Success of slimy sculpin reintroductions in Minnesota trout streams: influence of feeding and diets.
Abstract: Slimy sculpin Cottus cognatus are being reintroduced into coldwater streams in southeastern Minnesota to restore native biotic diversity and provide forage for large trout, but success has been variable. We examined slimy sculpin diets and prey consumption in a series of field and laboratory experiments to assess the potential role of invertebrate prey in affecting reintroduction success. Sculpin consumed 35 different types of prey in the field, but frequency of occurrence was highest and preferences (Ivlev's selectivity index) strongest for Diptera larvae and Amphipoda. Benthic samples indicated that preferred prey items were abundant in most streams. In 24-h feeding experiments, sculpin typically consumed 5 to 15 prey/day, and exhibited selective feeding for Amphipoda (Gammarus), Isopoda (Asellus), and Ephemeroptera (Baetis) while rejecting Trichoptera (Brachycentrus) and Gastropoda (Physella). Sculpin are euryphagous and flexible to varying prey availability in different streams, demonstrating both mixed diets and multiple prey preferences that allow fish to maximize their consumption when confronted with differing prey assemblages. Our data indicate that preferred prey taxa are not limiting and that lack of suitable prey is not a factor in the limited success of sculpin reintroductions in some streams.
Article Type: Report
Subject: Sculpin (Environmental aspects)
Sculpin (Distribution)
Fish populations (Environmental aspects)
Fish populations (Distribution)
Population biology (Research)
Authors: Mundahl, Neal D.
Mundahl, Darcy E.
Merten, Eric C.
Pub Date: 07/01/2012
Publication: Name: The American Midland Naturalist Publisher: University of Notre Dame, Department of Biological Sciences Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Earth sciences Copyright: COPYRIGHT 2012 University of Notre Dame, Department of Biological Sciences ISSN: 0003-0031
Issue: Date: July, 2012 Source Volume: 168 Source Issue: 1
Topic: Event Code: 690 Goods & services distribution; 310 Science & research Advertising Code: 59 Channels of Distribution Computer Subject: Company distribution practices
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 296160027
Full Text: INTRODUCTION

Slimy sculpin Cottus cognatus Richardson are being reintroduced to coldwater streams in southeastern Minnesota (MN DNR, 2003). These streams are naturally species-poor (often <5 species offish per stream; Lyons et al., 1996; Mundahl and Simon, 1998), and many have lacked sculpin since extensive land use changes (conversion to row-crop agriculture and intensive livestock grazing) caused severe stream degradation (burying of coarse substrates by fines, creating wider and shallower channels, stream warming) a century ago (Thorn et al., 1997). Now that stream conditions have improved dramatically because of improved watershed management, channel and instream habitat restorations and protection of riparian buffers (Thorn et al., 1997), sculpin reintroductions are attempting to reestablish a native, ecologically important nongame species (Dineen, 1951; MN DNR, 2003; Adams and Schmetterling, 2007), coinciding with reintroductions of native brook trout Salvelinus fontinalis Mitchill and expanding management for wild brown trout Salmo trutta Linnaeus (Thorn et al., 1997).

Success of sculpin reintroductions to date has been mixed in the 10 streams where they were stocked beginning in 2003. Stocked sculpin survived and reproduced in all 10 streams, but only in five streams have sculpin populations increased significantly in number (population estimates ranging from 895-3147 fish) beyond initial 150 fish founder populations and expanded away from initial stocking sites (Huff, 2010; Vaughn Snook, MN DNR-Lanesboro, unpub, data). Genetic problems with reintroduced sculpin (Huff et al., 2010, 2011), recent 1000-y stream flooding (Keillor, 2010), predation by brown trout (Dineen, 1951) and variations in amount of preferred sculpin habitats (Anderson, 1985; Mundahl et al., 2012) have been speculated as reasons for reduced success of sculpin in some systems. To date, only outbreeding depression resulting from the mixed-source reintroductions (reintroduced sculpin obtained from three separate stream populations in three distinct watersheds) has been implicated in reducing the relative fitness of sculpin offspring at reintroduction sites (Huff et al., 2011).

A growing body of evidence suggests that availability of benthic prey is considerably more important to the success of sculpin populations than are other physical and biotic factors (e.g., Brocksen et al., 1968; Petty and Grossman, 1996; Ruetz et al., 2004; Grossman et al., 2006; Zimmerman and Vondracek, 2006, 2007a, 2007b; Petty and Grossman, 2007, 2010). Sculpin may select habitat patches on the basis of high macroinvertebrate abundance, irrespective of the patches' physical attributes (Petty and Grossman, 1996), competing directly with conspecifics for habitats with the most invertebrate prey (Grossman et al., 2006; Petty and Grossman, 2007, 2010). Low prey densities, especially of larger prey, can result in reduced rates of food consumption and poorer growth in sculpin (Brocksen et al., 1968; Zimmerman and Vondracek, 2006), ultimately influencing individual fitness and population success.

Many invertebrate taxa are preyed on by sculpin in Minnesota streams, with diets often differing among streams and seasons (Dineen, 1951; Zimmerman and Vondracek, 2007a, 2007b). Amphipods, caddisfly larvae, mayfly nymphs, blackfly and midge larvae, and snails typically comprise large proportions of sculpin diets, with mayflies, amphipods, and blackflies preferentially selected by sculpin in one stream (Zimmerman and Vondracek, 2007a, 2007b). Meanwhile, streams in southeastern Minnesota display wide variation in production and diversity of benthic invertebrates (Troelstrup and Perry, 1989; Waters, 2000), with variation frequently correlated with subsurface geology (via effects on nutrients and alkalinity in spring water, water temperature, and flow stability) and riparian land-use (e.g., row-crop agriculture, pasture, urban landscape, forest, protected buffer) practices (Troelstrup and Perry, 1989; Muck and Newman, 1992).

The present study was initiated to examine the possible influence of benthic invertebrate availability (both abundance and type) on sculpin diets and the success of sculpin reintroductions. We assessed benthos availability, sculpin diets and prey selectivity in streams with native and introduced populations of slimy sculpin, as well as benthos availability in streams where sculpin may be reintroduced in the future. We also conducted 24-h feeding experiments in the laboratory to assess selective feeding and maximum rates of food consumption by sculpin at various temperatures for comparison to field-fed fish.

METHODS

STUDY AREA

Our study area included 16 coldwater streams in the counties of Olmsted, Wabasha, and Winona (Fig. 1). Seven of these streams support healthy populations of slimy sculpin, and three of them (Garvin Brook, Beaver Creek, Cold Spring Brook) have been used as sources of fish for reintroduction to other streams. Five streams presently are lacking sculpin, and four others recently had sculpin reintroduced to them. Slimy sculpin were reintroduced into Latsch Creek and Sugarloaf Creek in 2004 and Big Trout Creek and Little Trout Creek in 2005. The Latsch Creek and Little Trout Creek reintroductions have been termed successful, the Big Trout Creek reintroduction unsuccessful, and the Sugarloaf Creek reintroduction undetermined (V. Snook, MN DNR, pers. comm.).

[FIGURE 1 OMITTED]

SCULPIN DIETS

Feeding habits of slimy sculpin were examined in fish collected from eight streams in southeastern Minnesota during three time periods: winter 1992-1993 (Gilmore, Garvin), summer 2006 (Gilmore, Garvin, Beaver, Cold Spring, S. Br. Whitewater, Trout Run), and summer 2007 (Gilmore, Garvin, Latsch, Big Trout). We collected approximately 20 fish at each site/date, spanning the complete size range of fish present. We specifically chose to examine the diets of fish from two reintroduction streams in 2007: Latsch Creek (a successful reintroduction) and Big Trout Creek (least successful reintroduction). Fish were collected with a Smith-Root backpack electrofisher, over-anesthetized immediately with MS222, fixed in 10% formalin for at least 24 h and then preserved in 70% ethanol. No regurgitation of food items was observed with this procedure.

Fish were weighed, measured, and then dissected to remove preserved stomachs. An incision was made down the side of the stomach and stomach contents were flushed into a watch glass with water. Prey items in the stomach contents were viewed, identified, and counted under a dissecting microscope (8-40 X magnification). We quantified diet items by frequency of occurrence (presence-absence in individual fish) and as percent composition by number (Bowen, 1996). Ingested foods were then transferred to a pre-weighed plastic dish, dried for 24 h in a drying oven (35 C) and weighed (nearest 0.1 mg) to determine the total dry weight of foods (2006 and 2007 samples only). Dry weights of foods were standardized for fish size (mg food dry weight/g fish wet weight) and compared among streams for each year with single-factor analysis of variance. No attempt was made to determine separate dry weights for different prey taxa.

We examined patterns among fish diets in 2006 (six streams, all with native populations of sculpin) using non-metric dimensional scaling (NMS) with the statistical software R (available online at http://www.r-project.org/). All analyses used Bray-Curtis dissimilarity indices to characterize differences in diet composition between individual sculpin. We first examined a scree plot to determine the number of dimensions (using a stress value of 0.2 as the criterion), and then used the "metaMDS" function in R to conduct the NMS analysis. We also used the "ordiellipse" function to determine 95% confidence limits for the diet data originating from each stream. Additionally, we examined whether fish size (total length or wet weight) influenced diet composition, using the "envfit" function in R with 1000 permutations. For each environmental variable (i.e., total length or wet weight) the "envfit" function determined the vector with the greatest [r.sup.2] value relative to the projection of points in the NMS ordination.

BENTHIC INVERTEBRATE AVAILABILITY

We assessed the availability of sculpin prey during 2006 and 2007 at each of the same eight stream sites where sculpin diets were assessed during those 2 y (six streams with native sculpin populations and two with reintroduced populations). Benthic invertebrates were collected using triplicate Hess or D-frame kick net samples (0.1 [m.sup.2]) from erosional habitats with coarse (gravel: 2-64 mm, rubble: 64-250 mm, boulder: >250 mm) substrates (Rabeni, 1996). In addition, similar samples were collected from five streams where sculpin may be introduced in the future (non-sculpin streams), two additional streams where sculpin were recently reintroduced, and one more stream that supported large populations of sculpin. Samples were preserved in 70% ethanol and returned to the lab for analyses. Each invertebrate sample was sorted and specimens were identified (most insects to genus or family, most non-insects to order or class) and counted. Triplicate samples from each site were individually assessed for total invertebrate abundance, taxa richness, EPT (Ephemeroptera-Plecoptera-Trichoptera) taxa richness, and benthic IBI (Wittman and Mundahl, 2003) score and rating. Assessments were averaged to produce single values for each stream site, and comparisons were made among native sculpin streams (n = 7), non-sculpin streams (n = 5) and reintroduction streams (n = 4) with four single-factor ANOVAs (total invertebrate abundance, taxa richness, EPT taxa richness, and benthic IBI as response variables). To assess benthos quality in reintroduction streams in more detail, total invertebrate abundance, taxa richness, EPT taxa richness, and benthic IBI values for individual triplicate benthos samples at each site were compared among reintroduction sites with four, single-factor ANOVAs.

PREY SELECTIVITY

Using the benthic invertebrate samples and sculpin diet data gathered in 2006 and 2007, we compared prey availability with prey consumption by sculpin in six streams (Gilmore, Garvin, Beaver, Cold Spring, S. Br. Whitewater, Trout Run) with native sculpin populations and two streams (Latsch, Big Trout) with reintroduced populations. Invertebrates were grouped into major taxonomic categories (e.g., insect orders), and proportional abundances of the various categories were compared between benthic samples and sculpin diets (stream by stream) with the Ivlev selectivity index (Krebs, 1989) to determine whether various invertebrates were preferentially selected or rejected by slimy sculpin (Bowen, 1996). The use of the term "rejected" here only means that prey were consumed in proportions less than they were present in the environment, regardless of what mechanism(s) or prey attribute (s) may have caused that under-consumption.

24-HOUR FEEDING EXPERIMENTS: CONSUMPTION RATES

We assessed food consumption rates of slimy sculpin at different water temperatures with a series of laboratory experiments. Sculpin of all sizes available were collected from Garvin Brook and Gilmore Creek in Jun, and Jul. 2006 and 2007 (water temperatures = 12 to 16 C) with a Smith-Root backpack electrofisher and dip nets. Fish were transported immediately to the lab and acclimated to a constant water temperature of 7 (2007 only), 12, 17, or 22 C for 24 h prior to testing. This was done because these temperatures span the range of summer water temperatures in streams inhabited by sculpin in this region (N. Mundahl, unpub. data), and we wanted to determine if temperature differences may result in different rates of prey consumption. Because of equipment limitations, testing could be done at only one temperature at a time, and new fish were used at each temperature. A refrigerated, recirculating acclimation aquarium (150 L) was aerated and filtered, and water was changed between experiments. Sculpin received natural lighting and a photoperiod of 15 h light : 9 h dark via large laboratory windows. No food was provided to sculpin during the 24-h acclimation period.

Fish were tested individually in rectangular, 1-L translucent plastic testing chambers. Openings (approximately 75 x 100 mm) were cut in two opposite sides of each container and covered with screening (1.4-mm mesh) to allow for water circulation. Each container was closed with a translucent, snap-on lid.

Sculpin were placed into testing chambers along with amphipods Gammarus pseudolimnaeus Bousfiel numbering 10-30 depending on fish size. Amphipods were collected from a vegetated habitat in Burns Valley Creek (a local coldwater stream lacking sculpin) for use as a natural, live food source during feeding experiments. Preliminary diet studies had indicated that sculpin often fed heavily on amphipods when they were available and abundant. Sculpin in individual testing chambers (12-19 per experiment) were placed into the acclimation aquarium at 7, 12, 17, or 22 C and fish were allowed to feed for 24 h. After each experiment, the number of amphipods consumed was determined and wet weight and total length of the fish were measured. Representative samples (10-15 individuals) of amphipods were weighed individually (nearest mg dry weight after 24 h in a drying oven at 35 C) each year to estimate prey biomass consumed by sculpin in the lab. As in field-caught fish, consumption was standardized by fish weight (mg food dry weight/g fish wet weight).

Comparisons of sculpin food consumption were made among fish feeding at different temperatures in the laboratory, and between laboratory-fed and stream-caught fish. Standardized amphipod consumption rates in the laboratory were compared among test temperatures each year with single-factor analysis of variance. Standardized food consumption rates and a 2-factor analysis of variance (lab vs. field, 2006 vs. 2007) were used to compare consumption between lab-fed and stream-caught sculpin.

24-HOUR FEEDING EXPERIMENTS: PREY SELECTION

Prey selection experiments were conducted in the laboratory during Oct. 2008 and 2009. Sculpin were electrofished from Garvin Brook and acclimated to laboratory conditions (12 h light : 12 h dark photoperiod, 12 C refrigerated, 400-L recirculating stream tank) for 48-72 h prior to feeding trials. Fish were not fed during acclimation.

Sculpin were tested individually in the 1-L test chambers described above with varying combinations of five different types of prey (amphipods Gammarus pseudolimnaeus, isopods Asellus sp., caddisflies Brachycentrus occidentalis Banks, snails Physella sp., and mayflies Baetis sp.). Test chambers were submerged in the stream tank and fish were allowed to feed for 24 h. Six different trials were conducted, with each fish offered 10 individual prey of each of two, three, or four different prey types (20-40 individual prey/fish). Amphipods and caddisflies were used in all trials, isopods in five trials, and snails and mayflies in one trial each.

Possible selective feeding was assessed for each trial separately by comparing prey numbers offered versus those consumed with chi-square contingency table tests. Ivlev selectivity values (Krebs, 1989) also were calculated for each prey type in each trial to quantify selected versus rejected prey types.

RESULTS

SCULPIN DIETS

Sculpin collected for diet examination ranged in size from 0.1 to 30.0 g in weight, and from 20 to 116 mm in length (Table 1). Fish average sizes and size ranges differed among sites and years, largely the result of differences in sampling season (average fish larger during winter 1992-1993) and availability and abundance of different age classes of fish [average fish smaller in Jul. 2007 because of presence of many year-of-young (YOY) fish]. All comparisons of sculpin wet weights and total lengths among sites and among years indicated significant (all ANOVA P < 0.05) differences.

The 272 sculpin examined contained 2674 individual prey items, and only 10 (3.7%) fish had empty stomachs. Numbers of prey in stomachs varied widely within and among stream sites and seasons (Table 1). The mean ([+ or -] SD) number of prey per fish at the stream sites ranged from 4 ([+ or -] 2) to 31 ([+ or -] 21), with an overall mean of 10 ([+ or -] 9) diet items per fish. None of the fish from the reintroduction streams (Latsch, Big Trout Creek) in 2007 had empty stomachs (Table 1), and fish from reintroduction streams contained numbers of prey (mean [+ or -] SD: 14 [+ or -] 16) similar ([t.sub.(73)] = 0.01, P = 0.99) to fish from native sculpin streams (15 [+ or -] 16).

Standardized dry weights of prey contained in sculpin stomachs displayed considerable variation among sites and between the 2 y of sampling, with site means ([+ or -] SD) ranging from 0.6 ([+ or -] 6) to 10.1 ([+ or -] 5.0) mg prey dry weight/g fish wet weight (Fig. 2). Standardized prey weights were significantly different (ANOVA [F.sub.(3,50)] = 4.06, P = 0.01) among sites in 2007, but not (ANOVA [F.sub.(6,120)] = 1.65, P = 0.14) in 2006. Standardized weights of prey from sculpin collected during afternoon hours were frequently reduced (Fig. 2), but not significantly ([t.sub.(125)] = 1.87, P = 0.12) less than those from sculpin collected during morning hours during 2006. Similarly, standardized weights of prey were slightly, but not significantly ([t.sub.(179)] = 2.42, P = 0.07), higher in 2007 than in 2006 (Fig. 2). The standardized weights of prey from sculpin in the two reintroduction streams differed significantly between reintroduction sites but were similar (Big Trout) to those from fish from the native sculpin streams or significantly higher than those at one native site (Latsch > Garvin) (Tukey HSD tests, Fig. 2).

Throughout all the sites and years, there were 35 different types of prey consumed by sculpin (Table 2). Aquatic organisms dominated sculpin diets in all streams, with terrestrial prey items (e.g., Arachnida, Lepidoptera larvae) rarely consumed. Sculpin diets were least diverse in Cold Spring Brook, with only eight different types of organisms consumed, and most diverse in Beaver Creek, with 21 different prey types consumed. The average number of different prey types consumed for all sites was 13. However, most (94%) individual sculpin stomachs examined contained four or fewer different types of prey.

The most common types of prey found in stomachs were fly larvae (Diptera) and Amphipoda (Fig. 3). Fly larvae, especially Chironomidae and Simuliidae, were important components of the diet in nearly all streams (especially in 2007), whereas amphipods were more variable in their importance. These two taxa had the highest frequency of occurrence of any prey consumed (Table 3). Other organisms widely consumed, both as a percentage of the diet and as frequency of occurrence, included mayflies (Ephemeroptera), caddisflies (Trichoptera), and snails (Gastropoda) (Fig. 3, Table 3). Organisms that were relatively rare in stomach contents included beetles (Coleoptera), horsehair worms (Nematomorpha), segmented worms (Oligochaeta), terrestrial arthropods, and fish. Isopoda seldom comprised a large percentage of the diet (Fig. 3) but were consumed by a significant proportion of sculpin in some streams (Table 3). Sculpin in the two reintroduction streams (Latsch, Big Trout) had diets dominated by Diptera, with fish from Big Trout Creek consuming more Gastopoda, Emphemeroptera, and Trichoptera than fish from Latsch Creek (Fig. 3, Table 3).

NMS revealed significant differences in sculpin diets among the six streams examined in 2006 (Fig. 4). NMS was conducted using three dimensions (based on the scree plot) ; stress of the final solution was 0.164. The 95% confidence ellipses indicated that diet composition at two streams (Gilmore Creek and Cold Spring Brook) were both different from two other streams (Trout Run and Beaver Creek) and that no other significant differences existed among streams. Diet composition at Gilmore Creek and Cold Spring Brook included more Isopoda, Amphipoda, and Coleoptera, whereas composition at Trout Run included more Oligochaeta and Diptera and at Beaver Creek included more Ephemeroptera and Trichoptera. Fish size (length or weight) had no significant effect on diet composition (both P > 0.2).

[FIGURE 2 OMITTED]

BENTHIC AVAILABILITY

In benthic invertebrate samples collected from the 16 streams, Amphipoda, Diptera, Trichoptera, and Ephemeroptera were the most common organisms present (Fig. 5). Every stream site contained Amphipoda and one or more types of Diptera larvae. The least common organisms were Plecoptera, Gastropoda, Ostracoda, Oligochaeta, and Nematomorpha. Benthic community composition, although highly variable from stream to stream, was similar among streams with and without sculpin, and those where sculpin were reintroduced (Fig. 5). Benthic composition in sculpin reintroduction streams was highly variable among streams, especially for Amphipoda, Ephemeroptera, and Trichoptera (Fig. 6). In addition, invertebrate community measures (abundance, taxa richness, EPT richness, benthic IBI score) differed significantly among reintroduction streams (Table 4). However, differences in these benthic community parameters did not clearly separate the two successful reintroduction streams from the two unsuccessful streams (Table 4). When these same community measures were compared among the reintroduction streams, native sculpin streams, and streams lacking sculpin, no significant differences were observed (Table 5). Among the streams assessed, invertebrate densities ranged from 93 to 668 organisms/0.1 [m.sup.2], taxa richness ranged from six to 16, EPT richness from one to seven and benthic IBI score from 8 (very poor) to 52 (good).

[FIGURE 3 OMITTED]

PREY SELECTIVITY

Slimy sculpin demonstrated diet preferences by selecting for some organisms and rejecting others (Fig. 7). Fly larvae represented the only group in both years that always comprised more than 5% of the diet and/or benthos and was preferred by sculpin in every stream examined. Nematomorpha were always preferred prey, but seldom were a major proportion of the diet or benthos. Amphipoda, Isopoda, Gastropoda, and Ephemeroptera were important taxa that were selected for in some streams and selected against in others. Trichoptera and Coleoptera always were rejected by sculpin, except in one stream in 2006. Sculpin in both reintroduction streams (Latsch, Big Trout) selected for Diptera and against Amphipoda and Trichoptera, but differed in selectivity of Isopoda (Fig. 7).

24-HOUR FEEDING EXPERIMENTS: CONSUMPTION RATES

Slimy sculpin used for the 24-h consumption rate feeding experiments ranged in size from 1.6 to 22.7 g wet weight, and from 94 to 115 mm total length (Table 6). Fish representative of the entire size range were used in feeding trials at each temperature.

Feeding trials conducted in the two years produced generally similar results. In the first trial, sculpin consumed slightly, but not significantly (ANOVA F = 2.97, P = 0.06), more Amphipoda (mg amphipods/g fish) at 17 C than at either 12 C or 22 C (Table 6). In the second trial, consumption rates also did not differ (ANOVA F = 0.11, P = 0.95) among the four temperatures tested, but fish in the second trial consumed significantly (t = 3.44, P < 0.001) more amphipods than did fish in the first trial (trial 1 overall mean = 6.6 mg amphipod dry weight/g sculpin wet weight; trial 2 overall mean = 10.4 mg amphipod dry weight/g sculpin wet weight).

[FIGURE 4 OMITTED]

Consumption rates were higher for sculpin in laboratory feeding trials than in field-caught fish (two-factor ANOVA, lab vs. field F = 29.49, P < 0.001; Fig. 8). Higher rates during 2007 for both lab-fed and field-caught fish (two-factor ANOVA, 2006 vs. 2007 F = 25.86, P < 0.001) did not alter this relationship.

24-HOUR FEEDING EXPERIMENTS: PREY SELECTION

Slimy sculpin (n = 72) used for the 24-h prey selection feeding experiments ranged in size from 0.8 to 15.4 g wet weight, and from 40 to 102 mm total length. Both juvenile and adult fish were used in each of the six feeding trials.

Fish consumed 954 prey items during the feeding trials, averaging 13 prey/fish (range six-24 prey/fish). In none of the trials did sculpin consume the different prey types in the same proportions that they were offered, instead exhibiting selective feeding behavior (all contingency table P values <<0.001). Amphipoda were favored (selected for) in five of the six trials, Trichoptera were rejected (selected against) in all six trials and Isopoda were favored in four of five trials (Fig. 9). Gastropoda were rejected in the single trial tested, whereas Ephemeroptera were favored in the single trial (also the same trial where Amphipoda and Isopoda were selected against; Fig. 9).

[FIGURE 5 OMITTED]

DISCUSSION

This study highlighted several aspects of slimy sculpin diets, sculpin feeding selectivity, and benthic invertebrate assemblages in coldwater streams in southeastern Minnesota. First, sculpin had flexible diets, consuming a variety of aquatic invertebrate prey, with diets varying significantly among streams. Secondly, sculpin consistently fed selectively on specific taxa in the field and laboratory, typically consuming ~10 prey/sculpin/day. Third, potential invertebrate assemblages varied among streams, but all streams contained preferred prey in quantifies sufficient to support sculpin populations in densities typical for the region. Taken together, these aspects largely eliminate food resources as a factor contributing to failures of recent sculpin reintroductions in southeastern Minnesota.

Despite the important role of invertebrate prey in habitat patch selection, rates of food consumption and growth, and survival of sculpin (e.g., Brocksen et al., 1968; Petty and Grossman, 1996, 2007, 2010; Ruetz et al., 2004; Grossman et al., 2006; Zimmerman and Vondracek, 2006, 2007a, 2007b), sculpin populations can be successful in streams that vary greatly in their invertebrate communities (Dineen, 1951; Zimmerman and Vondracek, 2007a, 2007b, present study). Slimy sculpin in the present study were euryphagous, consuming an average of 13 different taxa of prey at each stream site. Sculpin in most of the study streams consumed primarily Diptera larvae and Amphipoda, but Ephemeroptera, Trichoptera, and Gastropoda also contributed significantly to the diet in some streams. Previous studies have reported similar, mixed diets for sculpin in Minnesota (Dineen, 1951; Petrosky and Waters, 1975; Ruetz et al., 2004; Zimmerman and Vondracek, 2007b), Alaska (Hershey, 1985; Cuker et al., 1992), Oklahoma (Tumlison and Cline, 2002), and Lake Ontario (Owens and Dittman, 2003), although various methods for reporting diets (frequency of occurrence, number, biomass) complicate direct comparisons among study streams.

[FIGURE 6 OMITTED]

Variation in sculpin diets among nearby streams, or even within the same stream during different seasons or years, was likely the result of differences in prey availability and accessibility among streams (Cuker et al., 1992; Tumlison and Cline, 2002). Sculpin also may exhibit ontogenetic changes in diets (Brandt, 1986; Tabor et al., 2007), at least in part due to habitat shifts (Koczaja et al., 2005; Petty and Grossman, 2010) that result from intraspecific competition between juvenile and adult fish (Petty and Grossman, 2007). Sculpin generally are insensitive to even major changes in prey assemblages, shifting diets to compensate for declines in important prey items (Owens and Dittmann, 2003; Hondorp et al., 2005) with no observable effect on sculpin population parameters (O'Brien et al., 2005). Fish with such mixed, flexible diets are well equipped to cope with changes in prey availability resulting from environmental change (Pratchett et al., 2004), species invasions (Pothoven et al., 2001; Truemper and Lauer, 2005; Pothoven and Nalepa, 2006), or disturbance (Wilson et al., 2006; Emslie et al., 2011), and consequently can thrive in a variety of systems where fishes with more specialized diets may be affected disproportionately (Schoener, 1971; Dill, 1983; Munday, 2004; Berumen and Pratchett, 2008). Because of their diet flexibility, slimy sculpin can be successful in reintroduction streams in southeastern Minnesota even if these streams vary considerably in their invertebrate prey assemblages (Troelstrup and Perry, 1989).

Slimy sculpin selected for certain types of prey while rejecting others in both field and laboratory studies. Selective predation has been documented previously for several species of sculpins under both field and laboratory conditions, with selection for or against prey items generally attributed to prey size, behavior, and/or accessibility (Newman and Waters, 1984; Hershey and Dodson, 1985; Kohler and McPeek, 1989; Tumlison and Cline, 2002; Miyasaka et al., 2003; Fairchild and Holomuzki, 2005; Zimmerman and Vondracek, 2007b). Because of their relatively large gape (Foote and Brown, 1998), slimy sculpin can access successfully all but the largest individuals (e.g., crayfishes, large stoneflies) in benthic invertebrate assemblages and can prey selectively on the larger, more energy-efficient sizes of food organisms (Newman and Waters, 1984; Sparkes, 1996). This suggests that sculpin might select for taxa of larger prey, as long as reduced attack success and longer handling times do not negate the energy efficiencies that come with larger size prey (Kratz and Vinyard, 1981). However, slimy sculpin selected for the relatively small Diptera larvae (usually Chironomidae) in every stream examined. Because sculpin often rely on their mechanosensory lateral line to locate and orient to prey in low-light environments (Hoekstra and Janssen, 1985; Kanter and Coombs, 2003), selection by sculpin may be more related to prey movement than to other factors (Kratz and Vinyard, 1981). Consequently, what are generally interpreted as being prey preferences of sculpin based on diet analyses (Bowen, 1996; Begon et al., 2006) may largely be a function of detection and encounter rates of foraging sculpin with moving prey (Kratz and Vinyard, 1981). "Selected" prey are those that are abundant enough to be encountered regularly by feeding sculpin and whose movements produce vibrations of the type that can be detected readily by a sculpin's lateral line system (Kantner and Coombs, 2003). "Rejected" prey are those that do not move (Kratz and Vinyard, 1981; Moore and Williams, 1990) or can otherwise escape detection and capture (Feltmate and Williams, 1989).

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

Rates of prey consumption (numbers and biomass of prey) by slimy sculpin observed in laboratory and field-caught fish were in agreement with those reported for sculpin in various systems in Minnesota, Alaska, Washington, and other locations in North America (Brocksen et al., 1968; Hershey, 1985; Greenberg and Holtzman, 1987; Foote and Brown, 1998; Moss, 2001; Zimmerman and Vondracek, 2007a, b). Most field-caught and laboratory-fed sculpin averaged ~10 prey items per fish, although individual fish may consume >50 smaller prey in a single day (Foote and Brown, 1998; present study). Under normal sculpin densities (<1 fish/[m.sup.2]; WI DNR, 1978; Anderson, 1985; Gibson et al., 2004), these rates of prey consumption may produce an observable effect on the density of benthic invertebrates (Brocksen et al., 1968; Cuker et al., 1992; Gibson et al., 2004), although this observation is not universal (Miyasaka et al., 2003; Ruetz et al., 2003; Zimmerman and Vondracek, 2007a, b). The presence of potential predators (e.g., large trout; Chivers et al., 2001) or competitors of sculpin may alter sculpin feeding behavior, lessening their effects on invertebrate assemblages (Freeman and Stouder, 1989; Grossman et al., 1995; Chivers et al., 2001). With benthic invertebrate densities ranging from 1500 to 6000 organisms/[m.sup.2], and with no significant difference in invertebrate densities between streams with and without sculpin, streams in southeastern Minnesota appear to have adequate prey to support populations of slimy sculpin. Even after a 1000-y flood event in Aug. 2007 reduced benthic invertebrate densities in many streams in southeastern Minnesota by >95% (Mundahl and Hunt, 2011), slimy sculpin retained normal feeding rates, maintained or improved condition throughout fall and winter months, and spawned successfully the following spring (N. Mundahl, unpub, data).

In conclusion, slimy sculpin reintroductions into southeastern Minnesota streams have not been compromised by a lack of available suitable prey. Although benthic invertebrate communities varied among streams, all streams contained adequate abundances of prey that were selectively preferred by sculpin, and sculpin were highly flexible to varying prey assemblages. There were no consistent differences in either the diets of sculpin collected from successful and unsuccessful reintroduction streams, or in the prey communities available to sculpin in these streams. These streams also have adequate sculpin habitat (Mundahl et al., 2012) and recent, massive flooding had no observable effects on sculpin populations (N. Mundahl, unpub. data). The genetics of the reintroduced sculpin may be responsible for some reintroduction failures, as heterozygosity is lower than expected and some degree of outbreeding depression may be occurring at some sculpin reintroduction sites (Huff et al., 2010, 2011). Although few brown or brook trout consume sculpin in these small streams (<7% of 292 fish stomachs examined contained sculpin; Dineen, 1951), the possible effects of salmonid predation on reintroduced sculpin should be examined to determine if increasing and expanding brown trout populations may be suppressing sculpin populations and hindering reintroduction efforts.

Acknowledgments.--We thank A. Bilicki, K. Nelson, E. Mundahl, N. Zwonitzer, and I. Phelps for assisting with field collections and lab set up. The Minnesota Department of Natural Resources Fisheries research office issued fish collecting permits for the study streams, and the Winona State University IACUC approved the lab feeding studies. A portion of this study was funded by an award to D. Mundahl from the Minnesota Academy of Science Research and Engineering Apprenticeship Program.

LITERATURE CITED

ADAMS, S. B. AND D. A. SCHMETTERLING. 2007. Freshwater sculpins: phylogenetics to ecology. Trans. Am. Fish. Soc., 136:1736-1741.

ANDERSON, C. S. 1985. The structure of sculpin populations along a stream size gradient. Env. Biol. Fishes, 13:93-102.

BEGON, M., C. R. TOWNSEND, AND J. L. HARPER. 2006. Ecology: from individuals to ecosystems. Blackwell, Malden, Massachusetts.

BERUMEN, M. L. AND M. S. PRATCHETT. 2008. Trade-offs associated with dietary specialization in corallivorous butterflyfishes (Chaetodontidae: Chaetodon). Behav. Ecol. Sociobiol, 62:989-994.

BOWEN, S. H. 1996. Quantitative description of the diet, p. 513-532. In: B. R. Murphy and D. W. Willis (eds.). Fisheries techniques, 2nd ed. American Fisheries Society, Bethesda, Maryland.

BRANDT, S. B. 1986. Ontogenetics shifts in habitat, diet, and diel-feeding periodicity of slimy sculpin in Lake Ontario. Trans. Amer. Fish. Soc., 115:711-715.

BROCKSEN, R. W., G. E. DAVIS, AND C. E. WARREN. 1968. Competition, food consumption, and production of sculpins and trout in laboratory stream communities. J. Wildl. Mgmt., 32:51-75.

CHIVERS, D. P., R. S. MIRZA, P.J. BRYER, AND J. M. KIESECKER. 2001. Threat-sensitive predator avoidance by slimy sculpins: understanding the importance of visual versus chemical information. Can. J. Zool., 79:867-873.

CUKER, B. E., M. E. MCDONALD, AND S. C. MOZLEY. 1992. Influences of slimy sculpin (Cottus cognatus) predation on the rocky littoral invertebrate community in an arctic lake. Hydrobiologia, 240:83-90.

DILL, L. 1983. Adaptive flexibility in the foraging behavior of fishes. Can. J. Fish. Aquat. Sci., 40:398-408.

DINEEN, C. F. 1951. A comparative study of the food habits of Cottus bairdii and associated species of Salmonidae. Am. Midl. Nat., 46:640-645.

EMSLIE, M. J., M. S. PRATCHETT, AND A. J. CHEAL. 2011. Effects of different disturbance types on butterflyfish communities of Australia's Great Barrier Reef. Coral Reefs, 30:461-471.

FAIRCHILD, M. P. AND J. R. HOLOMUZKI. 2005. Multiple predator effects on microdistributions, survival, and drift of stream hydropsychid caddisflies. J. N. Am. Benthol. Soc., 24:101-112.

FELTMATE, B. W. AND D. D. WILLIAMS. 1989. A test of crypsis and predator avoidance in the stonefly Paragnetina media (Plecoptera: Perlidae). Anim. Behav., 37:992-999.

FOOTE, C.J. AND G. S. BROWN. 1998. Ecological relationships between freshwater sculpins (genus Cottus) and beach-spawning sockeye salmon (Oncorhynchus nerka) in Iliamna Lake, Alaska. Can. J. Fish. Aquat. Sci., 55:1524-1533.

FREEMAN, M. C. AND D. J. STOUDER. 1989. Intraspecific interactions influence size specific depth distribution in Cottus bairdi. Env. Biol. Fish., 24:231-236.

GIBSON, C. A., R. E. RATAJCZAK, AND G. D. GROSSMAN. 2004. Patch based predation in a southern Appalachian stream. Oikos, 106:158-166.

GREENBERG, L. A. AND D. A. HOLTZMAN. 1987. Microhabitat utilization, feeding periodicity, home range and population size of the banded sculpin, Cottus carolinae. Copeia, 1:19-25.

GROSSMAN, G. D., R. E. RATAJCZAK, AND M. K. CRAWFORD. 1995. Do rock bass (Ambloplites rupestris) induce microhabitat shifts in mottled sculpin (Cottus bairdi)? Copeia, 1995:343-353.

--, --, J. T. PETTY, M. D. HUNTER, J. T. PETERSON, AND G. GRENOUILLET. 2006. Population dynamics of mottled sculpin (Pisces) in a variable environment: information theoretic approaches. Ecol. Monogr., 76:217-234.

HERSHEY, A. E. 1985. Effects of predatory sculpin on the chironomid communities in an arctic lake. Ecology, 66:1131-1138.

-- AND S. I. DODSON. 1985. Selective predation by a sculpin and a stonefly on two chironomids in laboratory feeding trials. Hydrobiologia, 124:269-273.

HOEKSTRA, D. AND J. JANSSEN. 1985. Non-visual feeding behavior of the mottled sculpin, Cottus bairdi, in Lake Michigan. Env. Biol. Fish., 12:111-117.

HONDORE, D. W., S. A. POTHOVEN, AND S. B. BRANDT. 2005. Influence of Diporeia density on diet composition, relative abundance, and energy density of planktivorous fishes in southeast Lake Michigan. Trans. Am. Fish. Soc., 134:588-601.

HUFF, D. D. 2010. Examining genetic diversity, outhreeding depression, and local adaptation in a native fish reintroduction program. Doctoral dissertation, University of Minnesota. 116 p.

--, L. M. MILLER, C.J. CHIZINSKI, AND B. VONDRACEK. 2011. Mixed-source reintroductions lead to outbreeding depression in the second generation descendents of a native North American fish. Molec. Ecol., 20:4246-4258.

--, --, AND B. VONDRACEK. 2010. Patterns of ancestry and genetic diversity in reintroduced populations of the slimy sculpin: implications for conservation. Conserv. Genet., 11:2379-2391.

KANTER, M.J. AND S. COOMBS. 2003. Rheotaxis and prey detection in uniform currents by Lake Michigan mottled sculpin (Cottus bairdi). J. Exp. Biol., 206:59-70.

KEILLOR, L. 2010. Fishing after the flood. Minnesota Conserv. Volunt., 73(431):8-17.

KOCZAJA, C., L. McCALL, E. FITCH, B. GLORIOSO, C. HANNA, J. KYZAR, M. NIEMILLER, J. SPIESS, A. TOLLEY, R. WYCKOFF, AND D. MULLEN. 2005. Size-specific habitat segregation and intraspecific interactions in banded sculpin (Cottus carolinae). Southeast. Nat., 4:207-218.

KOHLER, S. L. AND M. A. MCPEEK. 1989. Predation risk and the foraging behavior of competing stream insects. Ecology, 70:1811-1825.

KRATZ, K. AND G. L. VINYARD. 1981. Mechanisms of prey selectivity in the Piute sculpin, Cottus beldingi. CalNeva Wildl. Trans., 17:11-18.

KREBS, C.J. 1989. Ecological methodology. Harper & Row, New York.

LYONS, J., L. WANG, AND T. D. SIMONSON. 1996. Development and validation of an index of biotic integrity for coldwater streams in Wisconsin. N. Am. J. Fish. Manage., 16:241-256.

MINNESOTA DNR. 2003. Southeast Minnesota sculpin reintroduction plan. Minnesota DNR, St. Paul.

MIYASAKA, H., M. GENKAI-KATO, N. KUHARA, AND S. NAKANO. 2003. Predatory fish impact on competition between stream insect grazers: a consideration of behaviorally- and density-mediated effects on an apparent coexistence pattern. Oikos, 101:511-520.

MOORE, K. A. AND D. D. WILLIAMS. 1990. Novel strategies in the complex defense repertoire of a stonefly (Pteronarcys dorsata) nymph. Oikos, 57:49-56.

MOSS, J. H. H. 2001. Development and application of a bioenergetics model for Lake Washington prickly sculpin (Cottus asper). M.S. Thesis, University of Washington.

MUCK, J. A. AND R. M. NEWMAN. 1992. The distribution of amphipods in southeastern Minnesota and their relation to water quality and land use. J. Iowa Acad. Sci., 99:34-39.

MUNDAHL, N. D. AND A. M. HUNT. 2011. Recovery of stream invertebrates after catastrophic flooding in southeastern Minnesota, USA. J. Freshw. Ecol., 26:445-457.

-- AND T. P. SIMON. 1998. Development and application of an index of biotic integrity for coldwater streams of the upper midwestern United States, p. 383-415. In: T. P. Simon (ed.). Assessing the sustainahility and biological integrity of water resource quality using fish assemblages. Lewis Publishers, Boca Raton, Florida.

--, K. NELSON THOMAS, AND E. D. MUNDAHL. 2012. Selected habitats of slimy sculpin in coldwaree tributaries of the upper Mississippi River in Minnesota. Am. Midl. Nat., 168:144-161.

MUNDAY, P. L. 2004. Habitat loss, resource specialization, and extinction on coral reefs. Glob. Change Biol., 10:1642-1647.

NEWMAN, R. M. AND T. F. WATERS. 1984. Size-selective predation on Gammarus pseudolimnaeus by trout and sculpins. Ecology, 65:1535-1545.

O'BRIEN, W.J., M. BARFIELD, N. BETTEZ, A. E. HERSHEY, J. E. HOBBLE, G. KIPPHUT, G. KLING, AND M. C. MILLER. 2005. Long-term response and recovery to nutrient addition of a partitioned arctic lake. Freshw. Biol., 50:731-741.

OWENS, R. W. AND D. E. DITTMAN. 2003. Shifts in the diets of slimy sculpin (Cottus cognatus) and lake whitefish (Coregonus clupeaformis) in Lake Ontario following the collapse of the burrowing amphipod Diporeia. Aquat. Ecosys. Health Manage. J., 6:311-323.

PETROSKY, C. E. AND T. F. WATERS. 1975. Annual production by the slimy sculpin population in a small Minnesota trout stream. Trans. Am. Fish. Soc., 104:237-244.

PETTY, J. T. 2007. Size-dependent territoriality of mottled sculpin in a southern Appalachian stream. Trans. Am. Fish. Soc., 136:1750-1761.

--. 2010. Giving-up densities and ideal pre-emptive patch use in a predatory benthic stream fish. Freshw. Biol., 55:780-793.

-- AND G. D. GROSSMAN. 1996. Patch selection by mottled sculpin (Pisces: Cottidae) in a southern Appalachian stream. Freshw. Biol., 35:261-276.

POTHOVEN, S. A. AND T. F. NALEPA. 2006. Feeding ecology of lake whitefish in Lake Huron. J. Great Lakes Res., 32:489-501.

--, ---, P.J. SCHNEEBERGER, AND S. B. BRANDT. 2001. Changes in diet and body condition of Lake Whitefish in southern Lake Michigan associated with changes in benthos. N. Am. J. Fish. Manage., 21:876-883.

PRATCHETT, M. S., S. K. WILSON, M. L. BERUMEN, AND M. I. McCORMICK. 2004. Sublethal effects of coral bleaching on an obligate coral feeding butterflyfish. Coral Reefs, 23:352-356.

RABENI, C. F. 1996. Invertebrates, p. 335-352. In: B. R. Murphy and D. W. Willis (eds.). Fisheries techniques, 2nd ed. American Fisheries Society, Bethesda, Maryland.

RUETZ, C. R., B. VONDRACEK, AND R. M. NEWMAN. 2004. Weak top-down control of grazers and periphyton by slimy sculpins in a coldwater stream. J. N. Am. Benthol. Soc., 23:271-286.

SCHOENER, T. W. 1971. Theory of feeding stategies. Ann. Rev. Ecol. Sys., 2:369-404.

SPARKES, T. C. 1996. Effects of predation risk on population variation in adult size in a stream-dwelling isopod. Oecologia, 106:85-92.

TABOR, R. A., E.J. WERNER, K. L. FRESH, B. A. FOOTEN, AND J. R. CHAN. 2007. Ontogenetic diet shifts of prckly sculpin in the Lake Washington Basin, Washington. Trans. Am. Fish. Soc., 136:1801-1813.

THORN, W. C., C. S. ANDERSON, W. E. LORENZEN, D. L. HENDRICKSON, AND J. W. WAGNER. 1997. A review of trout management in southeast Minnesota trout streams. N. Am. J. Fish. Manage., 17:860-872.

TROELSTRUP, N. H. AND J. A. PERRY. 1989. Water quality in southeastern Minnesota streams: observations along a gradient of land use and geology. J. Minn. Acad. Sci., 55:6-13.

TRUEMPER, H. A. AND T. E. LAUER. 2005. Gape limitation and piscine prey size-selection by yellow perch in the extreme southern area of Lake Michigan, with emphasis on two exotic prey items. J. Fish Biol., 66:135-149.

TUMLISON, R. AND G. R. CLINE. 2002. Food habits of the banded sculpin (Cottus carolinae) in Oklahoma with reference to predation on the Oklahoma salamander (Eurycea tynerensis). Proc. Okla. Acad. Sci., 82:111-113.

WATERS, T. F. 2000. Wildstream: a natural history of the free flowing river. Riparian Press, St. Paul. 608 p.

WILSON, S. K., N. A. J. GRAHAM, M. S. PRATCHETT, G. P. JONES, AND N. V. C. POLUNIN. 2006. Multiple disturbances and the global degradtion of coral reefs: are reef fishes at risk or resilient? Glob. Change Biol., 12:2220-2234.

WISCONSIN DNR. 1978. Distribution and density of sculpins in a Wisconsin coulee stream. Wisconsin Department of Natural Resources Research Report 98.

WITTMAN, E. AND N. MUNDAHL. 2003. Development and validation of a benthic index of biotic integrity (B-IBI) for streams in southeastern Minnesota. Floruit, 1:41-48.

ZIMMERMAN, J. K. H. AND B. VONDRACEK. 2006. Interactions of slimy sculpin (Cottus cognatus) with native and nonnative trout: consequences for growth. Can. J. Fish. Aquat. Sci., 63:1526-1535.

-- AND --. 2007a. Interactions between slimy sculpin and trout: slimy sculpin growth and diet in relation to native and nonnative trout. Trans. Am. Fish. Soc., 136:1791-1800.

-- AND --. 2007b. Brown trout and food web interactions in a Minnesota stream. Freshw. Biol., 52:123-136.

SUBMITTED 23 NOVEMBER 2010

ACCEPTED 31 JANUARY 2012

NEAL D. MUNDAHL (1) AND DARCY E. MUNDAHL

Department of Biology and Large River Studies Center, Winona State University, P.O. Box 5838, Winona, Minnesota 55987

AND

ERIC C. MERTEN

Department of Biology, University of Wisconsin-Eau Claire, Eau Claire 45702

(1) Corresponding author: Telephone: (507) 457-5695; e-mail: nmundahl@winona.edu
TABLE 1.--Mean ([+ or -]SD) wet weights (g), total lengths (mm),
and prey consumption (number per stomach) of slimy sculpin
collected from streams in southeastern Minnesota in 1992-1993,
2006, and 2007

                              Number       Wet weight
  Season/Year/Stream         of fish         (range)

Winter 1992-1993

  Gilmore Creek (Dec.)          18       8.5 [+ or -] 5.3
                                           (1.8-30.0)

  Garvin Brook (Dec.)           11      12.1 [+ or -] 5.0
                                           (0.7-18.5)

  Garvin Brook (Mar.)           42      6.3 [+ or -] 6.3
                                           (0.5-23.2)

Summer 2006

  Garvin Brook (Jun.)           37       4.6 [+ or -] 3.0
                                           (1.6-12.9)

  Beaver Creek (Jun.)           18       5.6 [+ or -] 3.7
                                           (2.6-17.3)

  Cold Spring Brook (Jul.)               5.3 [+ or -] 2.9
                                            (1.9-9.3)

  S.Br. Whitewater (Jul.)                9.6 [+ or -] 3.9
                                           (3.6-15.9)

  Gilmore Creek (Jul.)          18       5.1 [+ or -] 4.3
                                           (1.4-19.4)

  Trout Run (Jul.)              18       5.3 [+ or -] 2.7
                                           (2.0-10.0)
Summer 2007

  Garvin Brook (Jun.)            7       8.3 [+ or -] 3.8
                                           (2.8-12.6)

  Gilmore Creek (Jul.)          18       1.3 [+ or -] 2.0
                                            (0.2-6.9)

  Latsch Creek (Jul.)           26       1.6 [+ or -] 2.9
                                            (0.1-8.9)

  Big Trout Creek (Jul.)        23       5.3 [+ or -] 8.4
                                           (0.7-24.0)

                              Total length    Number of prey
  Season/Year/Stream            (range)          (range)

Winter 1992-1993

  Gilmore Creek (Dec.)       67 [+ or -] 14    5 [+ or -] 4
                                (52-116)          (1-17)

  Garvin Brook (Dec.)        92 [+ or -] 21    7 [+ or -] 6
                                (38-111)          (0-18)

  Garvin Brook (Mar.)        71 [+ or -] 22   10 [+ or -] 13
                                (39-116)          (0-69)

Summer 2006

  Garvin Brook (Jun.)        67 [+ or -] 12    7 [+ or -] 5
                                (50-97)           (0-21)

  Beaver Creek (Jun.)        73 [+ or -] 12   11 [+ or -] 15
                                (60-102)          (2-62)

  Cold Spring Brook (Jul.)   73 [+ or -] 13    8 [+ or -] 8
                                (56-90)           (0-30)

  S.Br. Whitewater (Jul.)    85 [+ or -] 11    4 [+ or -] 2
                                (66-100)          (2-10)

  Gilmore Creek (Jul.)       69 [+ or -] 16    9 [+ or -] 7
                                (48-112)          (1-24)

  Trout Run (Jul.)           71 [+ or -] 11    7 [+ or -] 7
                                (57-88)           (0-28)
Summer 2007

  Garvin Brook (Jun.)        77 [+ or -] 13   31 [+ or -] 21
                                (55-90)           (1-56)

  Gilmore Creek (Jul.)       40 [+ or -] 15    8 [+ or -] 7
                                (27-76)           (1-25)

  Latsch Creek (Jul.)        36 [+ or -] 23   19 [+ or -] 21
                                (20-85)           (1-81)

  Big Trout Creek (Jul.)     57 [+ or -] 27   10 [+ or -] 5
                                (38-112)          (2-23)

                               Number of      Total
  Season/Year/Stream         empty stomachs   prey

Winter 1992-1993

  Gilmore Creek (Dec.)             0            87

  Garvin Brook (Dec.)              1            73

  Garvin Brook (Mar.)              4           423

Summer 2006

  Garvin Brook (Jun.)              1           270

  Beaver Creek (Jun.)              0           198

  Cold Spring Brook (Jul.)         1           148

  S.Br. Whitewater (Jul.)          0            79

  Gilmore Creek (Jul.)             0           165

  Trout Run (Jul.)                 3           123

Summer 2007

  Garvin Brook (Jun.)              0           219

  Gilmore Creek (Jul.)             0           142

  Latsch Creek (Jul.)              0           480

  Big Trout Creek (Jul.)           0           267

TABLE 2.--Taxa of prey consumed by slimy sculpin in streams of
southeastern Minnesota

    Phylum            Class           Order          Family (Genus)

Chordata
  (slimy
  sculpin, fish
  eggs)
Mollusca
                  Gastropoda
                    (Physella,
                    snail eggs)
Nematomorpha
Annelida
                  Oligochaeta
                  Hirudinea
Arthropoda
                  Ostracoda
                  Isopoda
                    (Asellus)
                  Amphipoda
                    (Gammarus)
                  Arachnida
                  Acari
                  Insecta
                                  Lepidoptera
                                  Plecoptera
                                  Megaloptera
                                                  Sialidae
                                                    (Sialis)
                                  Diptera
                                                  Chironomidae
                                                  Simuliidae
                                                    (Simulium)
                                                  Ceratopogonidae
                                                  Empididae
                                                  Tipulidae (Tipula,
                                                    Limnophora,
                                                    Antocha, Hexatoma)
                                  Ephemeroptera
                                                  Ephemerellidae
                                                    (Ephemerella)
                                                  Baetidae (Baetis)
                                                  Leptohyphidae
                                                    (Tricorythodes)
                                                  Heptageniidae
                                                    (Stenonema)
                                  Trichoptera
                                                  Hydropsychidae
                                                    (Hydropsyche,
                                                    Cheumatopsyche)
                                                  Brachycentridae
                                                  (Brachycentrus,
                                                    Micrasema)
                                                  Limnephilidae
                                                    (Limnephilus)
                                                  Glossosomatidae
                                                    (Glossosoma)
                                                  Helicopsychidae
                                                    (Helicopsyche)
                                  Coleoptera
                                                  Elmidae
                                                    (Optioserous,
                                                    Stenelmis)
                                                  Dytiscidae (Agabus)

TABLE 3--Frequency of occurrence (percent of stomachs examined)
of major prey taxa of slimy sculpin from streams in southeastern
Minnesota, 1992, 1993, 2006, and 2007

                            Amphipoda   Isopoda   Gastropoda
Year/stream                 Gammarus    Asellus    Physella

1992-1993
  Garvin Brook ('92)           27          0          73
  Garvin Brook ('93)           12          2           7
  Gilmore Creek ('92)          94         61           6
2006
  Garvin Brook (AM)            78          0           6
  Garvin Brook (PM)            37          5           5
  Beaver Creek                  6          0          17
  Cold Spring Brook            83          0          11
  S. Br. Whitewater River      44          0          22
  Gilmore Creek                89         50           6
  Trout Run                    22          0          17
2007
  Garvin Brook                 29          0           0
  Gilmore Creek                22         11           0
  Latsch Creek                 12         30           0
  Big Trout Creek              22         35           9
Median                         28          1           7
Average                        41         14          13

                            Ephemeroptera   Trichoptera     Diptera
Year/stream                    Baetis       Hydropsyche   Chironomidae

1992-1993
  Garvin Brook ('92)              9             18             55
  Garvin Brook ('93)             29             26             74
  Gilmore Creek ('92)             0             22             33
2006
  Garvin Brook (AM)               6             17             72
  Garvin Brook (PM)               0             47             84
  Beaver Creek                   22             17             44
  Cold Spring Brook              17              0             61
  S. Br. Whitewater River        33             17             67
  Gilmore Creek                   0              6             39
  Trout Run                      17             11             78
2007
  Garvin Brook                    0              0             86
  Gilmore Creek                   6             50             72
  Latsch Creek                   12             23             96
  Big Trout Creek                48             61             83
Median                           11             18             72
Average                          14             23             67

TABLE 4.--Assessments of benthic invertebrate communities in four
streams where sculpin have been reintroduced: two streams where
reintroductions are considered successful, two where introductions
have been unsuccessful. Values are means, with SD in parentheses.
Abundance = number/0.1 [m.sup.2]. Within each row, values followed
by different letters are significantly different from one another
(Tukey HSD tests)

                        Successful

Parameter       Little Trout      Latsch

Abundance        612 (46) A     266 (22) B
Taxa richness   13.3 (2.3) A   12.3 (1.5) A
EPT richness     6.0 (0.0) A    4.7 (0.6) A
BIBI score      26.7 (2.9) A   31.7 (11.5) A
BIBI rating      poor/fair       poor/fair

                       Unsuccessful                 ANOVA

Parameter        Big Trout       Sugarloaf       F       P

Abundance        294 (97) B     164 (142) B     11.8   <0.001
Taxa richness   13.5 (1.8) A    6.7 (2.1) B     25.4   <0.001
EPT richness     4.7 (0.5) A    2.1 (1.1) B     27.5   <0.001
BIBI score      30.0 (7.1) A   12.5 (6.5) B     14.9   <0.001
BIBI rating      poor/fair     very poor/poor

TABLE 5.--Assessments of benthic invertebrate communities in
streams with native populations of slimy sculpin (n = 7), streams
where sculpin have been reintroduced (n = 4), and non/sculpin
streams (n = 5). Values are means, with SD in parentheses.
Abundance = number/0.1 [m.sup.2]

                            Stream type                     ANOVA

Parameter         Native      Introduced   Non-sculpin    F      P

Abundance        344 (151)    284 (165)    311 (237)     0.23   0.80
Taxa richness   11.7 (2.7)    11.2 (3.0)   12.3 (3.9)    0.17   0.84
EPT richness     4.9 (1.4)    4.2 (1.5)    4.1 (1.4)     0.70   0.51
BIBI score      28.3 (12.9)   24.3 (8.2)   28.0 (14.6)   0.24   0.79
BIBI rating      poor/fair    poor/fair     poor/fair

TABLE 6.--Mean ([+ or -] SD) wet weights (g), total lengths (mm) and
consumption rates (mg amphipod dry weight/g sculpin wet weight/
24 h) of slimy sculpin in 24h feeding experiments at constant
temperatures. Values in parentheses are ranges. N values are
sample sizes

Temperature   n      Wet weight      Total length   Consumption rate

Trial 1-2006

  12C         19  6.1 [+ or -] 5.2  76 [+ or -] 18   5.4 [+ or -] 5.5
                     (1.6-22.7)        (52-115)        (0.0-17.3)

  17C         12  6.8 [+ or -] 2.6  80 [+ or -] 9    9.2 [+ or -] 2.9
                     (3.7-10.6)        (24-70)         (4.9-14.9)

  22C         14  6.7 [+ or -] 5.5  77 [+ or -] 17   5.9 [+ or -] 3.8
                     (2.0-20.1)        (57-110)        (1.3-12.1)

Trial 2-2007

  7C          12  8.2 [+ or -] 6.3  79 [+ or -] 17   9.8 [+ or -] 4.2
                     (3.0-21.2)        (61-109)        (4.8-17.0)

  12C          6  7.0 [+ or -] 6.8  76 [+ or -] 19  10.7 [+ or -] 3.6
                     (2.9-20.6)        (61-111)        (7.4-17.6)

  17C         12  8.8 [+ or -] 7.1  81 [+ or -] 22  10.5 [+ or -] 6.7
                     (2.0-23.0)        (53-118)        (4.4-25.5)

  22C          6  7.3 [+ or -] 6.9  75 [+ or -] 23  11.3 [+ or -] 6.7
                     (2.1-20.6)        (53-115)        (5.4-24.3)
Gale Copyright: Copyright 2012 Gale, Cengage Learning. All rights reserved.