2013 NORTHEASTERN NATURALIST 20(1):37–48 Diet of the Eastern Mudminnow (Umbra pygmaea DeKay) from Two Geographically Distinct Populations within the North American Native Range Frank M. Panek1,* and Judith S. Weis2 Abstract - Umbra pygmaea (Eastern Mudminnow) is a freshwater species common in Atlantic slope coastal lowlands from southern New York to northern Florida and is typical of slow-moving, mud-bottomed, and highly vegetated streams, swamps, and small ponds. We examined its seasonal food habits at the Great Swamp National Wildlife Refuge (NWR), NJ and at the Croatan National Forest, NC. A total of 147 Eastern Mudminnow from 35–112 mm TL and 190 Eastern Mudminnow from 22–89 mm TL were examined from these sites, respectively. At both locations, we found it to be a bottom-feeding generalist that consumes cladocerans, ostracods, chironomid larvae, coleopteran larvae, and other insects and crustaceans. Ostracods were most common in the diet at the Great Swamp NWR and occurred in 62% ± 2.5% of the stomachs with food. At Croatan National Forest, chironomid larvae were most common and occurred in 66.7% ± 15.8% of the stomachs. There were no statistically significant differences in diet composition between the sites during the winter, summer, and fall. However, when compared on an annual basis, Jaccard’s Index (θJ = 0.636, P = 0.05) suggested that the diet at the two study sites was significantly different. While we identified the same major food groups at both locations, the utilization of these food groups varied seasonally. Detritus was a major stomach content at both locations throughout the year. We also documented cannibalism during the summer season at both locations. The seasonal diet of the Eastern Mudminnow was similar to that of Umbra limi (Central Mudminnow) and Umbra krameri (European Mudminnow). Our findings here are the first quantitative examinations of seasonal differences in the diet of the Eastern Mudminnow within its native North American range. Introduction Umbra pygmaea DeKay (Eastern Mudminnow) is one of four species of Umbridae in North America (Wilson and Veilleux 1982). It is a freshwater species common in Atlantic slope coastal lowlands from southern New York to northern Florida and is typical of slow-moving, mud-bottomed, and highly vegetated streams, swamps, and small ponds (Lee et al. 1980). Aspects of the Eastern Mudminnow life history have been described by Panek and Weis (2012) for a North American population and by Verreycken et al. (2010) for populations in France, Belgium, Germany, and The Netherlands, where it is introduced and highly invasive. Aspects of the life history of the Eastern Mudminnow are similar to those more extensively described for Umbra limi Kirtland (Central Mudminnow) by 1US Geological Survey, Leetown Science Center, 11649 Leetown Road, Kearneysville, WV 25420. 2Rutgers - The State University, Department of Biological Sciences, 195 University Avenue, Newark, NJ 07102. *Corresponding author - fpanek@usgs.gov. 38 Northeastern Naturalist Vol. 20, No. 1 Peckham and Dineen (1957). However, little is known of the foods and feeding of the Eastern Mudminnow, and most published observations are anecdotal. In this study, we quantitatively examine the seasonal food habitats of the Eastern Mudminnow at two locations within its native range and contrast these with those reported for the Central Mudminnow. Field Site Descriptions Fish were sampled seasonally for one year from two locations within the native range of the Eastern Mudminnow: at the Great Swamp National Wildlife Refuge in north-central New Jersey within the Black Creek system of the Passaic River Basin as described by Panek and Weis (2012), and at Black Swamp Creek within the Croatan National Forest near New Bern and Morehead City, NC (Fig. 1). The Croatan National Forest is a Mid-Atlantic coastal forest ecosystem bordering the Neuse and White Oak rivers and the coastal waters of Bogue Sound. The forest ecosystem is dominated by Pinus palustris Mill. (Long-leaf Pine), evergreen-shrub bogs, and raised swamps or pocosins. Soils and associated vegetation in the forest were described by Christensen et al. (1988). Throughout the basin, soils contain a high percentage of organic material, and waters exhibit a dark brown or tea color as a result of the leaching of tannic acids, which also reduces the pH of waters in the area to 5.8 to 6.5. The freshwater fish fauna within Croatan National Forest was described by Rohde et al. (1979), and the habitat at Black Swamp Creek was described by Ross and Rohde (2003). Figure 1. Locations of sampling sites at the Great Swamp NWR, NJ and the Croatan National Forest, NC. 2013 F.M. Panek and J.S. Weis 39 Materials and Methods Fish were collected with minnow traps, seines, and hand nets from several streams and swamp drainage ditches feeding into Black Swamp Creek within Croatan National Forest during 1977 and at Black Brook in the Great Swamp National Wildlife Refuge in 1978–1979. Fish were euthanized with a lethal dose of quinaldine sulfate, and the total length (mm) and the weight (g) were measured for each fish. Stomachs were excised and then fixed in 10% formalin in the field, and the contents later removed in the laboratory and stored in 70% ethanol for processing. All samples were processed and food items identified within six months of collection, and voucher specimens saved. Identifications of aquatic biota were made under a dissecting microscope and according to Pennak (1978), Thorp and Covich (1991), and Merritt and Cummins (1996). Food items were sorted and counted, and the volume visually estimated as a percentage of the total stomach contents (Westman 1941). Estimates of percent volume were made after sorting the contents on a gridded Petri plate and visually estimating the relative percentage of the total volume for each food group. Percent frequency of prey occurrence (%f), percent number (%η), and the percent composition by volume (%v) were determined by season. Faunal similarities in diet composition between sampling locations and seasons were compared using taxonomic presence and absence. These comparisons were made at the family or higher level of taxonomic organization, which we refer to herein as food groups. Family-level or higher organization helps minimize the effects of differing subsample size (Gerritsen et al. 2000) and has been found to be sufficient in broad-scale bioassessment programs (Hewlitt 2000). The same level of taxonomic identify was maintained between sampling locations to foster this simplified assessment. The faunal similarity in seasonal diets at each sampling location and between sampling sites was described by Jaccard’s index of similarity (Jaccard 1912): θJ = C / (N1+ N2 - C), where, θJ = faunal similarity, C = number of shared taxa, and N1 and N2 = number of taxa in assemblages 1 and 2, respectively. This index is most frequently used to compare species overlap between samples (Lydy et al. 2000) and treats each taxon with equal weight. Tables of statistical significance at P = 0.05 of lower and upper critical values (Real 1999) were used to assess significance in the observed values of faunal similarity. For these faunal comparisons we assumed that there were any possible distributions of taxa in assemblages N1 and N2 and that Jaccard’s index was dependent upon the number of taxa present in either of the two assemblages. Results Great Swamp National Wildlife Refuge A total of 147 Eastern Mudminnow collected averaged 62.0 mm TL (range = 35–112 mm), of which 96 had stomach contents (Table 1). The percent of 40 Northeastern Naturalist Vol. 20, No. 1 empty stomachs ranged from a low of 28.3% in the fall to a high of 40.7% during the spring. The principal stomach contents included detritus, dipterans, ostracods, coleopterans, and molluscs. The percent frequency of occurrence (%f), percent number (%η), and percent volume (%v) varied considerably by season (Table 1). Overall, the single dominant stomach content throughout the year was detritus. Detritus was found in 74.4% of the fall samples and in as few as 40% of the summer samples and ranged from 24.2% by volume in winter to 38.3% in the summer. Ostracods were the other major food throughout the year, occurring in 61.9% ± 2.5% of the stomachs with food and representing 30.7% ± 14.2% of the volume. During the winter months, 47 stomachs with food contained principally detritus, ostracods, and dipterans (Table 1). Dipterans were represented by several genera of chironomid larvae including Chironomus, Orthocladius, Pseudocladius, and Stictochironomus (Table 2). Collectively, dipterans occurred in 57.4% of stomachs and represented 73.9% of the number of food items and 31.2% of stomach volume (Table 1). Of these, Chironomus was dominant and occurred in 48.1% of samples (Table 2). Ostracods were present in 61.7% of stomachs and totaled 22.0% of stomach volume. They were numerous in the diet (24.7%) and were second only to dipterans in number. Other components of the winter diet included amphipods, copepods, coleopterans, odonates, and unidentified aquatic snails and pelecypods. Plant material in the form of unidentified filamentous algae and Lemna (duckweed) Table 1. Percent frequency of occurrence (%f), percent compositi on by number (%η), and percent composition by volume (%ν) of major food groups in the diet of the Eastern Mudminnow (Umbra pygmaea) from the Great Swamp National Wildlife Refuge, NJ. Winter (n = 68) Spring (n = 32) Summer (n = 30) Fall (n = 17) (Jan–March) (April–June) (July–Sept) (Oct–Dec) Dietary group %f %η %ν %f %η %ν %f %η %ν %f %η %ν Detritus 51.1 - 24.2 53.0 - 33.4 40.0 - 38.3 74.4 - 36.9 Amphipoda 10.6 0.1 4.2 3.1 0.1 3.1 3.3 0.1 3.0 6.9 1.4 2.6 Cladocera - - - 6.2 4.8 1.1 - - - 4.7 1.8 0.1 Copepoda 14.9 0.9 1.0 6.3 34.2 3.8 - - - 6.9 21.2 1.6 Ostracoda 61.7 24.7 22.0 65.6 50.4 28.2 60.0 97.4 50.6 60.5 53.8 19.5 Diptera 57.4 73.9 31.2 21.9 5.5 9.0 - - - 18.6 10.1 4.5 Coleoptera 19.2 0.1 10.6 15.6 0.8 6.4 3.3 0.1 3.0 27.9 5.3 17.9 Odonata 4.3 0.1 2.3 6.3 0.4 5.3 - - - 9.3 1.6 3.5 Trichoptera - - - 9.4 3.2 3.4 - - - - - - Nematoda - - - - - - 3.2 1.5 0.2 - - - Oligochaetea - - - 3.1 0.4 2.5 3.3 0.1 1.7 - - - Pelecypoda 6.4 0.1 0.9 3.1 0.1 1.3 - - - 11.6 2.4 6.3 Gastropoda 4.3 0.1 2.2 6.2 0.1 2.5 - - - 9.3 2.4 6.3 U. pygmaea - - - - - - 3.2 0.8 3.2 - - - Duckweed 4.3 na 1.2 - - - - - - 9.3 na 1.0 Filamentous algae 2.1 na 0.2 - - - - - - 9.3 na 0.6 % empty stomachs 30.9 40.7 37.5 28.3 2013 F.M. Panek and J.S. Weis 41 was also present in winter samples but was generally insignificant in volume (Table 1). For winter, we identified 10 taxonomic categories in the diet in addition to detritus. Taxonomic richness in the diet was greatest in the spring, with 11 identified taxonomic food groups in addition to detritus (Table 1). Ostracods were the most prevalent items in spring, and occurred in 65.6% of stomachs examined and represented 50.4% of the number of food items and 28.2% of the volume (Table 1). Dipterans represented by the chironomids Bezzia, Chironomus, and Tanytarus (Table 2) were the second most common food group, found in 21.9% of stomachs and representing 9.0% of the volume (Table 1). However, they were no longer numerically dominant, representing only 5.5% of the organisms consumed. Other food groups included amphipods, cladocerans, copepods, and a variety of insects including some coleopterans, odonates, and trichopterans. Of these groups, the coleopterans and the trichopterans occurred most often but were few in number (Table 1). In summer, the number of food groups in the diet was reduced and was clearly dominated by ostracods. These totaled 97.4% by number and 50.6% by volume and occurred in 60% of stomachs (Table 1). In total, we identified only six taxonomic food groups plus detritus, the fewest in any season at either study location. Dipterans, which were most common in the winter diet, were absent, as were many of the other insect groups. Coleopterans persisted, albeit as a minor dietary component, representing only 3.0% of food volume. Summer was the only period in which fish was a component of the diet, consisting of juvenile U. pygmaea, suggesting cannibalism. In summer, 37.5% of the stomachs were empty. The principal food groups in the fall were ostracods, coleopterans, and molluscs (Table 1). Ostracods were once again clearly the most common, occurring in 60.5% of the stomachs and totaling 19.5% of food volume. Second Table 2. Percent frequency of occurrence of chironomid larvae in the diet of Umbra pygmaea at the Great Swamp NWR, NJ and the Croatan National Forest, NC. Great Swamp NWR, NJ Croatan National Forest, NC Genus Winter Spring Summer Fall Winter Spring Summer Fall Ablabesmyia - - - - - 22.0 12.9 - Bezzia - 6.3 - 2.3 17.6 1.8 3.2 - Chironomus 48.1 3.1 - 11.6 5.9 16.6 9.7 7.1 Corynoneura - - - - - 1.8 - - Labrundinia - - - - - - 3.2 - Micropsectra - - - - - - 6.4 - Orthocladius 2.1 - - - 8.8 1.8 - 14.3 Parachironomus - - - - 2.9 - 6.4 7.1 Pentaneura - - - - - - 3.2 - Polypedilum - - - - - 3.7 - 7.1 Procladius - - - - - 1.8 - - Psectrocladius 2.1 - - - 14.7 5.6 - 21.4 Stictochironomus 2.1 - - - - - 3.2 7.1 Tanytarsus - 9.4 - - - - - - 42 Northeastern Naturalist Vol. 20, No. 1 was coleopteran larvae, which while present in the diet throughout the year, were most common in the fall (Table 1). Those few dipterans present occurred in 18.6% of stomachs and represented 4.5% of volume. Chironomus was the most prevalent dipteran, occurring in 11.6% of the stomachs (Table 2). Aquatic snails and fingernail clams (Pelecypoda) were also important in terms of volume (12.6%), with each occurring in 2.4% of the samples. Plant material in the form of duckweed and unidentified filamentous algae was also present, but was limited. Croatan National Forest Collections at Croatan yielded 190 Eastern Mudminnow averaging 50.2 mm TL (range = 22–89 mm), of which 133 had stomach contents. Empty stomachs ranged from a low of 10.5% in winter to a high of 38.0% during summer and averaged 25.6% ± 13.7% for the year (Table 3). The principal foods were dipterans, coleopterans, amphipods, and cladocerans. Detritus was also a major component and seasonally ranged from 25.8–47.2% of stomach volume and occurred in 71.4–88.2% of stomachs. Table 3. Percent frequency of occurrence (%f), percent composition by number (%η), and percent composition by volume (%ν) of major food groups in the diet of the Eastern Mudminnow (Umbra pygmaea) from the Croatan National Forest, NC. Winter (n = 38) Spring (n = 85) Summer (n = 50) Fall (n = 17) (Jan–March) (April–June) (July–Sept) (Oct–Dec) Dietary group %f %η %ν %f %η %ν %f %η %ν %f %η %ν Detritus 88.2 - 25.8 87.0 - 47.2 87.0 - 35.0 71.4 - 37.8 Amphipoda 14.7 0.9 10.4 9.3 2.1 4.2 9.7 3.0 5.7 21.4 4.7 16.8 Cladocera 23.5 79.6 9.3 22.0 26.4 4.2 9.6 20.7 0.5 7.1 0.9 0.4 Copepoda 14.1 0.7 1.8 9.2 4.6 2.2 6.5 6.1 1.1 14.3 14.2 2.4 Ostracoda 5.7 0.5 0.6 5.6 3.4 0.3 19.4 9.7 4.0 14.3 4.7 2.5 Diptera 52.9 15.3 23.5 55.6 44.1 13.2 87.1 37.1 9.1 71.4 62.0 10.0 Coleoptera 17.6 0.8 15.7 33.3 5.9 17.8 22.5 6.7 15.8 28.6 5.7 20.0 Odonata - - - 1.8 0.4 0.9 9.7 3.0 6.2 - - - Trichoptera - - - 7.4 3.8 2.9 - - - 7.1 0.9 5.7 Lepidoptera - - - 3.2 0.8 0.6 - - - - - - Ephemeroptera - - - - - - 9.7 3.0 6.2 - - - Collembola 2.9 0.1 0.1 - - - 3.2 0.7 1.0 - - - Heteroptera 11.7 1.0 7.1 7.4 3.4 4.8 6.4 1.5 2.3 - - - Hydrachnidae 5.9 0.2 1.0 1.8 0.4 0.1 - - - 21.4 4.7 1.2 Formicidae - - - 3.7 0.8 0.3 6.4 2.3 1.3 - - - Nematoda - - - - - - 3.2 1.5 0.2 - - - Oligochaetea 8.8 0.8 3.8 5.6 1.7 0.4 3.2 0.8 0.1 - - - Pelecypoda - - - - - - 3.2 0.8 1.8 - - - Gastropoda - - - - - - 6.4 1.5 5.9 - - - U. pygmaea - - - - - - 3.2 0.8 3.2 - - - Unidentified 0.1 0.9 0.3 1.5 - - - 2.2 3.3 % empty stomachs 10.5 36.5 38.0 17.6 2013 F.M. Panek and J.S. Weis 43 Food during winter was dominated by amphipods, cladocerans, dipterans, coleopterans, and larval heteropterans (Table 3). Collectively, they represented 66.0% of the volume of food. Amphipods were found in 14.7% of stomachs and represented 10.4% of volume. Cladocerans were found in 23.5% of stomachs and represented the majority by number (79.6%) of organisms from stomachs with food. Dipterans were dominated by the chironomids Bezzia and Psectrocladius (Table 2). Collectively, the dipterans were present in 52.9% of stomachs and totaled 23.5% of stomach volume. Larval coleopterans represented 15.7% of the volume and were found in 17.6% of stomachs (Table 3). In total, 10 taxonomic categories of food items in addition to detritus were observed in the winter. Detritus was a major component, occurring in 88.2% of the stomachs with food and representing 25.8% of the stomach volume (Table 3). Detritus continued to be a major component of the diet during spring, occurring in 87.0% of the stomachs and comprising 47.2% of the volume. Other major items by volume included coleopteran and dipteran larvae. These two food groups collectively represented 31.0% of the stomach contents by volume (Table 3). Cladocerans occurred in 22% of the stomachs and totaled 4.2% of the volume. Several genera of dipterans were identified (Table 2), of which Ablabesmyia was dominant, occurring in 22.0% of stomachs and representing 5.0% of volume. In total, 13 taxonomic categories of food items were identified in the spring diet. During summer, 17 taxa of foods were identified from 31 stomachs. Chironomid larvae were the most frequently consumed, occurring in 87.1% of stomachs with food, but their significance in terms of food volume (9.1%) was less than in previous seasons (Table 3). Eight genera were present (Table 2), with Ablabesmyia the most common and found in 12.9% of the stomachs. Coleopteran larvae were the second most frequent food, present in 22.5% of stomachs and representing 15.8% of volume. Mayfly nymphs (Ephemeroptera) occurred in 9.7% of stomachs and totaled 6.2% of the volume. Snails (Gastropoda) and fingernail clams (Pelecypoda) were both observed only in summer, and collectively occurred in 9.6% of stomachs and represented 7.7% of food by volume. As with our finding at Great Swamp, we also documented summer feeding on juvenile U. pygmaea (Table 3). Detritus was found in 87.0% of the stomachs with food and represented 35.0% of the food volume (Table 3). The principal food groups of the Eastern Mudminnow in the fall were aquatic coleopterans, amphipods, and dipterans, with these three groups comprising 46.8% of the food volume (Table 3). Larval coleopterans occurred in 28.6% of stomachs and, while numerically few (5.7%), they represented 20.0% of food volume. The dipterans, primarily represented by the chironomids Psectrocladius and Orthocladius (Table 2), were found in 71.4% of stomachs but only represented 10.0% of volume (Table 3). Consumption of amphipods was greater in the fall than during any of the previous seasons. Amphipods comprised 16.8% of food volume and were found in 21.4% of stomachs. Overall, we documented eight taxa of food groups in addition to the presence of detritus in the fall diet (Table 3). Detritus was found in 71.4% of the stomachs with food and totaled 37.8% of the volume (Table 3). 44 Northeastern Naturalist Vol. 20, No. 1 Faunal similarities in diet Similarities in the diet composition of the Eastern Mudminnow at Great Swamp and at Croatan were compared seasonally between sites (Table 4) and seasonally within sites (Table 5). Jaccard’s index (θJ) ranged from a low of 0.375 in winter to a high of 0.625 in spring for the between-site seasonal comparisons (Table 4). There were no statistically significant differences in diet composition between sites during the winter, summer, and fall. However, this was not the case during the spring, when the two diets were most dissimilar (θJ = 0.625, P = 0.05). When compared across all seasons, the diet of the Eastern Mudminnow was statistically different (θJ = 0.636, P = 0.05) between the two sampling locations. This result was due to the larger number of different food groups consumed during the spring. When examined seasonally, the diet of the Eastern Mudminnow at Great Swamp was significantly different in WI-SP, WI-FA, and SP-FA comparisons (Table 5). WI-SU, SP-SU, and FA-SU comparisons were not significantly different. During summer, the diet consisted largely of ostracods and detritus. These two foods comprised 88.9% of food volume (Table 1). Other taxa were few and mostly insignificant, resulting in considerable diet overlap in food items with the other seasons. At Croatan, the diet was more variable and seasonally exhibited a greater number of taxa (Table 3). Comparisons of seasonal differences in Jaccard’s index Table 5. Jaccard’s index of similarity in the diet of Umbra pygmaea between seasons at the Great Swamp NWR (GSNWR), NJ and at the Croatan National Forest (CNF), NC (WI = winter, SP = spring, SU = summer, FA = fall, ni = number of major taxa in assemblage I, nj = number of major taxa in assemblage J, Ci ● j = number of taxa in common between assemblage I and J and θ = Jaccard’s index of similarity) (*significantly different at P = 0.05). Location WI-SP WI-SU WI-FA SP-SU SP-FA SU-FA GSNWR nwi = 11 nwi = 11 nwi = 11 nwi = 12 nsp = 12 nsu = 7 nsp = 12 nsu = 7 nfa = 12 nsu = 7 nfa = 12 nfa = 12 Cwi●sp = 10 Cwi●sp = 4 Cwi●fa = 11 Cwi●su = 4 Csp●fa= 10 Csu●fa = 4 θ = 0.769* θ = 0.286 θ = 0.917* θ = 0.267 θ = 0.714* θ = 0.267 CNF nwi = 11 nwi = 11 nwi = 11 nwi = 14 nwi = 14 nwi = 17 nsp = 14 nsu = 17 nfa = 9 nsu = 17 nfa = 9 nfa = 9 Cwi●sp = 10 Cwi●su = 10 Cwi●fa = 8 Csp●su = 11 Csp●fa= 9 Csu●fa = 7 θ = 0.667* θ = 0.556* θ = 0.667* θ = 0.550* θ = 0.643* θ = 0.368 Table 4. Jaccard’s index of similarity in seasonal diet of Umbra pygmaea at the Great Swamp NWR (GSNWR), NJ and at the Croatan National Forest (CNF), NC, where C = number of taxa in common between assemblages and θJ = Jaccard’s Index (*significantly different at P = 0.05). Number of taxa observed in diet Location Winter Spring Summer Fall All seasons GSNWR 11 12 7 12 16 CNF 11 14 17 9 20 C 6 10 7 7 14 θJ 0.375 0.625* 0.412 0.500 0.636* 2013 F.M. Panek and J.S. Weis 45 (θJ) ranged from a low of 0.368 for the SU-FA comparison to 0.667 for the WI-SU and WI-FA comparisons. WI-SP, WI-SU, WI-FA, SP-SU, and SP-FA comparisons were significantly different, suggesting a high diversity of food items. In contrast to the diet at Great Swamp, the diet for the species at Croatan was most diverse in the summer (Table 3). Discussion Seasonal food habits of Eastern Mudminnow at both the Great Swamp NWR and Croatan National Forest indicate that the species is a bottom-feeding generalist highly dependent upon ostracods, chironomid larvae, coleopteran larvae, and other insects and crustaceans. This dietary description is generally what has been reported for the species from small numbers of specimens from various geographic locations (Jenkins and Burkhead 1994, Smith 1985). Declerck et al. (2002) examined the diet mass composition of the Eastern Mudminnow at the “De Maten” nature reserve in Gent, Belgium, where the species was introduced. They found that the diet mainly consisted of chironomid larvae, ephemeropterans, asellid isopods, odonates, and coleopteran larvae, which is generally consistent with our findings. Overall, the diet appears similar to that described for other Umbridae. The Central Mudminnow feeds primarily on aquatic insects, amphipods, and molluscs (Peckham 1955, Westman 1941) and it has been described as a dietary generalist feeder (Paszkowski 1985). A detailed quantitative food analysis of the Central Mudminnow was completed by Peckham and Dineen (1957) on a population in Judy Creek, IN. There, the Central Mudminnow fed on ostracods, copepods, chironomid larvae and pupae, and other aquatic insects. The diet of the Eastern Mudminnow is also similar to that of Umbra krameri Walbaum (European Mudminnow), which feeds mainly on small crustaceans and chironomid larvae as juveniles and mostly on benthic items such as amphipods, isopods, and snails as adults (Wanzenbock 1995). The wide variety of prey items of the Eastern Mudminnow indicates diet flexibility in varying environments. This characteristic was most pronounced at Black Swamp Creek in Croatan, where Jaccard’s index indicated statistically significant differences in diet among five of the six seasonal comparisons (Table 5). Seasonal dissimilarities were not as evident at Great Swamp, where diets were statistically different in only three of the six seasonal comparisons (Table 5). When compared across all seasons, the diet of the Eastern Mudminnow was statistically different (θJ = 0.636, P = 0.05) between the two sampling locations (Table 4), suggesting local adaptability to habitat and food item availability. There are some highly successful fishes, especially catostomids and cyprinids, which utilize benthic microplant material and detritus as primary foods (Ahlgren 1990). The role of amorphous detritus has been shown to be an important dietary component in juvenile Catostomus commersonii Lacépède (White Sucker) (Ahlgren 1990), Dorosoma cepedianum (Le Sueur) (Gizzard Shad) (Mundahl and Wissing 1988) and Fundulus heteroclitus (L.) (Mummichog) (Prinslow et al. 1974). Likewise, Johnson and Dropkin (1995) found stomachs from Pimephales 46 Northeastern Naturalist Vol. 20, No. 1 notatus (Rafinesque) (Bluntnose Minnow) in the Juniata River, PA to contain mostly detritus (75.5%). In our investigation, detritus was a major stomach content at both study locations throughout the year. For Great Swamp, detritus averaged 33.3% ± 6.3% by volume across all seasons and was found to be most prevalent during the summer when the Eastern Mudminnow fed largely on ostracods. Likewise, at Croatan, detritus represented 36.4% ± 8.8% by volume across all seasons. It is unlikely that this consistent and significant consumption of detritus was incidental to feeding on benthic organisms. However, detritus is less digestible and provides less protein than benthic invertebrates and is assimilated less efficiently (Bowen 1983). Detritus as a valuable food by the Eastern Mudminnow is a hypothesis that warrants additional investigation. We documented cannibalism at both sampling locations during the summer, when it occurred in approximately 3% of the fish collected at both locations. No other species of fish was prey. Cannibalism is well documented in 36 teleost families (Smith and Reay 1991) and has been observed in the closely related Esocidae (pikes and pickerels), where it may serve as a population-regulating mechanism (Persson et al. 2006). Our note of cannibalism in the feeding ecology of the Eastern Mudminnow warrants further study. Acknowledgments Sincere appreciation is expressed to the late James D. Anderson for his support and guidance during the early stages of this work and to Claire E. Ryan-Panek for her support with the fieldwork. The cooperation of the New Jersey Division of Fish and Wildlife, the US Fish and Wildlife Service, and the staffs at the Great Swamp NWR and Croatan National Forest is greatly appreciated. We also would like to thank Steve W. Ross and Fred C. Rohde for collections in Croatan National Forest. Several scientists and the editor made valuable comments and suggestions during manuscript review. This work was supported in part by a Grant-in-Aid of Research from Sigma Xi - The Scientific Research Society. Disclaimer: Use of trade, product, or firm names does not imply endorsement by the US Government. Literature Cited Ahlgren, M.O. 1990. Diet selection and the contribution of detritus to the diet of the juvenile White Sucker (Catostomus commersoni). Canadian Journal of Fisheries and Aquatic Science 47:41–48. Bowen, S.H. 1983. Detritivory in neotropical fish communities. Environmental Biology of Fishes 9(2):137–144. Christensen, N.L., R.B. Wilbur, and J.S. McLean. 1988. Soil-vegetation correlations in the pocosins of Croatan National Forest, North Carolina. Fish and Wildlife Service, Washington, DC, Biological Report 88(28), September 1988. 97 pp. Declerck, S., G. Louette, T. De Bie, and L. De Meester. 2002. Patterns of diet overlap between populations of non-indigenous and native fishes in shallow ponds. Journal of Fish Biology 61:1182–1197. Gerritsen, J., J. Burton, and M. Barbour. 2000. A stream-condition index for West Virginia wadeable streams. Tetra Tech, Inc., Owens Mills, MD. 80 pp. 2013 F.M. Panek and J.S. Weis 47 Hewlitt, R. 2000. Implications of taxonomic resolution and sample habitat for stream classification at a broad geographic scale. Journal of North American Benthological Society. 19(2):352–361. Jaccard, P. 1912. The distribution of the flora in the alpine zone. New Phytology 11:37–50. Jenkins, R.E., and N.M. Burkhead. 1994. Freshwater Fishes of Virginia. American Fisheries Society, Bethesda, MD. 1079 pp. Johnson, H.J., and D.S. Dropkin. 1995. Diel feeding chronology of six fish species in the Juniata River, Pennsylvania. Journal of Freshwater Ecology 10(1):11–18. Lee, D.S., C.R. Gilbert, C.H. Hocutt, R.E. Jenkins, D.E. McAlliser, and J.R. Stauffer. 1980. Atlas of the North American Freshwater Fishes. North Carolina State Museum of Natural History, Raleigh, NC. 867 pp. Lydy, M.J., C.G. Crawford, and J.W. Frey. 2000. A comparison of selected diversity, similarity, and biotic indices for detecting changes in benthic-invertebrate community structure and stream quality. Archives of Environmental Contamination and Toxicology 39:469–479. Merritt, R.W., and K.W. Cummins. 1996. Aquatic Insects of North America. 3rd Edition. Kendall/Hunt Publishing Company, Dubuque, IA. 862 pp. Mundahl, N.D., and T.E. Wissing. 1988. Selection and digestive efficiencies of Gizzard Shad feeding on natural detritus and two laboratory diets. Transactions of the American Fisheries Society 117:480–487. Panek, F.M., and J.S. Weis. 2012. Age, growth, and reproduction of the Eastern Mudminnow (Umbra pygmaea) at the Great Swamp National Wildlife Refuge, New Jersey. Northeastern Naturalist 19(2):217–228. Paszkowski, C.A. 1985. The foraging behavior of the Central Mudminnow and Yellow Perch: The influence of foraging site, intraspecific and interspecific competition. Oecologia 66:271–279. Peckham, R.S. 1955. Ecology and life history of the Central Mudminnow, Umbra limi (Kirtland). Ph.D. Dissertation. University of Notre Dame, South Bend, IN. 71 pp. Peckham, R.S., and C.F. Dineen 1957. Ecology of the Central Mudminnow, Umbra limi (Kirtland). American Midland Naturalist 58:222–231. Pennak, R.W. 1978. Fresh-water Invertebrates of the United States. The Ronald Press Company, New York, NY. 769 pp. Persson, L., A. Bertolo, and A.M. De Roos. 2006. Temporal stability in size distributions and growth rates of three Esox lucius (L.) populations. A result of cannibalism? Journal of Fish Biology 69(2):461–472. Prinslow, T.E., I. Valiela, and J.M. Teal. 1974. The effect of detritus and ration size on the growth of Fundulus heteroclitus (L.). Journal of Experimental Marine Biology and Ecology 16(1):1–10. Real, R. 1999. Tables of significant values of Jaccard’s index of similarity. Miscellania Zoologica 22(1):29–40. Rohde, F.C., G.H. Burgess, and G.W. Link, Jr. 1979. Freshwater fishes of the Croatan National Forest, North Carolina, with comments on the zoogeography of coastal plain fishes. Brimleyana 2:97–118. Ross, S.W., and F.C. Rohde. 2003. Life history of the swampfish from a North Carolina stream. Southeastern Naturalist. 2(1):105–120. 48 Northeastern Naturalist Vol. 20, No. 1 Smith, C.L. 1985. The inland fishes of New York State. New York State Department of Environmental Conservation. Albany, NY. Smith C., and P. Reay. 1991. Cannibalism in teleost fish. Reviews in Fish Biology and Fisheries 1:41–64. Thorp, J.H., and A.P. Covich. 1991. Ecology and Classification of North American Freshwater Invertebrates. Academic Press, San Diego, CA. 911 pp. Verreycken, H., C. Geeraerts, C. Duvivier, and C. Belpaire. 2010. Present status of the North American Umbra pygmaea (DeKay, 1842) (Eastern Mudminnow) in Flanders (Belgium) and in Europe. Aquatic Invasions 5(1):83–96. Wanzenbock, J. 1995. Current knowledge on the European Mudminnow, Umbra krameri Walbaum, 1792. Annalen des Naturhistorischen Museums in Wien 97(B):439–449. Westman, J.R. 1941. A consideration of population life-history studies in their relation to the problems of fish management research, with special reference to the Small-mouthed Bass, Micropterus dolomieu Lacepede, the Lake Trout, Cristivomer namaycush (Walbaum), and the Mudminnow, Umbra limi (Kirtland). Ph.D. Thesis. Cornell University, Ithaca, NY. 182 pp. Wilson, C.M., and P. Veillieux. 1982. Comparative osteology and relationships of the Umbridae (Pisces:Salmoniformes). Zoological Journal of the Linnean Society 76:21–352.