Southeastern Naturalist 61 T. Seaborn and K. Catley 22001166 SOUTHEASTERN NATURALIST Vo1l5.( 115):,6 N1–o7. 51 Abiotic Microhabitat Parameters of the Spruce–fir Moss Spider, Microhexura montivaga Crosby and Bishop (Araneae: Dipluridae) Travis Seaborn1,2,* and Kefyn Catley1 Abstract - The Spruce–fir Moss Spider (Microhexura montivaga) is a federally endangered species found only in the high-elevation southern Appalachian spruce–fir forests. Little is known about the basic ecology of the spider. The goal of this project was to determine the temperature and humidity parameters of the microhabitat around known spider locations. iButton temperature and humidity data loggers were placed at sites on Mt. Lyn-Lowry, Browning Knob, Whitetop Mountain, and Mt. Rogers (a range that encompasses all metapopulations). No statistically significant (P > 0.05) differences in humidity between positive and negative presence sites, among metapopulations, or individual sites were found. Temperature data showed varied results. This research provides a number of applications for the conservation and management of the Spruce–fir Moss Spider, such as understanding metapopulation variation, better husbandry techniques, and using collected data to determine conversion factors/models for temperature data between microhabitat measurements and larger-scale measuring methods. Introduction The endangered southern Appalachian endemic Microhexura montivaga Crosby and Bishop (Spruce–fir Moss Spider) is the world’s smallest and northernmost member of the family Dipluridae, more commonly known as the funnel-web tarantulas. Diplurids are generally found worldwide within the tropics, with most species found in South and Central America and Australia, although they can also be found in India and Africa. Microhexura is the northernmost genus found in the temperate zone. There are a total of 24 genera with 181 species in Dipluridae (Platnick 2014). Raven (1985) describes 3 diagnostic characters for the Dipluridae: lowered caput and elevated thoracic region, elongated lateral spinnerets composed of 3 sections, and widely separated sections of the spinnerets. Using Microhexura as an informative outgroup to the rest of the diplurids may be indicated. However, due to the high level of derived specialized characters and unique habitat requirements, its utility as an outgroup is debatable (Coyle 1995). The Spruce–fir Moss Spider ranges in size from 2.5 mm to 5.6 mm (Coyle 1981) and is restricted to the southern Appalachian Mountains. Although listed as endangered since 1994 (Fridell 1994), minimal research has been done on the basic ecology of the Spruce–fir Moss Spider. The overarching purpose of this project was to define habitat correlates of the Spruce–fir 1Department of Biology, Western Carolina University, Cullowhee NC 28723. 2Current address - School of Biological Sciences, Washington State University, Pullman, WA 99163. *Corresponding author - travis.seaborn@email.wsu.edu. Manuscript Editor: Richard Baird Southeastern Naturalist T. Seaborn and K. Catley 2016 Vol. 15, No. 1 62 Moss Spider and fill in knowledge gaps that are preventing proper management of this endangered species, such as proper husbandry techniques. Spruce–fir Moss Spider webs of are confined to bryophyte mats and appear as messy tangles of flat tubes and sheets in the interstitial space between the rock substrate and the bryophyte mat. Although their diet has not been confirmed, springtails (Collembola) and mites (Acari) are assumed to play a role due to their great abundance in leaf-litter/bryophyte habitats in general (Coyle 1981) . Spruce–fir Moss Spiders attain maturity in 2–3 years, with females laying eggs in June and spiderlings emerging in September (Coyle 1981). Mating occurs in the fall; once males have completed their last molt, they leave their webs in search of females and die that winter (Coyle 1981). Male mating behavior is triggered by the presence of a female’s web (Coyle 1985). Dispersal strategies, which might play a vital role in the biology of this species, are still somewhat debated. Microhexura idahoana Chamberlin and Ivie 1945, the sister species found in the western United States, has been reported from snowfields, giving rise to the idea that ballooning may occur (Coyle 1981). However, because millipedes have also been collected from snowfields, presence there does not mean that aerial dispersal is required (Crawford and Edwards 1986). If dispersal is not aerial, the very small size of this animal suggests that movement, even across a single mountainside, or from one rock outcrop to another, may prove very infrequent. Genetic flow among and between metapopulations is currently being studied by Dr. Marshall Hedin of San Diego State University. However, previous work from a small sample size suggested that the metapopulations were in fact isolated populations with minimal gene flow (Martens 2005). As part of its conservation efforts, the US Fish and Wildlife Service (USFWS) would like to determine the validity of artificially increasing gene flow either in the field or lab, but it is imperative to fully understand the habitat requirements before moving any individuals to a new environment . Originally collected and described by Crosby and Bishop (1925), the Spruce–fir Moss Spider was added to the Federal List of Endangered and Threatened Wildlife and Plants by the USFWS in 1994 (Fridell 1994). The reason was two-fold: the relatively low abundance of the species and the rapid deterioration of suitable habitat. The known Spruce–fir Moss Spider population is separated into 6 metapopulations: Whitetop and Pine Mountains in Virginia, and Grandfather Mountain, Roan Mountain, the Black Mountains, the Great Smoky Mountains, and the Plott Balsams of North Carolina. The total number of mountain peaks the Spruce–fir Moss Spider is known to inhabit is limited to 22, all in the southern Appalachians, resulting in its endemic status. Although past surveys showed possible decreases in abundance (US Fish and Wildlife Service 1998), it appears that all populations outside of Clingman’s Dome (Coyle 2009) may be stable, but the total number of individuals over all populations remains unknown, so final conclusions of population health should be made with caution. When dealing with such small organisms, it is important to consider the scale of the landscape they experience; for example, when considering different species, the characterization of the same habitat can shift from continuous to fragmented as body Southeastern Naturalist 63 T. Seaborn and K. Catley 2016 Vol. 15, No. 1 size shifts from large to small (Borthagaray et al. 2012). Research on soil-dwelling spiders has found that environmental variables result in similar spider assemblages across spatial scales (Ziesche and Roth 2008). In mite species, it has been shown that particular microhabitats, such as the presence of dead wood, significantly increased species diversity on the forest floor (Madej et al. 2011). Although the habitat of the Spruce–fir Moss Spider appears well known, the actual parameters have yet to be documented, which is one of the primary goals of this research. The importance of recording microhabitat parameters and predictive mapping is reflected in the goals of the Recovery Plan for the Spruce–fir Moss Spider (USFWS 1998). The current research contributes to task 1.3, characterization of the species’ habitat requirements, by collecting temperature and humidity data, and to task 2, the search for additional populations and/or habitat suitable for reintroduction, by providing abiotic guidelines for potential locations. If current populations become more imperiled, it will be important to understand as much as possible to decrease chances of mortality of individuals and increase overall success rates of re-establishment. Description of the natural microhabitat also assists in developing artificial holding and propagation techniques, which is task 3 of the Recovery Plan. A clearer understanding of habitat requirements will enhance the effectiveness of captive breeding efforts, which have proven problematic. At Louisville Zoological Park, populations were maintained but not well enough for breeding activity to occur (US Fish and Wildlife Service 1998). Knowledge of habitat requirements will also aid in determining requirements for establishing new populations, which is another important goal. The endangered status of the Spruce–fir Moss Spider is one of the key driving points of this research. More information will aid in bettering the future prospects of the species. The species’ highly specific habitat requirements and loss of that habitat makes conservation of this spider urgently important. The extensive loss of Abies fraseri (Pursh) Poir. (Fraser Fir) in the spider’s habitat is a direct result of infestation by Adelges piceae (Ratzeburg) (Balsam Woolly Adelgid), which is an exotic species that was introduced in 1956. Mature Fraser Fir death within 5–7 years of infestation is possible though not inevitable (White et al. 1993), resulting in loss of canopy, an increase in heat and light, decrease in moisture, and consequently desiccation of the bryophyte mats that are vital for the Spruce–fir Moss Spider (Coyle 1997). It is anticipated that a decline of bryophyte mats would lead directly to a decline in the spider population (US Fish and Wildlife Service 1998), and indeed the entire and largely unknown high-elevation bryophyte-mat community. The loss of Fraser Fir would not only be detrimental to the Spruce–fir Moss Spider, but also to other southern Appalachian endemic arthropod species, such as Dasycerus bicolor Wheeler and McHugh, a staphylinid beetle, and Sisicottus montigenus Crosby and Bishop, a linyphiid spider, which have also shown sharp declines correlated with declines of the fir (Sharkey 2001, Zujiko-Miller 1999). Declines of Fraser Fir due to the Woolly Adelgid seem to have subsided, with some areas showing regeneration (McManamay et al. 2011); however, some have postulated that a second wave of Southeastern Naturalist T. Seaborn and K. Catley 2016 Vol. 15, No. 1 64 infestation is likely to occur (Moore et al. 2008). Climate change may also potentially negatively influence the habitat, as a reduction in cloud cover would decrease water availability (Berry et al. 2014). It is important to determine as much information on the ecology of the Spruce–fir Moss Spider and the status of its current populations as possible to aid in predicting the viability of current and future populations in the face of the these 2 ecosystem stressors. Field-Site Description All metapopulations of the Spruce–fir Moss Spider are defined by some shared characteristics. Populations are restricted to high elevations (1600–2042 m) in spruce–fir forests (Coyle 1981). Spruce–fir forests in this area are dominated by Fraser Fir and Picea rubens Sarg. (Red Spruce) (Spira 2011). In addition, the species is only known from rock outcrops and boulders that serve as substrate for bryophyte mats. These mats are generally 1–4 cm thick and moderately drained— neither dry nor soggy. the Spruce–fir Moss Spider’s sensitivity to desiccation also restricts it to north-facing slopes (Coyle 1981). The bryophyte genera most often associated with the Spruce–fir Moss Spider include Bazzania (liverwort), Dicranodontium (moss), and Polytrichum (moss). Methods We placed temperature and humidity data loggers at 2 sites at the farthest known north and south metapopulations. The north metapopulation encompasses the Mt. Rogers area, Grayson and Smyth counties, VA, and the southern metapopulation encompasses the Plott Balsams area, Haywood and Jackson counties, NC (Fig. 1). The distributions of these as metapopulations are presumed to be defined by limited dispersal, not only between mountain slopes within a mountain range, but also between appropriate microhabitats such as rock structures and bryophyte mats on a single mountain. A positive presence location was chosen within each metapopulation: Whitetop Mountain and Mt. Lyn Lowry, respectively. Known negative presence sites were Mt. Rogers and Browning Knob, respectively. Exact GPS points of positive and negative sites were provided by the USFWS. Surveys were performed in 2009 by Dr. Fred Coyle. We used iButton DS1920 loggers for humidity measurements and iBCod50 G loggers for temperature data at each site. We placed 3 iButton DS1920 and 4 iBCod50 G loggers at each site and averaged their data. Minimum and maximum temperatures were recorded for each day at all sites for the period June 2013 to November 2013, and we calculated the daily average and difference of those minimum and maximum temperatures. We read data from loggers in the field using a USB clamp in September, November, and April. We set the data loggers to record measurements every hour with a resolution of 0.5 for relative humidity (% RH) and temperature (°C) for June through November. From November to April, measurements were recorded every 1.5 hours. Due to the inaccessibility of the sites during winter, we increased measurement increments to prevent data Southeastern Naturalist 65 T. Seaborn and K. Catley 2016 Vol. 15, No. 1 loss due to logger memory limitations. The time period of the study encompassed the hottest months of the year, which may be a critical period given future climate change. We used ANOVA with post-hoc Tukey’s pairwise comparison to analyze the following variables: differences within metapopulations and between metapopulations, presence of spiders, and difference between maximum and minimum temperature (isothermality). We incorporated 2 additional data sets in the final statistical analysis. These were provided by the Mt. LeConte weather station and from a HOBO data logger deployed at Mt. Lyn Lowry by the USFWS. Mt. LeConte is within the Great Smoky Mountains National Park metapopulation. Data were provided through National Oceanic and Atmospheric Administration (NOAA) and accessed through the National Climatic Data Center for the same period of time as the iButton deployment. The Mt. Lyn Lowry HOBO data logger was mounted in a tree near the iButton site by USFWS throughout the duration of the iButton deployment. These 2 additional data sets allowed for investigation of the effect of the bryophyte mats on humidity and temperature. In addition to comparisons with all iButton loggers, modeling analysis was used to determine correlation of the HOBO logger's measured daily Figure 1. Distribution map of the 6 metapopulations of M. montivaga (Spruce–fir Moss Spider). Zones created by buffering 7.5 km from known positive presence locations. Black regions denote metapopulations for which we placed data-loggers, while the grey region, represents the metapopulation for which data was obtained from the Mt. Leconte weather station. Temperature data associated with the other metapopulations (white-filled regions) that lie in the middle of the northeast–southwest georgraphic range of the species, were not collected for this study. Southeastern Naturalist T. Seaborn and K. Catley 2016 Vol. 15, No. 1 66 maximum and minimum temperature values and the iButton's measured temperature values at the Mt. Lyn Lowry site. Results Temperature measurements Temperatures were taken at both Whitetop and Mt. Rogers (274 measurements), while 262 measurements were taken at both Lyn Lowry and Browning Knob. The maximum temperature recorded was 19.8 °C, while the minimum temperature recorded over the study period was -17.8 °C. ANOVA results show maximum daily temperatures were significantly different for the pairwise comparison of Lyn Lowry–Browning Knob (P < 0.01), Whitetop–Lyn Lowry (P < 0.01), Whitetop–Mt. Rogers (P < 0.01), and Mt. Rogers–Lyn Lowry (P = 0.01), but not for Mt. Rogers– Browning Knob (P = 0.77). Significant differences among maximum temperatures within individual sites and within metapopulations were also indicated. Minimum daily temperatures were significant for pairwise comparisons of Lyn Lowry– Browning Knob (P < 0.01), Mt. Rogers–Browning Knob (P < 0.01), Whitetop–Lyn Lowry (P < 0.01), and Whitetop–Mt. Rogers (P < 0.01), but not significantly different for Whitetop–Browning Knob (P = 0.63) nor Mt. Rogers–Lyn Lowry (P = 0.99), indicating some differences of minimum temperatures within individual sites or within metapopulations. Average minimum and maximum temperatures were significant for pairwise comparisons of Lyn Lowry–Browning Knob (P < 0.01), Mt. Rogers–Browning Knob (P = 0.01), Whitetop–Lyn Lowry (P < 0.01), and Whitetop– Mt. Rogers (P < 0.01), but not significantly different for Whitetop–Browning Knob (P = 0.27) or Mt. Rogers–Lyn Lowry (P = 0.76) (Fig. 2A). The average daily maximum, average daily minimum, and daily average of minimum and maximum temperatures varied by less than 25% within metapopulations (Table 1). The highest percent change difference was between the averages of the daily minimum of Lyn Lowry–Browning Knob, at 24.6%. It should be noted, however, that the average of the minimums was 4.4 °C and 3.5 °C, respectively, so the percentage does not reflect a large change in actual degrees. Isothermality, the difference between the maximum and minimum temperatures, was significantly different (P < 0.03) for all pairs except Lyn Lowry–Browning Knob (P = 0.99). Table 1. Percent change comparison of temperature data averages within metapopulations, between metapopulations, between presence status, and between all sites and USFWS logger on LynLowry and the LeConte weather station. * indicates correlation is signific ant at 0.05 level (two-tailed). Avg high Avg. low Avg. isothermality temp. temp. Avg. temp. value Lyn Lowry–Browning Knob -13.6%* -24.6%* 17.8%* 4.1% Whitetop–Mt. Rogers 14.3%* 10.3%* -5.0%* -53.5%* Plott Balsam–Virginia -13.9%* -0.4% -8.6%* -38.2%* Positive–Negative Presence -12.8% -10.5% -6.2% -22.7%* Lyn Lowry–USFWS -42.3%* 35.0%* -26.6%* -137.4%* All–USFW -64.5%* 40.1%* 1.2%* -128.3%* All–Leconte -86.4%* 103.9%* 12.0%* -265.9%* Southeastern Naturalist 67 T. Seaborn and K. Catley 2016 Vol. 15, No. 1 Maximum daily temperature was significantly different between the Virginia (most northern) and Plott Balsams (most southern) sites ( -13.9%, P < 0.01). Conversely minimum daily temperature was not significantly different between the Virginia and Plott Balsams sites (-0.4%, P = 0.78). Average difference of the minimum and maximum, the isothermality value, was also significantly different (-8.6%, P = 0.03; Fig. 2B). Isothermality was significantly different (P < 0.01) between metapopulations, with 38.2% change measured between the Plott Balsams and Virginia sites (Table 1). Figure 2. Temperature data logger summary: (A) Average of daily maximum and minimum temperature for Browning Knob, Lyn Lowry, Whitetop, and Mt. Rogers; (B) Average of daily maximum and minimum temperature for the 2 metapopulations studied, in the Plott Balsams and Virginia; and (C) Average of daily maximum and minimum temperature for positive- versus negative-presence sites. Positive-presence sites were Lyn Lowry and Whitetop; negative sites were Mt. Rogers and Browning Knob. Southeastern Naturalist T. Seaborn and K. Catley 2016 Vol. 15, No. 1 68 Daily maximum temperature values did not differ significantly between locations for which the Spruce–fir Moss Spider has and has not been documented (P = 0.41). Likewise, daily minimum temperature values did not differ significantly between positive- and negative-presence sites (P = 0.88). Average of the minimum and maximum daily temperature was also not different (P = 0.92; Fig. 2C). Isothermality was significantly different (P < 0.01) between positive- and negativepresence sites by -22.7%, equating to a dif ference of 0.5 °C. Daily maximum and minimum temperature values differed significantly between data loggers placed under the bryophyte mats, the HOBO logger in the tree, and the LeConte weather station (P < 0.001). The average daily maximum and minimum did differ significantly between our data loggers across all sites, the HOBO USFWS data logger, and the Mt. LeConte weather station (P < 0.01; Fig. 3A). The difference of the average of the maximum and minimum was 1.2% and 12% different Figure 3. Comparisons with HOBO and weather station data. (A) Average of the daily maximum and minimum temperature values from all iBCod50 G data loggers placed under bryophyte mats at Mt. Rogers, Whitetop, Lyn Lowry, and Browning Knob (“All”) compared to the values for the Leconte weather station and the USFWS HOBO logger mounted in a tree near the Lyn Lowry iBCod50 G loggers. (B) Difference of the daily maximum and minimum temperature values for the same scenarios as (A). Southeastern Naturalist 69 T. Seaborn and K. Catley 2016 Vol. 15, No. 1 between the iButton loggers and the HOBO logger and the LeConte weather station, respectively. Overall, isothermality under bryophyte mats was, on average, 128.3% lower compared to the USFWS logger and 265.9% lower compared to the LeConte weather station (Table 1, Fig. 3B). Direct comparisons of the iBCod50 G data loggers found at Lyn Lowry to the tree-deployed USFWS data logger showed significant differences in daily maximum (-42.3%), minimum (35.0%), average of maximum and minimum temperature (-26.6%), and isothermality calculation (-137.4%) (P < 0.001). A direct percent change between the Lyn Lowry iButton loggers and the USFWS HOBO logger gave a difference in maximum temperature of -14.8%, in minimum temperature of 8.4%, in average of maximum and minimum of -4.2%, and in isothermality value of -136.6%. Linear regression analysis of the temperature data collected from the iButton data loggers and the HOBO logger at Lyn Lowry showed a significant relationship (P < 0.001) between the recorded daily maximum and minimum temperature. Both models reported an adjusted R2 value of 0.94. The equation for the daily maximum was iButton = 0.968*HOBO - 2.1, while the equation for the daily minimum was iButton = 0.82*HOBO + 1.7. Humidity measurements No loggers were recovered from Browning Knob; they all went missing (assumed stolen). Three were recovered and still operational from Lyn Lowry for the period of June to September, taking a total of 1808 measurements. Three were recovered from Mt. Rogers: 1 for June–Sept (1808 measurements) and 2 for June–November (3251 measurements). Three were recovered from Whitetop: 2 for June–September (1808) and 1 for June–November (3559 measurements). The USFWS HOBO logger recorded from June to November (6427 measurements). Primary cause of failure for recovery was battery failure due to oversaturation. The majority of measurements were greater than 100% RH, so percentage of measurements below 100 was used to calculate differences. Whitetop–USFWS were significantly different (P = 0.04); all other pairwise comparisons were non-significant. There were no significant differences between metapopulations (P = 0.13) or presence of the Spruce–fir Moss Spider (P = 0.98) (Fig. 4). Discussion Previous research shows that microclimate is of great importance to a wide range of taxa. For the spider Anelosimus studiosus (Hentz), temperature in the web can drive the success of solitary or multifemale colonies and is also a key factor in the maturation process (Jones and Reichert 2008, Jones et al. 2007). In aquatic Diptera, emergence time and flight period are influenced by temperature in the Plitvice Lakes (Cmrlec et al. 2013). In vertebrate taxa, microhabitat parameters buffer and reduce vulnerability in frogs and determine growth and size in avian offspring (Dawson et al. 2005, Scheffers et al. 2013a). In the Philippines, microclimate habitats have been found to increase in temperature by a range of 0.11–0.66 °C while the macroclimate changes by 1 °C, also providing evidence Southeastern Naturalist T. Seaborn and K. Catley 2016 Vol. 15, No. 1 70 that microclimates can buffer and ameliorate the ambient macro-level temperature (Scheffers et al. 2013b). Although daily maximum, minimum, and average of the maximum and minimum may show small differences, the difference between the minimum and maximum values remains much more stable under the bryophyte mats in the spruce–fir forest when compared to ambient temperatures measured by the HOBO data logger on Mt. Lyn Lowry and the Mt. LeConte weather station. This contrast was most apparent Figure 4. Humidity data logger summary. (A) Percentage of days measured below 100% RH for each site. Tukey’s pairwise comparison showed that Whitetop–USFWS were significantly different (P = 0.04), but all other pairwise comparisons were not significant (P > 0.05). USFWS logger was mounted in tree, whereas all other sites were iButton DS 1920 loggers placed down near bryophyte mats. (B) There were no significant differences between metapopulations (P = 0.13) or (C) presence of M. montivaga (Spruce–fir Moss Spider; P = 0.98). Bars are standard deviation. The USFWS logger data were not included in the metapopulations and presence analysis. Southeastern Naturalist 71 T. Seaborn and K. Catley 2016 Vol. 15, No. 1 when considering data from loggers placed underneath bryophyte mats at Lyn Lowry compared to data from the logger mounted up in a tree within 2 m of the same mats. The difference between the LeConte weather station and the bryophyte mat loggers and tree-mounted logger also brought to light the importance of realizing that macroclimate records may not accurately reflect the microhabitat conditions actually experienced by the Spruce–fir Moss Spider. Sequential differences were observed as distance from the ground was increased. Although the LeConte weather station was not located in the same range, it is expected that the increases in isothermality, and similar average temperatures, would be observed elsewhere with similar monitoring techniques. Isothermality, the difference between the maximum and minimum temperature, was found to be 1 of the 5 most important variables when creating distribution models of Procapra przewalskii Büchner (Przewalski’s Gazelle; Hu and Jiang 2010). In addition, lower isothermality values correlate with rates of spider endemism (Goncalves-Souza et al. 2014). Further research may show, as in that study, that the fluctuation of the temperature over the course of the day may be a key driver in the distribution patterns of the Spruce–fir Moss Spider. Indeed, temperature stabilization provided by bryophyte mats, and especially the reduction of maximum daily temperatures during summer, may in fact be 2 of the defining features of the Spruce–fir Moss Spider ’s ecological requirements. The minor differences in temperature and humidity between and among the metapopulation sites, and between the positive and negative presence sites, are not surprising. All sites were previously considered to be within the defined habitat of the Spruce–fir Moss Spider; thus, general characteristics, such as slope, canopy species, and aspect, were similar across all sites. However, such lack of differences may be important in consideration of the USFWS stated goals, including identifying potential locations of new populations and sites for future populations. It is hoped that by knowing specific temperature and humidity ranges, potential artificial migration events may be more successful. Temperature and humidity are also very important in respect to captive-breeding conservation efforts. As stated previously, Louisville Zoological Park successfully maintained populations, but not well enough for breeding to occur (US Fish and Wildlife Service 1998). Hopefully, the abiotic parameters of wild populations provided by this study will aid in the success of potential breeding efforts. In conjunction with this, studies of captive maintained populations could be informative on the thermal tolerances of the species, allowing for more accurate monitoring in the wild by using thermal tolerances derived from captive individuals as the bounds for potential new locations and assessing the quality of current locations of populations. Under bryophyte mats, the maximum temperature was lower, the minimum temperature was higher, the average of the minimum and maximum values was lower, and isothermality was lower. The percentages given in Table 1 should be used when considering the habitat of the Spruce–fir Moss Spider. Areas of higher maximum temperatures may be more vulnerable to climate change. The percentage conversion factors and linear model could also provide a clearer understanding of the potential effects of climate change, which are of particular concern to the spruce–fir Southeastern Naturalist T. Seaborn and K. Catley 2016 Vol. 15, No. 1 72 forests (Spira 2011). Climate change may be buffered by certain microhabitats because it appears that microclimates may moderate the macroclimate shifts recorded in forests (De Frenne et al. 2013). Although large-scale climate models are convenient and may enlighten general patterns, microclimates must still be considered if accurate prediction of the imperiled status of a species is to be realized. However, a large-scale monitoring goal can be achieved by using the calculated conversion factors and the linear-model protocols to determine more accurate measurements. Large-scale monitoring using the regression analysis for the temperature data can be done without having to actually measure the temperatures underneath the bryophyte mats. The feasibility of large-scale temperature monitoring with data loggers under the bryophyte mats would be a massive undertaking from a financial and labor stand point. This highlights the importance of the model, which allows for more accurate monitoring while using preexisting infrastructure . Although collecting data for the southern and northern metapopulations may be a good starting point, future consideration should be given to all metapopulations of the Spruce–fir Moss Spider to fully solidify the knowledge of this spider’s ecology. These additional measurements across all metapopulations would also help provide knowledge of the basic biology that is lacking for this endangered species. Although new data presented here provide a strong baseline for conservation efforts, there is still much work to be done. Temperature data from multiple positive-presence sites within every metapopulation still needs to be collected. Additional comparisons between iButton and HOBO data loggers may also add support to the provided model as direct site comparisons are considered. A similar methodology could use iButton loggers and the correlation of temperature data with the Mt. LeConte weather station. It should also be noted that a single site was used at each location; therefore, expansion to incorporate multiple rock outcrops at each location would be helpful as well. Long-term monitoring will be key to ensuring the survival of this species. In addition to using the linear model, it may be beneficial to take the percent-change calculations between the Lyn Lowry iButton loggers and the HOBO logger and use it as a conversion factor by multiplying the percent change and non-microclimate measurements, allowing USFWS to use the tree-mounted HOBO loggers; this would provide much more efficient data collection due to increased storage capacity and durability. One fruitful area of research could focus on the differences (both abiotic and biotic) between eastern and western species of Microhexura. Microhexura idahoana is found in a greater range of habitat, including an expanded elevation range and is not restricted to moss substrate for web construction (Coyle 1981). If minimal differences are found, lab studies on the non-endangered species may be helpful to better understand the physiology and behavior of the Spruce–fir Moss Spider. Conversely, if differences are found, it would provide justification for working directly on the endanger ed species. Efforts to conserve nationwide biodiversity will need to be focused on those ecosystems that harbor endemic species. The Spruce–fir Moss Spider may prove to be very important in monitoring the system within which it is found. As a potential key species within the microhabitat, the presence of the Spruce–fir Moss Spider Southeastern Naturalist 73 T. Seaborn and K. Catley 2016 Vol. 15, No. 1 could indicate the health of the bryophyte mats. In turn, these bryophyte mats may well be indicators of the overall health of the high-altitude forest. As forest canopy decreases, bryophyte mats also decline (Baldwin et al. 2011). The loss of an entire ecosystem will not only endanger the known endemic spiders like the Spruce–fir Moss Spider and S. montigenus that reside there, but also any other species that may be adapted for that entire ecosystem. The potential for a complete loss of an entire suite of taxa needs to be seriously contemplated and assessed to encourage intensive conservation and habitat restoration. Acknowledgments T. Seaborn would like to express his deep gratitude to his thesis adviser and co-author Dr. Kefyn Catley for introducing him to this fascinating little mygalomorph and giving him copious advice over the course of his time at Western Carolina University; Dr. Beverly Collins for her knowledge of the tools and ecological concepts of microhabitat research; and the rest of his thesis committee: Dr. Ron Davis and Greg Adkison. Sue Cameron at USFWS also deserves recognition for helping throughout the study. Last but most importantly, we thank Dr. Fred Coyle for doing almost all previous research and surveys and for overall guidance. We would are also grateful to our funding sources, the Western Carolina University Biology Department, Western Carolina University Graduate School, and Highlands Biological Station Grant in Aide of Research. Literature Cited Baldwin, L.K., C.L. Petersen, G.E. Bradfield, W.M. Jones, S.T. Black, and J. Karakatsoulis. 2011. Bryophyte response to forest-canopy treatments within the riparian zone of highelevation small streams. Canadian Journal of Forest Research 42 :141–156. Berry, Z.C., N.M. Hughes, and W.K. Smith. 2014. Cloud immersion: An important water source for spruce and fir saplings in the southern Appalachian Mountains. Oecologia 174:319–326. Borthagaray, A.I., M. Arim, and P.A. Marquet. 2012. Connecting landscape structure and patterns in body-size distributions. Oikos 121(5):697–710. Cmrlec, K., M. Ivkovic, P. Semnicki, and Z. Mihaljevic. 2013. Emergence phenology and microhabitat distributions of aquatic diptera community at the outlets of barrage lakes: Effect of temperature, substrate, and current velocity. Polish Journal of Ecology 61(1):135–144. Coyle, F.A. 1981. The Mygalomorph genus Microhexura (Aranae, Dipluridae). Bulletin American Museum of Natural History 170:64–75. Coyle, F.A. 1985. Observations on the mating behavior of the tiny mygalomorph spider Microhexura montivaga. Bulletin of the British Arachnological Society 6(8):328–330. Coyle, F.A. 1995. A revision of the funnel web mygalomorph spider subfamily Ischnothelinae (Araneae, Dipluridae). Bulletin of the American Museum of Natural History 226:3–133. Coyle, F.A. 1997. Status survey of the endangered spruce–fir moss spider, Microhexura montivaga Crosby and Bishop, on Mount LeConte. Report to the US Department of the Interior, Fish and Wildlife Service, Asheville, NC. 8 pp plus Appendix, Tables 1 and 2, and Figures 1–13. Southeastern Naturalist T. Seaborn and K. Catley 2016 Vol. 15, No. 1 74 Crawford, R.L., and J.S. Edwards. 1986. Ballooning spiders as a component of arthropod fallout on snowfields of Mount Rainier, Washington, USA. Artic and Alpine Research 18(4):429–437. Crosby, C.R., and S.C. Bishop. 1925. Two new spiders from the Blue Ridge Mountains of North Carolina (Araneina). Entomology News 35:142–146. Dawson, R.D., C.C. Lawrie, and E.L. O’Brien. 2005. The importance of microclimate variation in determining size, growth, and survival of avian offspring: Experimental evidence from cavity-nesting passerine. Oecologia 144(3):499–50 7. De Frenne, P., F. Rodriguez-Sanchez, D.A. Coomes, L. Baeten, G. Verstraeten, M. Vellend, M. Bernhardt-Romermann, C.D. Brown, J. Brunet, J. Cornelis, G.M. Decocq, H. Dierschke, O. Eriksson, F.S. Gilliam, R. Hedl, T. Heinken, M. Hermy, P. Hommel, M.A. Jenkins, and D.L. Kelly. 2013. Microclimate moderates plant responses to macroclimate warming. Proceedings of the National Academy of Sciences of the United States of America 110(46):18561–18565. Fridell J. 1994. Endangered and threatened wildlife and plants; Spruce–fir Moss Spider determined to be endangered. Federal Register 60(24):6968–6974. Hu, J., and Z. Jiang. 2010. Predicting the potential distribution of the endangered Przewalski's Gazelle. Journal of Zoology 282(1):54–63. Goncalves-Souza, T., G.Q. Romero, and K. Cottenie. 2014. Metacommunity versus biogeography: A case study of two groups of neotropical vegetation-dwelling arthropods. PLoS ONE 9(12):e115137. Jones, T.C., and S.E. Reichert. 2008. Patterns of reproductive success associated with social structure and microclimate in a spider system. Animal Behaviour 76(6):2011–2019. Jones, T.C., S.E. Reichert, S.E. Dalrymple, and P.G. Parker. 2007. Fostering model explains environmental variation in levels of sociality in a spider system. Animal Behaviour 73(1):195–204. Madej G., G. Barczyk, and I. Gawenda. 2011. Importance of microhabitats for preservation of species diversity, on the basis of Mesostigmatid mites (Mesostigmata, Arachnida, Acari). Polish Journal of Environmental Studies 20(4):961–968. Martens, M. 2005. A preliminary phylogeographic study of the Diplurid genus Microhexura. M.Sc. Thesis. Western Carolina University, Cullowhee, NC. 50 pp. McManamay, R.H., L.M. Resler, J.B. Campbell, and R.A. McManamay. 2011. Assessing the impacts of Balsam Woolly Adelgid (Adelges piceae Ratz.) and anthropogenic disturbance on the stand structure and mortality of Fraser Fir (Abies fraseri (Pursh) Poir.) in the Black Mountains, North Carolina. Castanea 76(1):1–19. Moore, P.T., H. Van Miegroet, and N.S. Nicholas. 2008. Examination of forest recovery scenarios in a southern Appalachian Picea–Abies forest. Forestry 81(2):183–194. Platnick, N.I. 2014. The world spider catalog, version 15. American Museum of Natural History, New York, NY. Available online at http://research.amnh.org/iz/spiders/catalog. Accessed 1 August 2014. Raven, R.J. 1985. The spider infraorder mygalomorphae (Araneae): Cladistics and systematics. Bulletin of American Museum of Natural History 182(1):1–180. Scheffers, B.R., D.P. Edwards, A. Diesmos, S.E. Williams, and T.A. Evans. 2013a. Microhabitats reduce animal’s exposure to climate extremes. Global Change Biology DOI:10.111/gcb.12439. Scheffers, B.R., R.M. Brunner, S.D. Ramirez, L.P. Shoo, A. Diesmos, and S.E. Williams. 2013b. Thermal buffering of microhabitats is a critical factor mediating warming vulnerability of frogs in the Philippine biodiversity hotspot. Biotro pica 45(5):628–635. Southeastern Naturalist 75 T. Seaborn and K. Catley 2016 Vol. 15, No. 1 Sharkey, M.J. 2001. The all taxa biological inventory of the Great Smoky Mountains National Park. The Florida Entomologist 84(4):556–564. Spira, T. 2011. Wildflowers and Plant Communities of the Southern Appalachian Mountains and Piedmont. The University of North Carolina Press, Chapel Hill, NC. US Fish and Wildlife Service (USFWS). 1998. Recovery plan for the Spruce–fir Moss Spider. Prepared by J. Harp and J.A. Fridell. Atlanta, GA. 22 pp. White, P.S., E. Buckner, J.D. Pittillo, and C.V. Cogbill. 1993. High-elevation forests: Spruce–fir forests, northern hardwood forests, and associated communities. Pp. 305– 337, In W.H. Martin, S.G. Boyce, and C. Echternacht (Eds.). Biodiversity of the Southeastern United States. John Wiley and Sons, Inc, New York, NY. Ziesche, T.M., and M. Roth. 2008. Influence of environmental parameters on small-scale distribution of soil-dwelling spiders in forests: What makes the difference, tree species or microhabitat? Forest Ecology Management 255:738–752. Zujko-Miller, J. 1999. On the phylogenetic relationships of Sisicottus hibernus (Araneae, Linyphiidae, Erigoninae). Journal of Arachnology 27(1):44–52.