2010 NORTHEASTERN NATURALIST 17(3):387–414 Freshwater Mussel (Unionidae: Bivalvia) Distributions and Densities in French Creek, Pennsylvania Tamara A. Smith1,2 and Darran Crabtree3,* Abstract - The freshwater mussel (Unionidae: Bivalvia) communities of the French Creek watershed are nationally recognized for their importance to biodiversity. The goal of this research was to gather more information on the distribution and densities of these species throughout the river. This study utilized two-phase sampling with timed searches to characterize mussel species richness in a large number of sites and to calculate catch-per-unit-effort values. The results of the timed search were used to select a subset of sites for quantitative mussel surveys to estimate density and abundance. Starting in New York, the main stem of French Creek was divided into 5.6-km lengths, and one site was randomly chosen within each of those lengths, favoring habitat (large riffle/runs) for rare species and high species diversity. Snorkelers collected as many unionid individuals as possible, with a target search rate of 0.5 m2 min-1. Thirty-two main-stem sites were surveyed with timed searches. Mean species richness was 11.8 (SE = 0.94), and mean CPUE was 59.5 mussels per person-hour (SE = 9.32). Quantitative sampling was performed at ten sites using a double sampling design. Mean density estimates ranged from 0 to 27.98 m-2. Abundance estimates range from 0 to 69,848 live mussels per site. For the main stem, we calculated regression models to estimate densities and abundances at qualitatively sampled sites based on the CPUE at quantitatively sampled sites. Extrapolation yields approximately 22 million mussels in the 2.04 million m2 of large riffle-run habitat in the main stem of French Creek. Introduction French Creek, originating in western New York and flowing 188 km to its confluence with the Allegheny River at Franklin, PA, is perhaps the most ecologically significant waterway in either state, containing more species of fish and freshwater mussels (Unionidae: Bivalvia) than any other similarsized stream in the northeast United States. Over 80 native species of fish and 29 native species of freshwater mussels have been documented in the watershed along with various other wildlife and plant species (Bier 1994, Ortmann 1919, WPC and FCP 2002). Two of the mussels found in French Creek are presently listed as Endangered under the US Endangered Species Act and the Pennsylvania Fish Code: Epioblasma torulosa rangiana (Northern Riffleshell) and Pleurobema clava (Clubshell). Villosa fabalis (Rayed Bean), is a candidate for federal listing and is designated as endangered in Pennsylvania (US Fish and Wildlife Service 1991). Epioblasma triquetra (Rafinesque) (Snuffbox), Fusconaia subrotunda (Longsolid), and Quadrula 1Pennsylvania Natural Heritage Program, Western Pennsylvania Conservancy, 11881 Valley Road, Union City, PA 16438. 2Current address - US Fish and Wildlife Service, Twin Cities Field Office, 4101 American Boulevard East, Bloomington, MN 55425. 3The Nature Conservancy, Allegheny College, Meadville, PA 16335. *Corresponding author - dcrabtree@tnc.org. 388 Northeastern Naturalist Vol. 17, No. 3 cylindrica (Rabbitsfoot) are all designated as endangered in Pennsylvania (PA Natural Heritage status rank) and have global status of either imperiled or vulnerable. Thirteen other mussel species found in French Creek are considered rare, threatened, endangered, or possibly extirpated in Pennsylvania (Table 1; NatureServe 2008). Threats to the mussel population in the watershed include siltation and pollution due to improper agriculture and timbering practices, mineral extraction, water extraction, development, and wastewater treatment plants (see the review in Western Pennsylvania Conservancy and French Creek Project 2002). Other threats to the viability of freshwater mussels include dams and stream channel alteration (see the review in Watters 2000), and invasive species such as Dreissena polymorpha Pallas (Zebra Mussel) (Biggins et al. 1995, Ricciardi et al. 1998, Strayer and Malcom 2007). Our study was designed to yield spatially explicit information on the densities and abundances of imperiled freshwater mussel species on a river-wide Table 1. Global and Pennsylvania State ranks for each unionid species historically found in French Creek. Key to global ranks: G5 = secure, G4 = apparently secure, G3 = vulnerable, G2 = imperiled, G1 = critically imperiled, Q = questionable taxonomy, T = Infraspecific taxon (subspecies). Key to state ranks: S5 = secure, S4 = apparently secure, S3 = vulnerable, S2 = imperiled, S1 = critically imperiled, SNR = not ranked, SX = Presumed extirpated. Ranks according to NatureServe as of October 21, 2008. Species Global rank PA State rank Actinonaias ligamentina (Lamarck) G5 S4 Alasmidonta marginata (Say) G4 S4 Amblema plicata (Say) G5 S2 S3 Anodontoides ferussacianus (Lea) G5 S2 S3 Cyclonaias tuberculata (Rafinesque) G5 SX Elliptio dilatata (Rafinesque) G5 S4 Epioblasma torulosa rangiana (Lea)1 G2 T2 S2 Epioblasma triquetra (Rafinesque) G3 S1 Fusconaia subrotunda (Lea) G3 S1 Lampsilis cardium (Rafinesque) G5 S4 Lampsilis fasciola (Rafinesque) G5 S4 Lampsilis ovata (Say) G5 S3 S4 Lampsilis siliquoidea (Barnes) G5 S4 Lasmigona complanta (Barnes) G5 S1 Lasmigona compressa (Lea) G5 S2 S3 Lasmigona costata (Rafinesque) G5 S4 Ligumia nasuta (Say) G4 S1 S3 Ligumia recta (Lamarck) G5 S3 S4 Pleurobema clava (Lamarck)1 G2 S1 S2 Pleurobema sintoxia (Rafinesque) G4 G5 S2 Ptychobranchus fasciolaris (Rafinesque) G4 G5 S4 Pyganodon grandis (Say) G5 S4 Quadrula cylindrica cylindrica (Say) G3 G4 T3 S1 Simpsonaias ambigua (Say) G3 S1 Strophitus undulatus (Say) G5 S4 S5 Toxolasma parvus (Barnes) G5 S1 S3 Utterbackia imbecillis (Say) G5 S3 S4 Villosa fabalis (Lea)2 G2 S1 S2 Villosa iris (Lea)1 G5 Q S1 1Federally endangered species. 2Candidtate for federal listing. 2010 T.A. Smith and D. Crabtree 389 basis. Ultimately, we used this information combined with size distribution as a measure of the overall health of the imperiled mussels. We believe evidence of recent recruitment, as determined from size distributions, is an indicator of population viability. However, both abundance estimates and subsequently derived size distributions may be dependent on the method of sampling. There are several methods, study designs, and types of sampling gear that can be implemented to estimate population sizes of mussels (see the review in Strayer and Smith 2003). Comparative studies have been conducted to evaluate quantitative vs. qualitative sampling methods and their effectiveness in assessing population density and detecting recruitment (Hornbach and Deneka 1996; Miller and Payne 1988, 1993; Obermeyer 1998; Strayer et al. 1997; Vaughn et al. 1997; Villella et al. 2004) with mixed results, and new methods to detect rare mussels continue to be developed (e.g., Pooler and Smith 2005, Smith 2006). This study utilized two methods to estimate mussel numbers and sizes: timed searches and double sampling (based on Smith et al. 2001). Timed searches were used to characterize mussel species richness in a given area and to calculate catch-per-unit-effort (CPUE) values. Quantitative double sampling surveys were then utilized to estimate densities and abundances. Through this study, we hoped to determine if it is possible to estimate the population size of mussel beds based on CPUE, if we could estimate population size for the riffle-run habitat in the whole river, and which survey method is better at detecting recent recruitment. Field-site Description French Creek is part of the Allegheny River watershed and the greater Ohio River drainage. The entire French Creek watershed covers an area of approximately 3200 km2. Approximately 93% of the watershed is within Pennsylvania, and the remaining 7% is made up of headwater streams in New York. French Creek originates in Chautauqua County in New York State, then flows south through Pennsylvania’s Erie and Crawford counties, through the northeast corner of Mercer County, and finally into Venango County where it flows southeast to its confluence with the Allegheny River at Franklin, PA (Fig. 1). Methods In-stream mapping The main stem of French Creek was mapped using a Trimble GeoExplorer GPS unit. Stream reaches were categorized into one of 3 flow regimes—pool, run, riffle—or a combination of these regimes, based on Hankin and Reeves (1988). Substrate types, defined by particle sizes in a modified Wentworth scale, were visually estimated and recorded as percent area of each substrate type (silt, sand, gravel, cobble, boulder, bedrock, and organic debris) in each reach. Gravel-sized substrate in riffle-and-run flow regimes make up what is believed to be essential habitat to many freshwater mussels (Butler 2003, 2006; Ortmann 1919; Parmalee and Bogan 1998; US Fish and Wildlife Service 1994) and fish (Etnier and Starnes 1993, Jenkins and Burkhead 1994) of special 390 Northeastern Naturalist Vol. 17, No. 3 concern in this river. We used ArcGIS to calculate the total area of large (>100 m in length) riffle-run habitats. Approximately 2.18 million m2 of the main stem was classified as “large” riffle-run habitat. Sensitivity buffer Because of the sensitivity of the species highlighted in this manuscript, it was decided to buffer exact site locations (C. Copeyon, US Fish and Wildlife Service, State College, PA, pers. comm.; C. Urban and N. Welte, Pennsylvania Fish and Boat Commission, Bellefonte, PA, pers. comm.). Sites are represented as whole number river kilometers (RKM), measured upstream Figure 1. Locations of survey sites on French Creek, PA. 2010 T.A. Smith and D. Crabtree 391 from the confluence of French Creek with the Allegheny River. River kilometers were rounded to the nearest whole number. So, a site listed as 23 RKM, for example, can be found anywhere between RKM 22.5 and RKM 23.5. Timed searches Given our time and funds, we estimated we would be able to survey approximately 25 sites during the first field season. Starting at the Pennsylvania/ New York border, the main stem of French Creek was divided into twenty-five sections approximately 5.6 km in length. One site was randomly chosen within each section by numbering all the large riffle-run habitats (<100 m in length) within each section and using a random number generator to select sites. If no riffle-run habitat was found within a 5.6-km section, then pool habitat was chosen in a similar manner for that section. Three additional sites were added to the survey scheme by randomly selecting three 5.6-km sections and then randomly choosing an additional site within those sections (RKMs 74, 125, and 142). Twenty-four riffle-run and 3 pool sites were surveyed using timed searches in 2003. Two sites (RKMs 1 and 29) that were chosen for 2003 sampling were not actually surveyed until 2005, due to high water. Because of funding constraints, sites in the New York portion of French Creek were not surveyed until 2005, and were selected in a similar manner as the Pennsylvania sites. In total, we conducted timed-search surveys at 29 riffle-run sites and 3 pool sites. All timed surveys occurred between late June and late August in their respective years. Snorkelers collected as many unionid individuals as possible in equal areas for a specified amount of time. Since downstream sites were wider than upstream sites, we standardized the search area to 2500 m2 at each site, surveying from bank to bank. The total search time at each site was 300 minutes, which was divided equally among observers. SCUBA was used at the 3 pool sites, as water depths there exceeded 1.5 m. Due to technical difficulties, one pool site (RKM 76) was only sampled for a total of 1 person-hour (p-h). Sampling started at the downstream end of the study section, and observers moved in an upstream, zigzag direction in equally sized transects (cells), covering the entire stream width. Observers used a combination of tactile and visual methods, with a target search rate of 0.5 m2 min-1. Although most of the mussels collected were visible at the surface, observers periodically brushed away sediment, flipped over non-embedded rocks, and did some light raking during each search. The effective sampling fraction, or the percentage of area searched thoroughly, was assumed to be 0.06 for all timed searches, which was similar to that used in Smith et al. (2001). The effective sampling fraction is a basis for standardizing timed searches. Search time (ST) was calculated by: ST = (effective sampling fraction x survey area/target search rate), or ST = (0.06 x survey area /0.5 m2 min-1). Live mussels were identified, counted, measured, and returned to the substrate in their cell of origin. CPUE was calculated as number of unionid individuals collected divided by p-h spent sampling. 392 Northeastern Naturalist Vol. 17, No. 3 Quantitative surveys Location and number of quantitative sampling sites were determined from the initial 24 riffle-run sites surveyed in 2003. To gain a more precise river-wide population estimate than simple random sampling, and to allocate more effort to sites of interest (i.e., high-density areas and rare species occurrences), principles of stratification were utilized to partition sites into strata (Villella and Smith 2005). Sites were separated into two strata based on our surface counts from our timed searches: high CPUE (≥50 mussels/p-h) and low CPUE (<50 mussels/p-h, which is slightly higher than similar studies in West Virginia; Villella and Smith 2005). To allocate sampling effort to spend more time at high-density sites, our goal was to sample 60–80% of our high- CPUE sites and 20–40% of our low-CPUE sites (Villella and Smith 2005). Sampling 60–80 % of our 16 high-CPUE sites gave 9–12 sites to be sampled quantitatively, and sampling 20–40% of the twelve low-CPUE sites gave 2–5 sites to be sampled quantitatively. To allow time for stream recovery, quantitative sampling occurred in the summer of 2004. Because of an unusually wet summer, we were only able to quantitatively sample seven randomly selected high-CPUE sites and 3 randomly selected low-CPUE sites. Quantitative sampling was performed using protocol developed by Smith et al. (2001). We used a double-sampling design, systematically placed with multiple random starts and 0.25-m2 quadrats. Replicate samples (quadrats) in each site are needed to get a desired level of precision. The total number of quadrats sampled depends on mussel density, variance in density, the desired precision, and resources. Generally, the lower the density, the more samples needed to attain the ideal precision. Based on studies in the Allegheny River (Smith et al. 2001, Villella and Smith 2005), we set our goals at attaining a coefficient of variation (CV) of 0.30 as our desired level of precision. Previous mussel data from two locations on French Creek provide a wide range of densities (0.006–2.327 mussels m-2 per species) depending on species and site. Mean total densities for these two sites were roughly 0.38 and 0.19 mussels m-2 (Environmental Science 2002a, b). We consulted Smith et al. (2001) to determine sample size, given these densities and desired coefficient of variation. For densities as low as 0.006 m-2, our sample sizes would have to be well over 700 to attain a CV of 0.30. More liberal density estimates of 0.2 mussels m-2 allow a sample size of 400 to get a CV of 0.33. Given our limited resources and concern with disturbing mussel resources, we sampled approximately 400 quadrats per site. To get a sample size of 400 quadrats within our average study area (2500 m2), we used transects equally spaced parallel to shore and divided them into 4- x 4-m cells (Fig. 2). Three pairs of random numbers were generated between 0 and 4 m (across), and 0 and 4 m (upstream), to use as coordinates for three quadrat samples within each cell. We used a double-sampling design, such that excavation to a depth of 10–15 cm occurred at a random subset of the sites. Double sampling increased sampling precision of density estimates by using total catch from a random subset of quadrats to calibrate the surface counts for all quadrats. The protocol used to determine the proportion of excavated quadrats followed Smith et al. (2001). To determine the proportion of quadrats to excavate, we excavated a 2010 T.A. Smith and D. Crabtree 393 random sub-sample of the total quadrats (5%, i.e., 20 quadrats), keeping the surface catch separate from the buried catch for each quadrat. Buried mussels were defined as those mussels that cannot be detected at the surface after lightly brushing away surface silt and/or loose rocks. The number of exposed vs. buried mussels were calculated to determine the excavation intensity for the remaining quadrats: if >60% are exposed, then excavate 25%; if 50–60%, then excavate 33%; if 40–50%, then excavate 50%; and excavate 100% of the quadrats if less than 40% of the mussels are detected at the surface. Material removed during excavation was placed in a 0.63-cm mesh sieve and inspected for mussels. After excavation, the quadrat was visually examined for any remaining mussels. A regression estimator based on the relationship between surface counts and total counts on the excavated quadrats was used to calibrate the surface counts on the remaining quadrats to arrive at total density for each sample (US Geological Survey Mussel Estimation Program, Version 1.4.3). All mussels were placed in an underwater mesh bag, identified, measured with vernier calipers (total length to the nearest 0.1 mm), and returned to the substrate in the original quadrat. Linear regression was used to evaluate the relationship between the surface count data from the times searches and the density estimates from quantitative sampling (Villella and Smith 2005). All quantitative surveys took place between June 16 and October 11, 2004. We utilized our length data as simple means to investigate recruitment by choosing reasonable length cut-offs to get rudimentary numbers of recruits. We Figure 2. Systematic sampling design illustration. The close-up on right depicts an example of 3 pairs of random start points between 0 and 4 m: (0.5 m, 3.0 m), (2.0 m, 2.0 m), and (2.5 m, 1.0 m), which were used as coordinates for three quadrat samples within each 4 m by 4 m cell. The remaining quadrats were placed at standardized distances (4 m) upstream from the 3 start points. Illustration modified from Environmental Science, Inc. (2002a). 394 Northeastern Naturalist Vol. 17, No. 3 chose this simple method, rather than other methods such as examining external annuli, for its ease of implementation and consistency in the field. For most species, we defined recent recruits as individuals less than or equal to 30 mm in total length, which is the approximate size used in other studies (i.e., Mohler et al. 2006, Obermeyer 1998). We used a cut-off of 20 mm for 2 naturally smaller species, Rayed Bean and Snuffbox (Cummings and Mayer 1992). Caution should be noted when interpreting sex ratios for some species in studies where the sex of a large proportion of individuals was not recorded. Results Timed searches Total number, percent relative abundance, number of sites, and CPUE for timed searches are given in Tables 2a and 2b. A total of 8710 individual live mussels representing 24 species were found at the 32 main-stem sites sampled using timed searches. Five-hundred and fifteen fresh-dead and 408 weathereddead shells were also found; however, no additional species were detected as only dead shells. Actinonaias ligamentina (Mucket) was the most abundant and widely distributed species, found at 29 sites and accounting for about 42.5% of the total number of mussels found in our timed searches. The second-most abundant species was Ptychobranchus fasciolaris (Kidneyshell), which was found at 28 sites and had a relative abundance of 13.4%. The three next-most abundant species were Elliptio dilatata (Spike) (8.2%, 25 sites), Lasmigona costata (Flutedshell) (5.9%, 26 sites), and Rayed Bean (5.1%, 19 sites). Northern Riffleshell was found at 11 sites and had a relative abundance of 3.2%. Clubshell was found at 5 sites, and had a relative abundance of approximately 0.2%. Total live mussels ranged from 0 to 946 live mussels at each site. The mean number of mussels found per site was 272.3 (SE = 44.55). CPUE ranged from 0–189 mussels/p-h. The mean CPUE was 58.0 mussels/p-h (SE = 9.13). CPUE was relatively low in the upper reaches except for RKM 146, which had a CPUE of 148.8 mussels/p-h, and RKM 125, with a CPUE of 102.8 mussels/p-h (Fig. 3). CPUE was highest at RKM 74, with a CPUE of 189.2 mussels/p-h. There was no significant linear trend between RKM and CPUE (r2 = 0.134, Fstatistic = 0.4073, df = 30, P = 0.5281). Species richness ranged from 0 to19 species per site, with a mean of 11.3 (SE = 0.94) across all sites. Although the results from a simple linear regression give us no significant linear relationship between RKM and species richness (r2 = 0.051, F-statistic= 1.619, df = 30, P = 0.213), generally, species richness was low in the upper river, sharply rising to 17–19 species between RKM 109–74 (with the exception of the 2 pool sites at RKM 82 and 76), then dropping gradually to 6–9 species in the lower reaches. Evidence of recent recruitment was found at 26 study sites. Out of the 8710 live individuals found in timed searches, only 3.05% were considered recent recruits. Table 3 gives the numbers of recruits for each species, including length data and the number of sites with recruits for each species. Female and males of all sexually dimorphic species were found (Table 3). 2010 T.A. Smith and D. Crabtree 395 Quantitative surveys Three low-CPUE sites and 7 high-CPUE sites were surveyed using quantitative methods in 2004. Overall, we documented 12,733 live, 216 fresh-dead, and 3220 weathered-dead individuals representing 23 unionid species. Two species found during timed searches, Utterbacki imbecillis (Paper Pondshell) and Villosa iris (Rainbow Mussel), were not found in quantitative surveys. Lasmigona complanata (White Heelsplitter) was found in quantitative searches but not in timed searches. The Mucket was the most abundant and widely distributed species, found at 9 of the 10 quantitative sites and accounting for about 43.8% of the total number of mussels found; its highest density was at RKM 74 (estimated density = 11.3 mussels /m-2, SE = 0.41; Table 4). The highest density of Northern Riffleshell was also at RKM 74 (6.65 mussels m-2, SE = 0.37). Although we did not find any Zebra Mussels in our timed searches, a total of 10 live Zebra Mussels were found among 5 sites (RKMs 74, 68, 52, 23, and 19) during our quantitative surveys. Zebra Mussels ranged in size from 14.5–34 mm in length, with a mean of 20.7 mm (SE = 1.68). Out of the 12,733 live unionid individuals found in quadrat surveys, 10.52% were recruits, and evidence of recruitment was observed at all 9 quantitative sites where mussels were found. Table 5 gives the numbers of recruits for each species, including length data and the number of sites with recruits for each species. Female and males were found for all sexually dimorphic species (Table 5). Paired t-tests were used to test if there was a significant difference between methods in detecting recruitment at sites where both quadrat and timed searches Figure 3. Percentage of individuals of each unionid species found in surface searches vs. subsurface (excavation) in the quantitative surveys. Numeric labels on each bar indicate the actual numbers found. 396 Northeastern Naturalist Vol. 17, No. 3 Table 2a. Numbers and relative abundance of each species found at each site during timed searches in French Creek. Also given is the CPUE, total abundances, and species richness per site. Zeros represent species found only as dead shells at a particular site. Sites given as river kilometers (RKM) measured upstream from the mouth. RKM Species 157 146 142 140 138 132 127 125 123 115 109 107 98 93 89 82 Actinonaias ligamentina (Mucket) 1 205 1 1 18 100 20 14 93 35 383 170 155 4 Alasmidonta marginata (Elktoe) 4 1 1 11 8 16 24 27 21 8 17 Amblema plicata (Three Ridge) 1 1 1 2 2 17 Anodontoides ferussacianus (Cylindrical Papershell) 7 19 36 1 2 2 Elliptio dilatata (Spike) 175 7 12 55 11 1 1 2 31 15 26 Epioblasma torulosa rangiana (Northern Riffleshell) 0 0 4 E. triquetra (Snuffbox) 2 20 0 0 1 1 0 3 3 1 Fusconaia subrotunda (Longsolid) 2 3 14 6 12 Lampsilis cardium (Plain Pocketbook) 9 2 3 2 3 2 1 2 6 1 L. fasciola (Wavy-rayed Lampmussel) 1 1 1 5 21 3 3 L. ovata (Pocketbook) 1 2 3 4 15 9 9 13 66 L. siliquoidea (Fatmucket) 2 55 9 0 1 65 79 101 4 4 5 1 2 2 Lasmigona compressa (Creek Heelsplitter) 3 1 1 1 1 L. costata (Flutedshell) 2 79 21 8 29 7 4 27 6 10 1 14 Ligumia recta (Black Sandshell) 2 5 2 0 0 Pleurobema clava (Clubshell) 3 3 3 1 3 P. sintoxia (Round Pigtoe) 35 0 1 12 5 0 4 1 17 5 12 1 Ptychobranchus fasciolaris (Kidneyshell) 71 31 1 19 202 66 20 70 33 70 34 58 3 Pyganodon grandis (Giant Floater) 0 10 1 1 2 0 0 Quadrula cylindrica (Rabbitsfoot) 2 3 1 2 6 Strophitus undulatus (Creeper) 13 88 90 1 2 2 0 7 4 7 Utterbackia imbecillis (Paper Pondshell) Villosa fabalis (Rayed Bean) 7 3 20 17 8 8 V. iris (Rainbow Mussel) Unidentified 1 1 Total number 25 744 199 1 0 15 131 515 221 76 283 140 595 286 421 18 CPUE (#/p-h) 5.0 148.8 39.8 0.2 0.0 3.0 26.2 102.8 44.2 15.2 56.6 27.8 119.2 57.2 84.2 3.6 Species richness 5 12 10 1 0 5 11 12 9 12 19 18 17 17 19 6 2010 T.A. Smith and D. Crabtree 397 Table 2b. Numbers and relative abundance of each species found at each site during timed searches in French Creek. Also given is the CPUE, total abundances, and species richness per site. Zeros represent species found only as dead shells at a particular site. Sites given as river kilometers (RKM) measured upstream from the mouth. RKM Species 76 74 72 68 63 58 52 44 41 35 29 23 19 11 5 1 Actinonaias ligamentina 479 231 193 118 23 102 221 270 10 35 218 418 173 6 3 Alasmidonta marginata 35 18 16 10 2 20 36 3 1 23 20 3 0 1 Amblema plicata 0 117 22 0 3 4 6 Anodontoides ferussacianus 3 1 1 1 Elliptio dilatata 68 13 45 41 11 18 6 7 1 44 36 80 5 5 Epioblasma torulosa rangiana 184 11 14 4 0 1 0 1 9 24 23 6 E. triquetra 6 0 0 3 1 3 4 2 2 0 1 0 Fusconaia subrotunda 1 0 1 Lampsilis cardium 8 1 1 1 1 L. fasciola 1 0 0 3 17 3 2 5 4 1 1 L. ovata 18 21 3 4 3 7 8 1 4 11 4 0 1 L. siliquoidea 3 2 5 1 1 0 1 0 3 3 Lasmigona compressa 1 0 1 1 L. costata 27 21 16 16 3 19 54 32 7 14 53 38 1 1 Ligumia recta 3 4 0 0 2 0 0 1 3 2 1 Pleurobema clava 0 P. sintoxia 1 1 0 6 1 2 2 0 0 3 Ptychobranchus fasciolaris 36 17 56 133 28 60 30 14 3 1 15 46 41 1 6 Pyganodon grandis 1 4 Quadrula cylindrica 6 0 7 3 1 0 3 0 3 4 Strophitus undulatus 37 10 6 20 1 13 15 5 0 9 5 4 Utterbackia imbecillis 1 Villosa fabalis 30 11 8 24 1 9 58 58 15 42 43 76 0 1 V. iris 1 1 Unidentified Total number 0 946 362 367 382 75 255 452 526 69 41 386 675 464 15 25 CPUE (#/p-h) 0 189.2 70.6 76.2 76.4 15 51 90.4 105.2 13.8 8.2 77.2 135 92.8 3 5 Species richness 0 18 15 12 12 11 12 13 15 13 7 12 16 16 6 9 398 Northeastern Naturalist Vol. 17, No. 3 Table 3. Total numbers, percent relative abundance, minimum length, maximum length, mean length and standard error (SE), number of recruits, number of sites where each species was found, number of sites with recruits of each species, number of male (M), number of female (F), number of unreported sex (U), and female-to-male ratio for each species for timed searches. Percent Minimum Maximum Mean SE No. sites Total relative length length length mean No. No. with F:M numbers abundance (mm) (mm) (mm) length (mm) recruits sites recruits F M U ratio Actinonaias ligamentina 3700 42.48 10.6 165.0 92.3 1.1 75 29 18 3700 Alasmidonta marginata 326 3.74 19.5 104.0 64.7 0.8 4 23 3 326 Amblema plicata 176 2.02 36.5 135.0 93.8 2.2 0 10 0 176 Anodontoides ferussacianus 73 0.84 24.0 73.1 46.1 1.2 4 10 2 73 Elliptio dilatata 716 8.22 14.0 124.9 78.1 0.8 9 25 8 716 Epioblasma t.rangiana 281 3.23 14.0 78.1 45.6 0.8 40 11 7 148 120 13 1.2:1 E. triquetra 53 0.61 15.5 67.0 43.3 1.5 2 15 2 36 12 5 3.0:1 Fusconaia subrotunda 39 0.45 11.0 136.0 84.8 7.1 5 7 2 39 Lampsilis cardium 43 0.49 27.0 135.0 90.3 4.8 2 15 1 6 12 25 0.5:1 L. fasciola 72 0.83 17.0 121.3 56.2 1.6 1 16 1 16 47 9 0.3:1 L. ovata 207 2.38 13.0 146.1 104.3 2.1 11 21 6 43 21 143 2.0:1 L. siliquoidea 349 4.01 33.0 121.2 76.2 0.9 0 21 0 179 159 11 1.0.1 Lasmigona compressa 10 0.11 60.5 89.0 72.9 3.3 0 8 0 10 L. costata 510 5.86 12.8 134.9 96.8 0.8 7 26 4 510 Ligumia recta 25 0.29 81.0 155.0 128.0 4.2 0 7 0 8 5 12 1.6:1 Pleurobema clava 13 0.15 17.5 58.5 33.3 4.3 5 5 3 13 P. sintoxia 109 1.25 21.5 141.0 74.6 4.3 11 17 7 109 Ptychobranchus fasciolaris 1165 13.38 17.0 129.0 81.2 0.8 15 28 10 1165 Pyganodon grandis 19 0.22 24.5 124.0 61.1 7.2 3 6 1 19 Quadrula cylindrica 41 0.47 37.0 140.0 91.3 4.5 0 12 0 4 37 Strophitus undulatus 339 3.89 33.5 99.0 62.8 0.6 0 20 0 339 Utterbackia imbecillis 1 0.01 58.3 58.3 58.3 NA 0 1 0 1 Villosa fabalis 439 5.04 10.5 41.8 26.9 0.3 61 19 14 217 219 3 1.0:1 V. iris 2 0.02 29.0 39.8 34.4 NA 1 2 1 2 Unidentified 2 0.02 22.0 22.0 22.0 NA 1 2 1 Total 8710 257 26 2010 T.A. Smith and D. Crabtree 399 Table 4. Estimated densities (#/m2) and standard error (SE) of each species found at the 9 quantitatively surveyed sites in French Creek, including the total estimate for all species at each site. Sites given as river kilometers (RKM) measured upstream from the mouth. The tenth quantitatively surveyed site (RKM 138) had zero mussels. RKM 127 123 98 89 74 68 52 23 19 Species #/m2 SE #/m2 SE #/m2 SE #/m2 SE #/m2 SE #/m2 SE #/m2 SE #/m2 SE #/m2 SE Actinonaias ligamentina 0.18 0.05 0.06 0.03 9.93 0.51 0.57 0.11 11.13 0.41 5.70 0.32 4.78 0.31 7.80 0.31 10.99 0.40 Alasmidonta marginata 0.07 0.03 0.02 0.02 0.59 0.10 0.06 0.03 1.46 0.14 0.58 0.10 0.62 0.08 0.49 0.08 0.79 0.09 Amblema plicata 0.29 0.06 0.04 0.02 0.01 0.03 0.10 0.04 0.11 0.04 Anodontoides ferussacianus 0.02 0.02 0.17 0.05 0.04 0.02 0.04 0.02 0.01 0.01 Dreissena polymorpha 0.02 0.01 0.02 0.02 0.01 0.01 0.02 0.02 0.02 0.02 Elliptio dilatata 0.74 0.09 0.04 0.02 1.39 0.15 0.20 0.06 2.43 0.18 2.50 0.22 1.43 0.13 2.62 0.21 2.99 0.22 Epioblasma t. rangiana 0.13 0.04 0.02 0.02 6.65 0.37 0.56 0.09 0.04 0.02 0.85 0.11 1.88 0.15 E. triquetra 0.46 0.07 0.04 0.02 0.15 0.04 0.02 0.02 0.11 0.04 0.06 0.03 0.14 0.04 0.06 0.02 0.12 0.04 Fusconaia subrotunda 0.22 0.06 0.04 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 Lampsilis cardium 0.05 0.02 0.11 0.04 0.04 0.02 0.09 0.03 0.06 0.03 0.13 0.03 0.03 0.02 0.02 0.01 L. fasciola 0.02 0.02 0.04 0.03 0.02 0.01 0.03 0.02 0.11 0.04 L. ovata 0.21 0.06 0.17 0.05 0.35 0.07 0.06 0.03 0.06 0.02 0.11 0.04 L. siliquoidea 1.60 0.12 0.15 0.04 0.03 0.02 0.01 0.01 0.17 0.05 0.04 0.02 0.09 0.03 0.07 0.03 L. complanata 0.01 0.01 0.01 0.01 0.03 0.01 Lasmigona compressa 0.01 0.01 0.01 0.01 0.11 0.04 0.02 0.02 0.01 0.01 0.01 0.01 L. costata 0.19 0.04 0.02 0.01 0.20 0.05 0.07 0.03 0.57 0.09 0.44 0.08 0.36 0.06 0.45 0.07 1.13 0.12 Ligumia recta 0.03 0.03 0.02 0.02 0.09 0.03 0.01 0.01 0.01 0.01 0.06 0.02 0.08 0.03 Pleurobema clava 0.21 0.06 0.02 0.03 P. sintoxia 0.03 0.02 0.61 0.10 0.05 0.03 0.05 0.03 0.04 0.03 0.02 0.01 0.04 0.02 Ptychobranchus fasciolaris 0.76 0.09 0.37 0.07 2.75 0.22 0.41 0.09 1.95 0.18 2.74 0.21 3.36 0.21 1.21 0.12 2.31 0.17 Pyganodon grandis 0.01 0.01 Quadrula cylindrica 0.07 0.03 0.02 0.02 0.12 0.04 0.14 0.05 0.02 0.01 0.01 0.02 0.12 0.04 Strophitus undulatus 0.01 0.01 0.15 0.04 0.63 0.09 0.10 0.04 0.06 0.02 0.14 0.04 0.21 0.06 Villosa fabalis 1.99 0.19 0.05 0.03 1.75 0.17 0.76 0.18 1.08 0.11 1.69 0.17 3.02 0.22 Unidentified 0.01 0.01 0.01 0.01 Total density estimate 4.10 0.22 0.71 0.11 19.24 0.82 1.82 0.23 27.98 0.70 13.89 0.50 12.26 0.60 15.87 0.50 24.04 0.50 400 Northeastern Naturalist Vol. 17, No. 3 Table 5. Total numbers, percent relative abundance, minimum length, maximum length, mean length and standard error (SE), number of recruits, number of sites where each species was found, number of sites with recruits of each species, number of male (M), number of female (F), number of unreported sex (U), and female-to-male ratio for each species for quantitative surveys. Percent Minimum Maximum Mean SE No. sites Total relative length length length mean No. No. with F:M numbers abundance (mm) (mm) (mm) length (mm) recruits sites recruits F M U ratio Actinonaias ligamentina 5586 43.84 7.0 163.0 92.3 0.4 333 9 8 5586 Alasmidonta marginata 508 3.99 16.0 116.0 61.8 0.7 14 9 7 508 Amblema plicata 63 0.49 3.0 138.0 77.0 4.3 6 5 3 63 Anodontoides ferussacianus 31 0.24 34.5 83.0 62.4 2.2 4 5 0 31 Dreissena polymorpha 10 0.08 14.5 34.0 20.7 1.7 NA 5 NA 10 Elliptio dilatata 1449 11.37 3.0 129.0 67.4 0.6 137 9 7 1449 Epioblasma t. rangiana 1052 8.26 14.0 147.0 41.1 0.4 325 7 7 677 347 28 0.5:1 E. triquetra 117 0.92 9.0 66.5 36.3 0.9 3 9 2 80 29 8 0.4:1 Fusconaia subrotunda 32 0.25 19.0 108.0 56.9 4.1 3 6 2 32 Lampsilis cardium 54 0.42 15.0 147.0 90.5 4.1 2 8 2 22 9 23 0.4:1 L. fasciola 43 0.34 39.0 77.0 53.4 1.4 0 7 0 10 5 28 0.5:1 L. ovata 119 0.93 11.0 153.0 104.3 2.5 2 7 2 44 42 33 1.0:1 L. siliquoidea 226 1.77 18.0 123.0 69.7 1.3 12 9 1 88 113 25 1.3:1 L. complanata 2 0.02 98.0 100.5 99.3 1.3 0 3 0 2 Lasmigona compressa 17 0.13 28.0 110.0 58.9 5.6 2 6 2 17 L. costata 393 3.08 10.0 135.0 95.7 1.0 8 9 5 393 Ligumia recta 31 0.24 44.0 165.0 127.1 5.2 0 7 0 3 1 27 0.3:1 Pleurobema clava 24 0.19 19.0 81.0 43.8 2.7 2 2 1 24 P. sintoxia 92 0.72 14.0 150.0 89.1 3.5 1 7 1 92 Ptychobranchus fasciolaris 1633 12.81 5.0 147.0 68.9 0.7 229 9 9 1633 Pyganodon grandis 1 0.01 97.0 97.0 97.0 NA 0 1 0 1 Quadrula cylindrica 57 0.45 12.0 139.0 88.5 3.8 2 7 2 57 Strophitus undulatus 141 1.11 31.0 101.0 64.7 1.1 0 8 141 Villosa fabalis 1060 8.32 7.0 60.0 25.2 0.2 256 7 7 550 462 48 0.8:1 Unknown 2 0.02 2 2 2 2 Total 12,743 1343 9 9 2010 T.A. Smith and D. Crabtree 401 were conducted. Table 6 gives recruitment data for each site that was sampled using both methods. We found a significant difference between the two sampling methods (t = 2.3372, df = 9, P = 0.0442), showing that our quantitative sampling methods were more effective at finding recent recruits at these sites. Over 60% of individuals of most species found in quantitative surveys were found during the surface search (Fig. 3), with a few exceptions: Snuffbox (41% surface), Lampsilis siliquoidea (Fatmucket) (54% surface), and Kidneyshell (59% surface). We used a two-tailed, paired t-test to evaluate if there were significant differences in the relative proportion of individuals of each species found in the surface versus the subsurface (excavation) portion of the quantitative surveys at each site by contrasting the numbers at each site. A similar two-tailed, paired t-test was used to test sex-specific contrasts for sexually dimorphic species. Significantly higher surface counts were found for 3 species: Mucket, Lampsilis ovata (Pocketbook), and Quadrula cylindrica (Table 7). Similarly, significant sex-specific differences were only found for Pocketbook and only for the males (male surface/sub-surface = 8.50, P = 0.004; and for comparison, female surface/subsurface = 6.167, P = 0.064). Low-CPUE sites. Three low-CPUE sites, at RKMs 138, 127, and 123, were surveyed quantitatively in 2004. Zero live or dead shells were found at RKM 138. Total estimated density at RKM 127 was 4.10 mussels m-2 (SE = 0.225), and estimated abundance was 10,244 (SE = 561.4). Of all our sites, this site had the highest densities of Snuffbox (0.46 mussels m-2, SE = 0.068). Three species were found in quantitative survey that were not found in timed search at RKM 127: Lampsilis fasciola (Wavy-rayed Lampmussel), Lampsilis cardium (Plain Pocketbook), and Strophitus undulatus (Creeper). Two species were found in the timed search at RKM 127 that were not found in quantitative survey: Pocketbook and Pyganodon grandis (Giant Floater). Low densities may explain the absence of those species in quadrat surveys; only 3 individual Table 6. Recruitment data for sites surveyed using both quantitative and timed searches methods. Total numbers (live) found, number of recent recruits found, total numbers (live) of animals, and percentage of recruits are given for each site. Sites given as river kilometers (RKM) measured upstream from the mouth. Timed searches Quantitative surveys Number of Total Percent Number of Total Percent RKM recruits number recruits recruits number recruits 138 0 0 0.00 0 0 0.00 127 2 131 1.53 29 418 6.94 123 2 221 0.90 2 84 2.38 98 7 596 1.17 94 2,000 4.70 89 13 255 5.10 13 179 7.26 74 50 946 5.29 304 2,858 10.64 68 5 381 1.31 96 1,234 7.78 52 3 255 1.18 465 1,412 32.93 23 13 386 3.37 156 1,886 8.27 19 6 675 0.89 180 2,662 6.76 Total 101 3846 2.63 1339 12,733 10.52 402 Northeastern Naturalist Vol. 17, No. 3 Pocketbook and 1 Giant Floater were found in the initial timed search at RKM 127, although other species with low densities were detected. Total density estimate at RKM 123 was 0.71 m-2 (SE = 0.115), and total abundance was 1783 individuals (SE = 287.1). The number of species detected in quadrat surveys versus quantitative surveys decreased at RKM 123; neither Plain Pocketbook nor Northern Riffleshell were found in quantitative surveys, though live Snuffbox mussels were found. The overall mean density estimate at low-CPUE sites was 1.60 mussels m-2 (SE = 1.264) and the mean abundance estimate was 4009 (SE = 3159.7). High-CPUE sites. Densities estimates at high-CPUE sites ranged from 1.82 to 27.98 mussels m-2, and abundance estimates range from 4540 to 69,848 live mussels per site. The mean density estimate at high-CPUE sites was 16.44 mussels m-2 (SE = 3.226), and the mean abundance estimate was 41,110 (SE = 8064.5). The highest densities and abundances were detected at RKM 74, with a density estimate of 27.98 mussels m-2 (SE = 0.697) and abundance estimate of 69,960 mussels (SE = 1714.9). Lower overall densities and abundances were observed at RKM 89 than at one site (RKM 127) in the low-CPUE strata. The number of species (live) found at high-density sites ranged from 8 to 21 per site. The number of species detected in quadrat surveys versus Table 7. Number of individuals of each mussel species found in surface vs. subsurface (excavation) by length (less than 30 or ≥30 mm) and sex (F = female, M = male, and U = undetermined sex). Also given are the P-values from the two tailed paired t-tests (df = 8) of the numbers of surface individuals (Q) to number of subsurface individuals (Qb). * = too few to calculate Surface (Q) Subsurface (Qb) F M U F M U Species <30 ≥30 <30 ≥30 <30 ≥30 <30 ≥30 <30 ≥30 <30 ≥30 P Actinonaias ligamentina 64 4326 269 922 0.04 Alasmidonta marginata 5 391 9 101 0.06 Amblema plicata 4 48 2 9 0.14 Anodontoides ferussacianus 28 2 0.18 Elliptio dilatata 52 841 85 469 0.24 Epioblasma t. rangiana 32 232 183 317 4 15 18 65 80 90 8 1 0.18 E. triquetra 10 9 25 1 3 3 16 23 23 2 2 0.56 Fusconaia subrotunda 2 24 1 5 0.24 Lampsilis cardium 5 14 1 13 4 1 7 9 0.44 L. fasciola 6 2 20 4 3 8 0.34 L. ovata 38 1 36 28 6 1 4 5 0.01 L. siliquoidea 44 60 17 1 42 6 46 5 3 0.72 L. complanata 2 * Lasmigona compressa 11 2 4 0.27 L. costata 5 337 3 48 0.11 L. recta 2 21 1 1 6 0.30 Pleurobema clava 1 16 1 6 0.30 P. sintoxia 1 78 12 0.32 Ptychobranchus fasciolaris 72 890 158 510 0.54 Pyganodon grandis 1 * Quadrula cylindrica 1 50 1 5 0.05 Strophitus undulatus 116 23 0.10 Villosa fabalis 254 36 224 119 15 9 162 9 160 45 19 2 0.28 2010 T.A. Smith and D. Crabtree 403 quantitative surveys stayed the same, but species composition changed slightly at RKM 89, and increased at the other 6 high-CPUE sites (RKMs 98, 74, 68, 52, 23, and 19). The highest density and abundance estimates of Northern Riffleshell were at RKM 74, where Northern Riffleshell was the second-most abundant species, next to Mucket. It was estimated that there were 16,633 (SE = 915.4) live animals at this site. Northern Riffleshells were found farther upstream (RKM 98) during our quantitative sampling than in timed searches (RKM 89). RKM 98 and RKM 89 are the only sites in our study known to house both federally endangered species. RKM 98 had our highest densities of Clubshell, with estimated densities of 0.21 mussels m-2 (SE = 0.055) and estimated abundance of 531 (SE = 148.3). RKM 98 also had the highest relative density and abundance of Longsolid (density estimate = 0.22 mussels m-2, SE = 0.056). Estimating densities from CPUE The relationship between the density estimates from quantitative sampling and CPUE from timed searches was linear (density estimate = 0.1594*CPUE + 0.4147, r2 = 0.91, F-statistic 80.04, d.f. = 8, P = 0.000; Fig. 4), as was the relationship between abundance estimates and CPUE (abundance estimate = 0.398.553*CPUE + 1036.998, r2 = 0.91, F-statistic = 80.04, d.f. = 8, P = 0.000; Fig. 5). Coefficient of variations were low, ranging from 0.02 to 0.16, giving our estimates a low margin of error and a high probability of encountering Figure 4. Linear relationship between timed search sampling catch-per-unit effort and abundance estimates from quantitative sampling. Abundance estimate = 0.398.553*CPUE + 1036.998, r2 = 0.91, F-statistic = 80.04, d.f. = 8, P = 0.000. Points are labeled with river kilometers corresponding to each site. 404 Northeastern Naturalist Vol. 17, No. 3 Figure 5. Linear relationship between timed search sampling catch-per-unit effort and density estimates from quantitative sampling. Density estimate = 0.1594*CPUE + 0.4147, r2 = 0.91, F-statistic 80.04, d.f. = 8, P = 0.000. Points are labeled with river kilometers corresponding to each site. individuals (Smith et al. 2001). Using these linear relationships, we were able to extrapolate densities and abundances for riffle-run sites surveyed only with timed searches (Table 8). Extrapolated values for sites where quadrat surveys were performed are also included in this table for comparison. Extrapolated densities and abundances at the quantitative sites generally fell within or slightly outside of the 90% confidence intervals of calculated values, with the exception of RKM 89 and RKM 123, which both had lower calculated densities and abundances than their respective extrapolated values. Estimating river-wide populations Using our 2002 mapping data and ArcGIS, we calculated the total large riffle-run area in the Pennsylvania portion of French Creek. Areas in the New York portion of the French Creek were not calculated, and therefore this analysis only pertains to the Pennsylvania portion of the stream. Each of our sites was approximately 2500 m2, so the calculated proportion of the stream studied (26 PA riffle-run sites = 65,000 m2) to the proportion of large riffle-run habitat in the stream (2,038,981 m2) is equal to 0.0318. Of the 26 randomly selected surveyed riffle-run habitats in Pennsylvania, 38.46% were low-CPUE sites and 65.38% were high-CPUE sites. To account for the proportion of total riffle-run sites that would be low density and the proportion of sites that would be high density in French Creek, we weighted the river-wide estimates accordingly to arrive at a total abundance estimate. The proportion of sites in each stratum was multiplied by the within-stratum estimate of abundance. These values were then combined to 2010 T.A. Smith and D. Crabtree 405 arrive at a weighted estimate of abundance for the large riffle-runs of French Creek. Abundance in high-CPUE stratum was estimated to be 41,110 (SE = 8064.5) and 4009 in the low-CPUE stratum (SE = 3159.7). Multiplying each estimate by its respective area of riffle-run habitat in French Creek gives an estimate of 20,633,376 (95% CI = 8,836,476–30,537,350) mussels in high-CPUE riffle-run habitat and 1,257,580 (95% CI = 0–6,705,697) in low-CPUE riffle-run, which add up to an estimated total of 21,890,957 (95% CI = 8,836,476, 37,243,047) mussels estimated for the large riffle-runs in the Pennsylvania portion of French Creek. Discussion Of the 66 mussels species documented in Pennsylvania, 29 of these have been documented in the French Creek watershed, with 26 of these species regularly occurring as of this study. French Creek still holds recent occurrences of Table 8. CPUE, extrapolated abundances, and density estimates of all species for sites surveyed only with timed searches, using the relationships between CPUE, densities, and abundances calculated from quadrat survey results. Also includes extrapolations for quantitative sites (*) to use as a comparison to calculated values (Table 4). Sites given as river kilometers (RKM) measured upstream from the mouth. Estimated density RKM CPUE (mussels/p-h) Estimated abundance (mussels m2) 157 5.0 3,030 1.2 146 148.8 60,342 24.1 142 39.8 16,899 6.8 140 0.2 1,117 0.4 *138 0.0 0 0.0 132 3.0 2,233 0.9 *127 26.2 11,479 4.6 125 102.8 42,008 16.8 *123 44.2 18,653 7.5 115 15.2 7,095 2.8 109 56.6 23,595 9.4 107 28.0 12,196 4.9 *98 119.0 48,465 19.4 93 57.2 23,834 9.5 *89 84.2 34,595 13.8 *74 189.2 76,443 30.6 72 70.6 29,175 11.7 *68 76.2 31,407 12.6 63 76.4 31,486 12.6 58 15.0 7,015 2.8 *52 51.0 21,363 8.5 44 90.4 37,066 14.8 41 105.2 42,965 17.2 29 8.2 4,305 1.7 *23 77.2 31,805 12.7 *19 135.0 54,842 21.9 11 92.8 38,023 15.2 5 3.0 2,233 0.9 1 5.0 3,030 1. 2 406 Northeastern Naturalist Vol. 17, No. 3 26 species of freshwater mussels, more than any other watershed in Pennsylvania or anywhere in the northeastern United States. Two species, Clubshell and Northern Riffleshell, are federally and state endangered, (US Fish and Wildlife Service 1994). In general, aquatic mollusks, including bivalves and gastropods, are a critically imperiled group throughout much of the world (Bogan 1993, Lydeard et al. 2004, Ricciardi et al. 1998, Williams et al. 1993). This fact makes this research extremely important and places special emphasis on the conservation of places like French Creek. The results of this study show that French Creek mussel populations remain relatively intact. In timed searches, we documented 24 species throughout the main stem, the same species which were documented in 1993 (Bier 1994), including the federally endangered Clubshell and Northern Riffleshell. We found one additional species in the quadrat surveys, White Heelsplitter, which was not detected in our timed searches or in 1993 surveys (Bier 1994). In total, we documented 25 of the 29 previously recorded freshwater mussels found in the French Creek watershed. Ligumia nasuta (Eastern Pondmussel), was not detected during this study; however, that species is known to inhabit only certain tributaries of French Creek, such as those recently documented in Conneaut Lake Outlet (Smith 2007). Toxolasma parvus (Liliput) and Cyclonaias tuberculata (Purple Wartyback) were documented in Ortmann’s early surveys of French Creek (1919); however, neither species were documented in a basin-wide survey in 1993 (Bier 1994), nor in later bridge-replacement surveys (Environmental Science 2002a, b) on the main stem of French Creek. The Purple Wartyback is now considered extirpated from Pennsylvania. Simpsonaias ambigua (Salamander Mussel), also collected by Ortmann (1919) in French Creek, has been found only as two dead shells recently—one collected in 1985 and another in 1995 (Bier 1994, Pennsylvania Natural Heritage Program files). No further evidence of the Salamander Mussel has been found since the discovery of those 2 shells (Bier 1994; Environmental Science 2002a, b; Mohler et al. 2006; Smith 2007; Smith and Crabtree 2005). Using our quantitative data, we calculated regression equations that enabled us to estimate the densities and abundances of mussels based on CPUE data from timed searches. Site density estimates ranged from 0–27.98 mussels m-2, and are consistent with 3 recently surveyed sites on French Creek (Environmental Science Inc. 2002a, b). Although the linear relationships between CPUE and estimated density and CPUE and estimated abundance were strong, there were 2 sites that had large discrepancies when comparing extrapolated densities with calculated densities: RKM 89 and RKM 123. The unusually high water and flooding events during 2004, combined with the predominance of very loose gravel and sandy substrate, may have contributed to a fair number of mussels washed downstream at RKM 89 and RKM 123, resulting in lower densities than what would have been observed during the previous year’s timed searches. These sorts of stream dynamics should be taken into account when interpreting stream-wide estimations. We documented evidence of recent recruitment for most species using both timed searches and quadrat surveys; however, we did find that quadrat 2010 T.A. Smith and D. Crabtree 407 sampling was more effective than timed searches in detecting recent recruits. These findings are consistent with other studies that have compared quadrat vs. timed searches (Hornbach and Deneka 1996, Vaughn et al. 1997), where the mean length of individuals collected differed significantly between quadrat and timed searches; however, some studies showed no difference in detection of smaller individuals when comparing methods (Miller and Payne 1993, Obermeyer 1998). Our results indicate that the increased detection of small individuals we observed in quadrat surveys may in part be due to increased effort, since proportions of recent recruits did not greatly differ from surface counts and excavations for most species. The presence of small individuals is one important measure of population viability. Another important measure of viability is the representation of both sexes within a population. We determined the sex ratios for sexually dimorphic species, and some differences were seen between survey methods. In timed searches, several species showed a 1:1 or slightly higher ratio of females to males, including Northern Riffleshell, Fatmucket, Ligumia recta (Black Sandshell), and Rayed Bean. Near 1:1 sex ratios were observed in quadrat surveys for the Rayed Bean and Fatmucket; however, some species showed quite different results between methods. For example, in our timed searches, we detected a 3:1 ratio of female to male Snuffbox, but quantitative surveys resulted in a female-to-male ratio of 0.4:1. Northern Riffleshell also showed a much lower female-to-male ratio in quantitative surveys (0.5:1) than in timed searches (1:1). The high number of females detected in timed searches is likely due to the high visibility of gravid females on the surface. In addition, some mussels may not begin to show female sex characteristics until they become sexually mature, which is thought to occur at age 3 (R. Villella, US Geological Survey, Biological Resources Division, Leetown Science Center, Kearneysville, WV., pers. comm.), which may explain the relatively high number of males recorded for small individuals. (See further discussion specific to Northern Riffleshell sex ratios in Crabtree and Smith [2009]). Interpreting sex ratios for certain species in our study should be done with caution since the sex for a large proportion of individuals from certain species was not recorded. For example, 44.8% of Pocketbook individuals did not have a sex recorded in our timed-search field notes. Sex ratios and recruitment in freshwater mussel populations need further study, especially in the wake of recent findings on the amount of endocrinedisrupting pharmaceuticals found in our nation’s waterways (Hayes 2005); for example, fluoxetine, an active ingredient in anti-depressants, has been linked to premature release of glochidia (ACS 2006). Trends in species richness were detected, particularly showing fewer species higher in the watershed and an area of high species richness between RKM 109 and RKM 74. Interestingly, very few mussels were detected at the three uppermost sites in Pennsylvania, as one might expect going up stream; however, the CPUE and species richness both rose in the New York portion of the creek. The habitat in these upper reaches seemed comparable to other riffle-run sites in our study (substrate, flow, pH, temperature, etc.). After comparing past and recent aerial photographs and talking to area residents, it appears that there is a new channel in this upper reach. Remnants of the 408 Northeastern Naturalist Vol. 17, No. 3 old meander still appear in an adjacent agricultural field and it is unclear as to whether the change was part of natural channel formation or the result of a diversion by humans. Regardless of its origin, there may not have been enough time for mussels to colonize this newer channel, resulting in their absence in our surveys. Previous investigations have identified Clubshell as having a limited range in the French Creek watershed, and the number of live individuals was very low wherever it was found, with no observed recent recruitment (Bier 1994). Similarly, in this study, live Clubshell individuals were found only in the upper part of the watershed and thus were separated from the main stem of the Allegheny River by at least 89 river kilometers. Using 13 microsatellite markers, Morrison et al. (2007) could not statistically distinguish Clubshell collected from RKM 98 in French Creek from Clubshell collected from the main stem of the Allegheny River. The genetic implications of this geographic distance are still being investigated (C.L. Morrison, US Geological Survey, Aquatic Ecology Branch, Leetown Science Center, Kearneysville, WV., pers. comm.). The main stem of French Creek may have viable population sources of Clubshell in tributary streams such as Muddy Creek (Mohler et al. 2006), Conneaut Outlet, and LeBoeuf Creek (Bier 1994, Smith 2007); however, connection to downstream populations is likely minimal. There appears to be a healthy population of Northern Riffleshell in French Creek, with evidence of recruitment at many sites and estimated densities and abundances that are among the highest in its current range (Zanatta and Murphy 2007). Northern Riffleshell exhibited a bimodal distribution in French Creek; the greatest numbers were found between RKMs 74 and 63, while the second mode was observed from RKM 29 to the bottom of the watershed. The range of Northern Riffleshell in French Creek was opposite of that of Clubshell. The bimodal distribution of Northern Riffleshell and the disjunct Clubshell population may be due to dramatic changes in the drainage before and after the last glacial event (Harrison 1980) and the influence of stream alterations (see further discussion in Crabtree and Smith 2009. Hydrologic and stream channel alterations are a general threat to the viability of freshwater mussels (see review in Watters 2000); however, most of the main stem of French Creek is free flowing. The Union City Dam, built in 1971 in Erie County, is the only major dam on the main stem of French Creek. This dam, located at RKM 71.5, may act as a barrier to upstream dispersal of freshwater mussels, disrupt nutrient spiraling, and increase erosion (WPC and FCP 2002). There is only one other USACE dam in the watershed, located on Woodcock Creek, and there are a number of smaller dams on other tributaries, all of which should be evaluated to determine if they significantly impact aquatic life (WPC and FCP 2002). Low numbers of mussels in certain portions of the stream may partially be due to poor in-stream and land-use practices. According to the US Environmental Protection Agency (2006), second to abandoned mine drainage, the major source of impairment to Pennsylvania streams is agriculture, which causes increased nutrients, siltation, and low dissolved oxygen levels. Improper agricultural activities produce the most significant amount of sediment 2010 T.A. Smith and D. Crabtree 409 entering streams (Henley et al. 2000, Waters 1995), and the resulting siltation causes reduced feeding, reduced growth rates, clogged gills, disrupted metabolic processes, limited burrowing activity and physically smothered freshwater mussels, all resulting in reduced freshwater mussel assemblages (Brim Box and Mossa 1999, Vannote and Minshall 1982, Watters 1995). Although a significant threat state-wide, French Creek has only a small number of streams listed as impaired due to agriculture sources, and thus this threat to mussels is localized. Although sparsely populated, the French Creek watershed is threatened by development encroaching from the City of Erie and a few small urban areas. Low freshwater mussel densities and diversity have been documented downstream of two urban areas, Meadville and Franklin, along the creek (Bier 1994), and similar findings were mirrored in our study. Freshwater mussels are susceptible to chlorine, a highly toxic chemical commonly found in discharge from wastewater treatment facilities (Valenti et al. 2006). There are 13 public wastewater facilities in the French Creek watershed; however, several have discontinued their use of chlorine (WPC and FCP 2002). In addition, the DEP reported low dissolved oxygen and high nutrient, metal, and fecal coliform levels below sewage treatment plants in French Creek (Haase 1992, 1994). Furthermore, 2 superfund sites have been documented on the main stem of French Creek (Burke 2007, Corbett 2002, Younis 1988), which may have resulted in increased PCB concentrations in portions of the creek. Water withdrawls for agricultural irrigation and municipal and industrial water supplies may also cause decreases in flow velocities and dissolved oxygen levels (Johnson et al. 2001). Ammonia has been shown to be lethal to freshwater mussels (Havlik and Marking 1987), and is a common pollutant from agricultural, municipal, and industrial sources, such as fertilizers and pesticides used in developed areas (Augspurger et al. 2003, Newton 2003). Non-native species invasions are another potential threat to the native biota in French Creek. Zebra Mussel is clearly a threat to aquatic ecosystems (Biggins et al. 1995, Ricciardi et al. 1998, Strayer and Malcom 2007). The first reports of Zebra Mussel in the watershed were reported from Edinboro Lake in 2000 (PA DEP 2000). Although we did not find any Zebra Mussel in our timed searches, several live specimens were found during our quantitative surveys. Given the relatively large size of these individuals, the low numbers, and the RKMs at which live individuals were found, it is highly likely that they were washed down from source populations upstream, rather than being offspring of actively reproducing populations in these portions of the stream (L. Parendes, Edinboro University of Pennsylvania. Edinboro, PA., pers. comm.). More monitoring needs to be done to test this theory, especially in the larger pool sections of French Creek and below source populations such as Edinboro Lake. Also, monitoring is needed to determine whether these individuals will become founding members of new, actively reproducing populations in these portions of the stream. However, given the amount of shallow, somewhat turbulent flowing habitats in French Creek, Zebra Mussel may never reach the high densities found in lakes or dammed rivers. Other exotic aquatic species in the watershed include Cyprinus carpio 410 Northeastern Naturalist Vol. 17, No. 3 (L.) (Common Carp), which may compete with other benthic feeders, as well as Salmo trutta (L.) (Brown Trout) and Oncorhynchus mykiss (Walbaum) (Rainbow Trout), both of which may have negative impacts on the native forage fish that play an essential role in mussel reproduction. The amount of suitable trout habitat in French Creek is limited, and the effects of these exotic aquatic species on freshwater mussel populations in this watershed, if any, are unknown. Corbicula fluminea (Muller) (Asiatic Clam) has also been documented in the French Creek drainage, in the West Branch of French Creek (Wellington 1994) and more recently in Conneaut Outlet (Smith 2007), although it was not recorded in our main-stem surveys. French Creek’s aquatic communities represent some of the last remaining relatively intact and continuous high-quality natural communities found anywhere in the Ohio River basin. There is a great need to fully understand these aquatic communities throughout their range, including their population genetics and viability. Furthermore, we need to better understand the threats unionids face from invasive species, improper land use, habitat degradation, and pollution (Bogan 1993, Ricciardi and Rasmussen 1999, Williams et al. 1993). To further underscore the importance of this project, the federally endangered Clubshell and Northern Riffleshell have been lost from over 95% of their historic world ranges (US Fish and Wildlife Service 1994). Both maintain relatively healthy populations in the French Creek watershed. Through this project, we have expanded the known ranges for these species and gained valuable recruitment data, which is essential information when assessing population viablilty and will ultimately aid in conservation efforts. Our approach of using system-wide surveys helps us understand the full range and integrity of populations within a system, rather than getting a snap shot of just a few sites, and could be applied to other watersheds. Building on this type of work, we can provide the public with defensible information about freshwater mussel populations and help target conservation efforts more effectively. Acknowledgments Funding was provided by the US Fish and Wildlife Service (USFWS) through State Wildlife Grants Program Grant T-2, administered through the Pennsylvania Game Commission and Pennsylvania Fish and Boat Commission. The Nature Conservancy (TNC) and the Western Pennsylvania Conservancy (WPC) provided additional funding. Work was conducted under a Scientific Collecting Permit Number 196 Type 1 granted by the Pennsylvania Fish and Boat Commission. WPC would like to thank staff from our partners at French Creek Project, TNC, Venango and Crawford County Conservation Districts, and USFWS. Todd Sampsell (WPC) secured funding for this project. Thanks to Laurie Parendes (Edinboro University), Rita Villella (US Geological Survey), and Cheryl Morrison (US Geological Survey) for comments. Thanks to Chris Schaffer (Allegheny College) for his GIS expertise. Thanks to Gerald Lang (TNC), Alan Wolf (TNC), Krystal Bastion (Allegheny College), Dima Haliwani, Carrie Altman (TNC), Lucas Mattera (St. Bonaventure), and Dave Zanatta (University of Toronto) for help with field work. Thanks to WPC field crewmembers Zachary Horn, Nathan Irwin, Erica Maynard, Elizabeth Peck, Todd Sampsell, Elizabeth Skinner-Meyer, Curtis Stumpf, and Erik Weber. Thanks to Student Conservation Association (SCA) volunteer crewmembers Megan Bradburn, 2010 T.A. Smith and D. Crabtree 411 Amy Bush, and Philip Kulkulski. This manuscript was substantially improved from the comments of 3 anonymous reviewers. Literature Cited American Chemical Society (ACS). 2006. Ingredient in Prozac increases risk of extinction for freshwater mussels. ScienceDaily 15 September 2006. Available online at http://www.sciencedaily.com/releases/2006/09/060914153812.htm. Accessed 24 March 2008. Augspurger, T., A.E. Keller, M.C. Black, W.G. Cope, and F.J. Dwyer. 2003. Derivation of water quality guidance for protection of freshwater mussels (Unionidae) from ammonia exposure. Environmental Toxicology and Chemistry 22:2569–2575. Bier, C. W. 1994. A survey of the mussels (Unionidae) in portions of the French Creek Basin of Northwestern Pennsylvania. Submitted to the US Fish and Wildlife Service. A publication of the Western Pennsylvania Conservancy. ii + 97pp. Biggins, R.G., R.J. Neves, and C.K. Dohner. 1995. Draft national strategy for the conservation of native freshwater mussels. US Fish and Wildlife Service, Washington, DC, 26 pp. Bogan, A.E. 1993. Freshwater bivalve extinctions (Mollusca: Unionoida): A search for causes. American Zoology 33:599–609. Brim Box, J., and J. Mossa. 1999. Sediment, land use, and freshwater mussels: Prospects and problems. Journal of the North American Benthological Society 18:99–117. Burke, J.J. 2007. Third five-year review report. Saegertown industrial area superfund site: Lord Corportation Property. US Environmental Protection Agency. Region III. Philadelphia, PA. Available online at http://www.epa.gov/superfund/sites/ fiveyear/f2007030001714.pdf. Accessed April 2007. Butler, R.S. 2003. Status assessment for the Rayed Bean, Villosa fabalis, occurring in the Mississippi River and Great Lakes systems. Unpublished report prepared by the Ohio River Valley Ecosystem Team Mollusk Subgroup, Asheville, NC. March 2003. 65 pp. Butler, R.S. 2006. Status assessment report for the Snuffbox, Epioblasma triquetra, a freshwater mussel occurring in the Mississippi River and Great Lakes Basins. A draft report for the Ohio River Valley Ecosystem Team, US Fish and Wildlife Service, Asheville, NC. 234 pp. Corbett, C. 2002. Five-year review report: Second five-year review report for Saegertown area industrial superfund site, borough of Saegertown, Crawford County, Pennsylvania. US Environmental Protection Agency, Region III, Philadelphia, PA. Available online at http://www.epa.gov/superfund/sites/fiveyear/f02-03018. pdf. Accessed January 2007. Crabtree, D.L., and T.A. Smith. 2009. Population attributes of an endangered mussel, Northern Riffleshell, Epioblasma torulosa rangiana, in French Creek and implications for its recovery in nearby watersheds. Northeastern Naturalist 16(3):339-354. Cummings, K.S., and C.A. Mayer. 1992. Field Guide to Freshwater Mussels of the Midwest. Illinois Natural History Survey. Manual 5. Champaign, IL. 194 pp. Environmental Science, Inc. 2002a. Mill Village truss bridge replacement biological assessment/freshwater mussel population survey. Unpublished October 3, 2002 report prepared for M.S. Consultants and PADOT Engineering District 1-0, Columbus, OH. Environmental Science, Inc. 2002b. S.R. 1002 Section B00, Venango Veterans Memorial Bridge replacement project biological assessment/freshwater mussel population survey. Unpublished October 3, 2002 report prepared for M.S. Consultants and PADOT Engineering District 1-0, Columbus, OH. 412 Northeastern Naturalist Vol. 17, No. 3 Etnier, D.A., and W.C. Starnes. 1993. The Fishes of Tennessee. University of Tennessee Press, Knoxville, TN. 681 pp. Hankin, D.G., and G.H. Reeves. 1988. Estimating total fish abundance and total habitat area in small streams based on visual estimation methods. Canadian Journal of Fisheries and Aquatic Sciences 45:834–844. Harrison, S. 1980. New survey provides key to Crawford County drainage puzzle. Pennsylvania Geology 11(2):10–12. Hasse, R.L. 1992. French Creek survey, Crawford County. Pennsylvania Department of Environmental Resources, Northwest Regional Office, Meadville, PA. Unpublished report 9 pp. (plus appendices). Hasse, R.L. 1994. French Creek survey, Cambridge Springs and Venango Boroughs. Pennsylvania Department of Environmental Resources, Northwest Regional Office, Meadville, PA. Unpublished report 7 pp. (plus appendices). Havlik, M.E., and L.L. Marking. 1987. Effects of contaminants on naiad mollusks (Unionidae): A review. US Fish and Wildlife Service, Washington, DC. Resource Publication 164. Hayes, T.B. 2005. Welcome to the revolution: Integrative biology and assessing the impact of endocrine disruptors on environmental and public health. Integrative and Comparative Biology 45:321–329. Henley, W.F., M.A. Patterson, R.J. Neves, and A.D. Lemly. 2000. Effects of sedimentation and turbidity on lotic food webs: A concise review for natural resource managers. Reviews in Fishery Science 8(2):125–139. Hornbach D.J., and T. Deneka. 1996. A comparison of a qualitative and quantitative collection method for examining freshwater mussel assemblages. Journal of the North American Benthological Society 15:587–596. Jenkins, R.E., and N.M. Burkhead. 1994. Freshwater Fishes of Virginia. American Fisheries Society, Bethesda, MD. xxiii + 1079 pp. Johnson, P.M., A.E. Liner, S.W. Golladay, and W.K. Michener. 2001. Effects of drought on freshwater mussels and instream habitat in Coastal Plain tributaries of the Flint River, southwest Georgia (July–October, 2000). Unpublished report, by the Jones Ecological Research Center for The Nature Conservancy, Tallahassee, fl. 45 pp. Lydeard, C., R.H. Cowie, W.F. Ponder, A.E. Bogan, P. Bouchet, S.A. Clark, K.S. Cummings, T.J. Frest, O. Gargominy, D.G. Herbert, R. Hershler, K.E. Perez, B. Roth, M. Seddon, E.E. Strong, and F.G. Thompson. 2004. The global decline of nonmarine mollusks. BioScience 54:321–330. Miller, A.C., and B.S. Payne. 1988. The need for quantitative sampling to characterize size demography and density of freshwater mussel communities. American Malacological Bulletin 6:49–54. Miller, A.C., and B.S. Payne. 1993. Qualitative versus quantitative sampling to evaluate population and community characteristics at a large-river mussel bed. American Midland Naturalist 130:133–145. Mohler, J.W., P. Morrison, and J. Haas. 2006. The mussels of Muddy Creek on Erie National Wildlife Refuge. Northeastern Naturalist 13:569–582. Morrison, C.L., T. King, M.S. Eackles, and R. Villella. 2007. Phylogeography and population genetics for enhanced conservation of the endangered Clubshell Mussel, Pleurobema clava. 2007 Freshwater Mollusk Conservation Society Symposium and Workshop. Little Rock, AR. NatureServe. 2008. NatureServe explorer: An online encyclopedia of life [web application]. Version 6.1. NatureServe, Arlington, VA. Available online at http:// www.natureserve.org/explorer. Accessed: 21 October 2008. Newton, T.J. 2003. The effects of ammonia and freshwater unionid mussels. Environmental Toxicology and Chemistry 22:2543–2544. 2010 T.A. Smith and D. Crabtree 413 Obermeyer, B.K. 1998. A comparison of quadrats versus timed snorkel searches for assessing freshwater mussels. American Midland Naturalist 139:331–339. Ortmann, A.E. 1919. A monograph of the naiads of Pennsylvania. Part III: Systematic account of the genera and species. Memoirs of the Carnegie Museum 8(1): xvi–384, 21pls. Parmalee, P.W., and A.E. Bogan. 1998. The Freshwater Mussels of Tennessee. The University of Tennessee Press, Knoxville, TN. 328 pp. Pennsylvania Department of Environmental Protection. 2000. Zebra Mussels found in Edinboro Lake. News release from PADEP. Available online at http://www. ahs.dep.state.pa.us/newsreleases/default.asp?ID=683&varQueryType=Detail. Accessed January 2007. Pooler, P.S., and D.R. Smith. 2005. Optimal sampling design for estimating spatial distribution and abundance of a freshwater mussel population. Journal of the North American Benthological Society 24:525–537. Ricciardi, A., and J.B. Rasmussen. 1999. Extinction rates of North American freshwater fauna. Conservation Biology 13:1220-1222. Ricciardi A., R.J. Neves, and J.B. Rasmussen. 1998. Impending extinctions of North American freshwater mussels (Unionoida) following the Zebra Mussel (Dreissena polymorpha) invasion. Journal of Animal Ecology 67:613–619. Smith, D.R. 2006. Survey design for detecting rare freshwater mussels. Journal of the North American Benthological Society 25:701–711. Smith, D.R., R.F. Villella, and D.P. Lemarie. 2001. Survey protocol for assessment of endangered freshwater mussels in the Allegheny River, Pennsylvania. Journal of the North American Benthological Society 20:118–132. Smith, T.A. 2007. Freshwater Mussel (Bivalvia: Unionidae), fish, and macroinvertebrate surveys in the tributaries of French Creek, Pennsylvania. Unpublished February 15 Progress Report submitted to the PA Department of Conservation and Natural Resources, Union City, PA. 14 pp + 22 pp. Smith, T.A., and D. Crabtree. 2005. Freshwater mussel (Unionidae) and fish assemblage habitat use and spatial distributions in the French Creek watershed: Reference for western Pennsylvania unionid protection and restoration. Unpublished July 30, 2005 Final Report submitted to the Pennsylvania Fish and Boat Commission, Union City, PA. iv + 180 pp. Strayer, D.L., and H.M. Malcom. 2007. Effects of Zebra Mussels (Dreissena polymorpha) on native bivalves: The beginning of the end or the end of the beginning? Journal of the North American Benthological Society 26:111–122. Strayer, D.L., and D.R. Smith. 2003. A guide to sampling freshwater mussel populations. American Fisheries Society Monograph 8. American Fisheries Society. Bethesda, MD. 103 pp. Strayer, D.L., S. Claypool, and S.J. Sprague. 1997. Assessing unionid populations with quadrats and timed searches. Pp. 163–169, In K.S. Cummings, A.C. Buchanan, C.A. Mayer, and T.J. Naimo. Conservation and Management of Freshwater Mussels II: Proceedings of a UMRCC Symposium.Upper Mississippi River Conservation Committee, Rock Island, IL. US Fish and Wildlife Service. 1991. Endangered and threatened wildlife and plants: Animal candidate review for listing as endangered or threatened species. Federal Register 56(225):58804–58836. US Environmental Protection Agency. 2006. Assessment data for the State of Pennsylvania year 2006. Available online at http://iaspub.epa.gov/waters10/ w305b_report_control.get_report?p_state–PA&p_cycle–#STREAM/CREEK/ RIVER_source_state_top_ten. Accessed April 2007. 414 Northeastern Naturalist Vol. 17, No. 3 US Fish and Wildlife Service. 1994. Clubshell (Pleurobema clava) and Northern Riffleshell (Epioblasma torulosa rangiana) recovery plan. US Fish and Wildlife Service, Hadley, MA. 68 pp. US Geological Survey. 2004. Mussel Estimation Program. Available online at http:// www.lsc.usgs.gov/aeb/2068/. Accessed January 2006. Valenti, T.W., D.S. Cherry. R.J. Currie, R.J. Neves, J.W. Jones, R. Mair, and C.M. Kane. 2006. Chlorine toxicity to early life stages of freshwater mussels (Bivalvia: Unionidae). Environmental Toxicology and Chemistry 25:2512–2518. Vannote, R.L., and G.W. Minshall. 1982. Fluvial processes and local lithology controlling abundance, structure, and composition of mussel beds. Proceedings of the National Academy of Sciences 79:4103–4107. Vaughn, C.C., C.M. Taylor, and K.J. Eberhard. 1997. A comparison of the effectiveness of timed searches vs. quadrat sampling in mussel surveys. Pp. 157–162, In K.S. Cummings, A.C. Buchanan, C.A. Mayer, and T.J. Naimo (Eds.). Conservation and management of freshwater mussels II: Initiatives for the future. Proceedings of an Upper Mississippi River Conservation Committee (UMRCC) symposium, 16–18 October 1995, St. Louis, MO. Upper Mississippi River Conservation Committee, Rock Island, IL. Villella, R., and D.R. Smith. 2005. Two-phase sampling to estimate river-wide populations of freshwater mussels. Journal of the North American Benthological Society 24:357–368. Villella, R., D.R. Smith, and D.P. Lemarie. 2004. Estimating Survival and Recruitment in a freshwater mussel population using mark-recapture techniques. American Midland Naturalist 151:114–133. Waters, T.F. 1995. Sediment in streams: Sources, biological effects, and control. American Fisheries Society Monograph 7. 251 pp. Watters, T.G. 2000. Freshwater mussels and water quality: A review of the effects of hydrologic and instream habitat alterations. Pp. 1–14, In P.D. Johnson and R.S. Butler. Proceedings of the Conservation, Captive Care, and Propagation of Freshwater Mussels Symposium, 1999. Ohio Biological Survey, Columbus, OH. Wellington, R.J. 1994. French Watershed Creek Stream Survey. Erie County PA Department of Health, Erie, PA. Western Pennsylvania Conservancy and French Creek Project (WPC and FCP). 2002. French Creek watershed conservation plan. Western Pennsylvania Conservancy, Union City, PA. 272 pp. Williams, J.D., M.L. Warren, K.S. Cummings, J.L. Harris, and R.J. Neves. 1993. Conservation status of freshwater mussels of the United States and Canada. Fisheries 18:6–22. Younis, P.C. 1988. Federal on-scene coordinator’s report for the nickel plate road site, Cochranton, Crawford County, PA. Available online at http://www.epa.gov/ reg3hwmd/super/sites/PAD982367369/index.htm. Accessed January 2007. Zanatta, D.T., and R.W. Murphy. 2007. Range-wide population genetic analysis of the endangered Northern Riffleshell mussel, Epioblasma torulosa rangiana (Bivalvia: Unionoida). Conservation Genetics 8:1393–1404.