Northeastern Naturalist Vol. 21, No. 3 D.M. Haan and R.S. Halbrook 2014 357 2014 NORTHEASTERN NATURALIST 21(3):357–368 Resting-Site Selection of American Minks in East-Central New York Damon M. Haan1,* and Richard S. Halbrook1 Abstract - Research related to resting-site selection among native Neovison vison (American Mink) populations in North America is scarce. We radio-tracked 35 American Minks from March 2008–September 2011 along the Hudson River, NY, and its tributaries to evaluate resting-site characteristics and selection. We located 583 resting sites, consisting primarily of boulder piles and riprap along the Hudson River (35.6%) and bank burrows along tributaries (29.0%). Results of logistic regression analysis revealed that the percentage of shoreline cover with woody debris best predicted resting-site selection <10 m from water and shoreline cover was the most important single-parameter model. Shorelines that provide abundant cover may help to conceal American Mink activity while also providing resting sites that are safe from predators. Introduction Neovison vison Schreber (American Mink) are ecologically important riparian carnivores that occupy a high trophic level and are widely distributed throughout much of North America. American Minks are also highly valued among fur trappers, likely due to stable market demand (Eagle and Whitman 1987, Linscombe 1994). Despite their value, research pertaining to native American Mink populations is scarce. (Larivière 2003, Stevens et al. 1997). As much as 53% of wetlands in the United States were drained from 1780 to 1980 (Dahl 1990) and riparian habitat remains threatened by degradation, development, and recreational activities (Dahl and Allord 1997, Racey and Euler 1983). It is therefore important to establish a better understanding of American Mink habitat use in North America so that impacts on American Minks can be better understood. Resting sites are an important component of habitat use for many species, which may serve to stabilize predator and prey interactions and provide shelter from harsh weather conditions (Berryman and Hawkins 2006, Dunstone 1993, Zabala et al. 2003). Thus, selection of resting sites directly relates to survival, especially among small carnivores that may be susceptible to predation (Zielinski et al. 2004). The few studies in North America that have evaluated American Mink ecology have only broadly characterized resting sites. Although American Minks appear to use a variety of resting sites, most studies have found that American Minks primarily use abandoned bank burrows that were initially excavated by species such as Ondatra zibethicus L. (Muskrats) (Arnold and Fritzell 1989, Marshall 1936, Schladweiler and Storm 1969, Stevens et al.1997). However, a study in Idaho showed that American Minks largely used logjams as resting sites (Whitman 1981). In Ontario, 1Cooperative Wildlife Research Laboratory, Department of Zoology, Mailcode 6504, Southern Illinois University, Carbondale, IL 62901. *Corresponding author - damon.haan@gmail.com. Manuscript Editor: Tom French Northeastern Naturalist 358 D.M. Haan and R.S. Halbrook 2014 Vol. 21, No. 3 Canada, American Mink den sites were primarily associated with cover provided by coniferous forests that had minimal impact from human development (Racey and Euler 1983). Quantitative research regarding microhabitat selection at American Mink resting sites in North America is lacking. American Minks have been more extensively researched outside of North America where they are considered invasive. Invasive American Minks in Europe also use a variety of resting sites distributed throughout their home range; these sites are typically located near water and in close proximity to foraging areas (Birks and Linn 1982, Garcia et al. 2010, Gerell 1970). In Spain, invasive American Mink resting sites were largely associated with areas of dense scrub (Zabala et al. 2007), whereas in the United Kingdom resting sites were primarily located in rabbit warrens (Yamaguchi et al. 2003). Furthermore, invasive American Mink research suggests that the availability of resting sites may limit populations in some areas (Halliwell and Macdonald 1996) and resting-site use may be influenced by the width or depth of adjacent waterbodies (Sidorovich et al. 1996, Zabala et al. 2007). Resting-site selection does not appear to differ between sexes (Zabala et al. 2007). Results of invasive American Mink studies may only be tenuously applied to native populations, and many of the studies have focused on evaluating measures to aid with control or eradication (Bonesi and Palazon 2007, Melero et al. 2008, Reynolds et al. 2010). The topic of habitat features surrounding American Mink resting sites has been relatively unexplored, especially among native populations. Because resting sites are an important feature in habitat use and their abundance may limit density, it is important to elucidate American Mink resting-site selection. The objectives of our study were to (1) quantitatively evaluate microhabitat characteristics surrounding American Mink resting sites compared with control resting sites, which we considered as unused and potentially available resting sites, and (2) broadly characterize and compare resting-site structures and use along the Hudson River and its comparatively smaller tributaries. Field-site Description We conducted this study in east-central New York along the Hudson River between Fort Edward and Mechanicville, extending along its tributaries up to 8.5 km waterway distance from the Hudson River (Fig. 1). The landscape consists of valleys and rolling foothills with moderate urbanization, abundant agriculture, and fragmented mixed and deciduous forests. The climate of the study area is northern temperate with annual rainfall and snowfall averages of 99.2 cm and 172.2 cm, respectively, and average temperatures ranging from 12.7 to 26.0 °C in summer and -11.6 to -0.1 °C in winter (NOAA 2013). The width of the Hudson River and its tributaries in the study area generally ranged 95–225 m and <1–15 m, respectively. The Hudson River is controlled by a series of locks and dams, including a dam upstream of Fort Edward that often caused water levels to fluctuate approximately 15–46 cm daily. Northeastern Naturalist Vol. 21, No. 3 D.M. Haan and R.S. Halbrook 2014 359 Methods Capture and handling We trapped American Minks during late February 2008–April 2011, September– mid November 2010, and late July–mid August 2010–2011 using 10–60 Tomahawk double-door live traps (Numbers 104, 105.5, 203; Tomahawk Live Trap, Tomahawk, WI). We set traps <2 m from water along the Hudson River and its tributaries and camouflaged the exterior and floor of each trap using natural materials such as leaves, grass, mud, and tree bark. We primarily relied on blind sets, but also occasionally used commercial lures and canned sardines. We checked traps 1–3 times per day, depending on weather conditions. We transported captured American Minks to a nearby veterinary clinic where they were transferred to an induction box and anesthetized using sevoflurane gas. Once American Minks were immobilized, we placed them on a surgery table and maintained anesthesia by continuously delivering sevoflurane gas at 4% into a facemask fitted on the American Minks throughout the following procedures. First, we prepped American Minks with povidone iodine and isopropyl alcohol. We then made a 1–2 cm incision dorsally between the scapulae and subcutaneously inserted a radio-transmitter (Lander et al. 2005; 9 g [Model# SI-2T] or 5 g [Model# SB- 2T], Holohil Systems Ltd., ON, Canada). We pushed radio-transmitters away from Figure 1. Trapping locations for American Minks along the Hudson River and its tributaries between Mechanicville and Fort Edward in east-central New York, 2008–2011. The Hudson River separates Saratoga County from Washington and Rensselaer counties. Northeastern Naturalist 360 D.M. Haan and R.S. Halbrook 2014 Vol. 21, No. 3 the incision to minimize wound dehiscence (Schulz et al. 1998) and then sutured and glued the incision closed. We also ear-tagged (Number 1005-3, National Band and Tag Co., Newport, KY) American Minks and administered an intramuscular injection of antibiotics (8 mg/kg bodyweight of cefovecin; ConveniaTM, Zoetis, Kalamazoo, MI). We sterilized radio-transmitters >24 hrs prior to use with ethylene oxide gas. Radio-transmitters had a 5–18-mo nominal battery life and a mortality sensor that corresponded to body temperature. Following the procedure, we removed the facemask and placed American Minks back inside the box traps. Once American Minks fully recovered from the anesthesia (typically 5–10 min), we immediately returned to their capture sites and released the animals. Trapping, capture, and handling procedures followed an approved Southern Illinois University, Carbondale, IL, animal care and use protocol, guidelines from the American Society of Mammalogists (Gannon et al. 2007), and collection permits issued by the New York State Department of Environmental Conservation, Albany, NY. Radiotelemetry We radio-tracked American Minks March–June 2008 and 2009 and March 2010–September 2011 between sunrise and sunset, when American Minks are least likely to be active (Arnold and Fritzell 1987). We used 3-element yagi antennas and handheld receivers (R-1000, Communications Specialists, Inc., Orange, CA) for all radio-tracking. After detecting a transmitter signal, we listened for variations in signal strength while standing stationary for 1–5 min to determine if American Minks were active or inactive (Arnold and Fritzell 1987). We used the homing technique described by Mech (1983) to identify the precise locations of American Minks. We did not consider locations as resting sites unless we were absolutely certain that the precise resting site had been located. Furthermore, visual clues including scat, tracks, and prey remains were often observed near resting sites. We used headphones to eliminate ambient receiver sound, and we were quiet to the greatest extent practicable while in close proximity to American Mink locations. As soon as resting sites were identified, we marked their locations using a global positioning system (Garmin eTrex; Garmin International, Inc., Olathe, KS). We defined resting sites as cavities or sheltered areas where American Minks rest when inactive. For all resting sites, we broadly characterized their structure, distance from water, and whether it was located closer to the Hudson River or tributaries. We also characterized mean resting-site use by determining number of resting sites, number of revisited resting sites, and number of revisits to resting sites for our most intensely monitored American Minks, arbitrarily considered as those with >30 telemetry locations on different days. Microhabitat evaluation During April–June 2008 and 2010, we evaluated microhabitat variables identified in the US Fish and Wildlife Service’s American Mink-habitat suitability-index model (Allen 1986) at randomly selected resting sites no more than 7 d after their use by American Minks. For resting sites <10 m from water, we established a 10 m x 2 m transect centered at the resting site and parallel to the shoreline where we visually estimated the percent cover to the nearest 10% for exposed roots, overhanging Northeastern Naturalist Vol. 21, No. 3 D.M. Haan and R.S. Halbrook 2014 361 vegetation, undercut banks, shrubs (1–5 m tall), woody debris and boulders (≥10 cm in diameter), and shoreline cover (all habitat features that would aid to conceal an American Mink). We established a second 10 m x 2 m transect also centered at the resting site, but perpendicular to the shoreline where we recorded only shrubs at 1 m intervals. Using a spherical concave densiometer, we recorded the percent canopy cover on opposite ends of both transects and directly above the resting site. For resting sites located >10 m from water, we established two 10 m x 2 m transects centered at the resting site, one running east–west and the other north–south, where we recorded canopy cover, ground cover, and shrub cover in the same manner as resting sites <10 m from water. For every evaluated resting site, we searched 10–150 m from the resting site for a control resting site. We considered control resting sites suitable for American Minks where the entrance diameter was ≥10 cm (Schladweiler and Storm 1969) and, based on visual inspection, it seemed the cavity would sufficiently house a 1.5-kg American Mink (Larivière 2003). If we were doubtful about a control resting site’s suitability for American Minks, then we searched for a dif ferent site. We evaluated the nearest suitable control resting site we could find. If the resting site was located <10 m from water, then we searched for a control resting site that was also <10 m from water along the shoreline, whereas for resting sites >10 m from water, we searched in a north–south and east–west direction to locate a control resting site that was also >10 m from water. Statistical analysis We used logistic regression, PROC LOGISTIC in SAS version 9.1 (SAS Institute, Inc., Cary, NC), to model microhabitat variables that best predicted American Mink resting sites in two classes: <10 m and >10 m from water. The binary response variables in both cases were resting sites and control resting sites, and the independent variables were the 8 microhabitat and 3 microhabitat variables for resting sites <10 m and >10 m from water, respectively. We used the Hosmer-Lemeshow goodness-of-fit test to evaluate models and withdrew those with P < 0.05. For resting sites <10 m from water, we ranked models by their Akaike information criterion (AIC) value and considered the lowest-ranked to be the top model. We used a corrected AIC (AICc) as a bias adjustment for resting sites >10 m from water due to small sample size (Burnham and Anderson 1998). For all models, we included the number of parameters (K), the change in AIC from the top model relative to subsequent models (ΔAIC), and each model’s weight (wi), which assesses the strength of evidence. We considered models with ≤2 ΔAIC as competitive with the top model. All microhabitat variables were individually modeled for each analysis (i.e., resting sites <10 m and >10 m from water), and we constructed multivariable models for variables believed most likely to influence resting-site selecti on. Results Capture We captured 41 (n = 27 male, 14 female) American Minks, 6 (n = 5 male, 1 female) of which were captured twice in 7786 trap-nights (0.6 American Minks/100 Northeastern Naturalist 362 D.M. Haan and R.S. Halbrook 2014 Vol. 21, No. 3 trap-nights). Waterway distance of captures from the Hudson River included 10 (24.4%) that were <0.1 km, 16 (39.0%) that were 0.1–1.0 km, 10 (24.4%) that were 1.0–3.0 km, 4 (9.8%) that were 3.0−6.0 km, and 1 (2.4%) that was 8.5 km (Fig. 1). We implanted radio-transmitters in 35 captured American Minks (n = 24 males, 11 females), with 7 males receiving 9-g radio-transmitters, and the remaining 28 receiving 5-g radio-transmitters. We did not implant radio-transmitters in 6 captured American Minks due to either trap mortality (n = 2 males, 3 females) or poor body condition (n = 1 male). Radio-transmitters comprised a mean of (± SE) 0.9 ± 0.05 % of body weight. Resting-site characteristics We located 583 different resting sites, including 180 along the Hudson River and 403 along tributaries. Boulder piles and riprap were the dominant resting-site structures used along the Hudson River (35.6%), whereas, bank burrows were the most frequently used structures along tributaries (29.0%) (Table 1). Mean restingsite distance from water was 4.0 ± 0.55 m along the Hudson River and 8.9 ± 1.04 m along tributaries. Along the Hudson River, 170 of 180 (94.4%) resting sites were <10 m from water and the farthest was 51 m, whereas along tributaries, 329 of 403 (81.6%) resting sites were <10 m from water and the farthest was 194 m. Mean number of resting sites among 12 American Minks that used the Hudson River with >30 telemetry locations on different days was 13.3 ± 3.0, whereas the 21 American Minks along tributaries with >30 telemetry locations on different days had a mean of 16.8 ± 3.7 resting sites (Fig. 2). Resting-site selection We randomly selected and evaluated 91 resting sites and 91 control resting sites (n = 182) <10 m from water to examine microhabitat use. We dropped 1 control resting site and its associated used resting site from analysis because an American Mink used the control resting site after it was evaluated for microhabitat Table 1. Resting-site structures used by all radio-monitored American Minks (n = 35) along the Hudson River and its tributaries. All 35 American Minks used resting sites along tributaries, whereas 19 American Minks also used resting sites along the Hudson River. Multiple visits to resting sites are not included. Percentages reflect the proportion of total restin g-site use for each respective column. Resting-site structure Hudson River Tributaries Overall Bank burrows 29 (16.1%) 117 (29.0%) 146 (25.0%) Brushpiles and logjams 23 (12.8%) 96 (23.8%) 119 (20.4%) Boulder piles and riprap 64 (35.6%) 4 (1.0%) 68 (11.7%) Dense vegetation 2 (1.1%) 64 (15.9%) 66 (11.3%) Cavities among exposed roots 17 (9.4%) 37 (9.2%) 54 (9.3%) Cavities in logs 13 (7.2%) 29 (7.2%) 42 (7.2%) Cavities in trees or snags 14 (7.8%) 20 (5.0%) 34 (5.8%) Undercut banks 7 (3.9%) 19 (4.7%) 26 (4.5%) Human-made structures 7 (3.9%) 4 (1.0%) 11 (1.9%) Refuse piles 1 (0.6%) 10 (2.5%) 11 (1.9%) Beaver lodges 3 (1.7%) 3 (0.7%) 6 (1.0%) Total 180 403 583 Northeastern Naturalist Vol. 21, No. 3 D.M. Haan and R.S. Halbrook 2014 363 characteristics. The model with the lowest AIC value for resting sites comprised shoreline cover + woody debris, which had moderate support (wi = 0.40) for predicting American Mink resting-site selection. There were 2 competitive models ≤2 ΔAIC from the top model including overhanging vegetation + shoreline cover + woody debris, and shoreline cover, a single-parameter model. The shoreline cover variable was included in all of the top 3 models and had the lowest AIC value among all 8 single-parameter models, while receiving moderate to low support (wi = 0.17) for predicting resting-site selection (Table 2). The 2 variables found in the top model and competing models, shoreline cover and woody debris, had positive model coefficients of β = 0.03. Overhanging vegetation, a variable in the second best model, had a positive model coefficient of β = 0.02. Both the shoreline cover and overhanging vegetation variables had significantly greater occurrence at resting sites than control resting sites (shoreline cover: t180 = 4.10, P < 0.0001; overhanging vegetation: t180 = 2.12, P = 0.034), and woody debris was nearly significant (t180 = 1.95, P = 0.052). For resting sites located >10 m from water, we randomly selected and evaluated microhabitat characteristics at 18 resting sites and 18 control resting sites Figure 2. Mean resting-site use for intensely monitored American Minks (i.e., >30 telemetry locations on different days) along the Hudson River (n = 160 sites, 12 American Minks), its tributaries (n = 352 sites, 21 American Minks), and overall (n = 512 sites, 21 American Minks). Revisits were occasions when American Minks vacated a resting site between consecutive daytime-telemetry locations before returning. Bars indicate SE. Northeastern Naturalist 364 D.M. Haan and R.S. Halbrook 2014 Vol. 21, No. 3 (n = 36). The model with the lowest AICc value was groundcover, a single-parameter variable, which had moderate support (wi = 0.30) for predicting American Mink resting-site selection (Table 3). None of the microhabitat variables evaluated >10 m from water had a significantly greater occurrence at resting sites relative to control resting sites, but groundcover appeared to have some association with resting sites (t36 = 1.12, P = 0.26) and had a positive model coefficient of β = 0.01. Canopy cover had a positive model coefficient of β = 0.004, and shrubs had a negative model coefficient of β = -0.02 (canopy cover: t36 = 0.36, P = 0.72; shrubs: t36 = -0.84, P = 0.40). Discussion Resting sites are an important component in American Mink habitat use that can limit density and distribution (Halliwell and Macdonald 1996) and may influence survival (Zielinski et al. 2004). In our study, the abundance of shoreline cover and woody debris were the best predictors for American Mink resting-site selection <10 m from water. Shoreline cover was the most important variable, and overhanging vegetation also appeared to be a desirable habitat feature for Ameri- Table 3. Logistic regression results for microhabitat characteristics influencing American Mink restingsite selection >10 m from water. The model with the lowest AICc (corrected) was considered the best predictor of American Mink resting-site selection and those with ≤2.0 ΔAICc were considered competitive with the top model. K is the number of parameters, and wi denotes the weighting factor for each model. Model K R2 AICc ΔAICc wi Groundcover 2 0.03 52.99 0.00 0.30 Shrubs 2 0.02 53.45 0.46 0.24 Groundcover + shrubs 3 0.07 54.01 1.02 0.18 Canopy cover 2 0.004 54.14 1.15 0.17 Groundcover + canopy cover 3 0.04 55.06 2.07 0.11 Table 2. Logistic regression results for microhabitat characteristics influencing American Mink resting- site selection along shorelines <10 m from water. The model with the lowest AIC was considered the best predictor of American Mink resting-site selection and those with ≤2.0 ΔAIC were considered competitive. K is the number of parameters, and wi denotes the weighting factor for each model. Model K R2 AIC ΔAIC wi Shoreline cover + woody debris 3 0.12 232.96 0.00 0.40 Shoreline cover + woody debris + overhanging vegetation 4 0.12 234.27 1.30 0.21 Shoreline cover 2 0.10 234.68 1.72 0.17 Shoreline cover + overhanging vegetation 3 0.10 235.86 2.89 0.09 Boulders + shoreline cover 3 0.10 236.10 3.14 0.08 Boulders + overhanging vegetation + shoreline cover 4 0.11 237.38 4.42 0.04 Boulders 2 0.04 247.12 14.16 0.0008 Overhanging vegetation 2 0.03 248.84 15.88 0.0004 Woody debris 2 0.02 249.28 16.32 0.0003 Shrubs 2 0.02 250.34 17.38 0.0002 Exposed roots 2 0.01 252.21 19.25 0.0001 Canopy cover 2 0.01 252.29 19.33 0.0001 Undercut banks 2 0.002 253.11 20.15 0.00004 Northeastern Naturalist Vol. 21, No. 3 D.M. Haan and R.S. Halbrook 2014 365 can Mink resting-site selection. These results are similar to Zabala et al. (2007), who determined that invasive American Mink resting sites were associated with areas of dense scrub cover. American Minks’ relatively small size makes them vulnerable to an array of avian and mammalian predators (Larivière 2003); therefore, resting sites in areas with abundant shoreline cover, where American Minks are inconspicuous, are important for survival. Predation believed to have been caused by Buteo spp. (hawks), Bubo virginianus Gmelin (Great Horned Owl), and Vulpes vulpes L. (Red Fox) accounted for 8 American Mink mortalities in our study, including 5 that occurred in areas with distinctly poor shoreline cover and 3 in areas with moderate–poor shoreline cover. Prey remains were often observed near the entrances of resting sites, indicating that American Minks also used resting sites as safe locations to consume prey, which is supported by Birks’ and Linn’s (1982) findings that invasive American Mink resting sites are generally located near foraging areas. None of the 3 microhabitat variables analyzed for resting sites >10 m from water were significantly different at resting sites relative to control resting sites. The lack of difference between the two site-classes may be attributable to our relatively low sample size (n = 36). Alternatively, as American Minks stray farther from shorelines, resting-site availability may be diminished to such an extent that American Minks become less selective, and areas with abundant cover are less important. The top 2 overall resting-site structures in our study were bank burrows, and brushpiles and logjams. These were also among the most frequently used structures reported by other native American Mink studies (Stevens et al. 1997, Whitman 1981). We chose to combine brushpiles and logjams into one variable because of their similar structural characteristics. American Minks along the Hudson River mostly used riprap as resting sites; riprap was found in much greater abundance along the Hudson River than it was along tributaries, thereby indicating an opportunistic aspect to resting-site use. Along tributaries, spring flooding and heavy rainfall created logjams and log piles that American Minks regularly used as resting sites. We also observed American Minks using human-made structures such as an abandoned shed, an active chicken coop, and several refuse piles, indicating they had a moderate tolerance of human-related habitat disturbance. Invasive American Minks have also been shown to use various human-made structures as resting sites including buildings, rock walls, and bridge parapets (Birks and Linn 1982, Garcia et al. 2010, Zabala et al. 2007). We determined that the abandoned shed and refuse piles in our study offered American Minks abundant foraging opportunities, namely small-mammal prey, as well as safe refuge from predators. American Mink resting sites in our study were usually located within a short distance from water. Resting-site distance from water is likely influenced by the availability of resting sites, cover, and proximity to foraging areas. In our study, the overall resting-site distance from water was similar to those reported for other native and invasive American Mink studies (Stevens et al. 1997, Whitman 1981, Zabala et al. 2007). Mean resting-site distance from water along tributaries was more than twice as far as those we recorded along the Hudson River. This difference Northeastern Naturalist 366 D.M. Haan and R.S. Halbrook 2014 Vol. 21, No. 3 was likely attributable to the greater availability of riprap along the Hudson River, which offered many resting-site opportunities in very close proximity to water. Moreover, human activity (e.g., roads and residences) was more prevalent along the Hudson River than along tributaries and may have restricted distances American Minks traveled away from water for resting sites. Habitat quality and prey availability likely have an important influence on how many resting sites American Minks require. American Minks in our study had a greater mean number of resting sites along tributaries than the Hudson River likely because none of the American Minks that used the Hudson River were solely dependent on it; rather, they also used reaches of its tributaries. Furthermore, 6 of 9 intensely monitored American Minks that never used the Hudson River (i.e., only located along tributaries) had at least 1 resting site <1.5 km (waterway distance) from the Hudson River, despite their capacity to travel ≥4.3 km in 1 night (Stevens et al. 1997). This finding suggests that American Minks may not have a preference for large rivers over smaller tributaries. There is scant evidence of territorial behavior in American Minks, and we doubt that it was a factor in our study (Larivière 2003). Shoreline cover, especially woody debris in the form of brushpiles, logjams, snags, and fallen logs affords American Minks ample resting and feeding sites while also providing protection from predators and harsh weather conditions. Burrowing animals living near shorelines, primarily Muskrats, provide American Minks with a source of food and resting sites. American Minks living along the Hudson River system appear to use a mean of at least 24 resting sites scattered throughout their home ranges and located near foraging areas. Currently, there is very little research regarding upland habitat use by American Minks, especially as it relates to resting-site selection. Upland habitat likely serves a number of important functions such as access to terrestrial prey. Therefore, future research should examine American Mink habitat use >10 m from water. There is also a need to elucidate American Mink tolerance levels of human disturbance as it relates to habitat quality and prey availability. Acknowledgments Funding was provided by the General Electric Company. We are tremendously grateful to the dedicated technicians who assisted with this research including H. Cimino, J. D’Agostino, A. Goldberg, K. Grmusich, J. Histed, S. McKenzie, A. VanVossen, and W. White. We are also appreciative of the assistance provided by T. McClenahan and P. Bernstein. We cannot give enough thanks to the Battenkill Veterinary Clinic staff, particularly G. Allen and M. Horn. This project would not have been possible had it not been for the many friendly landowners we encountered living near the Hudson River in Rensselaer, Saratoga, and Washington counties who granted us access to their properties. We also would like to extend a special thank you to C. Nielsen, J. Reeve, the Cooperative Wildlife Research Laboratory, the Department of Zoology, and the Graduate School at Southern Illinois University, Carbondale, IL for their assistance. Northeastern Naturalist Vol. 21, No. 3 D.M. Haan and R.S. Halbrook 2014 367 Literature Cited Allen, A.W. 1986. Habitat-suitability index models: Mink, revised. US Fish and Wildlife Service, Biological Report 82 (10.127). Fort Collins, CO. 23 pp. Arnold, T.W., and E.K. Fritzell. 1987. Activity patterns, movements, and home ranges of Prairie Mink. Prairie Naturalist 19:25–32. Arnold, T.W., and E.K. Fritzell. 1989. Spring and summer prey remains collected from male mink den sites in southwestern Manitoba. Prairie Naturalist 21:189–192. Berryman, A.A., and B.A. Hawkins. 2006. The refuge as an integrating concept in ecology and evolution. Oikos 115:192–196. Birks, J.D.S., and I.J. Linn. 1982. Studies of the home range of the feral mink, Mustela vison. Symposium of the Zoological Society of London 49:231–257. Bonesi, L., and S. Palazon. 2007. The American Mink in Europe: Status, impacts, and control. Biological Conservation 134:470–483. Burnham, K.P., and D.R. Anderson. 1998. Model Selection and Inference: A Practical Information-theoretic Approach. Springer-Verlag, New York, NY. 353 pp. Dahl, T.E. 1990. Wetland losses in the United States, 1780s to 1980s. US Department of the Interior, Fish and Wildlife Service, Washington, DC. 13 pp. Dahl, T.E., and G.J. Allord. 1997. Technical aspects of wetlands: History of wetlands in the conterminous United States. US Geological Survey Water-supply Paper 2425. Available online at http://water.usgs.gov/nwsum/WSP2425/history.html. Accessed 10 June 2013. Dunstone, N. 1993. The Mink. T. and A.D. Poyser Ltd. London, UK. 232 pp. Eagle, T.C., and J.S. Whitman. 1987. Mink. Pp. 615–624, In M. Novak, J.A. Baker, M.E. Obbard, and B. Malloch (Eds.). Wild Furbearer Management and Conservation in North America. Ontario Ministry of Natural Resources, Toronto, ON, Canada. 1150 pp. Gannon, W.L., S. Sikes, and the animal care and use committee of the American Society of Mammalogists. 2007. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. Journal of Mammalogy 88:809–823. Garcia, P., V. Arévalo, and M. Lizana. 2010. Characterisation of resting sites of American Mink, Neovison vison, in central Spain. Wildlife Biology 16:276–282. Gerell, R. 1970. Home ranges and movements of the Mink, Mustela vison Schreber, in southern Sweden. Oikos 21:160–173. Halliwell, E.C., and D.W. Macdonald. 1996. American Mink, Mustela vison, in the upper Thames catchment: Relationship with selected prey species and resting availability. Biological Conservation 76:51–56. Lander, M.E., M. Haulena, F.M.D. Gulland, and J.T. Harvey. 2005. Implantation of subcutaneous radio transmitters in the Harbor Seal (Phoca vitulina). Marine Mammal Science 21:154–161. Larivière, S. 2003. Mink. Pp. 662–671, In G.A. Feldhamer, B.C. Thompson, and J.A. Chapman (Eds.). Wild mammals of North America: Biology, Management, and Conservation. 2nd Edition. John Hopkins University Press, Baltimore, MD. 1216 pp. Linscombe, G. 1994. United States fur harvest (1970–1992) and fur value (1974–1992) statistics by state and region. Louisiana Department of Wildlife and Fisheries, Baton Rouge, LA. Marshall, W.H. 1936. A study of the winter activities of the mink. Journal of Mammalogy 4:382–292. Mech, L.D. 1983. Handbook of Animal Radio-tracking. University of Minnesota Press, Minneapolis, MN. 108 pp. Northeastern Naturalist 368 D.M. Haan and R.S. Halbrook 2014 Vol. 21, No. 3 Melero, Y., S. Palazón, E. Revilla, J. Martelo, and J. Gosàlbez. 2008. Space use and habitat preferences of the invasive American Mink (Mustela vison) in a Mediterranean area. European Journal of Wildlife Research 54:609–617. National Oceanic and Atmospheric Administration (NOAA). 2013. Glens Falls, NY, normals: 1981–2010. Available online at http://www.weather.gov/media/aly/Climate/KGFL _ Normals_ 1981-2010.pdf. Accessed 10 August 2013. Racey, G.D., and D.L. Euler. 1983. Changes in mink habitat and food selection as influenced by cottage development in Central Ontario. Journal of Applied Ecology 20:387–402. Reynolds, J.C., T.A. Porteus, S.M. Richardson, R.J. Leigh, and M.J. Short. 2010. Dectectability of American Mink using rafts to solicit field signs in a population control context. Journal of Wildlife Management 74:1601–1606. Schladweiler, J.L., and G.L. Storm. 1969. Den use by mink. Journal of Wildlife Management 33:1025–1026. Schulz, J.H., A.J. Bermudez, J.L. Tomlinson, J.D. Firman, and Z. He. 1998. Effects of implanted radiotransmitters on captive Mourning Doves. Journal of Wildlife Management 62:1451–1460. Sidorovich, V.E., B. Jedrzejewska, and W. Jedrzejewska. 1996. Winter distribution and abundance of mustelids and beavers in the river valleys of Bialowieza Primeval Forest. Acta Theriologica 41:155–170. Stevens, R.T., T.L. Ashwood, and J.M. Sleeman. 1997. Fall–early winter home-ranges, movements, and den use of male Mink, Mustela vison, in eastern Tennessee. Canadian Field-Naturalist 111:312–314. Whitman, J.S. 1981. Ecology of the Mink (Mustela vison) in west-central Idaho. M.Sc. Thesis. University of Idaho, Moscow, ID. 101 pp. Yamaguchi, N., S. Ruston, and D.W. Macdonald. 2003. Habitat preferences of feral American Mink in the Upper Thames. Journal of Mammalogy 84:1356–1373. Zabala, J., I. Zuberogoitia, I. Garin, and J. Aihartza. 2003. Landscape features in the habitat selection of European Mink (Mustela lutreola) in southwest Europe. Journal of Zoology, London 260:415–421. Zabala, J., I. Zuberogoitia, and J.A. Martínez-Climent. 2007. Winter habitat preferences of feral American Mink, Neovison vison, in Biscay, Northern Iberian Peninsula. Acta Theriologica 52:27–36. Zielinski, W.J., R.L. Truex, G.A. Schmidt, F.V. Schlexer, K.N. Schmidt, and R.H. Barrett. 2004. Resting-habitat selection by Fishers in California. Journal of Wildlife Management 68:475–492.