Southeastern Naturalist 649 C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 22001144 SOUTHEASTERN NATURALIST 1V3o(4l.) :1634,9 N–6o6. 24 Forage Selection of Native and Nonnative Woody Plants by Beaver in a Rare-Shrub Community in the Appalachian Mountains of North Carolina C. Reed Rossell, Jr.1,*, Scott Arico2, H. David Clarke2, Jonathan L. Horton2, Jennifer Rhode Ward2, and Steven C. Patch3 Abstract – Castor canadensis (Beaver) is a selective forager that can modify the species composition and structure of plant communities. However, no studies have examined the use of woody plants by Beaver in temperate forests that contain a dominant nonnative plant. We investigated foraging of woody plants by Beaver in a riparian shrub community that is dominated by both native and nonnative species, including the federally threatened shrub Spiraea virginiana (Virginia Spiraea). We established 48 random, 25-m transects along a 12-km reach of the Cheoah River in the Appalachian Mountains of North Carolina. We sampled woody plants every 5 m using a modified point-centered quarter method to estimate relative abundance and to quantify browsing by Beaver. We used a mixed linear model to determine Beaver forage selection on the 9 most abundant plant species and Virginia Spiraea. We recorded 984 plants of 58 woody species (55 native, 3 nonnative). Beaver browsed 24% of the woody species sampled and 8% of all stems. This finding suggests that the overall effects of browsing in this community were relatively low, likely because of the high gradient and turbulent nature of the Cheoah River. Relative stem abundance and location along the river did not differentially affect local levels of browsing. However, Beaver were selective foragers at both the species and individual-plant level. Of the 9 most abundant species, Carpinus carolinana (Musclewood), Liquidambar styraciflua (Sweetgum), and Alnus serrulata (Tag Alder) were selected most often; Lindera benzoin (Spicebush), Virginia Spiraea, Cornus amomum (Silky Dogwood), and Ligustrum sinense (Chinese Privet) were moderately selected. Least frequently selected species were Rhododendron maximum (Rosebay Rhododendron), Leucothoe fontanesiana (Doghobble), and Xanthorhiza simplicissima (Yellowroot). Browsing appeared to have a positive effect on both the invasive nonnative shrub, Chinese Privet, and the rare Virginia Spiraea by stimulating asexual reproduction and inducing plants to spread through suckering. This study demonstrates the importance of understanding the reproductive strategies of woody plants when gauging the community-wide effects of foraging by Beaver, particularly when an invasive plant species is present. Introduction Castor canadensis Kuhl (Beaver) is a generalist herbivore that feeds on most species of woody plants and numerous herbaceous and aquatic plants (Jenkins and Busher 1979, Rosell et al. 2005). They are selective foragers on woody plants (e.g., 1Department of Environmental Studies, University of North Carolina at Asheville, Asheville, NC 28804. 2Department of Biology, University of North Carolina at Asheville, Asheville, NC 28804. 3Department of Mathematics, University of North Carolina at Asheville, Asheville, NC 28804. *Corresponding author - crrossell@aol.com. Manuscript Editor: Renn Tumlison Southeastern Naturalist C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 650 Belovsky 1984, Busher 1996, Jenkins 1975, Svendsen 1980) and can modify the species composition and structure of plant communities (Raffel et al. 2009, Rosell et al. 2005). However, forage use by Beaver can vary widely depending on a variety of factors that affect plants, including geographic location, topography, and associated hydrology (Rosell et al. 2005). The effects of Beaver on plant communities are related to the assemblage of plant species that are present (Jones et al. 1997). Because they forage selectively, Beaver can alter habitats to favor certain types of plants (Raffel et al. 2009, Rosell et al. 2005). For example, in forested systems, browsing by Beaver may cause a shift in composition of woody plant communities towards early-successional species (Fryxell 2001, Pastor and Naiman 1992, Rosell et al. 2005). In addition, long-term foraging by Beaver can decrease the diversity of woody plants and alter the composition to less palatable species (Fryxell 2001, Raffel et al. 2009, Rosell et al. 2005). Disturbances related to Beaver may also increase the potential for invasion of exotic plant species (Anderson et al. 2005, Bandano et al. 2007), and thereby threaten rare species in a community (Lake and Leishman 2004). However, to our knowledge no studies have examined forage selection by Beaver in a temperate forest that contains a dominant, nonnative woody plant, and only one study has examined the impacts of Beaver on a rare woody plant (Rossell et al. 2013). To understand the impacts of Beaver on plant communities, it is important to determine which plants they select as browse (Rosell et al. 2005). Understanding the foraging patterns of Beaver in shrub dominated-communities presents unique challenges because each plant may have multiple stems at heights that are accessible to Beaver. The most common method for determining Beaver-forage selection of woody plants is by counting only browsed and unbrowsed basal stems (e.g., Belovosky 1984, Brzyski and Schulte 2009, Fryxell 2001, Jenkins 1975, Raffel et al. 2009, Rossell et al. 2013). However, this approach may be problematic when focusing on a shrub community because different plant growth forms (i.e., shrubs, trees) vary in the number of stems available as forage. Thus, considering only basal stems may underestimate the extent of foraging on shrubs and bias forage selectivity towards trees. Therefore, a method that counts both basal and lateral stems as available forage may more accurately reflect the forage selection of Beaver in a shrub community. Spiraea virginiana (Virginia Spiraea) is a rare shrub of the southern Blue Ridge and Appalachian Plateau physiographic provinces (Ogle 1991a). It has a limited range and occurs in only 7 states: Georgia, Kentucky, North Carolina, Ohio, Tennessee, Virginia, and West Virginia (USFWS 1992). Virginia Spiraea is currently listed as federally threatened and has a global ranking of G2, or globally imperiled (Buchanan and Finnegan 2010). The species is considered endangered in North Carolina (Buchanan and Finnegan 2010), with only 9 known populations (C. Wells, USFWS, Asheville, NC, pers. comm.). Virginia Spiraea is a disturbance-dependent species, occurring along high-gradient sections of second- and third-order streams and rivers (Anders and Murrell 2001, Ogle 1991b, USFWS 1992). This species Southeastern Naturalist 651 C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 is threatened for a variety of reasons including habitat loss and competition from exotic woody plants (Ogle 1991b, USFWS 1992). Rossell et al. (2013) reported that Virginia Spiraea is a preferred forage species of Beaver along the Cheoah River in North Carolina. However, that study was limited to a stretch of river where nonnative woody species were a minor component of the shrub community—a situation atypical of many alluvial forests of the southern Appalachian Mountains where nonnative woody species are often abundant (Drake et al. 2003, Merriam 2003). The purpose of our study was to use both basal and lateral stem counts to determine forage selection by Beaver of the most abundant woody species in a riparian shrub community that included Virginia Spiraea and other native and nonnative woody species as potential forage. Methods Study area We conducted our study along a 12-km reach of the Cheoah River in Graham County, NC (elevation range = 386–524 m; N 35°24', W 83°53'). This section of river is part of the Cheoah River Floodplain Significant Natural Heritage Area, designated nationally significant because it contains numerous rare plant and animal species (Schwartzman 2012). Virginia Spiraea was first documented along the Cheoah River in 1940 (North Carolina Natural Heritage Program [NCNHP], Raleigh, NC, unpubl. data). However, in that same year, the population was considered extirpated because of road building (USFWS 1992). Surveys in 1986 indicated that this population was still extirpated (Ogle 1991a), but Virginia Spiraea was rediscovered in 2000 along the river (NCNHP, unpubl. data). Currently, 23 subpopulations (defined as a grouping of plants, regardless of the number of genets; Brzyski and Culley 2011), which vary in size from 1 small plant to plants covering more than 20 m2, occur along the riparian zone of the river. The Cheoah River is a dam-controlled, high-gradient system, with large-boulder and pool habitats interspersed with Class III and Class IV rapids (NCDOC 2007). From 1928 to 2005, only minimal flows were maintained in the Cheoah River to maximize production of hydroelectric power (Dilts et al. 2005). In 2005, a more natural flow regime was established to restore the river’s ecological communities (Dilts et al. 2005). Current flow rates include monthly base flows of 1.1–2.8 m3s-1 (40–100 cfs) and approximately 20 annual high-flow events that peak around 28 m3s-1 (1000 cfs; FERC 2005). The study area’s average annual precipitation is 167.7 cm, average annual high temperature is 20.3 °C, and average annual low temperature is 6 °C (National Climatic Data Center 2011). The riparian corridor we studied consisted of 2 habitats. The downstream section (9.2 km) is narrow and adjacent to US Highway 129. It provided habitat for 19 subpopulations of Virginia Spiraea and consisted of a small alluvial floodplain, with a scour zone extending several meters up steep banks characterized by large boulders and rip-rap. The plant community was second-growth, Montane Alluvial Forest (Schafale and Weakley 1990), with a partially open canopy dominated by Liriodendron tulipifera (Tulip Poplar), Acer rubrum (Red Maple), Southeastern Naturalist C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 652 and Platanus occidentalis (American Sycamore). The shrub layer was sparse to moderately dense and dominated by species including Cornus amomum (Silky Dogwood), Alnus serrulata (Tag Alder), Carpinus caroliniana (Musclewood), and Xanthorhiza simplicissima (Yellowroot). The herb layer was sparse and included Osmunda regalis L. (Royal Fern), Polystichum acrostichoides (Michx.) Schott (Christmas Fern), and Lespedeza cuneata (Dum. Cours.) G. Don (Sericea Lespedeza). Vines were prevalent and included Toxicodendron radicans (L.) Kuntze (Poison Ivy), Pueraria lobata Willd. (Kudzu), and Celastrus orbiculatus Thunb. (Oriental Bittersweet). The upstream section (2.8 km), where 4 subpopulations of Virginia Spiraea occurred, generally had a wider floodplain, with a substrate comprised mostly of sand and gravel. It was adjacent to Joyce Kilmer Road and a US Forest Service road. The vegetation was similar to the downstream section, but had a higher abundance of species typical of Acidic Cove Forests (Schafale and Weakley 1990), including Betula lenta L. (Sweet Birch), Tsuga canadensis (Eastern Hemlock), Rhododendron maximum (Rosebay Rhododendron), and Leucothoe fontanesiana (Doghobble). The shrub layer was moderately dense, and Ligustrum sinense (Chinese Privet) and Arundinaria gigantea (Walt.) Muhl. (River Cane) dominated some areas. The herb layer was sparse to moderately dense and consisted of grasses and forbs. Little is known about the Beaver population along the Cheoah River. Beaver were thought to be extirpated in North Carolina by 1897 and were reintroduced to the western portion of the state in 1956 (NCWRC 2011). Beaver are currently considered common, inhabiting most watersheds in North Carolina (NCWRC 2011). At the time of our study, 1 lodge and 1 bank den occurred in our study area. Data collection During the summer of 2011, we used a modified point-centered quarter method (Cottam and Curtis 1956) to estimate the relative abundance of woody plants in the shrub layer and quantify the extent of browsing by Beaver. We divided the study area into seventy-two 167-m-long segments. Within each segment, we randomly established two 25-m transects along the center of the scour zone using a random numbers table to generate a downstream starting location. Because Beaver do not typically forage on the steep, rocky banks of the Cheoah River (Rossell et al. 2013), we located transects in areas where the slope less than 15%. Also, because plant composition compared to surrounding areas was different where the US Forest Service had conducted exotic pest control, we chose untreated locations for our transects. Using these criteria, we were able to establish 48 transects, covering approximately 9% of the study area. We used a 25-m tape to set up transects parallel to the river and sampled 6 points at 5-m intervals along each transect. At each sampling point, we established 4 quarters by centering a 5-m PVC pipe perpendicular to the measuring tape. Within each quarter, we located the nearest woody plant (less than 2.5 cm basal diameter) up to 2.5 m from the center point, identified it to species, and counted the number of browsed and unbrowsed stems ≤10 cm above the substrate (the approximate height Southeastern Naturalist 653 C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 that Beaver cut stems at the study site). This height limit allowed an estimate of the number of available stems (i.e., both basal and lateral stems) Beaver would typically consider as forage. If an individual plant occupied more than one quarter, we counted the number of stems (browsed and unbrowsed) in all quarters where it occurred, and recorded the totals in the quarter where the stems were closest to the center point. In the remaining quarters, we sampled the next closest individual plant. We recorded only 6 Virginia Spiraea plants using this method and subsequently decided to augment the data by sampling all plants of this species that occurred within 2.5 m of the transects, which added 4 plants to the sample size. Because plants that propagate via underground structures are difficult to delineate, we defined an individual plant as all basal stems that occurred within 50 cm of each other. We identified stems browsed by Beaver by sharp, angled cuts that usually included incisor marks (Lesica and Miles 2004). Data analysis Initially, we fit 2 generalized linear mixed models to the data. The response variable for each model was the proportion of stems browsed for each plant of the 9 most abundant species and Virginia Spiraea. We included only the 9 most abundant species in the analysis because each one had a sample size less than 30, and together represented the majority of the total stems available as forage. The models incorporated a binomial distribution for number of stems browsed with a logit-link function. To avoid pseudo-replication, each model included transect and plant as random effects to account for potential autocorrelation of browsing patterns within transects and stems within plants. One model had stem abundance and species as fixed effects, whereas the other model had only species as a fixed ef fect. The equation for the models was of the form: log (p / [1 - p]) = intercept + random effects + fixed effects + random error, where p represents the probability that an individual stem was browsed. Because the model with two fixed effects indicated stem abundance was not significantly related to browsing (P = 0.64), we used the model with species as the fixed effect for all subsequent analyses. For each species, we estimated the proportion of stems browsed by reverse-transforming the species-effect estimate using the equation: estimated proportion browsed = exp (species-effect estimate) / (1 + exp [species-effect estimate]). We then used the estimated proportions browsed to determine the forage-selection rank of each species. We used SAS version 9.2 (SAS Institute, Cary, NC) for all statistical analyses and considered results significant at alpha = 0.05. Results We recorded a total of 984 plants of 58 species (55 native, 3 nonnative; Appendix 1); 24% of the plants were browsed by Beaver. We assessed 7433 stems, of which 8% were browsed by Beaver. The 9 most abundant species represented 81% of the total stems and included 8 native and 1 nonnative species (Table 1). Virginia Spiraea (n = 6) ranked 30th in relative abundance and accounted for less than 1% of the total plants. Nonnative plants accounted for 19% of the stems. The nonnative species Chinese Privet was 3rd in relative abundance, equal in abundance to the native Southeastern Naturalist C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 654 species Musclewood (Table 1). Chinese privet represented 6% of the total plants and 16% (n = 1190) of the total stems. We also documented the non-native shrubs Rosa multiflora (Multiflora Rose), which accounted for 2.5% (n = 24) of the total plants, and Albizia julibrissin Durazz., which accounted for 0.001% (n = 1) of the total plants (Appendix 1). Relative amounts of foraging on stems differed among the 9 most abundant species and Virginia Spiraea (P < 0.0001), and varied among individual plants within species (estimated variance due to plant effect = 1.45, standard error of estimated variance = 0.21). Amounts of foraging were not affected by location of transect (estimated variance due to transect effect = 0.12, standard error of estimated variance = 0.12). After adjusting for plant and transect effects using the linear model, Musclewood, Sweetgum, and Tag Alder were the forage species Beaver selected most often; Spicebush, Virginia Spiraea, Silky Dogwood, and Chinese Privet were moderately selected. Least-often selected were Rosebay Rhododendron, Doghobble, and Yellowroot (Table 1). Discussion With the exception of Virginia Spiraea, the study area’s native woody plant species assemblage is typical of Montane Alluvial Forests in North Carolina (Schafale and Weakley 1990). Virginia Spiraea’s rarity is reflected in the finding that this species occurred only 6 times in our sampling and accounted for less than 1% of the total plants. The nonnative component of the community consisted of 3 species, which comprised 8% of the total woody plants. Chinese Privet tied for third in the list of the most abundant species in our samples. Nonnative species, particularly Chinese Privet and Multiflora Rose, have become increasingly common in riparian communities of the southern Appalachian Mountains because of their prolific reproductive abilities (Drake et al. 2003, Merriam 2003). Table 1. Estimated proportion of stems browsed by Beaver for the 9 most abundant woody species and Virginia Spiraea along the Cheoah River, Graham County, NC, summer 2011. Plants = number of plants recorded at 5-m intervals along forty-eight 25-m transects, PB = proportion browsed (total stems browsed of a species / total stems of a species x 100), APB = adjusted proportion browsed (calculated by reverse transforming the fixed effect [species] of the mixed logistic regression model), LCL = 95% lower confidence limit of APB x 100, UCL = 95% upper confidence limit of APB x 100. Mean # Selection Species Plants stems PB APB LCL UCL rank Carpinus caroliniana 58 2.3 24.8 17.4 10.2 28.0 1 Liquidambar styraciflua 36 2.2 19.2 17.0 8.5 31.2 2 Alnus serrulata 71 3.3 17.9 16.7 10.7 25.2 3 Lindera benzoin 31 3.9 20.7 14.4 7.0 27.3 4 Spiraea virginiana 10 17.2 23.3 11.5 3.9 29.3 5 Cornus amomum 258 4.2 10.1 8.3 6.2 11.1 6 Ligustrum sinense 58 20.3 9.4 6.4 3.6 11.0 7 Rhododendron maximum 32 7.8 9.2 5.7 2.6 11.9 8 Leucothoe fontanesiana 32 12.8 4.7 3.3 1.5 7.1 9 Xanthorhiza simplicissima 46 55.4 1.8 1.2 0.6 2.3 10 Southeastern Naturalist 655 C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 The high species richness (58 species) at the study site is similar to the richness (59 species) reported in Rossell et al. (2013); both results reflect the high diversity of woody plants in the southern Appalachian Mountains (Whitaker 1956). This high species richness is particularly evident compared to other communities inhabited by Beaver. For example, reported values of woody plant species richness include 38 in Louisiana bayous (Chabreck 1958), 35 in North Carolina wetland communities (Bartel et al. 2010), 15 in hardwood forests around southeastern Ohio impoundments (Voelker and Dooley 2008), and 14 in riparian forests around ponds at Isle Royale National Park in Michigan (Belovsky 1984). The impacts of Beaver along the Cheoah River were relatively low during our study. Twenty-four percent of all woody plants and 8% of all stems were browsed by Beaver, which is substantially less than levels reported in other studies. Northcott (1971) reported 73–76% of woody plants were browsed around Beaver colonies in Newfoundland, and Voelker and Dooley (2008) reported 55% and 36% of woody plants were browsed around active and abandoned Beaver impoundments in Ohio, respectively. Howard and Larson (1985) and Slough and Sadlier (1977) reported that high-gradient streams support fewer Beaver than low-gradient streams. Therefore, because of the high gradient and turbulent nature of the Cheoah River, it was not surprising that the extent of Beaver browsing at our study site was less than in communities adjacent to low-gradient streams and impoundments. Beaver are known to be selective foragers (e.g., Chabreck 1958, Jenkins 1975, Raffel et al. 2009, Shadle et al. 1943, Svendson 1980) and the significant species effect in our study confirmed this finding. That our estimated variance of the plant effect was almost 7 times the standard error also indicated that Beaver are selective at the individual-plant level. This selectivity may be related to various factors. For example, Beaver may prefer plants with larger stem diameters over plants with small stem diameters. This suggestion is supported by McGinley and Whitman (1985), who found that the mean diameter of Populus fremontii Watson (Fremont Cottonwood) stems cut by Beaver was 98% greater than uncut stems. Age and browsing history of plants also may affect Beavers’ forage preferences (Basey et al. 1988, 1990). In certain plants, browsing by Beaver is known to stimulate production of root suckers that contain higher levels of secondary compounds that deter Beaver foraging (Basey et. al. 1988, 1990; Bryant and Kuropat 1980). Studies have reported that variables such as proximity to a den (Fryxell and Doucet 1993, Jenkins 1980, McGinley and Whitham 1985), density of woody plants (Brzyski and Schulte 2009, Fryxell and Doucet 1993), stream width and gradient (Beier and Barrett 1987, Curtis and Jensen 2004, Howard and Larson 1985), and bank steepness and floodplain width (MacCracken and Lebovitz 2005) can affect local Beaver foraging activities. In contrast, our findings that Beaver browsing was not affected by transect location (estimated variance of transect effect was equal to the standard error) or relative stem abundance indicate that these attributes did not differentially affect local levels of browsing at our study site. However, Rossell et al. (2013) did report a transect effect. We are uncertain of the reason for this difference between the results of our studies because the design of both was similar; the Southeastern Naturalist C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 656 disparity may be related to the different methods used to quantify Beaver foraging (i.e., plant level versus stem level). The finding that the 9 most abundant species accounted for 81% of the total stems supports our use of the most abundant plants as an index to evaluate the effects of Beaver in this community. The selection ranks in our study are in general agreement with other studies representing the same species (Brenner 1962, Brzyski and Schulte 2009, Chabreck 1958, Northcott 1971, Jenkins 1979, Raffel et al. 2009, Rossell et al. 2013, Shadle et al. 1943, Svendson 1980, Voelker and Dooley 2008). The two species selected most often in our study, Musclewood and Sweetgum, are consistently reported among Beavers’ preferred forage species throughout the eastern US, including Louisiana (Chabreck 1958), Georgia (Bryzyski and Schulte 2009), Massachusetts (Jenkins 1979), New York (Shadle et al. 1943), and Pennsylvania (Brenner 1962). High concentrations of secondary chemicals in plants also can negatively affect forage selection by Beaver (Bryant and Kuropat 1980, Bryant et al. 1991). The low use of Yellowroot, Doghobble, and Rosebay Rhododendron support this suggestion because all 3 of these species can contain high levels of secondary metabolites that inhibit herbivory (Hughes 2011, Nielson 1980, Schmeller et al. 1997). It has been reported that browsing by Beaver can stimulate clonal growth in certain plants (Baker et al. 2005, McGinley and Whitman 1985). In our study, this effect is apparent in Chinese Privet and Virginia Spiraea, both of which had a relatively large number of stems per plant and were moderately browsed by Beaver (Table 1). Chinese Privet is highly invasive and known to vigorously respond to cutting through root sprouting (Drake et al. 2003). This species often forms dense stands (Drake et al. 2003, Miller 2003) and can reduce the richness and abundance of native woody plants when it is prevalent in an area (Wilcox and Beck 2007). Virginia Spiraea reproduces almost exclusively by asexual reproduction through rhizomatous growth (Anders and Murrell 2001, Ogle 1991b). Chinese Privet has been recognized as a threat to Virginia Spiraea (USFSW 1992) and other native flora (Wilcox and Beck 2007). Because browsing may exacerbate the threat of Chinese Privet by causing it to spread through increased root sprouting, we recommend that control measures be implemented to reduce the abundance of Chinese Privet, particularly in areas where this species co-occurs with rare plants such as Virginia Spiraea. Conclusions This study is the first to examine the use of woody plants for foraging by Beaver in a riparian shrub community dominated by both native and nonnative species and containing a rare woody species. The overall levels of browsing in this community were low compared to other communities inhabited by Beaver, likely because of the high gradient and turbulent nature of the Cheoah River. Our findings indicate that relative stem abundance and location along the river did not differentially affect local levels of browsing. Beaver were selective foragers at both the species and the individual-plant level. Of the 9 most abundant species, Musclewood and Southeastern Naturalist 657 C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 Sweetgum were the most utilized forage species, whereas Yellowroot, Doghobble, and Rosebay Rhododendron were the least utilized. Chinese Privet, a highly invasive nonnative shrub, and Virginia Spiraea, a federally threatened shrub, were both moderately utilized by Beaver. Browsing by Beaver appeared to have a positive effect on both Chinese Privet and Virginia Spiraea by stimulating asexual reproduction and inducing plants to spread through underground propagation. Our study demonstrates the importance of understanding the reproductive strategies of woody plants when gauging the effects of foraging by Beaver on a community, particularly when an invasive species is present. Acknowledgments We thank K. Keen, J. McKenna, and M. Searels for field assistance; E. Schwartzman for help with GIS; and I. Rossell for providing helpful comments that improved the manuscript. Funding was provided by a grant from the North Carolina Department of Transportation. Literature Cited Anders, M.A., and Z.E. Murrell. 2001. Morphological, molecular, and biogeographical variation within the imperiled Virginia Spiraea. Castanea 66:24–41. Anderson, C.B., C.R. Griffith, A.D. Rosemond, R. Rozzi, and O. Dollenz. 2005. The effects of invasive North American Beavers on riparian plant communities in Cape Horn, Chile: Do exotic Beavers engineer differently in sub-Antartic ecosystems? Biological Conservation 128:467–474. Badano, E.I., E. Villarroel, R.O. Bustamante, P.A. Marquet, and L.A. Cavieres. 2007. Ecosystem engineering facilitates invasions by exotic plants in high-Andean ecosystems. Journal of Ecology 95:682–688. Baker, B.W., H.C. Ducharme, D.C.S. Mitchell, T.R. Stanley, and H.R. Peinett. 2005. Interaction of Beaver and Elk herbivory reduces standing crop of willow. Ecological Applications 15:110–118. Bartel, R.A., N.M. Haddad, and J.P. Wright. 2010. Ecosystem engineers maintain a rare species of butterfly and increase plant diversity. Oikos 119:883–890. Basey, J.M., S.H. Jenkins, and P.E. Busher. 1988. Optimal central-place foraging by Beavers: Tree-size selection in relation to defensive chemicals of Quaking Aspen. Oecologia 76:278–282. Basey, J.M., S.H. Jenkins, and G.C. Miller. 1990. Food selection by Beavers in relation to inducible defenses of Populus tremuloides. Oikos 59:57–62. Belovsky, G.E. 1984. Summer-diet optimization by Beaver. American Midland Naturalist 111:209–222. Beier, P., and R.H. Barret. 1987. Beaver-habitat use and impact in Truckee River basin, California. Journal of Wildlife Management 51:794–799. Brenner, F.J. 1962. Foods consumed by Beavers in Crawford County, Pennsylvania. Journal of Wildlife Management 26:104–107. Bryant, J.P., and P.J. Kuropat. 1980. Selection of winter forage by subartic browsing vertebrates: The role of plant chemistry. Annual Review of Ecology and Systematics 11:261–285. Bryant, J.P., F.D. Provenza, J. Pastor, P.B. Reichardt, T.P. Clausen, and J.T. duToit. 1991. Interactions between woody plants and browsing mammals mediated by secondary metabolites. Annual Review of Ecology and Systematics 22:431–446. Southeastern Naturalist C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 658 Brzyski, J.R., and T.M. Culley. 2011. Genetic variation and clonal structure of the rare, riparian shrub Spiraea virginiana (Rosaceae). Conservation Genetics 12:1323–1332. Brzyski, J.R., and B.A. Schulte. 2009. Beaver (Castor canadensis) impacts on herbaceous and woody vegetation in southeastern Georgia. American Midland Naturalist 162:74–86. Buchanan, M.F., and J.T. Finnegan. 2010. Natural Heritage Program list of rare plant species of North Carolina. North Carolina Natural Heritage Program, Raleigh, NC. 125 pp. Busher, P.E. 1996. Food-caching behavior of Beavers (Castor canadensis): Selection and use of woody species. American Midland Naturalist 135:343–348. Chabreck, R.H. 1958. Beaver–forest relationships in St. Tammany Parish, Louisiana. Journal of Wildlife Management 22:179–183. Cottam, G., and J.T. Curtis. 1956. The use of distance measures in phytosociological sampling. Ecology 37:451–460. Curtis, P.D., and P.G. Jensen. 2004. Habitat features affecting Beaver occupancy along roadsides in New York state. Journal of Wildlife Management 68:278–287. Dilts, E., P. Leonard, and D. Hill. 2005. Development of an integrated flow-regime recommendation for the Cheoah River, NC. Proceedings of the 2005 Georgia Water Resources Conference. K.J. Hatcher (Ed.). Institute of Ecology, University of Georgia, Athens, GA. Drake, S.J., J.F. Weltzin, and P.D. Parr. 2003. Assessment of non-native invasive plant species on the United States Department of Energy Oak Ridge National Environmental Research Park. Castanea 68:15–30. Federal Energy Regulatory Commission (FERC). 2005. Tapoco Hydroelectric Project (FERC No. 2169), Cheoah River bypassed reach gravel-enhancement plan. Alcoa Power Generating, Tapoco Division, Badin, NC. 9 pp. Fryxell, J.M. 2001. Habitat suitability and source–sink dynamics of Beavers. Journal of Animal Ecology 70:310–316. Fryxell, J.M., and C.M. Doucet. 1993. Diet choice and functional response of Beavers. Ecology 74:1297–1306. Howard, R.J., and J.S. Larson. 1985. A stream-habitat classification-system for Beaver. Journal of Wildlife Management 49:19–25. Hughes, N.M. 2011. Winter-leaf reddening in “evergreen” species. New Phytologist 190:573–581. Jenkins, S.H. 1975. Food selection by Beavers. Oecologia 21:157–173. Jenkins, S.H. 1979. Seasonal and year-to-year differences in food selection by Beavers. Oecologia 44:112–116. Jenkins, S.H. 1980. A size–distance relation in food selection by Beavers. Ecology 61:740–746. Jenkins, S.H., and P.E. Busher. 1979. Castor canadensis. Mammalian Species 120:1–8. Jones, C.J., J.H. Lawton, and M. Shachak. 1997. Positive and negative effects of organisms as physical ecosystem engineers. Ecology 78:1946–1957. Lake, J.C., and M.R. Leishman. 2004. Invasion success of exotic plants in natural ecosystems: The role of disturbance, plant attributes, and freedom from herbivores. Biological Conservation 117:215–226. Lesica, P., and S. Miles. 2004. Beavers indirectly enhance the growth of Russian Olive and Tamarisk along eastern Montana rivers. Western North American Naturalist 64:93–100. MacCracken, J.G., and A.D. Lebovitz. 2005. Selection of in-stream wood structures by Beaver in the Bear River, southwestern Washington. Northwestern Naturalist 86:49–58. McGinley, M.A, and T.G. Whitman. 1985. Central-place foraging by Beavers (Castor canadensis): A test of foraging predictions and the impact of selective feeding on the growth-form of Cottonwoods (Populus fremontii). Oecologia 66:558–562. Southeastern Naturalist 659 C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 Merriam, R.W. 2003. The abundance, distribution, and edge associations of six non-indigenous, harmful plants across North Carolina. Journal of the Torrey Botanical Society 130:283–292. Miller, J.H. 2003. Nonnative plants of southern forests: A field guide for identification and control. USDA Forest Service, Southern Research Station General Technical Report SRS-62, Athens, GA. National Climatic Data Center. 2011. Annual precipitation and temperatures, Robbinsville, NC. Available online at http://www.ncdc.noaa.gov/oa/climate/climatedata.html. Accessed December 2011. Nielson, D.G. 1980. Strategy for minimizing insect damage to rhododendron. Pp. 305–318, In J.L. Lutegen and M.E. O’Brien (Eds.). Contributions toward a Classification of Rhododendron. Allen Press, Lawrence, KS. North Carolina Department of Commerce (NCDOC). 2007. The Cheoah River, Graham County, NC (brochure). Graham County Travel and Tourism Authority, Robbinsville, NC. 2 pp. North Carolina Wildlife Resources Commission (NCWRC). 2011. Beaver management in North Carolina. Available online at http://www.ncwildlife.org/wildlife_species_con/ wsc_BMNC.htm. Accessed December 2011. Northcott, T.H. 1971. Feeding habits of Beaver in Newfoundland. Oikos 22:407–410. Ogle, D.W. 1991a. Spiraea virginiana Britton: I. Delineation and distribution. Castanea 56:287–296. Ogle, D.W. 1991b. Spiraea virginiana Britton: II. Ecology and species biology. Castanea 56:297–303. Pastor, J., and R.J. Naiman. 1992. Selective foraging and ecosystem processes in boreal forests. American Naturalist 139:690–705. Raffel, T.R., N. Smith, C. Cortright, and A.J. Gatz. 2009. Central-place foraging by Beavers (Castor canadensis) in a complex lake habitat. American Midland Naturalist 162:62–73. Rosell, F., O. Bozser, P. Collen, and H. Parker. 2005. Ecological impact of beavers Castor fibre and Castor canadensis and their ability to modify ecosystems. Mammal Review 35:248–276. Rossell, C.R., Jr., K.S. Selm, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch. 2013. Impacts of Beaver on the federally threatened Virginia Spiraea (Spiraea virginiana) along the Cheoah River, North Carolina. Southeastern Naturalist 12:439–447. Schafale, M.P., and A.S. Weakley. 1990. Classification of the natural communities of North Carolina. North Carolina Natural Heritage Program, Division of Parks and Recreation, NC Department of Environment, Health, and Natural Resources, Raleigh, NC. 325 pp. Schmeller, T., B. Latz-Bruning, and M. Wink. 1997. Biochemical activities of berberine, palamatine, and sanguinarine mediating chemical defense against microorganisms and herbivores. Phytochemistry 44:257–266. Schwartzman, E. 2012. Site survey report: Choeah River floodplain (unpublished). North Carolina Heritage Program, Division of Parks and Recreation, NC Department of Environment, Health, and Natural Resources, Raleigh, NC. 2 pp. Shadle, A.R., A.M. Nauth, E.C. Gese, and T.S. Austin. 1943. Comparison of tree cuttings of six Beaver colonies in Allegany State Park, New York. Journal of Mammalogy 24:32–39. Slough, B.G., and M.F. Sadleir. 1985. A land-capability classification system for Beaver (Castor canadensis Kuhl). Canadian Journal of Zoology 55:1324–1335. Svendsen, G.E. 1980. Seasonal change in feeding patterns of Beaver in southeastern Ohio. Journal of Wildlife Management 44:285–290. Southeastern Naturalist C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 660 United States Fish and Wildlife Service (USFWS). 1992. Virginiana Spiraea (Spiraea virginiana Britton) recovery plan. Newton Corner, Newton, MA. 47 pp. Voelker, B.W., and J.L Dooley, Jr. 2008. Impact by North American Beaver (Castor canadensis) on forest-plant composition in the Wilds, a surface-mined landscape in southeastern Ohio. Ohio Journal of Science 108:9–15. Whitaker, R.H. 1956. Vegetation of the Great Smoky Mountains. Ecological Monographs 26:1–80. Wilcox, J., and C.W. Beck. 2007. Effects of Ligustrum sinense Lour. (Chinese Privet) on abundance and diversity of songbirds and native plants in a southeastern nature preserve. Southeastern Naturalist 6:535–550. Southeastern Naturalist 661 C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 Appendix 1. List of woody plant species sampled for Beaver browsing in forty-eight 25-m transects along the Cheoah River, Graham County, NC, summer 2011. Browsed by Species Common Name Number of plants Beaver Acer rubrum L. Red Maple 15 Yes Acer saccharum L. Sugar Maple 1 No Albizzia julibrissin Durazz. Mimosa 1 Yes Alnus serrulata (Ait.) Willd. Tag Alder 71 Yes Aralia spinosa L. Hercules’ Club 7 No Asimina triloba (L.) Dunal Pawpaw 10 Yes Betula nigra L. River Birch 1 No Calycanthus floridus L. Hairy Allspice 15 Yes Carpinus caroliniana Walt. Musclewood 58 Yes Carya cordiformis (Wang.) Koch Bitternut Hickory 1 No Carya glabra (Mill.) Sweet Pignut Hickory 12 Yes Carya tomentosa Nutt. Mockernut Hickory 1 No Carya sp. Hickory 1 No Cornus amomum Mill. Silky Dogwood 258 Yes Cornus florida L. Flowering Dogwood 2 No Corylus americana Walt. American Hazelnut 2 No Diospyros virginiana L. Persimmon 5 No Fraxinus americana L. White Ash 8 Yes Fraxinus pennsylvanica (Vahl) Fern Green Ash 5 Yes Halesia carolina L. Silverbell 25 Yes Hamamelis virginiana L. Witch-hazel 3 No Hydrangea arborescens L. Wild Hydrangea 12 Yes Ilex Montana T. & G. Mountain Holly 9 Yes Ilex opaca Ait. American Holly 7 Yes Itea virginica L. Sweet Spires 14 Yes Juglans nigra L. Black Walnut 1 No Kalmia latifolia L. Mountain Laurel 24 Yes Leucothoe fontanesiana Steud. Doghobble 32 Yes Ligustrum sinense Lour. Chinese Privet 58 Yes Lindera benzoin (L.) Blume Spicebush 31 Yes Liquidambar styraciflua L. Sweetgum 36 Yes Lirodendron tulipifera L. Tulip Poplar 8 Yes Magnolia fraseri Walt. Frazier Magnolia 1 No Magnolia tripetala L. Umbrella Magnolia 8 Yes Morus rubra L. Red Mulberry 1 No Nyssa sylvatica Marsh. Black Gum 25 Yes Pinus strobus L. White Pine 6 No Platanus occidentalis L. Sycamore 2 Yes Prunus serotina Ehrh. Black Cherry 1 No Pyrularia pubera Michx. Buffalo Nut 1 No Quercus alba L. White Oak 11 No Quercus coccinea Muenchh. Scarlet Oak 4 No Quercus rubra L. Red Oak 3 No Southeastern Naturalist C.R. Rossell, Jr., S. Arico, H.D. Clarke, J.L. Horton, J. Rhode Ward, and S.C. Patch 2014 Vol. 13, No. 4 662 Browsed by Species Common Name Number of plants Beaver? Rhododendron maximum L. Rhododendron 32 Yes Rhus copallina L. Winged Sumac 3 Yes Rhus glabra L. Smooth Sumac 7 Yes Robinia pseudoacacia L. Black Locust 3 No Rosa multiflora Thunb. Multiflora Rose 24 Yes Salix sp. Willow 2 Yes Sambucus canadensis L. Elderberry 2 No Sassafras albidum (Nutt.) Nees Sassafras 4 No Spiraea virginiana Britton Virginia Spiraea 6 Yes Tilia americana L. Basswood 1 Yes Tsuga canadensis (L.) Carr. Eastern Hemlock 1 No Vaccinium sp. Blueberry 2 Yes Viburnum dentatum L. Arrowwood 28 Yes Viburnum prunifolium L. Blackhaw 7 Yes Xanthorhiza simplicissima Marsh. Yellowroot 46 Yes