2012 NORTHEASTERN NATURALIST 19(2):229–248 Amphibian Sampling Techniques along Maryland Coastal-plain Streams Gabriel F. Strain*,1,2 and Richard L. Raesly1 Abstract - Amphibians and other herpetofauna may be useful in assessing the biological integrity of small streams, so determining which sampling technique maximizes encounters is important. Area-constrained surveys (ACS), used by the Maryland Biological Stream Survey, were tested against cover-board surveys, drift fences with pitfall and funnel traps, quadrat leaf-litter searches, leaf-litter bags, and electrofishing. Twenty sites within the coastal plain region of Maryland, west of the Chesapeake Bay, were sampled with each technique once a month from June 2006 through August 2006. Overall, ACS and electrofishing yielded significantly more taxa and total individuals than cover-board surveys, quadrat searches, and leaf-litter bags; drift fence captures were moderate between ACS and electrofishing and the other methods. Electrofishing and ACS collected both more taxa and more individuals more reliably through time than the other techniques used; therefore, efforts to use herpetofauna to monitor the health of small streams will benefit from incorporating these methods into a sampling protocol. Introduction Amphibians may be useful indicators of environmental conditions because they have permeable skin, are long-lived, and are intimately associated with aquatic systems (Blaustein 1994, Jung et al. 2000, Southerland et al. 2004, Welsh and Ollivier 1998). The dual life cycle of many amphibian species potentially exposes them to both terrestrial and aquatic disturbances and contaminants (Blaustein 1994, Blaustein and Johnson 2003, Fronzuto and Verrell 2000). Larval anurans are sensitive to increased concentrations of heavy metals and reduced pH levels (Jung and Jagoe 1995), and salamander relative abundance is inversely proportional to disturbed habitat (Willson and Dorcas 2003). Prior to 2000, malformations in at least 57 species of anurans had been reported across 44 states (Meteyer 2000), which may be attributed to anthropogenic causes such as contamination and increased UV-B radiation (Blaustein and Johnson 2003). Multiple quantitative methods have been developed to measure and evaluate these problems in amphibian communities. The Maryland Biological Stream Survey (MBSS) is a probability-based survey of Maryland’s non-tidal streams (Klauda et al. 1998) that uses metrics for fishes and benthic invertebrates to assess stream condition. Indices of biotic integrity (IBIs) using fish, however, are not useful for streams draining catchments of less than 300 acres, because fish species richness and abundance numbers are too low (Klauda et al. 1998; Southerland et al. 2000, 2004). Multiple geological and hydrological barriers often prevent fishes from entering these bodies of 1Department of Biology, Frostburg State University, Frostburg, MD 21532. 2Current address - West Virginia University, Division of Forestry and Natural Resources, Morgantown, WV 26506. *Corresponding author - gstrain54@yahoo.com. 230 Northeastern Naturalist Vol. 19, No. 2 water (Davic and Welsh 2004). In these smaller, sometimes ephemeral streams, amphibians may assume the role of top predators (Pauley 1995, Southerland et al. 2004) and play an important role in ecosystem processes (Davic and Welsh 2004). The total biomass of amphibians in some areas may equal that of small mammals and be twice that of birds (Burton and Likens 1975). Rocco and Brooks (2000) found that salamander assemblages along small streams are measurably affected by anthropogenic disturbances such as acidification and fragmentation. In these smaller catchments, therefore, surveys and development of metrics for amphibians may be valuable monitoring tools. Recently, a stream salamander IBI (SS-IBI) for use in Maryland watersheds was developed and tested for effectiveness (Southerland et al. 2004). If proven effective, the SS-IBI would be incorporated into the MBSS framework as an additional tool to assess the health of smaller, generally fishless streams. Southerland et al. (2004) used a combination of terrestrial and aquatic transects and quadrats to sample amphibians in streams throughout Maryland. All rocks and logs within each transect or quadrat were overturned to capture salamanders. Metrics such as the number of species and number of pollution-intolerant salamanders were then used to assign a score to each site (based on reference thresholds; Southerland 2004). The index was deemed effective when validated against benthic invertebrate IBIs (B-IBIs) for the same site; however, the average number of salamander species collected at a site was two, which may not be a large enough number “to discern convincing patterns” (Southerland et al. 2000). Southerland et al. (2000) also suggested “more intensive sampling (to identify more species) … would reduce the adverse effects of low metric numbers.” The MBSS currently employs an area-constrained survey (ACS) that consists of recording the presence of any herpetofauna at a site along the stream bank and in the channel during electrofishing sessions in the summer sampling period (Kazyak 2001). This procedure primarily consists of incidental observations made while walking from the vehicle to the stream and during electrofishing, but does not include additional methodology (e.g., area-constrained survey with cover object flipping). Although the combined use of ACS and electrofishing should encompass the majority of amphibian habitat at a site, uncertainty exists as to whether these methods provide representative data of the herpetofauna at each site (Southerland et al. 2000). Other methods such as cover-boards, leaf-litter bags, drift fences with pitfall and funnel traps, and quadrat leaf-litter searches (or some combination of methods) may be more suitable for characterizing the herpetofaunal assemblage. Study objectives The primary objective of this study was to determine whether ACS and electrofi shing are equivalent to other common survey techniques (cover boards, leaf-litter bags, drift fences with pitfall and funnel traps, and quadrat leaf-litter searches). The effect of increasing sampling effort was also evaluated for ACS and electrofishing. Our secondary objective was to determine the minimum number of methods required to adequately sample a typical Mid-Atlantic herpetofaunal assemblage. The focus was on salamanders, but we were interested in all herpetofauna encountered. In addition, peak activity periods for many species greatly depend on weather (Crump and Scott 1994, Duellman and Trueb 1994), and this may cause variable 2012 G.F. Strain and R.L. Raesly 231 sampling results from month to month. Another objective of this study, therefore, was to examine overall capture differences among months. Field-Site Description Nineteen sites were established in small, mostly headwater streams in the coastal plain region of Maryland, west of the Chesapeake Bay (Fig. 1). Sites were Figure 1. Coastal plain region of Maryland, west of the Chesapeake Bay, with locations of study sites. 232 Northeastern Naturalist Vol. 19, No. 2 located in the counties of Anne Arundel (3 sites), Calvert (1), Charles (6), Prince George’s (5), and St. Mary’s (4). Streams of this region generally have sand or gravel substrates, slow-moving water, low to moderate nitrate concentrations, a largely vegetated riparian zone, and low to moderate levels of dissolved oxygen (Boward et al. 1999). Streams in our study had an average width of 1.2 m and an average depth of 0.2 m. Methods At each site, 6 treatments were assigned: ACS, cover-board survey, leaflitter bags, quadrat leaf-litter searches, drift fences with pitfall and funnel traps, and electrofishing. The 6 treatments were ordered randomly within each study site to establish a randomized block design, with each study site representing a replicate. Each treatment was randomly assigned to either the left or the right bank of the stream. The cover boards, drift-fence arrays, and leaf-litter bags were established in late May/early June 2006. Each site was sampled once on three occasions: early June, early July, and early August. Although leaf-litter bags are designed to encounter larvae and not adults, we sampled them concurrently with the other methods because our two most commonly encountered species (Eurycea bislineata [Northern Two-lined Salamander] and Pseudotriton rubber [Red Salamander]) have larval periods lasting two or more years (Lannoo 2005), and so we expected to find larvae in streams throughout the year. During each sampling period, all sites were visited in approximately 15 days. All herpetofauna encountered were captured (if possible), identified to species, measured for snout–vent length (SVL), and released. Any anurans heard calling were also recorded. Sampling techniques Area constrained survey. Each ACS plot consisted of one 25- x 1-m transect (Crump and Scott 1994), with the transect running along the streambank parallel to and beginning at the stream margin. The transect was searched slowly, with two observers turning over all natural cover objects (e.g., rocks and logs) within the transect and returning those objects as close as possible to their original position to minimize bias due to habitat alteration. Cover-board survey. Cover boards consisted of pine boards of approximately 30 x 30 x 5 cm. Leaf litter and debris were cleared away so that the boards were placed upon bare mineral soil (Fellers and Drost 1994, Hyde and Simons 2001, Marsh and Goicochea 2003). Each plot contained 12 cover boards in a 6 x 2 pattern, with 1 row of 6 boards placed 2 m away from the stream margin, and 1 row of 6 boards placed 4 m away from the stream margin. Sampling consisted of the observer(s) walking a set path along the cover objects and lifting up the boards to check for amphibians. All amphibians were then released at the edge of the cover board to minimize exposure and desiccation (Fellers and Drost 1994). Leaf-litter bags. Leaf litter bag construction followed Jung and Pauley (2003) and Pauley and Little (1998). Bags were constructed of 61- x 40-cm plastic netting with 2.5-cm mesh and filled with leaf-litter debris and several small rocks. 2012 G.F. Strain and R.L. Raesly 233 The ends of the netting were drawn together and secured with a cable tie to form a bag, and the bags were then placed in the stream. Each site contained 5 bags, with 1 bag every 5 m. Yellow flagging was tied to each bag for greater visibility, and rocks were placed around or on top of each bag to prevent them from washing away in case of high-water events. In some cases, rocks were not available, so sticks were pushed through the bags and into the substrate to hold them in place. To check each bag, the observer(s) quickly pulled it out of the water and placed it in a white dishpan, shaking it to dislodge all salamanders. Because leaf-litter debris deteriorates, debris was replaced before returning the bag to the stream. Drift fences with pitfall and funnel traps. One drift-fence array was established at each site, and construction of arrays generally followed Corn (1994) and Enge (1997). The drift fences consisted of 50-cm-high aluminum flashing which was buried in the ground to a depth of 12–15 cm, yielding a 35–38-cm-high barrier. Two 5-m-long fences were used for each array, with one approximately 1 m away from the stream margin running parallel with the stream, and the other running perpendicular to and away from the stream forming an “L” shape. A third 1-m-long fence was placed running down to the stream margin opposite the perpendicular 5-m-long arm. Pitfall traps were constructed of 1.75-gallon restaurant buckets, approximately 21 cm in diameter and 19.5 cm deep, with holes drilled in the bottom for drainage. A wetted sponge was placed in each trap to prevent desiccation. One trap was buried at the end of the 5-m arms of the array, and one trap was buried at the junction of the three arms. The traps were buried flush with the ground, and a piece of masonite with 4–6-cm-long wooden legs attached was placed over each trap to provide shade and prevent rain from entering the trap. Funnel traps were constructed of aluminum window screening and based on the design for double-ended funnel traps described by Enge (1997). Two wetted sponges were placed in each trap to prevent desiccation. Four funnel traps were used in each array, with one trap placed on each side at roughly the midpoint of the two 5-m fence arms. Squares of 41- x 41-cm masonite were leaned against the fence over each funnel trap to provide shade. Leaf litter was pushed up into the mouth of each funnel to form a ramp that guided animals into the trap. The traps were opened for two nights and checked the following day. To prevent accidental captures when traps were not in use, the pitfall traps were sealed tightly with plastic lids, and the sponges in the funnel traps were placed in the mouth of each funnel. Quadrat leaf-litter searches. The 25- x 3-m section of each site receiving this treatment was divided into seventy-five 1- x 1-m quadrats numbered consecutively. For each site, 6 quadrats were randomly selected each month using a random numbers table, and quadrats were selected without replacement to minimize bias due to habitat alteration (Jaeger and Inger 1994). A 1- x 1-m frame constructed of PVC was placed over each selected quadrat, with the bottom of the frame as parallel to the stream margin as possible, and the leaf litter within the frame searched thoroughly down to bare soil. Ideally, captured amphibians would have been held and released at their capture location, but because surveys were most often performed 234 Northeastern Naturalist Vol. 19, No. 2 by 1 person, holding individuals was not feasible. Captured herpetofauna were released in an adjacent quadrat behind the observer to prevent recaptures during the search, and quadrats were sampled systematically across the site to minimize the chance of a released individual entering a quadrat that had not yet been sampled. However, due to the territorial nature of salamanders (Jaeger 1988), we do not recommend sampling in this manner. Individuals should be held until the end of sampling, and released at their capture locations. Electrofishing. Electrofishing was performed with a backpack unit (Smith- Root 12-B Electrofisher), with a single observer electrofishing and another netting. Each 25-m section was sampled thoroughly and continuously from bank to bank, including backwater areas, sloughs, and shallows (Kazyak 2001). We began electrofishing at the downstream end of each section and moved upstream. As organisms encountered the electric field, they either floated to the surface (most adult frogs) or sank to the bottom (most salamander larvae). The netter captured these individuals and placed them in a stream water-filled 5-gallon bucket. Error rate To estimate error rate for relatively commonly encountered taxa (those with captures comprising ≥1% of total captures in at least one month), we summed the number of sites where an individual of a particular species was detected by at least one method (sites where a species was not detected with any method were omitted because they may represent false absences). For each method, the error rate was then calculated as the number of sites where a species was detected with that method divided by the total number of sites where that species was detected by any method. Effect of increased effort Four area-constrained surveys and five electrofishing sessions were extended to examine the effect of an increase in sampling effort on species richness and total number of individuals. These randomly selected surveys were extended beyond the 25-m section of stream to sample 50-m- and 75-m-long sections. IBI comparison To further evaluate the efficacy of using stream salamanders in a monitoring program, we used the Non-Coastal Plain SS-IBI developed by Southerland et al. (2004) to calculate IBIs for sites that had corresponding benthic macroinvertebrate IBIs (there were six such sites). We used the Non-Coastal Plain SS-IBI because a SS-IBI for the coastal plain has not yet been developed and we were interested to examine how it would perform against the macroinvertebrate IBI (B-IBI; data publicly available at http://www.dnr.state.md.us/streams/MBSS. asp). The Non-Coastal Plain SS-IBI includes the following metrics: number of species, number of salamanders, number of intolerant salamanders, and number of adult salamanders. These metrics receive a score of 0, 5, or 10 depending on their value; for instance, if >3 species are recorded at a site, the site receives a 10 for number of species (a score of 5 for 2–3 species, and a 0 for less than 2 species). The scores are then averaged to produce the final score for the site. 2012 G.F. Strain and R.L. Raesly 235 Statistical analysis Because each site was re-sampled each month, an unknown amount of non-independence was inherent in the experimental design. Therefore, program PROC MIXED (SAS version 8e for Windows) was used to perform a repeated measures analysis. We ran a series of models with differing covariance structures in order to determine the best fit for repeated measures, and chose the model with the lowest AIC value. One-way ANOVA was used to test for differences among months, and counts of species and individuals were pooled across sampling methods to accomplish this. In comparing total species richness among months, any anuran species heard calling was included in the analysis. A significance level of 0.05 was used in testing all hypotheses. The means of differences under each hypothesis were compared using Tukey’s multiple range test. Results Twenty-eight species of amphibians and reptiles totaling 632 individuals were captured or encountered between June 6 and August 6 of 2006 (Table 1). Lithobates clamitans melanota (Green Frog), Eurycea bislineata (Northern Two-lined Salamander), and larvae of the genus Pseudotriton (red and mud salamanders) were the three most common taxa caught or encountered by all sampling methods and accounted for 72.4% of the total. Salamander taxa comprised 26.5% of the total. The 27 larval Ambystoma maculatum (Spotted Salamander), although not typical of stream habitats, were found in an isolated pool in a dried portion of the stream channel and were thus included in the analyses. Comparison of methods Taxa. Considering only salamander taxa, a significant difference existed in June only between drift-fence surveys, which yielded zero species, and electrofishing, which yielded a mean of 0.63 species (Fig. 2A, Table 2). A significantly greater number of species were captured with electrofishing than with all other methods in July. No significant differences among methods were detected in August. Considering all herpetofauna taxa encountered in the analysis, ACS and electrofi shing in both June and July yielded significantly more species than cover-board surveys, quadrat leaf-litter searches, and leaf-litter bags (Fig. 2B, Table 3). In August, significantly more species were captured with ACS than with cover-boards, quadrat leaf-litter searches, and leaf-litter bags, whereas electrofishing yielded significantly more species than only quadrat searches and leaf-litter bags. Total individuals. No significant differences existed among methods for total salamander individuals in June and August. In July, electrofishing yielded signifi- cantly more individuals than cover boards, drift-fence surveys, quadrat searches, and ACS (Fig. 2C, Table 4). Considering all herpetofauna individuals encountered in the analysis, signifi cantly more individuals were captured with electrofishing than with cover boards, drift-fence surveys, leaf-litter bags, and quadrat searches in June and July (Fig. 2D, Table 5). ACS yielded significantly more individuals than cover boards, leaf-litter bags, and quadrat searches in July and August. 236 Northeastern Naturalist Vol. 19, No. 2 Error rate. On average, the error rate for ACS was much lower than for the other techniques in all three months (Table 6). This finding means that when a species was present at a site, ACS detected it more often than the other methods. Individual error rates for each sampling method varied widely depending on taxa and month. Comparison of months No significant difference existed between the sampling months of June, July, and August 2006 for total taxa, salamander taxa, total individuals, and salamander individuals. Individually by sampling method, however, electrofishing in June and July produced significantly more salamander individuals than in August (Fig. 2C). Weather was similar during all sampling periods. Table 1. Number of individuals of each taxon encountered by sampling period. Sampling period Taxa June July August Total Anura Acris crepitans Baird (Eastern Cricket Frog) 9 2 4 15 Anaxyrus a. americanus (Holbrook) (Eastern American Toad) 5 4 5 14 A. fowleri (Hinckley) (Fowler’s Toad) 5 4 1 10 Hyla cinerea (Schneider) (Green Tree Frog) 2 - - 2 Lithobates catesbeianus (Shaw) (American Bullfrog) 2 5 5 12 L. clamitans melonata (Rafinesque) (Northern Green Frog) 109 123 100 332 L. palustris LeConte (Pickerel Frog) 9 4 10 23 L. sphenocephalus utricularius (Cope) (Southern Leopard Frog) 13 17 8 38 L. sylvaticus (LeConte) (Wood Frog) 1 2 - 3 Pseudacris crucifer (Wied-Neuwied) (Spring Peeper) - 2 - 2 Caudata Ambystoma maculatum (Shaw) (Spotted Salamander) 27 - - 27 A. opacum (Gravenhorst) (Marbled Salamander) - - 2 2 Desmognathus fuscus (Rafinesque) (Dusky Salamander) 1 - 1 Eurycea bislineata (Green) (Northern Two-lined Salamander) 39 26 16 81 Notophthalmus viridescens (Rafinesque) (Eastern Red-spotted Newt) - - 1 1 Plethodon cinereus Green (Eastern Red-backed Salamander) - - 1 1 Pseudotriton montanus Baird (Mud Salamander) - - 2 2 P. ruber (Sonnini de Manoncourt and Latreille) (Red Salamander) 4 - 4 8 P. spp. larvae 17 18 10 45 Sauria Cnemidophorus s. sexlineata (L.) (Eastern Six-lined Racerunner) - 1 - 1 Scincella lateralis (Say in James) (Little Brown Skink) 1 - - 1 Serpentes Coluber c. constrictor L. (Northern Black Racer) 1 - - 1 Nerodia s. sipedon (L.) (Northern Water Snake) - - 1 1 Pantherophis allegheniensis (Holbrook) ((Eastern Ratsnake) - - 2 2 Thamnophis sirtalis (L.) (Common Garter Snake) - 1 - 1 Testudines Chelydra serpentina (L.) (Snapping Turtle) - 1 1 2 Chrysemys p. picta (Schneider) (Eastern Painted Turtle) - 1 1 Terrapene carolina (L.) (Eastern Box Turtle) - 1 2 3 2012 G.F. Strain and R.L. Raesly 237 Figure 2. Mean number of salamander taxa (A), mean number of herpetofauna taxa (B), mean number of salamander individuals (C), and mean number of herpetofauna individuals (D) encountered with each technique during the three sampling periods in 2006 (ACS = area-constrained survey, CB = cover-board survey, LLB = leaf-litter bags, QU = quadrat leaf-litter search, DF = drift-fence survey, EL = electrofishing). Table 2. Salamander taxa minimum, maximum, and mean (± 1 SE) for each technique for each month. Superscripted letters represent differences within each month at the 0.05 significance level. Sampling method Month Min Max Mean SE Drift fenceA June 0 0 0.00 0.00 Leaf-litter bagA,B June 0 1 0.42 0.12 Cover-board surveyA,B June 0 1 0.05 0.05 Quadrat searchA,B June 0 2 0.16 0.11 ElectrofishingB June 0 2 0.63 0.17 Area-constrained surveyA,B June 0 2 0.37 0.14 Drift fenceA July 0 1 0.05 0.05 Leaf-litter bagA July 0 1 0.11 0.07 Cover-board surveyA July 0 1 0.05 0.05 Quadrat searchA July 0 1 0.05 0.05 ElectrofishingB July 0 2 0.58 0.18 Area-constrained surveyA July 0 1 0.05 0.05 Drift fenceA August 0 2 0.26 0.13 Leaf-litter bagA August 0 1 0.11 0.07 Cover-board surveyA August 0 2 0.21 0.12 Quadrat searchA August 0 1 0.05 0.05 ElectrofishingA August 0 2 0.31 0.13 Area-constrained surveyA August 0 1 0.21 0.10 238 Northeastern Naturalist Vol. 19, No. 2 Effect of increased effort The total number of individuals data from June illustrate that an increase in effort may increase yield for both ACS and electrofishing (Fig. 3). Increasing the ACS effort from 25 m to 50 m and 75 m increased the number of all herpetofaunal species, number of salamander species, number of individuals, and number Table 4. Salamander individuals minimum, maximum, and mean (± 1 SE) for each technique for each month. Superscripted letters represent differences within each month at the 0.05 significance level. Sampling method Month Min Max Mean SE Drift fenceA June 0 0 0.00 0.00 Leaf-litter bagA June 0 4 0.58 0.22 Cover-board surveyA June 0 3 0.16 0.16 Quadrat searchA June 0 2 0.16 0.11 ElectrofishingA June 0 27 3.11 1.53 Area-constrained surveyA June 0 2 0.37 0.14 Drift fenceA July 0 1 0.05 0.05 Leaf-litter bagA,B July 0 1 0.11 0.07 Cover-board surveyA July 0 1 0.05 0.05 Quadrat searchA July 0 1 0.05 0.05 ElectrofishingB July 0 14 1.95 0.90 Area-constrained surveyA July 0 1 0.05 0.05 Drift fenceA August 0 2 0.31 0.15 Leaf-litter bagA August 0 2 0.16 0.11 Cover-board surveyA August 0 4 0.31 0.22 Quadrat searchA August 0 1 0.05 0.05 ElectrofishingA August 0 9 0.79 0.49 Area-constrained surveyA August 0 2 0.31 0.15 Table 3. Herpetofauna taxa minimum, maximum, and mean (±1 SE) for each technique for each month. Superscripted letters represent differences within each month at the 0.05 significance level. Sampling method Month Min Max Mean SE Leaf-litter bagA June 0 1 0.42 0.12 Cover-board surveyA June 0 2 0.16 0.11 Quadrat searchA June 0 2 0.33 0.15 ElectrofishingB June 0 4 1.68 0.27 Area-constrained surveyB June 0 5 1.68 0.31 Drift fenceA,B July 0 2 0.84 0.17 Leaf-litter bagA July 0 1 0.16 0.09 Cover-board surveyA July 0 1 0.11 0.07 Quadrat searchA July 0 2 0.21 0.12 ElectrofishingB July 0 4 1.53 0.32 Area-constrained surveyB July 0 4 1.58 0.26 Drift fenceA,B,C August 0 3 1.05 0.22 Leaf-litter bagA August 0 1 0.11 0.07 Cover-board surveyA,C August 0 2 0.21 0.12 Quadrat searchA August 0 1 0.05 0.05 ElectrofishingB,C August 0 4 1.16 0.32 Area-constrained surveyB August 0 4 1.89 0.26 2012 G.F. Strain and R.L. Raesly 239 of salamander individuals. Increasing the electrofishing effort from 25 m to 50 m and 75 m increased the number of all herpetofaunal species and individuals. However, these patterns may be misleading because comparisons were based Table 5. Herpetofauna individuals minimum, maximum, and mean (± 1 SE) for each technique for each month. Superscripted letters represent differences within each month at the 0.05 significance level. Sampling method Month Min Max Mean SE Drift fenceA June 0 4 1.16 0.28 Leaf-litter bagA June 0 4 0.58 0.22 Cover-board surveyA June 0 4 0.26 0.21 Quadrat searchA June 0 2 0.31 0.15 ElectrofishingB June 0 27 6.42 1.60 Area-constrained surveyA,B June 0 9 3.79 0.71 Drift fenceA,C July 0 3 1.16 0.26 Leaf-litter bagA July 0 1 0.16 0.09 Cover-board surveyA July 0 1 0.11 0.07 Quadrat searchA July 0 2 0.21 0.12 ElectrofishingB July 0 22 5.11 1.55 Area-constrained surveyB,C July 0 16 4.26 0.96 Drift fenceA,B August 0 5 1.53 0.33 Leaf-litter bagB August 0 2 0.16 0.11 Cover-board surveyB August 0 4 0.31 0.22 Quadrat searchB August 0 1 0.05 0.05 ElectrofishingA,B,C August 0 18 3.42 1.18 Area-constrained surveyA,C August 0 24 4.53 1.23 Figure 3. Effect of increasing effort on number of individuals (+ 1 SE) during the month of June (ACS = area-constrained survey, EL = electrofishing). 240 Northeastern Naturalist Vol. 19, No. 2 Table 6. Error rate by sampling technique and month. Rates were based on detection of a particular species by at least one technique; only species whose captures comprised ≥1% of total captures in at least one month were included. See Table 1 for genus names of species. Species ACS CB LLB QU DF EL ACS CB LLB QU DF EL ACS CB LLB QU DF EL A. crepitans 0.00 1.00 1.00 1.00 0.75 0.75 0.00 1.00 1.00 1.00 1.00 1.00 0.33 1.00 1.00 1.00 1.00 0.67 A. maculatum 1.00 1.00 1.00 1.00 1.00 0.00 - - - - - - - - - - - - A. opacum - - - - - - - - - - - - 1.00 0.00 1.00 1.00 0.00 1.00 A. americanus 0.60 1.00 1.00 1.00 0.40 1.00 0.33 1.00 1.00 1.00 0.67 1.00 0.75 1.00 1.00 1.00 0.50 0.75 A. fowleri 0.00 1.00 1.00 1.00 0.00 1.00 0.67 1.00 0.67 1.00 0.67 1.00 1.00 1.00 1.00 1.00 0.00 1.00 E. bislineata 0.63 0.87 0.25 0.75 1.00 0.37 0.80 0.80 0.80 0.80 0.80 0.20 0.40 0.60 1.00 0.80 0.80 0.40 L. catesbeianus 1.00 1.00 1.00 1.00 1.00 0.00 1.00 1.00 1.00 1.00 0.80 0.20 0.60 1.00 1.00 1.00 0.40 0.80 L. clamitans 0.35 0.94 1.00 0.88 0.59 0.23 0.12 1.00 1.00 0.88 0.53 0.41 0.06 1.00 1.00 1.00 0.44 0.37 L. palustris 0.00 1.00 1.00 1.00 0.67 0.67 0.50 1.00 1.00 1.00 1.00 0.50 0.00 1.00 1.00 1.00 0.67 0.67 L. sphenocephala 0.57 0.86 1.00 1.00 0.86 0.57 0.29 0.86 1.00 1.00 0.71 0.71 0.40 1.00 1.00 1.00 1.00 0.60 P. montanus - - - - - - - - - - - - .50 1.00 1.00 1.00 0.50 1.00 P. ruber 0.00 1.00 1.00 0.67 1.00 1.00 - - - - - - 0.75 1.00 1.00 1.00 0.50 0.50 Pseudotrition spp. 1.00 1.00 0.75 1.00 1.00 0.13 1.00 1.00 0.86 1.00 1.00 0.00 0.50 1.00 0.50 1.00 1.00 0.50 S. allegheniensis - - - - - - - - - - - - 0.50 1.00 1.00 1.00 0.50 1.00 T. carolina - - - - - - 0.00 1.00 1.00 1.00 1.00 1.00 0.00 1.00 1.00 1.00 1.00 1.00 Mean 0.47 0.97 0.91 0.94 0.75 0.52 0.47 0.97 0.93 0.97 0.82 0.60 0.49 0.90 0.96 0.99 0.59 0.73 2012 G.F. Strain and R.L. Raesly 241 on data from only four sites for ACS and five sites for electrofishing, only one of which had a 75-m observation for each method. The standard error for most samples was relatively high and ranged from 0.11 to 2.5. July and August showed similar patterns. IBI comparison The SS-IBIs did not perform well compared to the B-IBIs. Only two of the 18 calculated SS-IBIs (one for each site and month) were non-zero: a site in June received a score of 1.25, and another site in July received a score of 2.50. In contrast, the mean B-IBI score for the six sites was 3.53, with two sites each having a score of 4.43. Discussion Area-constrained surveys and electrofishing were the most effective sampling methods for herpetofauna in small Maryland coastal plain streams. These methods collected both more taxa and more individuals more reliably through time than the other techniques used. Our findings are similar to other studies where ACS performed better than alternative methods (e.g., Fogarty and Jones 2003, Foley and Smith 1999, Hyde and Simons 2001, Paszowski et al. 2002). In this study, there often were no significant differences between ACS, electrofi shing, and drift-fence surveys; however, the cost in labor, maintenance, and time (drift-fence surveys require at least two visits for results) may cause the latter method to be less attractive than ACS and electrofishing (Corn 1994, Enge 1997). Also, Fogarty and Jones (2003) reported depredation of amphibians in pitfall traps by Procyon lotor L. (Raccoon), which may present a source of bias. In this study, a single Thamnophis sirtalis (Eastern Garter Snake), which appeared to have been bitten to death through the trap, possibly by a Raccoon individual, was found in a funnel trap. Electrofishing did not appear to distress amphibians or other animals more than momentarily (i.e., we retained each shocked animal in a water-filled bucket until normal behavior resumed; we observed no mortalities with this method). Cover-board surveys and leaf-litter bags also require at least two visits to establish and sample sites, and the low yield may not be worth the effort. Also, it has been suggested that snakes may become entangled in leaf-litter bags and die (Stuart and Watson 2001). Previous studies have shown leaf-litter bags to perform better than turning over rocks in the stream channel (Pauley and Little 1998); however, electrofishing unquestionably produced the most larval salamanders in this study, and a combination of electrofishing and ACS may be ideal. One of the disadvantages of cover boards is that their use may depend on the species and on the amount of available natural cover (Fellers and Drost 1994). This drawback may be a problem in interpreting the results of our comparison between ACS and cover boards. Due to the highly variable discharge of streams in the coastal plain, we could not establish cover boards near the edge of the water. This limitation resulted in ACS covering the wet floodplain area of the stream, and the cover boards covering the drier riparian zone. Species that tend to be found along the 242 Northeastern Naturalist Vol. 19, No. 2 streambank, such as Desmognathus fuscus (Northern Dusky Salamander) (Southerland 1986), were not found under cover boards; however, only a single Northern Dusky Salamander was encountered during the entire study. Individuals of Northern Two-lined Salamander were encountered in low but similar numbers with both methods, so the effect of the disparity in habitats sampled by these two methods is unclear. The decision of the most appropriate method to use may depend on the objectives of the study and the targeted species. A potential solution to this problem could be to secure cover boards to the substrate so that they would not wash away during high-water events. The area searched with ACS was roughly 4.2 times the size of the area sampled with quadrats, and when quadrat captures are extrapolated to the equivalent ACS area sampled, results are similar to those of ACS and electrofishing in some months. However, our objective was not to compare equal areas or equal amounts of time sampled, but rather to compare a method which has high coverage and a low level of search intensity (ACS) to one with low coverage but a high level of search intensity (quadrat leaf-litter search). Sampling more than six quadrats may increase the number of species and individuals encountered, but the drawbacks may preclude this option. This method takes at least fives times longer to perform than ACS (Strain et al. 2009) and is very destructive (Witham et al. 1993 in Monti et al. 2000), and the amount of habitat damage would quickly reach unacceptable levels. Sampling time must be considered in determining the most effective method to use in sampling herpetofauna, as survey duration contributes to the cost of a monitoring program. In our study, cover-board surveys were conducted in a mean of 0.11 person-hours (p-h), the shortest time of all methods, followed by ACS and leaf-litter bags (0.19 p-h), drift fence surveys (0.33 p-h), quadrat surveys (0.38 p-h), and electrofishing (0.86 p-h). In terms of capture rate, ACS had the highest mean catch per unit effort (cpu), with 24.46 individuals captured per person-hour. This rate was much higher than all other methods; electrofishing had the next highest (4.46 cpu), followed by drift-fence surveys (3.08 cpu), leaf-litter bags (0.87 cpu), cover-board surveys (0.66 cpu), and quadrat surveys (0.43 cpu). The combination of short survey duration and high yield makes ACS a very effective sampling technique. Although electrofishing took the longest to conduct, this method may be worth the effort because of the relatively high yield, especially if larval amphibians are targeted. These data also demonstrate that the sampling of amphibians and reptiles in Maryland’s coastal plain does not vary significantly among the months sampled. This result is consistent with data from a watershed in the highlands of Maryland (Strain et al. 2009). However, peak activity periods for the majority of temperate amphibians may occur in earlier months (Duellman and Trueb 1994, Stebbins and Cohen 1995), and an amphibian-monitoring program may benefit from sampling earlier. Also, sampling only during the summer months may not completely characterize the amphibian assemblage at a site, given different life histories and emergence times of some species (Stebbins and Cohen 1995). Smith and Grossman (2003) suggested that differences in larval abundance of Eurycea cirrigera (Southern Two-lined Salamander) from season to season were due to differences in seasonal microhabitat availability. 2012 G.F. Strain and R.L. Raesly 243 Temperature and rainfall may also influence the number of species encountered at a given site as well (Duellman and Trueb 1994), and differences among sites may be an artifact of weather or temperature. Some sampling methods detected species that others did not (Table 7), and this is consistent with previous studies (Fogarty and Jones 2003, Foley and Smith 1999). Drift-fence surveys detected 18 species total and 4 species that others did not, including 1 salamander species, Notophthalmus viridescens (Eastern Redspotted Newt). Cover boards detected 6 species total and 1 species that others did not, the salamander Plethodon cinereus (Eastern Red-backed Salamander). Electrofishing detected 11 species total and 1 species that others did not, the salamander Ambystoma maculatum (Spotted Salamander). ACS detected 19 species total and 6 species that others did not, including Desmognathus fuscus (Dusky Salamander). No unique taxa were detected by leaf-litter bags and quadrat leaflitter searches. Overall low captures of stream salamanders undoubtedly caused the low SSIBI scores. As previously mentioned, these scores were calculated with an IBI not designed for the coastal plain, and this result suggests that there may be a need for Table 7. Mean number of individuals (± 1 SE) of each taxon encountered with each technique (ACS = area-constrained survey, CB = cover boards, LLB = leaf-litter bags, QU = quadrat searches, DF = drift fences, EL = electrofishing). See Table 1 for genus names of species. Taxa ACS CB LLB QU DF EL A. crepitans 0.20 (0.07) - - - 0.02 (0.03) 0.03 (0.03) A. maculatum - - - - - 0.45 (0.45) A. opacum - 0.02 (0.02) - - 0.02 (0.02) - A. americanus 0.10 (0.05) - - - 0.10 (0.04) 0.02 (0.02) A. fowleri 0.05 (0.04) - 0.02 (0.02) - 0.10 (0.07) - A. sexlineata 0.02 (0.02) - - - - - C. serpentina 0.02 (0.02) - - - - 0.02 (0.02) C. picta 0.02 (0.02) - - - - - C. constrictor - - - - 0.02 (0.02) - D. fuscus 0.02 (0.02) - - - - - E. bislineata 0.20 (0.06) 0.12 (0.06) 0.17 (0.07) 0.07 (0.03) 0.05 (0.04) 0.80 (0.29) H. cinerea - - - 0.02 (0.02) 0.02 (0.02) - L. catesbeianus 0.03 (0.02) - - - 0.07 (0.03) 0.15 (0.06) L. c. melanotus 2.63 (0.42) 0.02 (0.02) - 0.07 (0.03) 0.63 (0.13) 2.37 (0.45) L. palustris 0.23 (0.10) - - - 0.03 (0.02) 0.07 (0.04) L. sphenocephala 0.38 (0.12) 0.05 (0.04) - - 0.05 (0.03) 0.15 (0.06) L. sylvaticus 0.02 (0.02) 0.02 (0.02) - 0.02 (0.02) 0.02 (0.02) - N. sipedon 0.02 (0.02) - - - - - N. viridescens - - - - 0.02 (0.02) - P. cinereus - - - - - P. crucifer 0.03 (0.03) - - - - - P. montanus 0.02 (0.02) - - - 0.02 (0.02) - P. ruber 0.07 (0.03) - - 0.02 (0.02) 0.02 (0.02) 0.03 (0.02) Psuedotriton spp. - - 0.07 (0.03) - - 0.72 (0.22) S. lateralis - - - - 0.02 (0.02) - S. allegheniensis 0.02 (0.02) - - - 0.02 (0.02) - T. carolina 0.05 (0.04) - - - - - T. sirtalis - - - - 0.02 (0.02) - 244 Northeastern Naturalist Vol. 19, No. 2 one. Rocco et al. (2004) found that high natural variability among streams across the Mid-Atlantic region resulted in poor classification efficiencies of their Stream Plethodontid Assemblage Response Index. They concluded that development of metrics for subdivided regions of more consistent habitat would improve metric performance. However, the low species diversity of stream salamanders along coastal-plain streams may preclude the development of a stream salamander-only IBI, and a more worthwhile direction to take may be an amphibian-IBI or a herpetofauna- IBI (which would include reptiles). Southerland et al. (2000) explored this possibility by developing IBIs that combined species richness metrics for different taxa, such as the number of all species (amphibian and reptile), number of frog and toad species, and the number of aquatic species. These combined metrics achieved classification efficiencies of 67.4%, but the authors ultimately deemed these metrics incapable of consistently discriminating between reference and non-reference condition in the coastal plain. Instead, they suggested that the numbers of each species encountered be used to further develop metrics. We agree, and suggest that metrics include a combination of species richness and abundance of both amphibians and reptiles, as this study and others have demonstrated the potentially high numbers of both taxa along small coastal-plain streams. Additional potential metrics could include indices that are commonly used in fish, such as biomass of certain taxa (stream salamanders, for instance) or total biomass (Miller et al. 1988), the Fulton condition factor, relative condition factor, and relative weight (Anderson and Neumann 1996). The results of this study demonstrate that area-constrained surveys and electrofi shing may be the most effective sampling techniques for targeting coastal plain herpetofauna in a long-term, large-scale monitoring program such as the MBSS. The results also demonstrate the abundance of herpetofauna along small headwater streams in this region and hence its potential for use in monitoring and maintaining the ecological integrity of those streams. Management recommendations As ACS and electrofishing were the best overall methods for maximizing species and individuals encountered, the sampling protocol that the MBSS currently has in place should be adequate, with a few suggestions. The MBSS Sampling Manual (Kazyak 2001) instructs crews to “collect/positively identify herpetofauna observed during electrofishing or other activities.” Dedicating time and a section of streambank exclusively to ACS would most likely maximize encounters by minimizing disturbance to herptiles before they are found. Also, great care should be taken to search all available microhabitats at a site (Foley and Smith 1999). Salamander larvae do not react to electrofishing in the same manner as most fishes. They tend to remain on the bottom, often blending in with the substrate unless they happen to roll over and expose their light bellies. Also, amphibians do not remain shocked as long as most fishes, usually recovering within seconds. Another difficulty is that smaller larvae easily escape through the 3.2-mm mesh of standard dipnets. In this study we switched to a dipnet with 2-mm mesh and had no further difficulty in this regard. The decrease in mesh size did not appear to interfere with sampling in any way. 2012 G.F. Strain and R.L. Raesly 245 Due to small sample sizes, it may be difficult to quantitatively assess population trends and distributions. Determining quantitatively whether or not an increase in the length of stream sampled increases the yield may be beneficial. However, estimates may be biased if they are calculated with the often incorrect assumption that species are equally detectable (ARMI 2006, Salvidio 2001). In order to be able to quantitatively assess salamander or other herpetofauna populations, it may be important to modify sampling protocols to estimate detection probabilities (ARMI 2006; Bailey et al. 2004; MacKenzie et al. 2002, 2003). Although sampling herpetofauna may be challenging, the prospects of using herpetofauna to monitor the health of small streams is good, especially in streams that may experience dry periods, thus preventing the sampling of other groups of organisms such as fishes and benthic macroinvertebrates. Herpetofauna, particularly stream salamanders, can be strong indicators of the biological integrity of streams (Rocco and Brooks 2000; Rocco et al. 2004; Southerland et al. 2000, 2004). Getting the most out of metrics that incorporate attributes of herpetofauna populations necessitates the use of methods which detect as many individuals as possible over a variety of habitat types. Electrofishing and area-constrained surveys are tools that may help accomplish this goal (Strain et al. 2009, this study). Acknowledgments This project was funded by a grant from the Maryland DNR, and we thank S. Stranko for the opportunity to work on it. We thank P. Bright (Mattowoman Natural Environment Area, Smallwood State Park) and C. Henderson (Doncaster Demonstration Forest) for permission to access the properties under their care. R. Chalmers, M. Southerland, E. Thompson, J. McCann, C. Swarth, R. Hilderbrand, S. Smith, R. Jung, and L. Smith assisted in developing the experimental design for this study. We thank E. Thompson for his assistance in identifying larval salamanders. We would like to thank the field crew, E. McGinley and J. Eells, for their long hours of work. J. Saville assisted with data entry and C. Saville created the map of the sites. Finally, we would like to thank R. Hilderbrand for his advice and assistance with statistical analysis of the data. We received approval from the Frostburg State University Institutional Animal Care and Use Committee to handle amphibians and reptiles. Literature Cited Amphibian Research and Monitoring Initiative (ARMI). 2006. The PAO Estimator. Available online at http://armi.usgs.gov/PAOEstimator.asp. Accessed 6 November 2006. Anderson, R.O., and R.M. Neumann. 1996. Length, weight, and associated structural indices. Pp. 447–482, In B.R. Murphy and D.W. Willis (Eds.). Fisheries Techniques, 2nd Edition. American Fisheries Society, Bethesda, MD. Bailey, L.L., T.R. Simons, and K.H. Pollock. 2004. Comparing population size estimators for plethodontid salamanders. Journal of Herpetology 38:370–380. Blaustein, A.R. 1994. Chicken Little or Nero’s fiddle? A perspective on declining amphibian populations. Herpetologica 50:85–97. Blaustein, A.R., and P.T. Johnson 2003. The complexity of deformed amphibians. Frontiers in Ecology and the Environment 1:87–94. 246 Northeastern Naturalist Vol. 19, No. 2 Boward, D., P. Kayzak, S. Stranko, M. Hurd, and A. Prochaska. 1999. From the mountains to the sea: The state of Maryland’s freshwater streams. Report EPA/903/R- 99/023. US Environmental Protection Agency,Washington, DC. Burton, T.M., and G.E. Likens. 1975. Salamander populations and biomass in the Hubbard Brook Experimental Forest, New Hampshire. Copeia 1975:541–546. Corn, P.S. 1994. Standard techniques for inventory and monitoring: Straight-line drift fences and pitfall traps. Pp. 109–117, In W.R. Heyer, M.A. Donnelley, R.W. McDiarmid, L.C. Hayek, and M.S. Foster (Eds.). Measuring and Monitoring Biological Diversity. Standard Methods for Amphibians, Smithsonian Institution, Washington, DC. Crump, M.L., and N.J. Scott, Jr. 1994. Standard techniques for inventory and monitoring: Visual encounter surveys. Pp. 84–92, In W.R. Heyer, M.A. Donnelley, R.W. McDiarmid, L.C. Hayek, and M.S. Foster (Eds.). Measuring and Monitoring Biological Diversity. Standard Methods for Amphibians, Smithsonian Institution, Washington, DC. 364 pp. Davic, R.D., and H.H. Welsh, Jr. 2004. On the ecological role of salamanders. Annual Review of Ecology, Evolution, and Systematics 35:405–434. Duellman, W.E., and L. Trueb. 1994. Biology of Amphibians. The Johns Hopkins University Press, Baltimore, MD. 670 pp. Enge, K.M. 1997. A standardized protocol for drift-fence surveys. Technical report number 14, Florida Game and Fresh Water Fish Commission, Tallahassee, FL. Fellers, G.M., and C.A. Drost. 1994. Supplemental approaches to studying amphibian biodiversity: Sampling with artificial cover. Pp. 146–150, In W.R. Heyer, M.A. Donnelley, R.W. McDiarmid, L.C. Hayek, and M.S. Foster (Eds.). Measuring and Monitoring Biological Diversity. Standard Methods for Amphibians, Smithsonian Institution, Washington, DC. 364 pp. Fogarty, J.H., and J.C. Jones. 2003. Pitfall trap versus area searches for herpetofauna research. Proceeding of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 57:268–279. Foley, D.H. III, and S.A. Smith. 1999. Comparison of two herpetofaunal inventory methods and an evaluation of their use in a volunteer-based statewide reptile and amphibian atlas project. Maryland Department of Natural Resources, Wildlife and Heritage Division, Wye Mills, MD. Fronzuto, J., and P. Verrell. 2000. Sampling aquatic salamanders: Tests of the efficiency of two funnel traps. Journal of Herpetology 34:146–147. Hyde, E.J., and T.R. Simons. 2001. Sampling plethodontid salamanders: Sources of variability. Journal of Wildlife Management 65:624–632. Jaeger, R.G. 1988. A comparison of territorial and non-territorial behaviour in two species of salamanders. Animal Behavior 36:307–310. Jaeger, R.G., and R.F. Inger. 1994. Standard techniques for inventory and monitoring: Quadrat sampling. Pp. 97–102, In W.R. Heyer, M.A. Donnelley, R.W. McDiarmid, L.C. Hayek, and M.S. Foster (Eds.), Measuring and Monitoring Biological Diversity. Standard Methods for Amphibians, Smithsonian Institution, Washington, DC. 364 pp. Jung, R.E., and C.H. Jagoe. 1995. Effects of low pH and aluminum on body size, swimming performance, and susceptibility to predation of Green Tree Frog (Hyla cinerea) tadpoles. Canadian Journal of Zoology 73:2171–2183. Jung, R.E., and T.K. Pauley. 2003. USGS Patuxent Wildlife Research Center, Manager’s monitoring manual: Leaf-litter bags. Available online at http://www.pwrc.usgs.gov/ monmanual/techniques/leaflitter.htm. Accessed 1 February 2005. 2012 G.F. Strain and R.L. Raesly 247 Jung, R.E., S. Droege, J.R. Sauer, and R.B. Landy. 2000. Evaluation of terrestrial and streamside salamander monitoring techniques at Shenandoah National Park. Environmental Monitoring and Assessment 63:65–79. Kazyak, P. 2001. Maryland Biological Stream Survey sampling manual. Maryland Department of Natural Resources, Annapolis, MD. Klauda, R., P. Kayzak, S. Stranko, M. Southerland, N. Roth, and J. Chaillou. 1998. Maryland biological stream survey: A state agency program to assess the impact of anthropogenic stresses on stream habitat quality and biota. Environmental Monitoring and Assessment 51:299–316. Lannoo, M. (Ed.). 2005. Amphibian Declines: The Conservation Status of United States Species. University of California Press, Berkeley, CA. MacKenzie, D.I., J.D. Nichols, G.B. Lachman, S. Droege, J.A. Royle, and C.A. Langtimm. 2002. Estimating site occupancy rates when detection probabilities are less than one. Ecology 83:2248–2255. MacKenzie, D.I., J.D. Nichols, J.E. Hines, M.G. Knutson, and A.B. Franklin. 2003. Estimating site occupancy, colonization, and local extinction when a species is detected imperfectly. Ecology 84:2200–2207. Marsh, D.M., and M.A. Goicochea. 2003. Monitoring terrestrial salamanders: Biases caused by intense sampling and choice of cover objects. Journal of Herpetology 37:460–466. Meteyer, C.U. 2000. Field guide to malformation of frogs and toads with radiographic interpretations. Biological Science Report USGS/BRD/BSR-2000-0005. US Geological Survey, Madison, WI. Miller, D.L., R.M. Hughes, J.R. Karr, P.M. Leonard, P.B. Moyle, L.H. Schrader, B.A. Thompson, R.A. Daniels, K.D. Fausch, G.A. Fitzhugh, J.R. Gammon, D.B. Halliwell, P.L. Angermeier, and D.J. Orth. 1988. Regional applications of an index of biotic integrity for use in water resource management. Fisheries 13:12–20. Monti, L., M. Hunter, Jr., and J. Witham. 2000. An evaluation of the artificial cover object (ACO) method for monitoring populations of the Redback Salamander, Plethodon cinereus. Journal of Herpetology 34:624–629. Paszkowski, C.A., G. Scrimgeour, B.A. Gingras, and S. Kendall. 2002. A comparison of techniques for assessing amphibian assemblages on streams in the western boreal forest. Canadian Field-Naturalist 116:116–119. Pauley, T.K. 1995. Aquatic salamanders. Pp. 14–22, In R.C. Reardon (Ed.). Effects of Diflubenzuron on Nontarget Organisms in Broadleaf Forested Watersheds in the Northeast, USDA Forest Service. FHM-NC-05-95. National Center of Forest Health Management, Morgantown, WV. 174 pp. Pauley, T.K., and M. Little. 1998. A new technique to monitor larval and juvenile salamanders in stream habitats. Banisteria 12:32–36. Rocco, G.L., and R.P. Brooks. 2000. Abundance and distribution of a stream Plethodontid salamander assemblage in 14 ecologically dissimilar watersheds in the Pennsylvania Central Appalachians. Report No. 2000-4, Penn State Cooperative Wetlands Center, PSU, University Park, PA. Rocco, G.L., R.P. Brooks, and J.T. Hite. 2004. Stream plethodontid assemblage response (SPAR) index: Development, application, and verification in the MAHA. Report No. 2004-1. Penn State Cooperative Wetlands Center, Forest Resources Laboratory, University Park, PA. Salvidio, S. 2001. Estimating terrestrial salamander abundance in different habitats: Effi ciency of temporary removal methods. Herpetological Review 32:21–23. 248 Northeastern Naturalist Vol. 19, No. 2 Smith, S., and G.D. Grossman. 2003. Stream microhabitat use by larval Southern Twolined Salamanders (Eurycea cirrigera) in the Georgia Piedmont. Copeia 2003:531– 543. Southerland, M.T. 1986. The effects of variation in streamside habitats on the composition of mountain salamander communities. Copeia 1986:731–741. Southerland, M.T., N. Roth, and G. Mercurio. 2000. Herpetofaunal indicator: Prospects using current and future data. Report to the Power Plant Research Program, Maryland Department of Natural Resources, Annapolis, MD. Southerland, M.T., R.E. Jung, D.P. Baxter, I.C. Chellman, G. Mercurio, and J.H. Volstad. 2004. Stream salamanders as indicators of stream quality in Maryland, USA. Applied Herpetology 2:23–46. Stebbins, R.C., and N.W. Cohen. 1995. A Natural History of Amphibians. Princeton University Press. Princeton, NJ. 316 pp. Strain, G.F., R.L. Raesly, and R.H. Hilderbrand. 2009. A comparison of techniques to sample salamanders assemblages along highland streams of Maryland. Environmental Monitoring and Assessment 156:1–16. Stuart, J.N., and M.L. Watson. 2001. Plastic netting: An entanglement hazard to snakes and other wildlife. Herpetological Review 32:162–164. Welsh, H.H., Jr., and L.M. Ollivier. 1998. Steam amphibians as indicators of ecosystem stress: A case study from California’s redwoods. Ecological Applications 8:1118– 1132. Willson, J.D., and M.E. Dorcas. 2003. Effects of habitat disturbance on stream salamanders: Implications for buffer zones and watershed management. Conservation Biology 17:763–71. Witham, J.W., E.H. Moore, M.L. Hunter, Jr., A.J. Kimball, and A.S. White. 1993. A longterm study of an oak-pine forest ecosytem: Techniques manual for the Holt Research Forest. Maine Agricultural Experiment Station, University of Maine, Orono, ME. Technical Bulletin 153. (Cited in Monti et al. 2000).