nena masthead
NENA Home Staff & Editors For Readers For Authors

Measuring Regal Fritillary Butterfly (Speyeria idalia) Habitat Requirements in South-Central Pennsylvania: Implications for the Conservation of an Imperiled Butterfly
Mark T. Swartz, Betty Ferster, Kevina Vulinec, and Gregory Paulson

Northeastern Naturalist, Volume 22, Issue 4 (2015): 812–829

Full-text pdf (Accessible only to subscribers. To subscribe click here.)

 

Access Journal Content

Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.



Current Issue: Vol. 30 (3)
NENA 30(3)

Check out NENA's latest Monograph:

Monograph 22
NENA monograph 22

All Regular Issues

Monographs

Special Issues

 

submit

 

subscribe

 

JSTOR logoClarivate logoWeb of science logoBioOne logo EbscoHOST logoProQuest logo

Northeastern Naturalist 812 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 22001155 NORTHEASTERN NATURALIST 2V2(o4l). :2821,2 N–8o2. 94 Measuring Regal Fritillary Butterfly (Speyeria idalia) Habitat Requirements in South-Central Pennsylvania: Implications for the Conservation of an Imperiled Butterfly Mark T. Swartz1,*, Betty Ferster2, Kevina Vulinec3, and Gregory Paulson4 Abstract - To understand the habitat components that contribute to the presence of populations of a rare butterfly, we examined the abundance of critical plant-components of old fields that support some of the last remaining Eastern Speyeria idalia (Regal Fritillary Butterfly) subpopulations at Fort Indiantown Gap (FTIG), a National Guard training facility in south-central Pennsylvania. We compared densities of larval-host plants (Viola spp. [violets]), adult-nectar plants (Asclepias spp. [native milkweeds] and Cirsium spp. [thistles]), and native, tussock-forming, warm-season bunch grasses that provide protective resting and pupation sites in fields occupied by the butterfly and in nearby fields that were unoccupied. We found no significant difference in violet density among sites. Fields with Regal Fritillary Butterfly populations had significantly more nectar-plant flowering structures and greater bunch-grass percent cover. Grassland habitat occupied by Regal Fritillaries was characterized by a violet density of at least 1.55 plants/m2 and particular varieties of flowering nectar-plants available throughout the June–September flight period. Bunch grasses were also important to persistence of Regal Fritillaries; occupied sites had 20–45% bunch-grass cover and tussock formation. Understanding the habitat needs of this rare butterfly in Pennsylvania is vital to its restoration and reintroductions of the eastern form in the mid-Atlantic and northeastern US. Introduction Butterfly populations have declined worldwide since the 1970s, largely due to habitat destruction (Gilbert and Singer 1975, Wallisdevries at al. 2012). One increasingly rare butterfly, Speyeria idalia (Drury) (Regal Fritillary), is a grassland endemic that until recently (Chazal 2014) persisted at 2 locations in the eastern US (Schweitzer 1984, 1993, 2000; Swengel 1993). The only known eastern population of this butterfly is at Fort Indiantown Gap National Guard Training Center (FTIG) in Annville, PA (Ferster et al. 2008). At one time, this species’ geographic range spanned across North America from New Brunswick to New England, south to northern Georgia, west to Colorado, and north to Manitoba (NatureServe 2014, Swengel 1993). It was listed as a Category II species under the US Endangered Species Act until that category was eliminated in 1996 (USFW 1996). Survival of the 1The Pennsylvania Department of Military and Veterans Affairs, Fort Indiantown Gap National Guard Training Center, Annville, PA 17003. 2Department of Biology, Gettysburg College, 300 North Washington Street, Gettysburg, PA 17325. 3Department of Agriculture and Natural Resources, Delaware State University, 1200 North DuPont Highway, Dover, DE 19901. 4Department of Biology, Shippensburg University, 1871 Old Main Drive, Shippensburg, PA 17257. *Corresponding author - markswartz@pa.gov. Manuscript Editor: John E. Rawlins Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 813 Regal Fritillary depends on 3 main habitat components: host plants for caterpillars, nectar plants for adults (Kelly and Debinski 1998), and native warm-season bunch grasses that provide protective sites for all life stages (Ferster and Vulinec 2010). Populations of this butterfly may have vanished as grassland-habitat components fell below critical density (Kelly and Debinski 1998). Pesticide use (Schildknecht 1986), genetic drift (Britten and Glasford 2002), and weather (Schweitzer 1993) can each also lead to population extinctions. The FTIG population may represent an evolutionarily distinct group that differs from western populations in a number of morphological and genetic characteristics (Keyghobadi et al. 2006). Habitat requirements may differ between western and eastern populations; thus, understanding the species’ habitat needs in Pennsylvania is vital to restoration and reintroductions of the eastern form in the mid-Atlantic and northeastern US. At FTIG, repeated disturbance from military activities maintains open habitats (Ferster and Vulinec 2010, Latham et al. 2007, Warren et al. 2007). By delaying succession of woody shrubs and trees, disturbance promotes the growth of herbaceous plants essential to the survival of the Regal Fritillary (Latham et al. 2007). Not all open habitats at FTIG are occupied by Regal Fritillaries, perhaps because there is between-site variation in critical-plant densities. Conservation efforts, including habitat restoration and reintroduction, will be successful only if habitat requirements are well understood. For this reason, we estimated the optimal density of each critical-plant component— larval host Viola spp. (violets), nectar plants, and bunch grasses—necessary to support a population of this once widespread, now vanishing species, in a series of managed grasslands in south-central Pennsylvania. Violets Violets are the larval-host plants for all speyerid butterflies (Cech and Tudor 2005, Ferris and Brown 1981, Klots 1951). At FTIG, the only documented larvalhost plant for the Regal Fritillary is Viola sagittata Ait. (Arrow-leaved Violet) (Ferster and Vulinec 2010), although other violets serve as host plants throughout the species’ range (Selby 2007, Swengel 1997). Several violet species occur at FTIG, and V. cucullata Aiton (Marsh Blue Violet), V. lanceolata L. (Bog White Violet), V. macloskeyi Lloyd (Small White Violet), V. pedata L. (Birdfoot Violet), Arrowleaved Violet, and V. sororia Willd. (Common Blue Violet) have all been found in occupied fields (TNC 2000a). Many violet species (including Arrow-leaved Violet) are early-successional species common in open areas (Rhodes and Block 2007) that do not compete well with taller plants. Ideal growing conditions for these violet species can be created and maintained by moderate soil disturbances and occasional wildfires that reduce competition (Latham et al. 2007). At FTIG, violet populations increased fourfold following vehicle disturbance (tracked M113 armored personnel carrier) and eightfold after a wildfire, but these changes were temporary (3 years and 1 year, respectively; Latham et al. 2007). Lack of food-plant availability for caterpillars may reduce genetic variability (by reducing population size), and increase local extinctions in butterflies (Frankham and Ralls 1998, Kelly and Debinski 1999, Saccheri et al. 1998). In midwestern fields, increased violet-population density was Northeastern Naturalist 814 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 positively correlated with adult Regal Fritillary female body size and abundance (Kelly and Debinski 1998). Bunch grasses Bunch-grass tussocks form as the leaves of warm-season grasses die and build up, leaving protective spaces within. The tussocks provide resting sites for adult Regal Fritillaries during the summer (M.T. Swartz and B. Ferster, pers. observ.), caterpillars in the winter and spring, and pupae in the late spring–early summer (Ferster and Vulinec 2010). At FTIG, bunch-grass species include Schizachyrium scoparium (Michx.) Nash (Little Bluestem), and to a lesser extent, Andropogon virginicus L. (Broomsedge) (Latham et al. 2007). Both species are early-sere grasses that rely on moderate disturbances for continued growth (Latham et al. 2007). The time interval required for suitable tussock formation has not b een determined. Nectar plants Adult Regal Fritillaries drink nectar from spring- and summer-flowering grassland forbs such as Asclepias syriaca L. (Common Milkweed), A. tuberosa L. (Butterfly Milkweed), A. incarnata L. (Swamp Milkweed), Cirsium pumilum Spreng. (Pasture Thistle), and C. discolor (Muhl. Ex Willd.) Spreng. (Field Thistle) (Ferster and Vulinec 2010, Latham et al. 2007). Assessing nectar-plant abundance is vital to understanding the ecology of Regal Fritillaries because nectar provides adult butterflies with essential sugars for daily activities (Baker and Baker 1983). Limitations on adult food resources are known to have adverse impacts on fecundity and fertility in Regal Fritillaries (Wagner 1995) as well as in other species of fritillaries (Boggs and Ross 1993, Hammond and McCorkle 1991). Elsewhere, declines in nectar plants have been associated with declines in butterfly abundance and diversity (Wallisdevries et al. 2012, Weber et al. 2008). We hypothesized that Regal Fritillaries occupy fields that contain the 3 necessary habitat components in sufficient quantities to support all life stages. We expected to find unoccupied fields lacking (absent or at a density below some critical value) one or more of these habitat components. Distribution and abundance patterns of the species the Regal Fritillary depends on might explain why this butterfly has disappeared from much of its northeastern and mid-Atlantic range. Comparing habitat components of occupied and nearby unoccupied fields could therefore help determine appropriate habitat-restoration efforts for reintroductions. We decided to examine one long-monitored area (C4) to generate explanations for low butterflycounts over years of population surveys there (Ferster and Vulinec 2010). If we found that critical-habitat components were lacking in this area, the observed butterflies would have likely been transients from nearby areas. These transients would in turn not be expected to remain or survive there. The goals of our surveys were to determine critical levels of Regal Fritillary’s 3 habitat components and to prescribe precise habitat conservation and restoration goals, and ultimately, to increase the probability of reintroduction success for this species. Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 815 Field-site Description FTIG is located in south-central Pennsylvania (40°26'13.15''N, 76°34'33.8''W), on the Appalachian Plateau physiographic region and within the northeastern deciduous forest biome (PAARNG 2006). The training areas consist of a mosaic of woodland and field patches. The fields are primarily Little Bluestem–Carex pensylvanica Lam. (Pennsylvania Sedge)-opening grassland communities (PAARNG 2006, Zimmerman et al. 2012) created by recent disturbances (Latham et al. 2007). Ferster et al. (2008) provide a detailed field site description. Some of the fields at FTIG have historically supported and/or currently support Regal Fritillary populations. Five fields have been regularly monitored since 1998 because they support relatively large, persistent subpopulations (Ferster and Vulinec 2010). During this time, 4 of these sites have been protected from most military activities by a memorandum of understanding and later by a management plan outlined in an environmental impact statement (PAARNG 2006 ). These fields have been maintained by a haphazard combination of occasional burns, vehicle disturbance, and manual stewardship (woody-plant removal and mowing; Ferster and Vulinec 2010). Methods Specific research sites for this study were located in the training corridor and cantonment portions of FTIG, within a 7.2-km-radius area (Fig. 1). Five of the 9 selected sites (B12, D1, D3, R23, C4) were inhabited by Regal Fritillaries (occupied) and monitored for their presence since 1998 (Ferster and Vulinec 2010). Four of these sites (B12, D1, D3, R23) had been protected from most military activity since that time. The remaining 4 sites (A1, A22, B7, B11) have not been studied in as much detail but fall well within the grassland definitions of Zimmerman et al. (2012). Sites that we considered unoccupied by the Regal Fritillary had historically harbored populations or now contain grasses, violets, and nectar plants, and are within dispersal distance of occupied fields (Ferster and Vulinec 2010). These unoccupied fields have been maintained through irregular mowing (when mowing is required for training, for haying in the fall, or for aesthetic purposes), occasional fires, and vehicle maneuvering. We chose sites A1, A22, B7, B11 for comparison because they were uninhabited by Regal Fritillaries (unoccupied) at the time of Ferster and Vulinec’s (2010) study, but at the time of our study, fell within the grassland definitions of Zimmerman et al. (2012). Historic reports of Regal Fritillary occupancy exist from 3 of these sites (A1, A22, and B7). Site B11 was forested before 2000, but when the site was then cleared of trees, grassland plants including grasses, violets, and some nectar plants appeared (M.T. Swartz, unpubl. data), indicating that this site may have once been grassland as well. All 4 of these sites were included in annual presence–absence field surveys for Regal Fritillaries established in 2000 (TNC 20 00b). We treated the occupied fields D1 and D3 as a single site for all analyses and fields B12 and C4 as a single site in some analyses. However, researchers Northeastern Naturalist 816 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 historically had treated these areas as 4 distinct sites because they were separated by unpaved roads and/or narrow tree-lines and were designated as different military training areas (Fig. 1). Distinguishing FTIG training fields D1 from D3, and B12 from C4 was a reasonable approach; Regal Fritillaries may not move between fields readily. Central Iowa prairie populations responded strongly to prairie-habitat edges—road, crop, field, and tree line—by turning toward the prairie patch (Ries and Debinski 2001). Individuals also responded positively to conspecifics by being more likely to stay in areas where population density was high (Ries and Debinski 2001). Prior to our study, 4 permanent population-monitoring transects had been set up to monitor each of the 4 FTIG sites over time (Ferster and Vulinec 2010). Later, Keyghobadi et al. (2006) demonstrated that the D1 and D3 populations were not genetically distinct from one another. Thus, although the field-site distinctions represented functional units to researchers and military personnel, they were not biologically meaningful. Ferster and Vulinec (2010), therefore, merged D1 with D3 in population-size analyses. We collected plant abundance and density data separately for D1 and D3, but combined these data for analysis (D). Ferster and Vulinec (2010) did not consider C4 in population-size analyses because of the small number of butterflies seen there; however, they noted the close proximity of C4 to B12 and that 2 of the 5 butterflies caught in C4 during the 2005 mark–recapture study had been marked in B12. Keyghobadi et al. (2006) were not Figure 1. Map of study sites at Fort Indiantown Gap, Annville, PA. Occupied sites are occupied by Regal Fritillary (Speyeria idalia). No Regal Fritillaries were observed in unoccupied sites during the 8 years (1998–2005) of monitoring populations. Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 817 able to analyze C4 butterflies because they caught no Regal Fritillary butterflies there for that study. Ferster and Vulinec (2010) and other researchers have questioned whether C4 represented habitat that supported few Regal Fritillaries or if it merely attracted adults from a nearby population (B12). We treated C4 data in 2 ways: (1) we merged B12 and C4 plant data for analysis because the site designations do not represent separate populations of the Regal Fritillary (see lumped designation in results below), and (2) we treated them as separate sites because of the possibility that some adult butterflies from B12 might simply wander over to C4 often enough to be seen by researchers, but not remain there among the separate population based in C4 (see unlumped designation in results). Violets and bunch grass Vegetation plots in occupied sites were originally sampled for violet and bunchgrass densities during 5 May–25 August 2001 (Latham et al. 2007) and then again during 28 April–19 August 2004. We assigned sampling points throughout sites using a random-point generator in ArcView GIS 3.2 (ArcView 3.2 ESRI GIS and Mapping Software, Redlands CA). We generated 1 point per 0.40 ha for each site. We used a Trimble Total Station 5700 GPS unit (Trimble Navigation Ltd., Sunnyvale, CA) to locate points in the field. We placed a 2-m2 census quadrat at each point and calculated the collective rosette density for all violet species. We considered the violet species collectively because Regal Fritillary larvae reportedly consume multiple species of violets (Selby 2007, Swengel 1997), although we found very few species other than Arrow-leaved Violet at FTIG sites (M.T. Swartz and B. Ferster, pers. observ.). We visually estimated bunch-grass abundance as a proportion of groundcover and estimated cover as 0%, 1%, or the nearest multiple of 5%. In 2004, we re-evaluated occupied sites and established new sample-points for unoccupied sites not sampled in 2001. There was a difference in sampling effort in occupied areas over time. In 2001, 244 of 265 plots were sampled and in 2004 all sites were sampled. To compensate for the difference in number of sites sampled between years, we only used the 244 plots sampled in both years for time-series comparisons (2001 vs 2004). We sampled a total of 107 plots in unoccupied sites in 2004. Nectar-plant identity Nectar plants utilized by Regal Fritillaries in occupied fields were identified at FTIG over 8 years of butterfly population surveys from 1998–2005 (Ferster and Vulinec 2010). Researchers established permanent survey routes—Pollard-walk transects (Pollard and Yates 1993)—in each occupied site and walked them weekly during the adult flight period each year to generate Regal Fritillary population estimates and to identify nectar plants favored by butterflies. Researchers noted the gender and location of each adult Regal Fritillary sighted as they walked routes at a steady pace during a narrow range of environmental conditions. Methods and most results of the Regal Fritillary population surveys are reported elsewhere (Ferster and Vulinec 2010). When we encountered Regal Fritillaries foraging on flowers during our surveys, we recorded the plant species and report the results here (Table 1). Northeastern Naturalist 818 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 Nectar-plant abundance At each site, we counted and determined the number of flowering stems or heads/ha (density) for each of the 5 primary nectar plants—Common Milkweed, Butterfly Milkweed, Swamp Milkweed, Pasture Thistle, and Field Thistle. Surveyors were spaced 2 m apart and they moved through the field in a line at a similar pace while counting nectar plants. We used flagging tape to mark the edges of each pass through a field to avoid double counting. For milkweeds, we counted flowering stems, but we counted flowering heads on each thistle because we thought that this method would more accurately estimate nectar abundance of perennial plant species with very different types of inflorescences (e.g., umbels vs. composite heads) and large variation in the number of flowering heads/plant in a population. In order to catch peak flowering of both the early-flowering Common Milkweed and Butterfly Milkweed and the later-flowering Swamp Milkweed, Pasture Thistle, and Field Thistle, we surveyed each field twice during the field season. We conducted nectar-plant surveys between 21 June and 20 September 2005. We counted plants over a total of 37 days. Analysis We used the Mann-Whitney U-test with the Dunn-Sidak correction for multiple comparisons (Sokal and Rohlf 1995) to detect the difference in violet density (rosettes/ m2), bunch-grass abundance (% cover), and nectar-plant density (# flowering stems and heads/ha) between occupied sites and unoccupied sites. We employed Table 1. Nectar-plant use by the Regal Fritillary (Speyeria idalia) as observed over 8 years (1998– 2005) of data collection along permanent Pollard-walk transects in S. idalia-occupied sites at Fort Indiantown Gap, Annville, PA. We made a total of 1032 S. idalia nectaring observations during which we identified 14 nectar-plant species. Three Cirsium (thistle) species—C. discolor (Muhl.) Spreng., C. pumilum (Nutt.) Spreng., and the exotic C. vulgare (Savi) Ten.—were lumped here because early data collection did not distinguish the species. See text for information about our treatment of Wild Bergamot. # of nectaring Plant species Common name observations % nectaring Achillea millefolium L. Common Yarrow 4 0.39% Apocynum spp. Dogbane 12 1.16% Asclepias incarnata L. Swamp Milkweed 8 0.78% Asclepias syriaca L. Common Milkweed 50 4.84% Asclepias tuberosa L. Butterfly Milkweed 510 49.40% Centaurea nigrescens Willd. Knapweed 45 4.36% Chrysanthemum leucanthemum L. Ox-Eye Daisy 1 less than 0.01% Cirsium spp. Thistles 395 38.28% Cichorium intybus L. Blue Chicory 1 less than 0.01% Dianthus armeria L. Deptford Pink 2 less than 0.01% Monarda fistulosa L. Wild Bergamot n/a n/a Solidago spp. Goldenrod 2 less than 0.01% Vernonia noveboracensis (L.) Michx. New York Ironweed 1 less than 0.01% Pycnanthemum spp. Mountain Mint 1 less than 0.01% Total 1032 100.00% Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 819 the Wilcoxon signed-rank test to examine changes in violet density (rosettes/m2) and bunch-grass abundance (% cover) from 2001 to 2004. We chose non-parametric tests because each treatment (occupied and unoccupied) had ≤5 replicates. Therefore, we reported the median and inter-quartile range (Q1–Q3) rather than the mean and standard deviation. All data analyses were conducted in the statistical software SigmaPlot 11.0 (Systat Software, San Jose, CA). Results Violets We did not detect significant differences in violet density between occupied and unoccupied sites over time (2001 vs. 2004) (lumped: Z = -1.60, df = 2, P = 0.44; unlumped: Z = -1.83, df = 3, P = 0.24; Table 2) or in 2004 (lumped: U = 3.00, df = 1, P = 0.61; unlumped: U = 4.00, df = 1, P = 0.53; Table 3). In 2001, occupied sites had slightly higher violet densities than in 2004 (Table 2), regardless of site lumping. Site C4 showed the greatest decrease in violet density over time, dropping from 5.37 plants m2 in 2001 (Latham et al. 2007) to 1.82 plants m 2 in 2004. In 2004, an unoccupied site (B7) had the highest overall violet density (4.53 violet rosettes/m2; Table 3). Violet densities in the remaining unoccupied sites were below those of occupied sites. Violet density in occupied sites ranged from 1.55 to 2.67 violet rosettes/m2 (median = 1.60 violet rosettes/m2). Bunch grass We did not detect significant differences in bunch-grass abundance between occupied and unoccupied sites in 2004 when lumped (U = 0.00, df = 1, P = 0.10), but the difference was nearly significant when unlumped (U = 0.00, df = 1, P = 0.06) (Table 3). Bunch-grass abundance in C4 was well within the range of what we observed in the other occupied sites and was greater than the abundance recorded Table 2. Violet density (VD; rosettes/m2) and bunch-grass abundance (BG; % cover) in Regal Fritillary (Speyeria idalia)-occupied fields in 2001 (Latham et al. 2007) and 2004. One occupied site (C4) was known to attract individuals from a nearby occupied site (B12) (Ferster and Vulinec 2010) and, therefore, may be used as a single site by S. idalia. To examine this possibility, we considered these data two ways—(1) lumping (LUMP): sites C4 and nearby B12 into one replicate and (2) treating each site as separate replicates (UNLUMP). Ranges shown are all quar tile 1–quartile 3. Violet density (rosettes/m2) Bunch-grass cover (% cover) P-value Site (n = 244) 2001 2004 2001 2004 (VD/BG) C4 (n = 42) 5.37 1.82 16.79 19.32 B12 (n = 35) 3.99 3.69 13.32 20.11 B12/C4LUMP (n = 77) 4.73 2.67 15.21 19.68 R23 (n = 56) 2.19 1.60 34.40 43.25 D (n = 111) 1.92 1.55 22.57 24.14 MedianLUMP 2.19 1.60 22.57 23.07 0.44/0.44 (range) (1.99–4.10) (1.48–2.40) (17.50–31.44) (20.53–38.21) MedianUNLUMP 3.09 1.71 19.68 21.59 0.24/0.24 (range) (2.06–4.68) (1.52–2.76) (15.06–28.49) (19.72–33.16) Northeastern Naturalist 820 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 Table 3 (continued on following page). Violet density (rosettes/m2), bunch-grass abundance (% cover), total nectar-plant density (flowering stems and heads/ha), and individual nectar-species densities (flowering stems or heads/ha) in Regal Fritillary (Speyeria idalia)-occupied (O) and -unoccupied (U) sites during the study period. One occupied site (C4) has been known to attract individuals from a nearby occupied site (B12) (Ferster and Vulinec 2010) and, therefore, may be considered by S. idalia as one site. To examine this possibility, these data were considered two ways: (1) lumping (L) sites C4 and nearby B12 into one replicate and (2) treating each site as separate replicates or “unlumped” (UL). Q = Quartile, n = # of plots. *denotes a significant difference. Monarda fistulosa (Wild Bergamot) was not included in total nectar-plant density analysis. Unoccupied (n = 107; total = 43.32 ha) Occupied (n = 265; total = 90.32 ha) Resource categories A1 A22 B11 B7 C4 B12 C4/B12L D R23 n 46 24 27 10 42 35 77 132 56 Size (ha) 18.62 9.72 10.93 4.05 17.00 14.16 31.16 53.41 22.66 Violet density 0.92 0.18 0.38 4.53 1.82 3.69 2.67 1.55 1.60 Bunch-grass abundance 4.04 0.01 0.06 2.53 19.32 20.11 19.68 24.14 43.25 Total nectar-plant density 37.54 32.62 0.00 72.62 1562.98 1790.55 1666.41 605.53 4174.62 Nectar plant density by individual species Common Milkweed (flowering stems/ha) 1.99 13.48 0.00 0.49 15.56 344.89 165.22 3.06 28.0 Butterfly Milkweed (flowering stems/ha) 0.70 13.89 0.00 13.59 27.43 123.85 71.25 19.95 22.55 Swamp Milkweed (flowering stems/ha) 0.05 3.91 0.00 11.36 1.89 9.81 5.49 0.09 0.07 Pasture Thistle (flowering heads/ha) 1.72 1.34 0.00 5.68 20.50 5.46 13.67 77.89 136.10 Field Thistle (flowering heads/ha) 33.08 0.00 0.00 41.50 1497.60 1306.58 1410.80 504.53 3987.85 Wild Bergamot (flowering stems/ha) 0.00 420.68 0.00 0.00 0.00 6.58 2.98 0.00 40.77 Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 821 Table 3, continued from previous page. MedianU MedianLO MedianULO P-value Resource categories (Q1–Q3) (Q1–Q3) (Q1–Q3) (L/UL) Violet density 0.65 (0.28–2.73) 1.60 (1.56–2.40) 1.71 (1.58–2.76) 0.61/0.53 Bunch-grass abundance 1.30 (0.04–3.29) 24.18 (20.81–38.48) 22.15 (19.72–33.72) 0.10/0.06 Total nectar-plant density 35.58 (16.81–55.08) 1666.41 (870.75–3547.57) 1676.79 (1084.26–2982.61) 0.06/0.03* Nectar plant density by individual species Common Milkweed (flowering stems/ha) 1.24 (0.25–7.74) 28.05 (9.31–130.93) 21.81 (9.31–186.47) 0.45/0.27 Butterfly Milkweed (flowering stems/ha) 7.15 (0.35–13.74) 22.55 (20.06–59.08) 24.99 (21.25–75.64) 0.23/0.14 Swamp Milkweed (flowering stems/ha) 1.98 (0.03–7.64) 0.09 (0.08–4.14) 0.99 (0.08–5.85) 1.00/1.00 Pasture Thistle (flowering heads/ha) 1.53 (0.67–3.70) 77.89 (29.73–121.55) 49.20 (12.98–107.00) 0.23/0.27 Field Thistle (flowering heads/ha) 16.54 (0.00–37.29) 1410.80 (731.10–3343.59) 1402.09 (905.46–2742.73) 0.23/0.14 Wild Bergamot (flowering stems/ha) 0.00 (0.00–105.17) 3.29 (0.00–15.13) 2.98 (1.49–21.88) Not tested Northeastern Naturalist 822 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 at any unoccupied site (Table 3). We found no significant change in bunch-grass density over time regardless of site lumping (Table 2). Nectar plants We observed 1032 adult Regal Fritillaries nectaring on 14 different floweringplant species over the 8 years of surveys (Table 1). Thistle species were not distinguished during early data collection, but later observations included 3 species— the native species Field Thistle and Pasture Thistle and less frequently, the exotic C. arvense (L.) Scop. (Canada Thistle)—which we lumped for our analyses. We rarely observed Regal Fritillaries feeding on most nectar-plant species (less than 0.01%); most documented nectaring was on the native thistle and milkweed species for which we later determined density. Median nectar-plant density was not significantly different between occupied and unoccupied sites when we lumped occupied sites (U = 0.00, df = 1, P = 0.06) but was significantly different when we analyzed occupied sites separately (unlumped) (U = 0.00, df = 1, P = 0.03) (Table 3). Site R23 had an overall nectar-plant density greater than twice that of the next-highest site (Table 3). Occupied sites had more than 45 times the number of total stems and heads/ha than unoccupied sites (Table 3). Site C4 had 1562.98 stems and heads/ha and the second-highest density of Field Thistle flowering heads (Table 3). All but 1 nectar-plant species were more abundant in occupied than in unoccupied sites (Table 3). The density of Field Thistle heads in unoccupied sites was only ~1% of what we observed in occupied sites, and the density of Pasture Thistle heads in unoccupied sites was only 2–3% of the density found in occupied sites. Butterfly Milkweed flowering stems in unoccupied sites comprised ~30% of the abundance of occupied sites, and Common Milkweed flowering-stem abundance in unoccupied sites was only 5–6% of that found in occupied sites. Only Swamp Milkweed was more abundant in unoccupied sites; however, these data were most likely skewed by the presence of this species in large numbers at one unoccupied site, and was nearly absent altogether from 2 occupied sites (Table 3). Researchers planted Monarda fistulosa L. (Wild Bergamot) in occupied sites (M.T. Swartz unpubl. data, TNC 2005) and it occurs naturally at one unoccupied site. Thus, we counted this species at each site, but did not include it in the statistical analysis (Table 3). Three unoccupied sites (A1, A22, and B7) contained nectar plants that never flowered; therefore we did not count them because butterflies did not use them. Site C4 (occupied) was more similar in flowering stem and head abundance to occupied areas than unoccupied areas for all nectar plant species. Discussion Occupied sites had nearly statistically significantly more flower structures (i.e., more nectar) and bunch grasses than unoccupied sites. Violet density was also higher in occupied sites than in 3 of the 4 unoccupied sites, though we found no statistically significant difference. Violets were most abundant in unoccupied site B7, suggesting that high violet abundance alone does not support a Regal Fritillary population. Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 823 The enigmatic site, C4, harbored all 3 critical Regal Fritillary habitat components— larval-host plants, adult-nectar plants, and protective bunch grasses—at abundance levels similar to occupied sites. C4 was in close proximity to occupied sites B12 and R23 (Fig. 1) and caterpillars have been found on violets there, yet this site harbored only a few individual adults each year of the long-term monitoring study that began in 1998 (Ferster and Vulinec 2010). Our data did not allow us to explain the low number of adult occurrences here. This site would be optimal as an experimental plot for future research to test population establishment in areas where habitat components have been restored and that are within dispersal distance from an existing population. Future studies may allow us to examine other environmental factors that impact survivorship of late juvenile stages, thereby decreasing adult population density. The effects of disease and microclimate must also be examined. Swamp Milkweed may not be essential for Regal Fritillary survival at FTIG where other nectar sources overlap in flowering phenology. The low abundance of Swamp Milkweed in D and R23 (Table 2) suggests that although Regal Fritillaries use this nectar plant (Table 1), a population can persist as long as other nectar sources are available. Swamp Milkweed was more abundant in old-fields at Gettysburg National Park, which historically supported Regal Fritillaries (Schildknecht 1986) and where reintroduction efforts failed in 2005 (Ferster 2005). Common Milkweed blooms in early June and is used by this butterfly at FTIG and elsewhere (Opler and Krizec 1984, Swengel 1993). Its flowering corresponds to the first male Regal Fritillary emergences each year at FTIG (Ferster and Vulinec 2010). Therefore, although in low abundance in some fields at FTIG, Common Milkweed is an important resource for the butterfly. We recommend that habitat restoration and reintroduction efforts include Common Milkweed as a nectar plant. Butterfly Milkweed, is a low-growing, perennial plant that flowers during July– August (Rhoads and Block 2007) and is used frequently by nectaring Regal Fritillaries (Table 1). It was abundant at 2 of the unoccupied sites, but apparently cannot support Regal Fritillary populations alone. Butterfly Milkweed is an important habitat component that provides nectar resources during mid-summer when other nectar plants are not blooming. Farmers, ranchers, and other land stewards often target thistles for removal. To that end, multiple exotic thistle species appear on noxious plant lists for many states (PDA 2003). However, 2 native species are important butterfly nectar-plants (Table 1). Pasture Thistle has a short flowering period at FTIG during the summer season (B. Ferster, pers. observ.), but Regal Fritillaries are frequently observed nectaring on these flowers when they are in bloom (Table 1). All but 1 occupied site (B12) had higher Pasture Thistle flower-head counts than the unoccupied sites. The occupied sites had the highest abundance of the 2 longest-flowering nectar plants, Common Milkweed and Butterfly Milkweed. Perhaps these species can act to bridge the gap in flowering phenology filled by Pasture Thistle in other fields. Stable Regal Fritillary populations (Ferster and Vulinec 2010) were associated with violet densities of at least 1.55 violets/m2, bunch grasses between 20% and 45% cover, and maintenance of milkweeds and thistles at species-specific densities Northeastern Naturalist 824 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 (Table 3). Common Milkweed is essential to butterfly survival but does not need to be present at high abundance. Butterfly Milkweed and Field Thistle comprise much of the nectar-plant community for adult butterflies. Wild Bergamot has also been reported to be an important Regal Fritillary nectar plant (Huebschman 1998). However, this nectar-plant species was not found in occupied fields at FTIG until land stewards planted it in some occupied fields in 2003, 2004, and 2005 in an effort to increase nectar-plant abundance (TNC 2005). Butterflies have been seen nectaring on this species; however, it represents a small proportion of nectar plants in these fields, and populations of Regal Fritillaries persisted at FTIG without Wild Bergamot. Although it is apparently not an obligate nectar plant for Regal Fritillaries, we recommend that Wild Bergamot be included in restorations because so few nectar-plant species are used by this rare butterfly; however, we cannot suggest an appropriate planting density. Keyghobadi et al. (2013) suggest that the remaining Pennsylvania Regal Fritillary population has a haplotype that is genetically distinct from the western haplotype that includes the isolated (and probably extinct) Virginia population (Chazal 2014). Their study, which included museum specimens collected from now-extinct populations and fresh specimens from surviving populations, supports earlier findings by Williams (2001a, b) who suggested subspecies designation for the 2 groups (S. i. idalia for the eastern subspecies, and S. i. occidentalis for the western subspecies), with a transition zone centered on western PA, OH, WV, and VA. Keyghobadi et al. (2013) found little evidence to support a prolonged period of isolation between the eastern and western groups. They suggested that genetic and morphological differences resulted from strong selection in different parts of the range where butterflies occupy different habitats. They cite Opler and Krizek (1984) and Swengel (1997) to support the hypothesis that Eastern Regal Fritillary populations preferred wet meadows and swampy areas. Swengel’s (1997) study examined the western form and habitat and host-plant partitioning among violet-feeding fritillaries (Euptoieta, Speyeria, Boloria). Although the study mentions populations “outside the prairie”, it does not indicate that this refers to eastern populations specifically. At FTIG, Regal Fritillary habitat consists mostly of dry fields where the necessary vegetative habitat components can be found. Arrow-leaved Violet, the violet used as the larval host plant at FTIG is a species of “dry woods, fields, and edges” throughout Pennsylvania according to Rhoads and Block (2007), who also report that Butterfly Milkweed is found most often in “dry fields, roadsides, and shale barrens.” Among the important nectar plants we have seen Regal Fritillaries feeding on at FTIG, only Swamp Milkweed is a plant of “swamps, floodplains, and wet meadows” (Rhoads and Block 2007). It is likely that adults will nectar at wet meadows, but it is unlikely they oviposit there, or that caterpillars could survive in this habitat. Reintroduction efforts in eastern locations should focus on dry fields for habitat restoration, and wet meadows should be restored to increase nectar-plant availability for populations established in nearby dry field sites. Indeed, habitat partitioning (Keyghobadi et al. 2013) should be reconsidered as an important selective pressure contributing to subspecific integrity. Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 825 Intact ecosystems provide goods and services (such as food, pollination, nutrient cycling, carbon sequestration, etc.) that diminish as these ecosystems are modified by human activity. Low levels of invertebrate diversity can be an indicator of ecosystem degradation because these organisms occupy a wide variety of functional niches (Kremen et al. 1993). The disappearance of the Regal Fritillary across much of its historic range likely indicates that land-use practices since the 1970s, or perhaps earlier, have degraded grassland ecosystems (Ferster and Vulinec 2010). The butterfly is even absent from grasslands that are protected and managed for other wildlife, perhaps an indication that management practices are insufficient for effectively maintaining biodiversity in these ecosystems. Many butterfly species have evolved in close association with their host plants and are mono- or oligophagous and thus highly dependent on particular plants (Ehrlich and Raven 1964, Fordyce 2010). Indeed, declines in larval host-plants and/or adult nectar-plants have been associated with many declining butterfly populations (Pleasants and Oberhauser 2013, Schultz and Dlugosch 1999, Severns and Warren 2008, Wallisdevries et al. 2012), and unless butterflies can switch host plants and/or nectar plants to more common species, they may be destined for extinction (Severns and Warren 2008). Conservation efforts aimed at butterflies must focus on maintaining the particular plant communities that butterflies rely on, or they will fail. Successful efforts to manage habitats for butterflies may contrast with methods used in conservation for species that are less tied to particular plants (i.e., mowing at 2–3 year intervals for grassland birds; Johst et al. 2006, Rothbart and Capel 2006). Mowing in unoccupied fields prevents tussocks from forming, decreases bunch-grass density and percent cover when it is being grown from seed, and promotes growth of weedy, cool-season grasses (Fedewa and Stewart 2011). Mowing regimes used in management for hayfields (i.e., twice per year) do not support most grassland-specialist butterflies (Johst et al. 2006; Swengel 1996, 1997). Mowing may also encourage cool-season grasses and other invasive vegetation by reducing competition from bunch grasses (Fedewa and Stewart 2011). Few butterflies may survive certain management and land-use strategies, reducing butterfly population persistence. In this study, we determined the densities of important plant species necessary for the persistence of Regal Fritillary populations. This information combined with the understanding of how these plants respond to various types of disturbance (Latham et al. 2007) can be used to formulate effective management strategies for both stewardship of existing populations and restoration of grasslands for successful Regal Fritillary reintroduction efforts in the mid-Atlantic and Northeast. The habitat that currently supports the rare Regal Fritillary Butterfly was accidently and haphazardly established and maintained, and only recently managed for butterflies and other biological elements (PAARNG 2006). Determining how best to mimic these conditions for conservation and reintroduction are questions we continue to explore and understand. Soil disturbance such as that caused by track vehicles (Latham et al. 2007), or perhaps bison (Ferster and Vulinec 2010), is more effective at increasing violet and bunch-grass densities than mowing. We know that 5 years after such a soil Northeastern Naturalist 826 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 disturbance, grasslands can support a Regal Fritillary Butterfly population (Ferster and Vulinec 2010), but we do not know how long this “good habitat” lasts. Given too much time without further management, violet populations, tussock-forming grasses, and nectar plants are shaded out as taller perennials and then woody plants become established and reduce essential habitat-components below some critical density. Mowing may extend the health of established grasslands that would otherwise succeed into forests by reducing taller perennials and woody plants, but the frequency of mowing and how long this treatment will successfully maintain habitat must both be established. Acknowledgments Fred Habegger, Denise Johnson Watts, Rattiford P.E. Jones, Andrew Mehring, Dr. Walter E. Meshaka Jr., Dave Zapotok, and Lindsay Zemba all enthusiastically assisted with fieldwork and data entry. Dr. Larry Klotz, Dr. Walter E. Meshaka Jr., Burnetta Swartz, Virginia Tilden, Dr. Richard Stewart, Ann B. Swengel, and anonymous reviewers provided invaluable comments to earlier drafts of this manuscript. The Pennsylvania Army National Guard (especially range-control personnel) granted safe access to fiel d sites and vehicles for field use. This project was sponsored by the PAARNG (Cooperative Agreement # DAHA36-01- 2-9001). Funding for this project was also provided by the Pennsylvania Department of Military and Veterans Affairs (DMVA). The information presented here does not necessarily reflect the position or the policy of the US government, and no official endorsement should be inferred. Literature Cited Baker, H.G., and I. Baker. 1983. Floral-nectar sugar constituents in relation to pollinator type. Pp. 117–141, In C.E. Jones and R.J. Little (Eds.). A Handbook of Experimental Pollination Biology. Van Nostrand-Reinhold, New York, NY. 558 pp. Boggs, C.L., and C.L. Ross. 1993. The effect of adult-food limitation on life-history traits in Speyeria mormonia (Lepidoptera: Nymphalidae). Ecology 74:433–441. Britten, H.B., and J.W. Glasford. 2002. Genetic population structure of the Dakota Skipper (Lepidoptera: Hesperia dacotae): A North American native-prairie obligate. Conservation Genetics 3:363–374. Cech, R., and G. Tudor, 2005. Butterflies of the East Coast: An Observer’s Guide. Princeton University Press, Princeton, NJ. 345 pp. Chazal, A.C. 2014. Surveys for rare Lepidoptera and Henslow’s Sparrow on the Radford Army Ammunition Plant. Natural Heritage Technical Report 14-10. Virginia Department of Conservation and Recreation, Division of Natural Heritage, Richmond VA. 10 pp. + plus appendix. Ehrlich, P.R., and P.H. Raven. 1964. Butterflies and plants: A study in coevolution. Evolution 18:586–608. Fedewa, C.A., and J.R. Stewart. 2011. Field establishment of Little Bluestem and Prairie Dropseed under managed conditions. Native Plants Journal 12:1 11–117. Ferris, C.D., and F.M. Brown. 1981. Butterflies of the Rocky Mountain States. University of Oklahoma Press, Norman, OK. 442 pp. Ferster, B. 2005. Regal Fritillaries: News from the eastern front. American Butterflies 13:4–12. Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 827 Ferster, B., and K. Vulinec. 2010. Population size and conservation of the last remnants of the Eastern Regal Fritillary, Speyeria idalia idalia (Drury) [Lepidoptera: Nymphalidae]: Implications for temperate-grassland restoration. Journal of Insect Conservation 14:31–42. Ferster, B., B.R. Leppo, M.T. Swartz, K. Vulinec, F. Habegger, and A. Mehring. 2008. Lepidoptera of Fort Indiantown Gap National Guard Training Center, Annville, Pennsylvania. Northeastern Naturalist 15:141–148. Fordyce, J.A. 2010. Host shifts and evolutionary radiations of butterflies. Proceedings of the Royal Society of Biological Sciences Series B 277:3735–3743 . Frankham, R., and K. Ralls. 1998. Inbreeding leads to extinctio n. Nature 392:441–442. Gilbert, L.E., and M.C. Singer. 1975. Butterfly ecology. Annual Review of Ecology and Systematics 6:365–397. Hammond, P.C., and D.V. McCorkle. 1991. 1991 introduction of the Oregon Silverspot butterfly (Speyeria zerene hippolyta) on Fairview Mountain. US Forest Service, Siuslaw National Forest, Corvalis, OR. 14 pp. Huebschman, J. 1998. The relationship between nectar sources and Regal Fritillary (Speyeria idalia Drury) butterfly populations. M.Sc. Thesis. University of Nebraska-Lincoln, Lincoln, NE. 62 pp. Johst, K., M. Drechsler, J. Thomas, and J. Settle. 2006. Influence of mowing on the persistence of 2 endangered large blue butterfly species. Journal of Applied Ecology 43:333–342. Kelly, L., and D.M. Debinski. 1998. Relationship of host-plant density to size and abundance of the Regal Fritillary Speyeria idalia Drury (Nymphalidae). Journal of the Lepidopterists’ Society 52:262–276. Kelly L., and D.M. Debinski. 1999. Effects of larval food-limitation on Vanessa cardui L. (Lepidoptera: Nymphalidae). American Midland Naturalist 141:315–322. Keyghobadi, N., K.P. Unger, J.D. Weintraub, and D.M. Fonseca. 2006. Remnant populations of the Regal Fritillary (Speyeria idalia) in Pennsylvania: Local genetic structure in a high-gene–flow species. Conservation Genetics 7:309–313. Keyghobadi, N., D. Koscinski, J.D. Weintraub, and D.M. Fonseca. 2013. Historical specimens reveal past relationships and current conservation status of populations in a declining species: The Regal Fritillary butterfly. Insect Conservation and Diversity 6:234–242. Klots, A.B. 1951. A Field Guide to the Butterflies of North America, East of the Great Plains. Houghton Mifflin Co., Boston, MA. 349 pp. Kremen, C., R.K. Colwell, T.L. Erwin, D.D. Murphey, R.F. Noss, and M.A. Sanjayan. 1993. Terrestrial arthropod assemblages: Their use in conservation planning. Conservation Biology 7:796–808. Latham, R.L., D. Zercher, P. McElhenny, P. Mooreside, and B. Ferster. 2007. Habitat restoration and management for the Eastern Regal Fritillary, Speyeria idalia idalia Drury (Nymphalidae) in Pennsylvania. Ecological Restoration 25:103–111. NatureServe. 2014. NatureServe Explorer: An online encyclopedia of life [web application]. Version 7.1. Arlington, VA. Available online at http://explorer.natureserve.org. Accessed 13 June 2014. Opler, P.A., and G.O. Krizek. 1984. Butterflies East of the Great Plains. Johns Hopkins University Press, Baltimore, MD. 294 pp. Pennsylvania Army National Guard (PAARNG). 2006. Final environmental impact statement for the transformation of the Pennsylvania Army National Guard 56th Brigade into a Stryker Brigade Combat Team at the National Guard Training Center, Fort Indiantown Gap, PA (3 vols.) (SBCT EIS). 1871 pp. Northeastern Naturalist 828 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 Vol. 22, No. 4 Pennsylvania Department of Agriculture (PDA). 2003. Pennsylvania noxious-weed control list. Pennsylvania Department of Agriculture. Available online at http://plants.usda.gov/ java/noxious?rptType=State&statefips=42. Accessed 4 June 2014. Pleasants, J.M., and K.S. Oberhauser. 2013. Milkweed loss in agricultural fields because of herbicide use: Effect on the Monarch Butterfly population. Insect Conservation and Diversity 6:135–144. Pollard, E., and T.J. Yates. 1993. Monitoring Butterflies for Ecology and Conservation. Chapman and Hall, London, UK. 292 pp. Rhodes, A.F., and T.A. Block. 2007. The Plants of Pennsylvania: An Illustrated Manual, 2nd Edition. University of Pennsylvania Press, Philadelphia, PA. 1042 pp. Ries, L., and D.M. Debinski. 2001. Butterfly responses to habitat edges in the highly fragmented prairies of Central Iowa. Journal of Animal Ecology 70:840–852. Rothbart, P., and S. Capel. 2006. Maintaining and restoring grasslands. Pp 14–27, In J.D. Oehler, D.R. Covell, S. Capel, B. Long (Eds.). Managing Grasslands, Shrublands and Young Forests for Wildlife. The Northeast Upland Habitat Technical Committee, Massachusetts Division of Fisheries and Wildlife. Westborough, MA. 154 pp. Saccheri, I., M. Kuussaari, M. Kankare, P. Vikman, W. Fortelius, and I. Hanski. 1998. Inbreeding and extinction in a butterfly metapopulation. Nature 39 2:491–494. Schildknecht, C.E. 1986. Eastern Regal Fritillary (S. idalia). News of the Lepidopterists’ Society. 28:69. Available online at http://images.peabody.yale.edu/lepsoc/ nls/1980s/1986/1986_v28_n5.pdf. Accessed 3 November 2015. Schweitzer, D.E. 1984. The Regal Fritillary, Speyeria idalia on the Dickinson-Lewis Preserve, Block Island. Unpublished report prepared for The Nature Conservancy, Block Island, RI. Schweitzer, D.E. 1993. Regal Fritillaries in the east. American Butterflies 1:9. Schweitzer, D.E. 2000. Element stewardship abstract for Speyeria idalia. Unpublished report prepared for The Nature Conservancy, Arlington, VA. Schultz, C.B., and K.M. Dlugosch. 1999. Nectar and hostplant scarcity limit populations of endangered Oregon butterfly. Oecologia 119:231–238. Selby, G. 2007. Regal Fritillary (Speyeria idalia Drury): A technical conservation assessment. USDA Forest Service, Rocky Mountain Region. Available online at http://www. fs.fed.us/r2/projects/scp/assessments/regalfritillary.pdf. Accessed 10 September 2009. Severns, P.M., and A.D. Warren. 2008. Selectively eliminating and conserving exotic plants to save an endangered butterfly from local extinction. Animal Conservation 1–8. Sokal, R.R., and F.J. Rohlf. 1995. Biometry, 3rd Edition. W.H. Freeman & Co., New York, NY. 937 pp. Swengel, A.B. 1993. Regal Fritillary: Prairie royalty. American Butterflies 1:4–9. Swengel, A.B. 1996. Effects of fire and hay management on abundance of prairie butterflies. Biological Conservation 76:73–85. Swengel, A.B. 1997. Habitat associations of sympatric violet-feeding fritillaries (Euptoieta, Speyeria, Boloria) (Lepidoptera: Nymphalidae) in tallgrass prairie. The Great Lakes Entomologist 30:1–17. The Nature Conservancy (TNC). 2000a. The 1999 flora and fauna Inventory for Fort Indiantown Gap National Guard Training Center, Annville, PA. Prepared for The PA Department of Military and Veterans Affairs, Fort Indiantown Gap, Environmental Section, Building 11-19, Annville, PA 17003-5002. 261 pp. TNC. 2000b. Population monitoring and life-history studies for the Regal Fritillary (Speyeria idalia) at the Ft. Indiantown Gap National Guard Training Center, PA. Final report prepared for The Department of Military and Veterans Affairs and the PA National Guard, Annville, PA. 25 pp. Northeastern Naturalist Vol. 22, No. 4 M.T. Swartz, B. Ferster, K. Vulinec, and G. Paulson 2015 829 TNC. 2005. Research summary: Population monitoring, life–history studies, and stewardship of the Eastern Regal Fritillary Speyeria idalia idalia (F: Nymphalidae) on the National Guard Training Center-Fort Indiantown Gap, Annville, PA. Unpublished final report prepared for The Department of Military and Veterans Affairs and the PA National Guard, Annville, PA. 46 pp. US Fish and Wildlife (USFWS). 1996. Final decision on identification of candidates for listing as endangered or threatened. Available online at https://www.fws.gov/endangered/ laws-policies/policy-final-decision.html. Accessed 25 May 2014. Wagner, D.L. 1995. Rearing regals for reintroduction: Playing the odds but still losing ground. American Butterflies 3:19–23. Wallisdevries, M.F., C.A.M. Van Swaay, and C.L. Plate. 2012. Changes in nectar supply: A possible cause of widespread butterfly decline. Current Zoology 58:384–391. Warren, S.D., S.W. Holbrook, D.A. Dale, N.L. Whelan, M. Elyn, W. Grimm, and A. Jentsch. 2007. Biodiversity and the heterogeneous disturbance-regime on military-training lands. Restoration Ecology 15:606–612. Weber, P.G., S. Preston, M.J. Dlugos, and A.P. Nelson. 2008. The effects of field mowing on adult-butterfly assemblages in Central New York State. Natural Areas Journal 28:130–143. Williams, B.E. 2001a. Patterns of morphological variation in Speyeria idalia (Lepidoptera: Nymphalidae) with implications for taxonomy and conservation. Annals of the Entomological Society of America 94:239–243. Williams, B.E. 2001b. Recognition of western populations in Speyeria idalia (Nymphalidae) as a new subspecies. Journal of the Lepidopterists’ Society 55:144–149. Zimmerman, E., T. Davis, G. Podniesinski, M. Furedi, J. McPherson, S. Seymour, B. Eichelberger, N. Dewar, J. Wagner, and J. Fike. 2012. Terrestrial and Palustrine Plant Communities of Pennsylvania, 2nd Edition. PA Natural Heritage Program, PA Department of Conservation and Natural Resources, Harrisburg, PA. Available online at http:// www.naturalheritage.state.pa.us/Communities.aspx. Accessed 31 December 2014.