2007 NORTHEASTERN NATURALIST 14(3):447–460 Habitat Use of Etheostoma maculatum (Spotted Darter) in Elk River, West Virginia Elizabeth A. Osier1 and Stuart A. Welsh2,* Abstract - Etheostoma maculatum (Spotted Darter) has a disjunct distribution within the Ohio River drainage. Researchers have generalized Spotted Darter habitat as large rocks in swift riffles. In West Virginia, Spotted Darters are known to occur only in the middle section of the Elk River system. Information on habitat use is lacking. Through direct observation (snorkeling), we examined microhabitat use of Spotted Darters in riffle and glide habitats at three sites in the Elk River. Spotted darters in the Elk River were observed primarily in glide habitats near large rocks and in moderate current velocities. In the Elk River, this species is a benthic-habitat specialist, making it highly vulnerable to habitat alterations such as sedimentation and substrate embeddedness. Given its habitat use and restricted distribution, further ecological studies are needed for conservation and management of the Spotted Darter population in the Elk River. Introduction Etheostoma maculatum Kirtland (Spotted Darter) has a disjunct distribution within the Ohio River drainage (Page and Burr 1991). Isolated populations occur in the Allegheny River watershed in Pennsylvania and New York (Raney and Lachner 1939, Stauffer et al. 1996), the Scioto River watershed in Ohio (Trautman 1981), the Blue River (Baker et al. 1985) and Wabash River watersheds (Etnier 1980) in Indiana, and the Green River (Kessler and Thorp 1993, Kessler et al. 1995) and North Fork Kentucky River watersheds (Burr and Warren 1986) in Kentucky. Range fragmentation of the Spotted Darter mirrors that of other species, such as Erimystax dissimilis (Kirtland) (Streamline Chub), Etheostoma tippecanoe Jordan and Evermann (Tippecanoe Darter), Etheostoma camurum (Cope) (Bluebreast Darter), and Percina evides (Jordan and Copeland) (Gilt Darter). Simons (2004) attributed this pattern to recent degradation and fragmentation of habitat following post-Pleistocene dispersal. Because Spotted Darter populations are often small and isolated, state agencies have listed the Spotted Darter as threatened, endangered, or as “species of special concern.” In West Virginia, this species occurs only in the middle section of the Elk River system (Cincotta et al. 1986, Stauffer et al. 1995), where populations are known from ten sites (Osier 2005). The Spotted Darter inhabits riffle habitats with relatively fast watercurrent velocity and large substrate in the Allegheny (Raney and Lachner 1West Virginia Cooperative Fish and Wildlife Research Unit, Division of Forestry, West Virginia University, PO Box 6125, Morgantown, WV 26506. 2US Geological Survey, West Virginia Cooperative Fish and Wildlife Research Unit, West Virginia University, PO Box 6125, Morgantown, WV 26506. *Corresponding author - swelsh@wvu.edu. 448 Northeastern Naturalist Vol. 14, No. 3 1939, Stauffer et al. 1996), Scioto (Trautman 1981), Green (Kessler and Thorp 1993, Kessler et al. 1995), and Kentucky (Burr and Warren 1986) river systems. Individuals are frequently observed under or near boulders or large cobbles (Kessler and Thorp 1993, Raney and Lachner 1939). No information has been published on habitat use of the Spotted Darter in Elk River. Habitat use in species of darters is closely linked to body shape (Page and Swofford 1984). Morphological variation among populations may indicate differences in habitat use. The Spotted Darter population from Elk River is being described as a new species by one of the authors because it differs morphologically from populations in other drainages (S.A. Welsh, unpubl. data). Our objective was to quantify microhabitat use of the Spotted Darter in the Elk River, WV, and compare these data to previous observations focused on other populations. Methods Study site The Elk River, located in central West Virginia, flows west 290 km, descending 631 meters before entering the Kanawha River (Fig. 1). The middle section of Elk River, which supports a population of Spotted Darter, has relatively moderate gradients and high sinuosity. In this study, we examined microhabitat use of the Spotted Darter in glide and riffle habitats at three sites adjacent to the towns of Spread, Whetstone, and Ivydale. The watershed area above the Spread site is approximately 2566 km2. We define glide habitat as a nonturbulent, moderate-velocity, Figure 1. Location of sampling sites (•) within the Elk River, WV. 2007 E.A. Osier and S.A. Welsh 449 low-gradient macrohabitat with a wide channel and no thalweg that occurs at the transition between a pool and a riffle (Arend 1999, Flosi et al. 1998). The glide habitat is distinguished from run habitat (nonturbulent with definite thalweg) and riffle habitat (moderately turbulent with a straight-to-convex channel profile; Arend 1999, Flosi et al. 1998). At Spread and Whetstone, glide habitats were just upstream of the head of riffles; a glide paralleled the riffle habitat (separated by a narrow island) at Ivydale. At each site, the size of the study area was determined by the size (area) of glide or riffle habitat (Table 1). Habitat availability In riffle and glide habitats, we estimated the availability of several habitat resources (current velocity, water depth, and substrate composition; hereafter termed “habitat availability”). We randomly selected 30 locations per site from a grid of numbered 1-m2 cells. At each location, we measured mean water-current velocity (60% of depth, cm sec-1), bottom water-current velocity (2 cm above substrate, cm sec-1), water depth (cm), and substrate composition. Water-current velocity was measured with a flow meter (Flowmate Model 2000, Marsh-McBirney, Frederick, MD). To measure substrate composition, a grid of twenty-five 5- x 5-cm cells was centered over each location, and the dominant substrate-size class for each cell was recorded. Substrate size, measured across the longest axis, was classified as the average value of 10 ranges: 0.032 mm (silt, range: 0.004–0.06 mm silt); 1.0 mm (sand, range: >0.06–2 mm); 0.5 cm (range: >0.02–1 cm); 2 cm (range: >1–3 cm); 4 cm (range: >3–5 cm); 7.5 cm (range: >5–10 cm); 12.5 cm (range: >10–15 cm); 17.5 cm (range: >15–20 cm); 22.5 cm (range: >20–25 cm); and 30 cm (>25 cm). The mean of the 25 scores (from the twenty-five 5- x 5-cm cell grid) produced a substrate-size index for each location. Substrate heterogeneity at each location was estimated by the standard deviation of the mean of the 25 substrate values (Bain 1985). Habitat use Habitat-use data were obtained from underwater observations (snorkeling) within glide and riffle habitats at Spread (13 and 20 September 2002), Whetstone (11 August 2004), and Ivydale (1 September 2004). While snorkeling in an upstream direction during daylight hours (9:00–15:00 h), we marked Spotted Darter locations using numbered weighted tags. Darter locations were not marked if the presence of divers noticeably altered fish behavior. Mean water current velocity, bottom water current velocity, water depth, and substrate composition were measured at each fish location. Data analysis We explored within- and among-site patterns of habitat availability and microhabitat use with principal components analysis (PCA). Specifically, 450 Northeastern Naturalist Vol. 14, No. 3 Table 1. Mean values of available habitat (A) and habitat used (U) by Spotted Darter listed by site (standard errors in parentheses). Substrate heterogeneity was estimated at each location by the standard deviation of the mean of 25 substrate values. Site dimensions Used/ Sample Velocity at 2 cm Mean velocity Substrate Substrate LxW (m) available size Depth (cm) (cm sec-1) (cm sec-1) size (cm) heterogeneity Glide sites Spread 60 x 60 A 56 37.06 (2.92) 14.58 (1.36) 35.78 (2.42) 11.20 (0.55) 6.54 (0.38) U 36 47.75 (2.90) 16.38 (1.81) 45.67 (3.11) 15.99 (0.51) 8.35 (0.44) Ivydale 60 x 16 A 31 35.96 (2.81) 3.85 (0.83) 13.86 (1.36) 14.46 (0.99) 8.65 (0.41) U 32 33.29 (1.47) 3.40 (0.68) 16.41 (1.87) 15.93 (0.67) 9.52 (0.35) Whetstone 53 x 50 A 28 33.09 (3.11) 12.16 (1.52) 31.12 (2.42) 9.85 (0.49) 5.44 (0.35) U 26 49.12 (1.37) 12.98 (1.94) 39.68 (1.19) 16.23 (0.76) 8.22 (0.42) Riffle sites Ivydale 44 x 24 A 28 34.40 (1.29) 17.80 (1.40) 40.41 (2.14) 7.79 (0.30) 4.37 (0.30) U 10 37.18 (1.19) 4.30 (1.44) 12.89 (3.06) 12.38 (0.59) 6.57 (0.40) Whetstone 22 x 52 A 29 26.80 (2.56) 19.20 (2.13) 36.24 (3.09) 10.80 (0.06) 5.92 (0.41) U 7 31.78 (3.93) 19.20 (5.47) 51.51 (4.63) 18.17 (1.73) 11.01 (0.83) 2007 E.A. Osier and S.A. Welsh 451 we compared microhabitats within and among riffle and glide habitats. Additionally, we examined relationships between microhabitat availability and use within riffle and glide habitats at each site. Non-normal data were log (x+1) transformed prior to PCA. We used varimax rotation to enhance interpretation of principal components (McGarigal et al. 2000). We used PCA plots to compare (1) habitat availability between riffle and glide sites, among glide sites, and between riffle sites, and (2) between habitat availability and habitat use within each site. For each PCA ordination plot, we used MANOVA with PC1 and PC2 scores as independent variables to determine if clusters of groups were significantly (P < 0.05) different following methods used by Stauffer et al. (1996), Welsh and Perry (1998), and van Snik Gray and Stauffer (1999). An ANOVA and Tukey-Kramer test (for >2 groups) or student’s t-test (for 2 groups) were used to determine differences among means of PC scores along each PC axis, if clusters were significantly different along one axis independent of the second axis. Means and standard errors of habitat variables aided interpretation of PCA. All statistical tests were conducted with SAS software (version 8.01; SAS Institute, Inc. 1999). Results Habitat availability Riffle and glide habitats differed significantly (MANOVA, F(2, 168) = 18.92, P < 0.05; Fig. 2, Table 1). Water current velocities in riffles Figure 2. PCA-ordination diagram of habitat availability data from glide (º) and riffle (•) habitats. 452 Northeastern Naturalist Vol. 14, No. 3 exceeded those in glide habitats (PC1, Student’s t-test: t(159) = 5.19, P < 0.05), and substrate size and substrate heterogeneity in glide habitats exceeded those of riffles (PC2, Student’s t-test: t(147) = 4.16, P < 0.05). Glide habitats differed significantly among sites (MANOVA: F(4, 222) = 16.85, P < 0.05; Fig. 3, Table 1). Water current velocities at Whetstone and Spread were significantly faster than those at Ivydale (PC1, ANOVA: F(2, 111) = 22.89, P < 0.05). Substrate size and heterogeneity values at Ivydale were significantly larger than those at Whetstone and Spread (PC2, ANOVA: F(2, 111) = 11.67, P < 0.05); however, mean water depth was similar among glide habitats across sites (Table 1). For riffle habitats, water depth at Whetstone was significantly shallower than that of Ivydale (Student’s t-test: t(58) = 2.00, P < 0.05; Table 1). Substrate size and heterogeneity within the Whetstone riffle exceeded that of the Ivydale riffle (PC1, Student’s t-test: t(48) = 3.96, P < 0.05, Fig. 4). Habitat use At Ivydale and Whetstone, we observed more Spotted Darters in glide habitats (n = 32 and n = 26) than in riffle habitats (n = 10 and n = 7) . At Spread, Spotted Darters were not observed in the riffle, but were observed (n = 35) in the glide habitat. Spotted darters were associated with larger rocks in glide habitats than in riffle habitats (Student’s t-test, t(74) =1.99, P < 0.05), but significant differences did not occur for substrate heterogeneity and current velocity. Large rocks were an important component of Spotted Darter habitat (Table 1); out of 111 total observations, Figure 3. PCA-ordination diagram of habitat availability data from glides; Ivydale (º), Whetstone (•), and Spread (▲). 2007 E.A. Osier and S.A. Welsh 453 69 individuals were near or under rocks > 25 cm diameter, and 31 individuals were near or under rocks between 20 and 25 cm diameter. Average rock sizes used by Spotted Darters among sites ranged from 12.4 to 18.2 cm (Table 1). Within-site habitat use versus habitat availability For glide habitats at Spread and Whetstone, the cluster of principal components of habitat use differed significantly from that of habitat availability (MANOVAs: F(2, 89) = 13.27, P < 0.05, and F(2, 51) = 35.65, P < 0.05, respectively) where substrate heterogeneity and substrate size associated with Spotted Darters exceeded that typical of the site as a whole (PC1, Student’s t-tests: t(88) = 5.11, P < 0.05, and t(46) = 5.36, P < 0.05, respectively; Figs. 5 and 6). Spotted darters used deeper areas and faster velocities relative to the mean value of available habitat in the glide at Whetstone (PC2, Student’s t-test: t(36) = 3.96, P < 0.05; Fig. 6). For the glide habitat at Ivydale, clusters of habitat use and habitat availability did not differ significantly (MANOVA: F(2, 59) = 2.09, P > 0.05; Fig. 7). Univariate analyses supported this interpretation of the PCA; however, in addition to larger substrate sizes and higher substrate heterogeneity, mean water velocity significantly exceeded that available within glide habitats (Student’s t-test: t(214) = 1.97, P < 0.05; Table 1). For riffle habitats at Whetstone and Ivydale, habitat use differed significantly from habitat availability (MANOVAs: F(2, 33) = 24.34, P < 0.05, and F(2, 35) = 50.05, P < 0.05, respectively), substrate heterogeneity Figure 4. PCA-ordination diagram of habitat availability data from riffles; Ivydale (º), and Whetstone (•). 454 Northeastern Naturalist Vol. 14, No. 3 Figure 6. PCA-ordination diagram of habitat use (•) of the Spotted Darter versus habitat availability (º) from the glide habitat at Whetstone, WV. Figure 5. PCA-ordination diagram of habitat use (•) of the Spotted Darter versus habitat availability (º) from the glide habitat at Spread, WV. 2007 E.A. Osier and S.A. Welsh 455 Figure 7. PCA-ordination diagram of habitat use (•) of the Spotted Darter versus habitat availability (º) from the glide habitat at Ivydale, WV. Figure 8. PCA-ordination diagram of habitat use (•) of the Spotted Darter versus habitat availability (º) from the riffle habitat at Whetstone, WV. 456 Northeastern Naturalist Vol. 14, No. 3 and substrate size associated with Spotted Darters exceeded that of habitat means (PC1, Student’s t-test: t(33) = 12.59, P < 0.05; and PC2, Student’s ttest: t(23) = 3.94, P < 0.05, respectively; Figs. 8 and 9). Also, water current velocities associated with Spotted Darters within the Ivydale riffle were significantly slower than those characteristic of the riffle as a whole (PC1, Student’s t-test:, t(11) = 4.73, P < 0.05). Discussion Darters are adapted morphologically for a wide range of substrates and velocities (Page 1983, Page and Swofford 1984). Habitat use has been documented for many species (Chipps et al. 1994, Hlohowskyj and Wissing 1986, Matthews 1985, Welsh and Perry 1998), including the Spotted Darter (Kessler and Thorp 1993, Stauffer et al. 1996). Within glide habitats of the middle section of the Elk River, large unembedded substrate (>20 cm), and moderate water current velocities (13 to 51 cm sec-1) were important components of habitat for Spotted Darters. Few Spotted Darters were observed in riffle habitats of Elk River. This finding, however, may not be inconsistent with reports of Spotted Darter use of riffle habitat in other river systems (Burr and Warren 1986, Kessler and Thorp 1993, Stauffer et al. 1996, Trautman 1981) because of differences in how riffle habitat is defined. Further, our study occurred during periods of low river flows within the Figure 9. PCA-ordination diagram of habitat use (•) of the Spotted Darter versus habitat availability (º) from the riffle habitat at Ivydale, WV. 2007 E.A. Osier and S.A. Welsh 457 months of August and September, and habitat use is undocumented for other months. Glide habitats within our study area had lower bottom and mean water- current velocities, larger rock size, and higher substrate heterogeneity than riffle habitats. Use of slower velocity glide habitats by Spotted Darters in the Elk River may be associated with availability of larger rocks, where individuals avoid riffles with smaller substrate. The Spotted Darter is associated with large rocks (>20 cm), a finding consistent among our results and those of previous studies (Kessler and Thorp 1993, Stauffer et al. 1996). Average sizes of rocks used by this species in Elk River were significantly larger than those of available habitat, except for the glide at Ivydale, where large rocks were uniformly distributed. We quantified habitat during low flows of summer and early fall; however, large substrate is also of primary importance for nest sites and egg attachment during spring spawning (Raney and Lachner 1939). Although researchers have also documented swift riffles as primary habitat of the Spotted Darter (Baker et al. 1985, Burr and Warren 1986, Kessler et al. 1995, Raney and Lachner 1939, Stauffer et al. 1996), we found most individuals in glide habitat, and few in riffle habitat. However, it is important to emphasize that previous researchers may not have recognized glide habitat, and may have included glide habitat in their definition of riffle habitat. Suitable glide habitats within the middle section of the Elk River were primarily located in the transition between slow pool and swift riffle habitat. The mean water-current velocity of Spotted Darter locations in glide habitats was generally higher than that of available habitat. In the Elk River, Spotted Darters may associate with relatively high velocity areas of glide habitats, in part, because of an absence of silt. Riffle habitats at our Elk River study sites, however, are also absent of silt, but did not contain larger rocks as found at the glide habitats. Kessler and Thorp (1993) reported that Spotted Darters were not associated with silt-covered substrates, possibly because the substrates may also be used as spawning sites. Darters with specific habitat requirements are often more threatened by habitat alterations than are habitat generalists (Connelly et al. 1999, Mattingly and Galat 2002). Specialized habitat use of the Spotted Darter often surpasses that of coexisting species (Kessler and Thorp 1993, Stauffer et al. 1996), thereby allowing it to be more easily displaced than those species considered to be habitat generalists. Kessler and Thorp (1993) noted that sedimentation reduces the availability of “large, loose, rough substrate” used by the Spotted Darter, and clean cavities under large rocks are spawning sites for this species (Raney and Lachner 1939). Large rocks also likely act as velocity shelters and as a refuge from predation (Harding et al. 1998), as individuals are observed under large rocks outside of the spawning season (Kessler and Thorp 1993). Further studies should address relationships between stream sedimentation and the seasonal use of benthic fishes, such as the Spotted Darter. 458 Northeastern Naturalist Vol. 14, No. 3 Sedimentation associated with land-use practices reduces available habitat of benthic fishes because it fills interstitial spaces (Mattingly and Galat 2002). Within the Elk River watershed, sedimentation results from many sources, including logging, coal mining, and oil and gas extraction and may degrade Spotted Darter habitat. Management actions such as protection of riparian areas (Jones et al. 1999) or entire watersheds (Freeman and Freeman 1994) can reduce negative impacts to sensitive darter species. Substrate specificity of the Spotted Darter in the Elk River supported by our observational study not only imparts management and conservation implications, but also provides a baseline for further experimental studies of Spotted Darter habitat use. Acknowledgments We thank L. Hedrick, H. Hildebrand, Z. Liller, K. Sheehan, and D. Wellman for assistance with data collection. Financial support for this project was provided by the West Virginia Division of Natural Resources and the US Fish and Wildlife Service. Reference to trade names does not imply government endorsement of commercial products. Literature Cited Arend, K.K. 1999. Macrohabitat identification. Pp. 57–74, In M.B. Bain and N.J. Stevenson (Eds.). Aquatic Habitat Assessment: Common Methods. American Fisheries Society, Bethesda, MD. 216 pp. Bain, M.B. 1985. Quantifying stream substrate for habitat analysis studies. North American Journal of Fish Management 5:499–506. Baker, C., B. Forsyth, T. Wiles, and D.B. Abrell. 1985. Rediscovery of the Spotted Darter, Etheostoma maculatum, in Indiana waters: Blue River, Crawford, Harrison, and Washington Counties; Ohio River Drainage, USA. Indiana Academy of Sciences 94:603–605. Burr, B.M., and M.L. Warren, Jr. 1986. Distributional Atlas of Kentucky Fishes. Kentucky Nature Preserves Commission, Frankfort, KY. 398 pp. Chipps, S.R., W.B. Perry, and S.A. Perry. 1994. Pattern of microhabitat use among four species of darters in three Appalachian streams. American Midland Naturalist 131:175–180. Cincotta, D.A., R.L. Miles, M.E. Hoeft, and G.E. Lewis. 1986. Discovery of Noturus eleutherus, Noturus stigmosus, and Percina peltata in West Virginia, with discussions of other additions and records of fishes. Brimleyana 12:101–121. Connelly, W.J., D.J. Orth, and R.K. Smith. 1999. Habitat of the Riverweed Darter, Etheostoma podostemone Jordan, and the decline of riverweed, Podostemum ceratophyllum, in the tributaries of the Roanoke Rover, Virginia. Journal of Freshwater Ecology 14:93–102. Etnier, D.A. 1980. Etheostoma maculatum Kirtland Spotted Darter. P. 664, In D.S. Lee, C.R. Gilbert, C.H. Hocutt, R.E. Jenkins, D.E. McAllister, and J.R. Stauffer, Jr. (Eds.). Atlas of North American Freshwater Fishes. North Carolina State Museum of Natural History, Raleigh, NC. 867 pp. 2007 E.A. Osier and S.A. Welsh 459 Flosi, G., S. Downie, J. Hopelain, M. Bird, R. Coey, and B. Collins. 1998. California salmonid stream habitat restoration manual, third edition. Technical Report. California Department of Fish and Game, Sacramento, CA. Freeman, B.J., and M.C. Freeman. 1994. Habitat use by an endangered riverine fish and implications for species protection. Ecology of Freshwater Fish 3:49–58. Harding, J.M., A.J. Burky, and C.M. May. 1998. Habitat preference of the Rainbow Darter, Etheostoma caeruleum, with regard to microhabitat velocity shelters. Copeia 1998:988–997. Hlohowskyj, I., and T.E. Wissing. 1986. Substrate selection by Fantail (Etheostoma flabellare), Greenside (E. blennioides), and Rainbow (E. caeruleum) Darters. Ohio Journal of Science 86(3):124–129. Kessler, R.K., and J.H. Thorp. 1993. Microhabitat segregation of the threatened Spotted Darter (Etheostoma maculatum) and closely related Orangefin Darter (E. bellum). Canadian Journal of Fisheries and Aquatic Sciences 50:1084–1091. Kessler, R.K., A.F. Casper, G.K. Weddle. 1995. Temporal variation in microhabitat use and spatial relations in the benthic fish community of a stream. American Midland Naturalist 134:361–370. Jones, E.B.D., G.S. Helfman, J.O Harper, and P.V. Bolstad. 1999. Effects of riparian forest removal on fish assemblages in southern Appalachian streams. Conservation Biology 13:1454–1465. Matthews, W.J. 1985. Critical current speeds and microhabitats of the benthic fishes Percina roanoka and Etheostoma flabellare. Environmental Biology of Fishes 12:303–308. Mattingly, H.T., and D.L. Galat. 2002. Distributional patterns of the threatened Niangua Darter, Etheostoma nianguae, at three spatial scales, with implications for species conservation. Copeia 2002:573–585. McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate Statistics for Wildlife and Ecology Research. Springer-Verlag New York, NY. 283 pp. Osier, E.A. 2005. Habitat use and distribution of the Crystal Darter (Crystallaria asprella) and Spotted Darter (Etheostoma maculatum) in the Elk River, West Virginia. M.Sc. Thesis. West Virginia University, Morgantown, WV. Page, L.M. 1983. Handbook of Darters. TFH Publications, Neptune City, NJ. 271 pp. Page, L.M., and B.M. Burr. 1991. A Field Guide to Freshwater Fishes of North America North of Mexico. The Peterson Field Guide Series. Houghton Mifflin, Boston, MA. 432 pp. Page, L.M., and D.L. Swofford. 1984. Morphological correlates of ecological specialization in darters. Environmental Biology of Fishes 11:139–159. Raney, E.C., and E.A. Lachner. 1939. Observations on the life history of the Spotted Darter, Poecilichthys maculatus (Kirtland). Copeia 1939:157–165. SAS Institute, Inc. 1999. SAS/STAT User’s Guide, Version 8.01 SAS Institute, Inc. Cary, NC. Simons, A.M. 2004. Phylogenetic relationships in the genus Erimystax (Actinopterygii: Cyprinidae) based on the cytochrome-b gene. Copeia 2004:351–356. Stauffer, J.R., Jr., J.M. Boltz, and L.R. White. 1995. The fishes of West Virginia. Proceedings of the Academy of Natural Sciences, Philadelphia 146:1–389. 460 Northeastern Naturalist Vol. 14, No. 3 Stauffer, J.R., Jr., J.M. Boltz, K.A. Kellogg, and E.S. van Snik. 1996. Microhabitat partitioning in a diverse assemblage of darters in the Allegheny River system. Environmental Biology of Fishes 46:37–44. Trautman, M.B. 1981. The Fishes of Ohio. Ohio State University Press, Columbus, OH. 782 pp. van Snik Gray, E., and J.R. Stauffer, Jr. 1999. Comparative microhabitat use of ecologically similar benthic fishes. Environmental Biology of Fishes 56:443–453. Welsh, S.A., and S.A. Perry. 1998. Habitat partitioning in a community of darters in the Elk River, West Virginia. Environmental Biology of Fishes 51:411–419.