Southeastern Naturalist M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 506 2015 SOUTHEASTERN NATURALIST 14(3):506–516 Small-Mammal Mortality Caused by Discarded Bottles and Cans along a US Forest Service Road in the Cherokee National Forest M. Kevin Hamed1,* and Thomas F. Laughlin2 Abstract - Discarded containers along primary and secondary roadways trap and kill small mammals, and recently, small-mammal remains were found in containers along roadside pull-offs in more remote areas. We investigated the effects of containers along 7.9 km of a graveled, 2-lane forest service road in a remote area of the Cherokee National Forest, TN. We examined 2997 containers, of which 107 containers had 202 small-mammal skulls representing 8 species of mammals including Sorex fumeus (Smoky Shrew), Sorex longirostris (Southeastern Shrew), and Synaptomys cooperi (Southern Bog Lemming), which are deemed species of greatest conservation need and in need of management by the Tennessee Wildlife Resource Agency. Our observation of Southern Bog Lemming mortality due to container debris is the first report of collection of this species from bottles. We recorded the first observation of shrew mortality caused by aluminum cans, where 1% of all aluminum cans contained shrew skulls. Unlike previous studies, we quantified all possible containers available to trap small mammals and found glass disproportionately trapped more small mammals than plastic or aluminum. Additionally, we determined the orientation of bottle and can openings for all available containers and found that containers oriented upslope (>15°) were significantly more likely to be associated with the mortality of small mammals than containers in other orientations. We estimated that a mean of 25.6 small mammals/km were killed in discarded containers. By using bottling dates on containers, we also estimated containers were in place along the road an average of 2.16 ± 0.37 years before discovery. Only using modern bottles with dates, we conclude that the minimal potential impact on small-mammal populations was at least 973 small mammals killed per year in container debris along graveled category 3 and 4 forest service roads in the Cherokee National Forest. Introduction Roads may markedly affect small-mammal populations by creating barriers to movement, increasing mortality from vehicles, and increasing opportunities for predation (Oxley et al. 1974). Discarded bottles along roadways and roadside pull-offs are a source of additional mortality by acting as traps for small mammals (Benedict and Billeter 2004, Brannon et al. 2010). Bottles containing rodent and shrew remains were first found along British roadways in the 1960s (Morris and Harper 1965), and similar findings were reported for discarded bottles in Italy (Debernardi et al. 1997). Numerous studies in the US have also demonstrated smallmammal mortality based on the presence of skulls in bottles (Benedict and Billeter 1Virginia Highlands Community College, PO Box 828, Abingdon, VA 24212. 2Department of Biological Sciences, East Tennessee University, Johnson City, TN 37614. *Corresponding author - khamed@vhcc.edu. Manuscript Editor: Roger W. Perry Southeastern Naturalist 507 M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 2004, Brannon et al. 2010, Gerard and Feldhamer 1990, Pagels and French 1987). Small mammals most likely become trapped in bottles and die by starvation or drowning (Benedict and Billeter 2004). Previous studies investigated the impact of bottles along well-traveled primary and secondary highways (Benedict and Billeter 2004, Gerard and Feldhamer 1990, Pagels and French 1987). Discarded containers killed up to 183 vertebrates/km along primary roads (Benedict and Billeter 2004). Recently, containers near roadside pull-off sites along primary and secondary roads in mountainous areas of the Blue Ridge Mountains were identified as sources of mortality, with small-mammal remains found in 4.7% of all open bottles (Brannon and Bargelt 2013). Ten percent of all US roadways are located within national forest lands and over 2500 km are within the Cherokee National Forest (CNF), but mortality associated with discarded containers along national forest roadways has not been investigated, and the potential impact to small mammals is unknown (Forman and Alexander 1998). Shrews from the genus Sorex prefer mature forested habitat or edges and may be especially vulnerable to becoming trapped in discarded, open containers along forest service roads that often pass through forest stands (Brannon and Bargelt 2013, Ford et al. 1997, Laerm et al. 1999). Additionally, the southern Appalachian Mountains support one of the most diverse Soricidae faunas in North America (Berman et al. 2007, Ford et al. 2005, Laerm et al. 1999). Six species of small mammals reported from the CNF and previously captured in bottles from areas within the southern Appalachian Mountains have conservation status as designated by the Tennessee Wildlife Resource Agency (TWRA), Tennessee Natural Heritage Program (TNHP), and/or US Forest Service (USFS; Table 1; TNHP 2009, TWRA 2005, USDA 2004). Sorex cinereus (Masked Shrew), S. longirostris (Southeastern Shrew), and S. fumeus (Smoky Shrew) are classified each as a species of greatest conservation need and deemed in need of management by the TWRA as well as an S4 species (uncommon, but not rare) species by the TNHP Table 1. Conservation status of rare small mammals collected from bottles within the southern Appalachian Mountains that also occur within the CNF (Brannon and Bargelt 2013, Brannon et al. 2010). TWRA GCN = Tennessee Wildlife Resource Agency designation of species of greatest conservation need, TWRA D = TWRA deemed in need of management, X = status granted. TNHP SRANK = Tennessee Natural Heritage Program (TNHP) state rank. USFS FRANK = US Forest Service (USFS) forest ranking (USFS 2004). S2 = very rare and imperiled, S3 = vulnerable, S4 = uncommon, but not rare, F1 = extremely rare on the forest unit, and F3 = rare and uncommon on the forest unit. The present study documented the first bottle mortality for the Southern Bog Lemming. TWRA TNHP USFS Species Common name GCN TWRA D SRANK FRANK Sorex fumeus (Miller) Smoky Shrew X X S4 - S. cinereus (Kerr) Masked Shrew X X S4 F3 S. longirostris (Bachman) Southeastern Shrew X X S4 - S. hoyi (Baird) Pygmy Shrew X - S2 - Parascalops breweri (Bachman) Hairy-tailed Mole - X S3 F1 Synaptomys cooperi (Baird) Southern Bog Lemming X X S4 F1 Ochrotomys nuttalli (Harlan) Golden Mouse X - - - Southeastern Naturalist M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 508 (Table 1; TNHP 2009, TWRA 2005). Bottles have been identified as a source of mortality for both Smoky Shrews and Southeastern Shrews, with trap rates as high as 46.4/km along primary roads (Benedict and Billeter 2004, Brannon et al. 2010). Mortality associated with discarded containers along forest service roads could impact populations of uncommon small mammals. Our objective was to determine the impact of bottle and can debris on small mammals along a graveled double-lane forest service road in the CNF. Additionally, we quantified bottle and can attributes to determine if material type or color, orientation, or distance from the road influenced small-mammal mortality from discarded containers. Field-site Description Our study site was located in the CNF (Sullivan County, TN) along Flatwoods Road (USFS Road 87) at elevations of 618–695 m. We sampled a total road length of 7.9 km beginning 0.6 km from the national forest property boundary (36.46824ºN, 82.12302ºW) and ending 100 m past the junction with Big Creek Road (USFS Road 87D; 36.4918ºN, 82.0689ºW). This road section had 2 lanes and a graveled surface and shoulders, with a mean (± 1 SE) width of 7.0 ± 0.3 m. It is classified as a USFS category 4 road (moderate degree of user comfort at moderate travel speeds) for 7.5 km, with a few sections (total 0.4 km) classified as category 3 road (maintained for travel by prudent drivers in standard passenger cars; USFS 2005). The mean width of the herbaceous vegetated area between the road shoulder and forest edge was 4.2 ± 0.7 m. The majority of the study area was forest-edge habitat, and the time since last logging was ~50 years. The forest was mixed deciduous and southern Appalachian oak (Quercus spp.) forest. A powerline right-of-way created additional edge habitat along Flatwoods Road at a single crossing. Methods We sampled containers along Flatwoods Road from 2 April to 30 April 2004 (4.2 km) and from 25 March to 22 April 2005 (3.7 km). We examined all open containers within 10 m of the road edge and excluded closed containers that prevented bottle access by small mammals. To locate containers, we walked along the road edge and searched for containers visually and also examined containers we stepped on that were hidden by leaf litter. We recorded the container material, color, size, and distance from the road shoulder. We listed container size as the volume printed on the container (e.g., 355 ml/12 oz), and classified container material as glass, plastic, aluminum, or steel. We also noted the orientation of the container opening in relation to a level plane (i.e., an upslope bottle was at an angle with the opening ≥15° higher than the bottom of the bottle and was facing upslope). In order to separate skulls and skull fragments from fur and other debris, we poured container contents through a sieve and removed bones. We returned skulls to the laboratory for identification, but left all other container contents (e.g., water, invertebrates, debris, fur, and other natural material) at the collection site. Southeastern Naturalist 509 M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 To determine container parameters, we retrieved a random subset of glass bottles (n = 10; 355-ml glass bottles) and all aluminum cans (n = 3; 355 ml) that contained small-mammal remains. We used SPI dial calibers (Wiha model 31-415-3, Schonach, Germany) to measure the widest opening of the container and an Ohaus Scout Pro scale (model SPE 123, Parsippany, NJ) to determine empty container mass. To estimate the length of time bottles and cans were present in the forest, we recorded production dates found on container labels with small-mammal remains and took the difference from when we found the container. We quantified the number of small mammals/km/year by using dated bottles and cans containing small-mammal skulls. Our estimate of container longevity was somewhat greater than the actual period because bottles were not placed in the forest on the production date, nor did small mammals enter the bottle on the discovery date, but we utilized this approach to standardize our approximation of deposition dates and species-mortality rates. Also, dates were only available on bottles produced since the mid-1990s; older bottles lacked production dates. Thus, our mortality-rate estimates should be considered the minimum possible mortality without the impact of older bottles. We identified skulls in the laboratory under a Nikon SMZ-2T stereomicroscope (Melville, NY). Some skulls required removal of tissue by Dermestes maculatus De Geer (Dermestid Beetles), but most were clean. We utilized small-mammal dichotomous keys developed to identify animals with whole bodies and skulls (Barbour 1974, Linzey 1998). Additionally, we used information from previous morphological studies and species descriptions to identify skulls or skull fragments with teeth (French 1980, George et al. 1986, Owens 1984, Smolen 1981). We employed chi-square goodness-of-fit tests to examine potential associations of small-mammal mortality with material composition, color, or the positioning of the container opening (IBM 2011). We used each km searched as our sampling unit to determine both containers and skulls/km. We established the mean number of individual species/bottle using only containers with skulls of each species. Finally, we used linear regression to determine if observed mortality was a function of distance from the road (IBM 2011). Results We examined a total of 2997 containers (381.6 ± 33.1 containers/km), and material composition was 84.8% glass, 10.3% aluminum, 4.8% plastic, and 0.1% steel (Table 2). Container volumes ranged from 162.7 ml (5.5 oz) to 1182.9 ml (40 oz). Brown glass was the most common container type (64.9%), clear glass comprised 14.9%, and green glass comprised 4.9% of all containers found ( Table 2). A total of 107 bottles and cans (3.6% of all containers found) contained the remains of 202 small mammals, with a mean of 1.9 ± 0.2 small mammals per container with remains (Table 3). We detected most skulls (95.5%) in glass containers (Table 3). We collected the largest number of skulls found in a single container—8 skulls: 7 Blarina brevicauda Say (Northern Short-tailed Shrew) and a single Smoky Shrew—from a 650.5-ml brown-glass bottle. We recorded skulls of the smallest shrews (Smoky and Southeastern Shrews) in aluminum cans, which is the first Southeastern Naturalist M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 510 reported mortality in this container type. Aluminum cans contained 2.5% of all skulls, and 1.0% of all aluminum cans found contained skulls (Table 2). Mortality associated with all containers resulted in an average of 25.7 ± 5.2 small mammals/km. We identified 8 species (3 shrews and 5 rodents; Table 3) among the 202 small-mammal skulls collected. The largest percentage of skulls found (66.3%; n = 134) were Northern Short-tailed Shrews, the majority (69.4%) of which were in brown-glass bottles (Table 3). Of bottles that contained Northern Short-tailed Shrew skulls, there was a mean of 1.8 ± 0.2 skulls/container, and containers often trapped more than 1 individual (range = 1–7 individuals/bottle). The second most commonly detected species was the Smoky Shrew (n = 29; 14.4% of all skulls; Table 3). Brown-glass bottles contained the majority of the Smoky Shrews (58.6%), whereas aluminum cans trapped 13.8%. There were 1.3 ± 0.1 skulls/container in cans and bottles containing Smoky Shrews, also reflecting the common occurrence of multiple captures within a single container for this species. Southeastern Shrew skulls were relatively uncommon (2.5% of skulls), and mortality of these animals had no apparent associations with container type or coloration. We infrequently Table 3. Number of mammals found per container type along 7.5 km of class 4 and 0.4 km of class 3 US Forest Service roads in the Cherokee National Forest, Sullivan County, TN, from 2 April–30 April 2004 and 25 March–22 April 2005. Glass Plastic Aluminum Species Brown Clear Green Clear Green Cans Total Northern Short-tailed Shrew 93 33 6 2 - - 134 Smoky Shrew 17 5 3 - - 4 29 Southeastern Shrew 2 1 1 - - 1 5 White-footed Mouse 14 3 - 1 1 - 19 Deer Mouse 4 - - - - - 4 Woodland Vole 7 - - - - - 7 Southern Bog Lemming - 2 - - - - 2 House Mouse 1 - - - - - 1 Unidentified - 1 - - - - 1 Total 138 45 10 3 1 5 202 Table 2. Characteristics of containers found along 7.5 km of class 4 and 0.4 km of class 3 US Forest Service roads in the Cherokee National Forest, Sullivan County, TN, from 2 April–30 April 2004 and 25 March–22 April 2005. Al. = aluminum. Glass Plastic Al. Brown Clear Green Blue Total Clear Green White Total Cans Steel # of containers found 1945 445 147 2 2539 95 47 4 146 310 2 Containers with skulls 73 22 6 0 101 2 1 0 3 3 0 # of skulls/container type 138 45 10 0 193 3 1 0 4 5 0 % of all containers found 64.9 14.9 4.9 0.1 84.8 3.2 1.5 0.1 4.8 10.3 0.1 % with skulls 3.8 4.9 4.1 0.0 - 2.1 2.1 0.0 - 1.0 0.0 % of all skulls found in 68.3 22.2 5.0 0.0 - 1.5 0.5 0.0 - 2.5 0.0 each container type # of skulls/container type/km 21.90 7.14 1.59 0.00 - 0.48 0.16 0.00 - 0.79 0.00 Southeastern Naturalist 511 M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 found remains of Peromyscus leucopus Rafinesque (White-footed Mouse) in containers (n = 19; 9.5% of all skulls; Table 3). Of containers with White-footed Mice, 73.7% were in brown-glass bottles (Table 3). We recorded smaller numbers of P. maniculatus Wagner (Deer Mouse; n = 4), Microtus pinetorum Le Conte (Woodland Vole; n = 4), and Mus musculus L. (House Mouse; n = 1), and all were all found in brown-glass bottles. Clear-glass bottles contained Southern Bog Lemmings (n = 2; Table 3) and an unidentified skull, which appeared to be a Peromyscus sp. Glass containers trapped the largest number of small mammals. More glass containers and fewer aluminum or plastic containers were related to mortality than expected based on container availability (χ2 = 18.48, P = 0.004). Thus, glass appears to be disproportionately associated with trapping and mortality of small mammals. However, glass color was not related to small-mammal mortality (χ2 = 4.05, P = 0.26). Only Northern Short-tailed Shrews, Smoky Shrews, and White-footed Mice were found in sufficient numbers to determine if their occurrence was associated with a container type and color. Northern Short-tailed Shrew mortality was associated with brown-glass and clear-glass bottles but not with aluminum cans (χ2 = 20.15, P = 0.0002). However, mortality of Smoky Shrews and White-footed Mice was not associated with any particular type of container (χ2 = 2.23, P = 0.47; χ2 = 3.26, P = 0.35, respectively). We did not detect an association of container type for all Peromyscus spp. combined (χ2 = 5.42, P = 0.37). Containers with skulls averaged 4.24 ± 0.19 m from the road. There was a significant positive relationship between distance from the road shoulder and the presence of small-mammal remains (r = 0.76, P < 0.01). Containers found with the opening oriented upslope contained significantly more skulls (77.8%) than those that were perpendicular to the slope (18.5%) or those pointing downslope (3.7%; χ2 = 360.31, P < 0.01). The estimated mean time that a subsample of bottles and cans were present in the forest was 2.16 ± 0.37 years, based on containers that had production dates (n = 13). We found a total of 21 skulls in 13 dated bottles and cans. Given the age of the bottles, the number of remains found in them, and the distance surveyed, we estimated a minimum of 1.21 small mammals killed/km/year in containers that had small-mammal remains in them. The mean weight of ten 354.9-ml (12 oz) brown-glass bottles that contained skulls was 188.65 ± 2.55 g. The mean diameter of brown-glass bottle openings was 19.39 ± 0.01 mm. The mean weight of three 354.9 ml aluminum cans was 14.13 ± 0.19 g, and the mean opening diameter was 11.23 ± 0.01 mm. Discussion Small-mammal mortality from discarded containers has been observed on well-traveled paved roads and pull-off areas (Benedict and Billeter 2004, Brannon and Bargelt 2013, Brannon et al. 2010, Morris and Harper 1965). We found that containers along a graveled national forest road (Flatwoods Road) also contributed to small-mammal mortality. Past studies did not detect small-mammal remains in aluminum cans (Benedict and Billeter 2004, Gerard and Feldhamer 1990, Pagels and French 1987); thus, aluminum cans were excluded from more Southeastern Naturalist M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 512 recent studies (Brannon and Bargelt 2013, Brannon et al. 2010). However, we documented both Smoky and Southeastern Shrews in aluminum cans, indicating this refuse may be an important source of mortality. Containers along Flatwoods Road captured uncommon mammals of conservation concern. In this study, approximately 18% of all captures were uncommon small mammals, including the first record of container-caused mortality of Southern Bog Lemming. Comparatively, a previous study along primary roads found less than 5% of all captures were uncommon species (Benedict and Billeter 2004). This difference is likely because most of Flatwoods Road passes through forested habitat and the roadassociated edge habitat, both of which are appropriate to support these species. Shrews in the southern Appalachians are habitat generalists or prefer either forest or edge habitats (Ford et al. 1997, Laerm et al. 1999, Linzey 1998). The species richness we observed in containers along a forest service road (8 species) was similar to findings along primary roads (7 species; Benedict and Billeter 2004). However, compared to our study, roadside pull-offs had greater species richness (11 species; Brannon et al. 2010), which was most likely due to a greater elevational range (255–1336 m) surveyed, encompassing ranges of both higher- and lower-elevation species (i.e., Sorex cinereus Kerr [Masked Shrew] and Southeastern Shrew; Brannon et al. 2010, Ford et al. 2005). In comparing small-mammal mortality along Flatwoods Road to other primary roads and pull-offs, the number of possible containers available to trap small mammals varied. Primary roads, including interstate ramps, averaged 2461 bottles/km (Benedict and Billeter 2004), secondary roads averaged 577 bottles/ km (Pagels and French 1987), and roadside pull-offs averaged 239.2 bottles/km (Brannon and Bargelt 2013). Flatwoods Road had considerably fewer bottles (381.6/km) than primary roads and was somewhat comparable to secondary roads and roadside pull-offs. Greater numbers of bottles provide more opportunities to trap small mammals, and any comparisons of mortality/km could be skewed for primary roads. Additionally, containers at roadside pull-offs were often clustered, which could also lead to an unequal comparison (Brannon et al. 2010). We believe the percentage of containers with small-mammal remains provides a better comparison between different road types. Along primary roads, 4% of containers were found with remains (Benedict and Billeter 2004) and 4.5% of containers at roadside pull-offs had remains (Brannon et al. 2010). Slightly fewer containers along Flatwoods Road contained remains (3.6%). However, we included aluminum cans, which reduced our percentage compared to other studies. Just counting bottles, 3.9% contained remains, which was still slightly less than primary roads and roadside pull-offs. As with primary roads and roadside pull-offs, containers along Flatwoods Road often trapped multiple individuals (mean 1.9 ± 0.2/container with small-mammal remains); mean skulls/bottle ranged from 1.9 to 3.7 in past studies (Benedict and Billeter 2004, Brannon and Bargelt 2013, Brannon et al. 2010, Gerard and Feldhamer 1990). The maximum number of skulls found in a single bottle along Flatwoods Road (8) was less than from bottles along primary roads (19; Benedict Southeastern Naturalist 513 M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 and Billeter 2004), secondary roads (39; Gerard and Feldhamer 1990), and roadside pull-offs (22; Brannon et al. 2010). Large bottles often contained greater numbers of skulls than smaller bottles (Benedict and Billeter 2004). Differences in studies’ findings could be due to fewer numbers of larger containers at our study site, but a comparison was not possible because other studies did not quantify container size (Brannon et al. 2010, Brannon and Bargelt 2013). There are 805.5 km of category 3 and 4 roads in the CNF. Assuming our mean mortality rate of 1.21/km/year in containers along these road types, containers have the potential to kill 973 small mammals/year in the CNF. We established our yearly estimate using only bottles and cans with dates. Older bottles have the potential to trap and kill large numbers of small mammals (Brannon et al. 2010), but without dates we were unable to estimate their impact. Thus, we acknowledge our estimates are conservative, but provide resource managers an approximation that may inform their conservation planning. Similar to Benedict and Billeter (2004), we found glass disproportionately killed small mammals versus other container types. The size of the container opening could be excluding larger species (Morris and Harper 1965), and we found aluminum- can openings to be 8.1 mm narrower than glass bottles. Smaller species, such as Smoky and Southeastern Shrews, can fit through aluminum-can openings, but Northern Short-tailed Shrews are too large for aluminum-can openings. Aluminum cans weighed only 7% of the mean weight of glass containers, potentially allowing shrews to reposition containers from the upslope position and escape (Benedict and Billeter 2004). Additionally, the height of bottle openings also might explain fewer remains of smaller shrews in glass bottles because their openings are much higher (25 mm) above the forest floor than those of aluminum cans (15 mm) (Gerard and Feldhamer 1990). Lastly, shrews have a behavior of squeezing through small holes and would be more likely than rodents to attempt to enter an opening just slightly larger than their skull width, which could explain a lack of rodents in aluminum cans (Morris and Harper 1965). The position of the discarded container proved to be an important factor in small-mammal mortality (Benedict and Billeter 2004, Brannon and Bargelt 2013, Gerard and Feldhamer 1990, Morris and Harper 1965). We found that containers with openings oriented upslope (≥15°) killed significantly more mammals (77.8%) than all other orientations, which is similar to Benedict and Billeter’s (2004) observations. Additionally, we observed that containers with multiple skulls were almost always oriented upslope at step angles (>45°) with the mouth of the container touching or close (<2.5 cm) to the ground. In captive trials, Cryptotis parva Say (Least Shrews), Apodemus sylvaticus L. (Wood Mouse), and Myodes glareolus Schreber (Bank Vole) could not escape a dry glass bottle if it was inclined at an angle >15°, and the minimum angle needed to prevent escape decreased to slightly above horizontal if the interior of the bottle was wet (Gerard and Feldhamer 1990, Morris and Harper 1965). Bottles oriented upslope were more likely to collect water and drown small mammals than those with other orientations (Benedict and Billeter 2004, Brannon and Bargelt 2013, Gerard and Feldhamer 1990). Southeastern Naturalist M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 514 Relative frequencies of small mammals found in containers along Flatwoods Road were similar to results from prior studies of containers along primary and secondary roads (Benedict and Billeter 2004, Gerard and Feldhamer 1990). The Northern Short-tailed Shrew was the most common mammal found in containers (66.6%), and this observation was similar to those made during previous studies in the eastern US (59.9–86.2%; Benedict and Billeter 2004, Gerard and Feldhamer 1990). The relative frequencies of both Smoky (14.4%) and Southeastern Shrews in containers (2.5%) were also comparable to other studies (4.8–20.6% and 1.1–4.3%, respectively; Brannon and Bargelt 2013, Gerard and Feldhamer 1990, Pagels and French 1987). However, our study and previous studies did not estimate the relative abundance of small mammals at study sites, which may have indicated if some species are more susceptible to bottle mortality due to population densities at the study site. We acknowledge that our conclusions are based solely on a single road, and sampling multiple roads would have provided a better understanding of the impact of discarded containers on small mammals. Traffic volumes, proximity to recreation facilities, and road-maintenance regimes could greatly impact the number of potential containers available to trap small mammals along USFS roads, and therefore multiple roads should be examined in the future. However, we found that containers along a single section of a graveled forest service road are trapping and killing small mammals at rates similar to those reported for developed areas (Benedict and Billeter 2004). Many of the species trapped along Flatwoods Road, such as Southern Bog Lemming, are considered rare; therefore, the impact of these containers could have negative consequences for small-mammal populations. Discarded containers along thousands of kilometers of national forest roads in the southern Appalachians will likely cause mortality as long as these containers are present. Small-mammal mortality created by discarded containers is one of the few anthropogenic effects that could be easily mitigated, e.g., through litter clean-ups, proactive enforcement of litter-control regulations, and education to reduce littering. Acknowledgments We are grateful to the CNF staff for project support and access to our study site. All sampling was approved by the TWRA (Scientific Salvage Permit #1888). We also thank students from Virginia Highlands Community College (VHCC) General Biology classes in spring 2004 and 2005 for their assistance in collecting bottles. Special thanks to VHCC students G. Blevins, A. McConnell, and J. Osborne for their help in collecting containers and cleaning skulls. We are grateful to M. Kennedy (University of Memphis), D. Linzey (Virginia Tech), R. Semlitsch (University of Missouri-Columbia), M. Chatfield (Unity College), and R. May (VHCC) for their advice and assistance. We also thank K. Powers and two anonymous reviewers for their suggestions on a previous version of this manuscript. Roger Perry and two anonymous reviewers improved our current manuscript and deserve thanks for their efforts. Southeastern Naturalist 515 M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 Literature Cited Barbour, R.W. 1974. Mammals of Kentucky. University of Kentucky Press, Lexington, KY. 386 pp. Benedict, R.A., and M.C. Billeter. 2004. Discarded bottles as a cause of mortality in small vertebrates. Southeastern Naturalist 3:371–7. Berman, J., T.S. McCay, and P. Scull. 2007. Spatial analysis of species richness of shrews (Soricomorpha: Soricidae) in North America north of Mexico. Acta Theriologiac 52:151–8. Brannon, M.P., and L.B. Bargelt. 2013. Discarded bottles as a mortality threat to shrews and other small mammals in the Southern Appalachian Mountains. Journal of North Carolina Academy of Science 129:126–9. Brannon, M.P., M.A. Burt, D.M. Bost, and M.C. Caswell. 2010. Discarded bottles as a source of shrew-species distribution data along an elevational gradient in the Southern Appalachians. Southeastern Naturalist 9:781–94. Debernardi, P., E. Patrtarca, A. Perrone, M. Cantini, and B. Chiarenzi. 1997. Small mammals found in discarded bottles in alpine and pre-alpine areas of NW Italy. Hystrix 9:51–5. Ford, W.M., J. Laerm, and K.G. Barker. 1997. Soricid response to forest-stand age in southern Appalachian cove-hardwood communities. Forest Ecology and Management 91:175–81. Ford, W.M., T.S. McCay, M.A. Menzel, W.D. Webster, C.H. Greenberg, J.F. Pagels, and J.F. Merritt. 2005. Influence of elevation and forest type on community assemblage and species distribution of shrews in the central and southern Appalachian Mountains. Pp. 303–315, In J.F. Merritt, S. Churchfield, R. Hutterer, and B.I. Sheftel (Eds.). Advances in the Biology of Shrews II. Special Publication of the International Society of Shrew Biologists No. 1, Powdermill Biological Station of the Carnegie Museum of Natural History, Pittsburg, PA. 488 pp. Forman, R.T., and L.E. Alexander. 1998. Roads and their major ecological effects. Annual Review of Ecology and Systematics 29:207–31. French, T.W. 1980. Sorex longirostris. Mammalian Species 143:1–3. George, S.B., J.R. Choate, and H.H. Genoways. 1986. Blarina brevicauda. Mammalian Species 261:1–9. Gerard, A.S., and G.A. Feldhamer. 1990. A comparison of two survey methods for shrews: Pitfall and discarded bottles. The American Midland Naturalist 124:191–4. IBM Corp. 2011. SPSS Statistics for Windows, Version 20.0. Armonk, NY. Laerm, J., W.M. Ford, T.S. McCay, M.A. Menzel, L.T. Lepardo, and J.L. Boone. 1999. Soricid communities in the Southern Appalachian. Pp. 177–93, In R.P. Eckerlin (Ed.). Proceedings of the Appalachian Biogeography Symposium. Virginia Museum of Natural History Special Publication No. 7, Martinsville, VA. 258 pp. Linzey, D.W. 1998. The Mammals of Virginia. The McDonald and Woodward Publishing Company, Blacksburg, VA. 459 pp. Morris, P.A., and J.F. Harper. 1965. The occurrence of small mammals in discarded bottles. Proceedings of the Zoological Society of London 145:148–53. Owens, J.G. 1984. Sorex fumeus. Mammalian Species 215:1–8. Oxley, D.J., M.B. Fenton, and G.R. Carmody. 1974. The effects of roads on populations of small mammals. The Journal of Applied Ecology 11:51–59. Pagels, J.F., and T.W. French. 1987. Discarded bottles as a source of small-mammal distribution data. The American Midland Naturalist 118:217–9. Southeastern Naturalist M.K. Hamed and T.F. Laughlin 2015 Vol. 14, No. 3 516 Smolen M.J. 1981. Microtus pinetorum. Mammalian Species 147:1–7. Tennessee Natural Heritage Program (TNHP). 2009. Rare Animal List. Tennessee Department of Environment and Conservation, Division of Natural Heritage, Nashville, TN. Tennessee Wildlife Resource Agency (TWRA). 2005. Tennessee’s Comprehensive Wildlife Conservation Strategy. Nashville, TN. 217 pp. US Department of Agriculture (USDA). 2004. Final environmental impact statement for the revised land resource management: Cherokee National Forest. US Department of Agriculture Management Bulletin R8-MB 114F, Cleveland, TN. USDA. 2005. Guidelines for road maintenance levels. Technology and Development Program 0577 1205-SDTDC, Washington, DC.