci. 95:262–266. Thayer, S.A., Haas, R.C., Hunter, R.D., and Kushler, R.H. 1997. Zebra mussel (Dreissena polymorpha) effects on sediment, other zoobenthos, and the diet and growth of adult yellow perch (Perca flavescens) in pond enclosures. Can. J. Fish. Aquat. Sci. 54:1903–1915. Tyson, J.T., and Knight, R.L. 2001. Response of yellow perch to changes in the benthic invertebrate commu- 113 nity of western Lake Erie. Trans. Am. Fish. Soc. 130: 766–782. Vitousek, P.M. 1990. Biological invasions and ecosystem processes: towards an integration of population biology and ecosystem studies. Oikos 57:7–13. ———, and Walker, L.R. 1989. Biological invasion by Myrica faya in Hawai’i: plant demography, nitrogen fixation and ecosystem effects. Ecol. Monogr. 59:247–265. ———, Walker, L.R., Whiteaker, L.D., Mueller-Dombois, D., and Matson, P.A. 1987. Biological invasion by Myrica faya alters ecosystem development in Hawai’i. Science 238:802–804. Werner, E.E., Gillian, J. F., Hall, D.J., and Mittelbach, G.G. 1983. An experimental test of the effects of predation risk on habitat use in fish. Ecology 64: 1540–1548. Witt, J.D.S., Hebert, P.D.N., and Morton, W.B. 1997. Echinogammarus ischnus: another crustacean invader in the Great Lakes basin. Can. J. Fish. Aquat. Sci. 54:264–268. Wu, L., and Culver, D.A. 1991. Zooplankton grazing and phytoplankton abundance: an assessment before and after invasion of Dreissena polymorpha. J. Great Lakes Res. 17:425–436. Submitted: 11 Mach 2003 Accepted: 7 November 2003 Editorial handling: Thomas F. Nalepa