2008 NORTHEASTERN NATURALIST 15(4):523–540 Spiders (Arachnida: Araneae) Associated with Seed Heads of Sarracenia purpurea (Sarraceniaceae) at Acadia National Park, Maine Daniel T. Jennings1,2,*, Bruce Cutler3, and Bruce Connery4 Abstract - We discovered that spiders use seed heads of Sarracenia purpurea (Northern Pitcher Plant) for moulting, nesting, and rearing of young. These associations represent only a few of the diverse interactions between spiders and pitcher plants. During July–August 2001 at Acadia National Park, seed heads (n = 301) of S. purpurea from four bog-heaths yielded spiders (n = 685) of four families (Theridiidae, Dictynidae, Clubionidae, Salticidae), 10 genera, and at least 11 species. Two additional spider families (Gnaphosidae, Thomisidae) were represented by cast exuviae. Jumping spiders (Salticidae) were the chief occupants, comprising 80.0% of species and 99.1% of individuals. The salticid Tutelina similis was the most common inhabitant, accounting for 63.8% of the overall spider fauna in these microhabitats. Spider foraging-guild presence favored hunters (99.7%) over web spinners; juveniles outnumbered adults almost 15 to 1, and females outnumbered males 43 to 1. Frequencies of spider webbing and retreats in seed heads were greater than expected (G-test, α = 0.05); however, spider occupancy was less than expected. Seed heads with multiple-spider occupants were more frequent than those with single-spider occupants; conspecific associations were more frequent than heterospecific associations. No evidence was found that spiders preferred either closed or open seed heads. Other associated arthropods included parasitic mites, spider-egg parasitoids, and insects. The identified taxa represent the first records of spiders inhabiting seed heads of S. purpurea in Maine. Introduction The carnivorous habits of pitcher plants in North America and elsewhere are well known (Folkerts 1999, Juniper et al. 1989). For example, the prey captured by leaves of Sarracenia purpurea L. (Northern Pitcher Plant) (Sarraceniaceae), consists chiefly of insects (e.g., Diptera, Coleoptera, Hymenoptera, Lepidoptera), but also includes spiders (Araneae) and mites (Acari) (Cresswell 1991, 1993; Judd 1959; Swales 1969). Not only do spiders serve as food for these plants, some species actively compete with pitcher plants for insectivorous prey (Cresswell 1991, 1993; Ellison 2005; Folkerts 1999); others apparently assist in prey capture by discarding distasteful food into the pitcher’s fluid (Bristowe 1939, Pocock 1898). 1USDA, Forest Service, Northern Research Station, 686 Government Road, Bradley, ME 04411. 2Current address - PO Box 130, Garland, ME 04939-0130. 3Department of Ecology and Evolutionary Biology, University of Kansas, 1200 Sunnyside Avenue, Lawrence, KS 66045-7534. 4USDI, National Park Service, Acadia National Park, PO Box 177, Bar Harbor, ME 04609-0177. *Corresponding author - Daniel. Jennings@umit.maine.edu. 524 Northeastern Naturalist Vol. 15, No. 4 Peucetia viridans (Hentz) (Oxyopidae) (Green Lynx Spider) often sits near the pitcher’s hood, where it ambushes potential prey attracted to the pitcher’s extrafloral nectaries; it also uses the pitchers as nesting sites for depositing egg sacs and guarding newly emerged spiderlings (Folkerts 1999). Less known interactions between spiders and pitcher plants include spiders as potential pollinators of these plants. This phenomenon was first hypothesized by Rebecca Austin in the late 1800s for Darlingtonia californica Torr. (California Pitcherplant). The hypothesis was weakly supported by Nyoka and Ferguson (1999), after finding spiders dusted with D. californica pollen. Pitcher plant pollen also may serve as food for spiders; nectarivory by spiders is receiving increased attention from investigators (Smith and Mommsen 1984, Taylor and Foster 1996). Spiders also prey on pitcher plant associates, including pollinators, herbivores, and frugivores (Ellison 2005, Folkerts 1999). In Indonesia, Misumenops nepenthicola Pocock (Thomisidae) (Crab Spider) lives in the pitchers of Nepenthes gracilis Korth (Slender Pitcher Plant (Nepenthaceae), where it steals captured prey (kleptoparasitism) from the pitcher’s liquid (Juniper et al. 1989, Pollard 2006), or discards distasteful prey (commensalism) into the liquid (Bristowe 1939). Despite these established spider-pitcher plant associations, little is known about the spider taxa associated with senescent flowers and fruits of S. purpurea. During studies of Endothenia daeckeana (Kearfott), an insect that attacks ovaries of the Northern Pitcher Plant in Québec and elsewhere, Hilton (1982) noted that spiders commonly reside in the protected area afforded by the enlarged, inverted umbrella that covers the ovary of pitcher plant flowers; however, the spiders occupying these protective shelters were not identified. In Massachusetts, Ellison (2005) observed sac spiders (Family Clubionidae) and crab spiders (Family Thomisidae) preying on specialist pollinators that visit Sarracenia flowers to collect nectar, thereby interrupting cross-fertilization. He hypothesized that predation by Clubiona obesa Hentz (Leafcurling Sac Spider) on adult flies attracted to these nectar sources may reduce fruit set and seed production. Here we describe the species of spiders found in seed heads (terminology of Slack 1980) of S. purpurea at four bog-heaths in Acadia National Park; compare spider-species compositions and abundances among sampled sites; denote the types of spider-plant associations (i.e., single vs. multiple inhabitants, conspecific vs. interspecific inhabitants) and their observed frequencies; and discuss possible spider-plant interactions. These observations represent the first recorded information about spider taxa associated with seed heads of the Northern Pitcher Plant in Maine. Methods Study sites We collected seed heads of S. purpurea during July–August 2001 at four study sites in Acadia National Park, Hancock County, ME. Three of the sites, Sunken Heath (44°22.8'N, 68°17.8'W), Duck Brook Heath, also known as New Mill Meadow Heath (44°23.5'N, 68°14.3'W), and Duck Pond Bog 2008 D.T. Jennings, B. Cutler, and B. Connery 525 (44°19.2'N, 68°22.8'W) are located on Mount Desert Island. The fourth site, Hanging Bog (44°20.4'N, 68°03.1'W), is farther east on Schoodic Peninsula. In addition to Sarracenia, plant-species compositions at each site are typical of northeastern bogs and heaths (e.g., species of Kalmia, Myrica, and Sphagnum). The vegetation at Hanging Bog represents a later successional stage with invasion of Larix laricina (Du Roi) K. Koch (Eastern Larch) and Thuja occidentalis L. (Northern White Cedar). Plant samples The floral parts of Sarracenia species were described in detail by Mc- Daniel (1971). Slack (1980) provides a diagrammatic cross-section of the flower and illustrates the seed head. The purplish-red flowers of S. purpurea are borne singly on a tall nodding scape. The 5-carpellate, 5-locular ovary is subtended by an umbrella-shaped style ca. 4–5 cm wide. After flowering, the scape becomes brittle, and some of the floral parts (petals, sepals, and stamens) wither and drop off. During senescence, the 5 lobes of the umbrella generally curl inward and turn from greenish-yellow to brown. The 5-chambered ovary dehisces, thereby releasing and spreading the laterally winged seeds (Folkerts 1999, McDaniel 1971). In Maine, S. purpurea generally flowers in late June and early July, followed by fruiting and senescence in July and August. Seed heads may persist over winter into the following year. On 14 July 2001, at Sunken Heath, Acadia National Park, we discovered that spiders inhabit senescent seed heads of S. purpurea. Apparently the curled lobes of the umbrella, and the vacant chambers of the ovary, provide protected shelters (i.e., refugia) for spiders. To explore this apparently unique spiderplant association, we collected samples of Sarracenia seed heads at Sunken Heath, and at three additional bog-heaths within the Park. At each study site, we collected seed heads ad libitum, i.e., as encountered. Most were found near the moist edges of the sampled bogs and heaths, a favored habitat of Sarracenia pitcher plants (McDaniel 1971). We excluded old, weathered seed heads devoid of umbrella bracts, but included seed heads with open, partially open, and closed umbrellas. The seed heads were taken by gently cutting or breaking-off the stiff scape ca. 4–5 cm below the nodding umbrella. Initially, we placed each sampled seed head into one of two large freezer bags (ca. 50/bag). Although convenient, this procedure proved to be inefficient because a few (<5) spiders emerged from their individual retreats before the seed heads could be dissected in the laboratory. These “loose” spiders, however, were readily identifiable and could be assigned to an individual seed head based on adult and corresponding juvenile identities. Subsequently, all field-collected seed heads were placed individually into a small plastic bag for transport to the laboratory. Sample sizes The number of collected seed heads varied depending on availability and observer time constraints. Sample sizes ranged from 30–102 per study site; 526 Northeastern Naturalist Vol. 15, No. 4 sampling dates in 2001 were: Sunken Heath-1st sampling, 14 July; Sunken Heath-2nd sampling, 8 August; Hanging Bog, 9 August; Duck Brook Pond, 10 August; Duck Pond Heath, 17 August. Sunken Heath was sampled twice; the remaining three sites were sampled only once. Seed-head dissections We dissected the collected seed heads within 0–2 days after collection. The Sunken Heath-1st samples were stored in a car trunk overnight, and dissected the following day at room temperature. Subsequent samples were stored in a cooler or refrigerator at ca. 5 ºC until dissected at room temperature. The seed heads were dissected individually, either as is or submerged in 70% ethanol. Because of spider mobility and concealment, submergence in alcohol proved to be more reliable for determining the exact location of adults and juveniles within their retreats. All specimens were preserved in 70% ethanol and stored in 2-dram vials with labels bearing locality, date, habitat, sample number, and collector. For each dissected seed head we recorded: the number of adult and juvenile spiders and their specific location within the seed head (e.g., under umbrella bract or inside ovary chamber); the presence or absence of spider-silk retreats (Fig. 1) and their location; presence or absence of spider egg sacs, eggs, and juveniles within the retreats; number and identity of non-resident spider exuviae and other seed-head associates (e.g., beetles, mites, parasitoids). In addition, we recorded the condition of each collected seed head; for example, umbrella mostly open or closed; umbrella partially open-closed; umbrella fragmented; ovary mostly open (dehisced) or closed; ovary partially open-closed; ovary fragmented (capsule walls missing); petals present or absent; and sepals present or absent. Spider identifications The collected spiders were examined submerged in 70% ethanol with a Leica™ MZ8 stereo-microscope (max. 80X) equipped with fiber-optic lighting. Adult spiders and their associated offspring were determined to species; juveniles not associated with an adult were determined to genus. In some cases, the probable species of juveniles was indicated based on characteristic color markings. Cast exuviae of juveniles were determined to genus if sufficient diagnostic parts (e.g., carapace, legs) were available; however, most exuviae were determined only to family. Spider identification manuals, keys, and revisionary works were consulted for all species determinations. For pertinent literature on spider families and genera in North America north of Mexico, see Ubick et al. (2005). Within each spider-foraging guild (i.e., web-spinner, hunter), enumeration of taxa generally follows that of Platnick (2008). Except for four specimens retained by B. Cutler, all collected specimens and their associated retreats are deposited in the Acadia National Park museum at Winter Harbor, ME. Data analyses Descriptive statistics (means ± SE) were calculated for count data. Observed proportions of seed heads with/without spider webbing, with/without 2008 D.T. Jennings, B. Cutler, and B. Connery 527 spiders, single vs. multiple occupancy, and conspecific vs. heterospecific associations were subjected to G-tests (Sokal and Rohlf 1981) at α = 0.05. For these comparisons, the null hypothesized expected frequency was set at 50% of n. Results Spider taxa Seed heads (n = 301) of the Northern Pitcher Plant collected at Acadia National Park during July–August 2001, yielded 685 spiders of 4 families, Figure 1. Nesting retreat of the jumping spider Eris militaris (Araneae: Salticidae) on a seed head of Sarracenia purpurea (Northern Pitcher Plant). SR = salticid-nesting retreat; F = frass of unknown lepidopterous larva. 528 Northeastern Naturalist Vol. 15, No. 4 10 genera, and at least 11 species (Table 1). Two additional families (Gnaphosidae, Thomisidae) were represented by cast exuviae, thus increasing the associated seed-head spiders at the Park to 6 families, 12 genera, and at least 13 species. This finding is a conservative estimate because juveniles not identified to species may represent more than one species. The associated spider taxa were unevenly distributed among the four sampled sites at Acadia National Park (Table 1). Jumping spiders (Salticidae) Table 1. Spiders (Araneae) associated with seed heads of Sarracenia purpurea at Acadia National Park, ME, July–August, 2001. Number of individuals given by spider life stage, sex, and remains, where M = male, F = female, J = juvenile, and Ex = exuviae. Number of individuals Spider taxaA SiteB M F J Ex Web spinners THERIDIIDAE (comb-footed spiders) Theridion sp. S 1 DICTYNIDAE Dictyna sp. S 1 Hunters CLUBIONIDAE (sac spiders) Clubiona bishopi Edwards S, H 2 Clubiona sp. S, B 2 1 GNAPHOSIDAE (ground spiders) Undet. genus, sp. S 1 THOMISIDAE (crab spiders) Misumenops sp. P 1 SALTICIDAE (jumping spiders) Eris militaris (Hentz) S, H, P 1 10 27 Eris sp. (prob. militaris (Hentz)) S, H, P, B 10 1 Eris sp. H 1 Evarcha hoyi (Peckham & Peckham) S, B 2 Pelegrina sp. (prob. proterva (Walckenaer)) S, B, P 18 Pelegrina sp. S, H, B 12 Phidippus clarus Keyserling S, B, P 4 54 Phidippus sp. (prob. whitmani Peckham & Peckham) S 1 Phidippus sp. S 1 Sitticus palustris (Peckham & Peckham)C S 1 37 Sitticus sp. (prob. palustris (Peckham & Peckham)) S 54 Synageles sp. H, B 11 Tutelina similis (Banks) S, B 24 179 Tutelina sp. (prob. similis (Banks)) S, B 234 Tutelina sp. S 3 Undetermined genus, sp. S, H, B, P 12 29 Totals 1 43 641 50 AWithin each spider foraging guild, enumeration of taxa generally follows Platnick (2008). BStudy site abbreviations and GPS coordinates: S = Sunken Heath, 1st and 2nd sampling, 44º22.8'N, 68º17.8'W; H = Hanging Bog, 44º20.4'N, 68º03.1'W; B = Duck Brook Heath, 44º23.5'N, 68º14.3'W; P = Duck Pond Bog, 44º19.2'N, 68º22.8'W. CPrószyński (1980) designated Sitticus palustris as a subspecies of the European Sitticus floricola (C.L. Koch); however, most North American salticid workers have not accepted this synonymy. 2008 D.T. Jennings, B. Cutler, and B. Connery 529 inhabited seed heads at all four study sites, whereas sac spiders (Clubionidae) inhabited seed heads at two sites, Sunken Heath and Hanging Bog. The remaining spider families (Theridiidae, Dictynidae, Gnaphosidae, and Thomisidae) were represented by individuals or exuviae at only one site each (Table 1). The number of spider species per sample site ranged from 4 to 10 (mean 6.75 ± 1.4, n = 4); Sunken Heath had the greatest number of associated species, Duck Pond Bog the fewest. The number of spider species per sampling date ranged from 4 to 8 (mean = 6.0 ± 0.8, n = 5); Sunken Heath-2nd and Duck Brook Heath yielded the most species per sampling date, and Duck Pond Bog the fewest. Species inhabiting seed heads at only one sampled site included Phidippus sp. (prob. whitmani) and Sitticus palustris, both at Sunken Heath (Table 1). Four species, Clubiona bishopi, Evarcha hoyi, Synageles sp., and Tutelina similis inhabited seed heads of Sarracenia purpurea at two sites each. A succession of associated spider species was evident at Sunken Heath as the season progressed. Sitticus palustris was found on 14 July during the first sampling at Sunken Heath, but absent during the second sampling on 8 August when other species, including Eris militaris, Pelegrina sp. (prob. proterva), were common. Females of Evarcha hoyi and Phidippus clarus Keyserling were absent during the first sampling at Sunken Heath, but coinhabited seed heads with their egg sacs and juvenile spiderlings during the second sampling in August. Adults and juveniles of Eris militaris occupied seed heads of Sarracenia purpurea at all four study sites, but most frequently at Sunken Heath (Table 1). Of the associated spider genera, only Phidippus was represented by more than one species; i.e., Phidippus sp. (prob. whitmani) in mid-July and P. clarus later in August. Spider abundances Spider abundances within S. purpurea seed heads differed by taxa, foraging guild, developmental stage, sex, season, and sampled locality. At each site, the Salticidae comprised >95% of the observed spider fauna; overall site mean = 98.6 ± 1.9%; range = 95.5–100.0%, where n = 685. Within the Salticidae (N = 679), rank-order abundances of represented genera were: Tutelina (64.4%), Sitticus (13.5%), Phidippus (8.7%), Eris (7.1%), and Pelegrina (2.6%). Collectively, the observed abundances of the remaining salticid genera (Evarcha, Synageles) made up <2% of the seed-head spiders. The Clubionidae was the second most abundant family; however, members of this sac spider family comprised <1% of the overall number of spiders found inhabiting seed heads at Acadia National Park during 2001. Spider-foraging guild representation among the collected seed heads clearly favored hunting spiders over web spinners; for hunters, n = 683 individuals, or 99.7% of all collected spiders; for web spinners, n = 2 individuals, or 0.3% of all collected spiders. At all sampled sites except Duck Pond Bog, juvenile spiders out-num530 Northeastern Naturalist Vol. 15, No. 4 bered adult spiders in the collected seed heads. The observed juvenileto- adult ratios were: Sunken Heath-1st, 6.6:1; Sunken Heath-2nd, 26.3:1; Hanging Bog, 1.4:1; Duck Brook Heath, 41.2:1; Duck Pond Bog, 0.7:1. The overall site ratio of juvenile to adult spiders was 14.6:1. The sex of adult spiders inhabiting seed heads was extremely skewed in favor of females; collectively, 43 females to 1 male over all sites. On 9 August, a single male of Eris militaris was found cohabiting with a penultimate female of the same species at Hanging Bog. The male had spun a large silk retreat on the flower umbrella, whereas the female was in a smaller silk retreat spun between dehisced walls of the senescent flower ovary. Temporal variability in spider abundance among seed heads was evident only at Sunken Heath; all remaining sites were sampled only once. At Sunken Heath, the first sampling of 102 seed heads on 14 July yielded 107 spiders; the second sampling of 100 seed heads on 8 August yielded 382 spiders, an increase of 275.0%. This increase was due chiefly to the emergence of Tutelina similis juveniles from eggs. Spider abundances among seed heads at the remaining three study sites were generally less than those observed at Sunken Heath, possibly related to smaller sample sizes (i.e., <40 seed heads/site). Seed head - spider association frequencies Despite unequal sample sizes, observed seed heads of S. purpurea with spider webbing or retreats were greater than expected (i.e., 50% of n) at three of the study sites, but not different from that expected at Sunken Heath-2nd or at Duck Pond Bog (Table 2). Collectively over all sites, the total observed frequency of seed heads with spider webbing or retreats was greater than expected (G = 23.2, P < 0.001, n = 192). These results indicate that spiders visit and often establish silken retreats within the seed heads of S. purpurea at Acadia National Park. Actual occupation frequencies, as evidenced by one or more spiders per sampled seed head, were less than expected at three of the sites, and not different from expected at Hanging Bog, and at Duck Brook Heath (Table 2). Collectively over all study sites, total occupation frequency of S. purpurea seed heads by spiders was substantially less than expected (G = 29.2, P < 0.001, n = 104). These results indicate that, although spiders may visit seed heads of S. purpurea, they are often not present within such microhabitats. The spiders found within seed heads of S. purpurea represent several categories of association. First, seed heads with spiders were grouped into single vs. multiple inhabitants (Table 3). Seed heads with single-spider occupants (n = 39) were encountered less frequently than those with multiplespider occupants (G = 6.57, P < 0.025, n = 65). Most of the single-occupant associations involved juveniles of Sitticus sp. at Sunken Heath-1st; juveniles of Pelegrina sp. and Eris sp. at Sunken Heath-2nd; adults or juveniles of Eris militaris at Hanging Bog and at Duck Pond Bog, and, juveniles of Eris sp. and Pelegrina sp. at Duck Brook Heath. Interestingly, six of seven re2008 D.T. Jennings, B. Cutler, and B. Connery 531 treats inhabited by single females of E. militaris also contained the remains of eggs, but no juvenile spiderlings. Likewise, at Duck Pond Bog, one of two retreats of E. militaris had a single female and remains of eggs, but no Table 2. Observed vs. expected frequencies of Sarracenia purpurea L. seed heads with spider webbing-retreats, and those with spiders, at four bog-heaths of Acadia National Park, 2001. G-tests, ∝ = 0.05 (Sokal and Rohlf 1981). N.s. = not significant. n Location (seed heads) Expected f Observed f G P Sunken Heath-1st 102 Webbing-retreats 51.0 69 12.98 <0.001 Spiders 51.0 27 23.51 <0.001 Sunken Heath-2nd 100 Webbing-retreats 50.0 49 0.04 N.s. Spiders 50.0 38 5.82 <0.05 Hanging Bog 39 Webbing-retreats 19.5 34 24.19 <0.001 Spiders 19.5 15 2.10 N.s. Duck Brook Heath 30 Webbing-retreats 15.0 26 18.03 <0.001 Spiders 15.0 19 2.16 N.s. Duck Pond Bog 30 Webbing-retreats 15.0 14 0.13 N.s. Spiders 15.0 5 14.56 <0.001 Over all sites 301 Webbing-retreats 150.5 192 23.19 <0.001 Spiders 150.5 104 29.21 <0.001 Table 3. Observed vs. expected frequencies of spiders associated with seed heads of Sarracenia purpurea L. at four bog-heaths in Acadia National Park, 2001. Compared associations are single vs. multiple spider occupants of seed heads. G-tests, α = 0.05 (Sokal and Rohlf 1981). N.s. = not significant. n Location (seed heads) Expected f Observed f G P Sunken Heath-1st 27 single 13.5 10 multiple 13.5 17 1.84 N.s. Sunken Heath-2nd 38 single 19.0 13 multiple 19.0 25 3.86 <0.05 Hanging Bog 15 single 7.5 7 multiple 7.5 8 0.07 N.s. Duck Brook Heath 19 single 9.5 6 multiple 9.5 13 2.64 N.s. Duck Pond Bog 5 single 2.5 3 multiple 2.5 2 0.20 N.s. Over all sites 104 single 52.0 39 multiple 52.0 65 6.57 <0.025 532 Northeastern Naturalist Vol. 15, No. 4 juvenile spiderlings. Apparently, the young spiderlings had emerged and dispersed from these female-inhabited shelters before the seed heads were collected. Seed heads with multiple-spider occupants (n = 65) included those inhabited by the same species of spider (i.e., conspecific associations), and those inhabited by different species of spiders (i.e., heterospecific associations). Conspecific associations (n = 56) were more frequent among the collected samples than heterospecific associations (n = 9); hence, the observed conspecific frequencies were greater than expected (G = 37.83, P < 0.001). Collectively over all sites, these multiple-conspecific associations included juveniles without adults (n = 24), adult females with eggs (n = 15), adult females with juveniles (n = 13), adult females with eggs and juveniles (n = 3), and adult male-penultimate female in cohabitation (n = 1). At Sunken Heath-1st, two adult females of Tutelina similis occupied the same seed head of S. purpurea, but in separate retreats; one with an egg sac containing eggs, the other with 15 post-embryonic juveniles. Heterospecific associations of spiders in seed heads of S. purpurea at Acadia National Park generally involved juveniles of Pelegrina, Eris, or Sitticus in the same seed head with adults and juveniles of T. similis. Apparently Pelegrina juveniles are attracted to and co-inhabit retreats previously spun by females of T. similis. We found two instances of single Pelegrina juveniles co-inhabiting nesting retreats spun by T. similis females; both retreats contained offspring (eggs or juvenile spiderlings) of T. similis. Three heterospecific associations involved single juveniles of Pelegrina associated with multiple juveniles of Tutelina, all in retreats devoid of adults. Four additional retreats inhabited by Tutelina juveniles had cast exuviae of Pelegrina juveniles adhering to their exterior surfaces. Seed-head condition and spider associations Several conditions were represented among the sampled seed heads; for example, both umbrella and ovary closed; umbrella closed, ovary open; umbrella open, ovary closed; both umbrella and ovary open. In addition, we occasionally encountered seed heads with some floral-fruit parts missing; for example, one or more lobes of umbrella missing; one or more carpels of ovary (capsule) missing. Because these conditions varied widely within and among study sites, we assigned each sampled seed head to one of two broad categories (open, closed) based on the preponderance of each condition. The closed category included seed heads with both umbrella and ovary closed, as well as, umbrella closed, ovary open. We found no evidence that spiders preferred any particular condition of S. purpurea seed heads with floral parts. Collectively over all study sites, the observed frequencies of seed heads with spider webbing or retreats did not differ from expected frequencies of 50% in each condition category; observed n = 98 closed, 94 open; G = 0.08, P > 0.90. Likewise, and collectively over all study sites, the observed frequencies of seed heads with spider occupants did not differ from expected frequencies of 50% in each condition 2008 D.T. Jennings, B. Cutler, and B. Connery 533 category; observed n = 58 closed, 46 open; G = 1.39, P > 0.50. Spider parasites, parasitoids, and other associates At Sunken Heath-2nd, seed heads of S. purpurea yielded three females of Tutelina similis, each infested with parasitic mites (Acari). One female spider had two mites; the other two females had one mite each. Mite attachment sites included legs, booklung, and abdomen of spider hosts. All three mite-infested females had broods of young spiderlings. At Hanging Bog, seed heads of S. purpurea yielded six jumping-spider retreats that were infested with tiny wasps. A sub-sample of these wasps was later identified as Idris sp. (Hymenoptera: Scelionidae), a known parasitoid of spider eggs. Most of the parasitoid-infested retreats contained numerous adult wasps and the remains of host-spider eggs, but few adult spiders. Other arthropods found within the seed heads at Acadia National Park included: mites (Acari); cadavers of flies (Diptera); psocids (Psocoptera); a plant bug (Homoptera); beetles (Coleoptera); live and dead ants (Hymenoptera: Formicidae), wasp cocoons, and live and dead wasps (Hymenoptera); lepidopterous larvae and pupae (Lepidoptera); and two unidentified, segmented larvae. Most of the arthropod associates were found in seed heads collected at Sunken Heath and Hanging Bog; fully 71% (n = 24) were associated with spiders or spider retreats. Several of the seed heads collected at Acadia National Park had been attacked by lepidopterous larvae, as evidenced by the presence of larvae, larval frass, and pupae. Apparently, the larvae bore into the fruits and feed on the seeds and other tissue. Such feeding activity produces copious frass (see Fig. 1) and plant debris, which we observed adhering to silk of associated spider retreats. We also observed seeds of S. purpurea adhering to spider silk, and especially to silk of retreats spun near dehisced ovaries. Discussion Spider taxa The species of spiders we found inhabiting seed heads of S. purpurea at Acadia National Park are widely distributed in the northeastern United States and Canada. All associated species have been recorded from other localities in Maine; all except Clubiona bishopi, Synageles sp., and possibly Pelegrina proterva have been recorded from the Mount Desert Region of Acadia National Park (Procter 1946). None of the identified species, however, has been recorded previously from the seed heads of S. purpurea in Maine or possibly elsewhere. Some of the spider families and genera found during this study are known to be associated with S. purpurea and other species of pitcher plants elsewhere (Ellison 2005, Folkerts 1999, Juniper et al. 1989, Nyoka and Ferguson 1999). In a Massachusetts bog, Ellison (2005) noted that sac spiders (Clubionidae, Clubiona) and crab spiders (Thomisidae) were associated with Sarracenia flowers; he also noted five species of sheet-web weavers (Linyphiidae) associated with Sarracenia pitchers. We did not examine 534 Northeastern Naturalist Vol. 15, No. 4 pitchers for associated spiders at Acadia National Park. Nyoka and Ferguson (1999) recorded spiders of 8 families, 9 genera, and at least 10 species dusted with pollen of Darlingtonia californica in an Oregon fen. This assemblage of spider species associated with D. californica differs from that associated with seed heads of S. purpurea in Maine. Such differences in faunal composition are most likely due to differences in sampling method and substrate sampled, i.e., spiders captured in or near flowers of D. californica vs. spiders dissected from seed heads of S. purpurea. Nonetheless, four spider families (Clubionidae, Theridiidae, Thomisidae, and Salticidae) are shared in common. Spider abundances The number of spiders we found inhabiting seed heads of S. purpurea at Acadia National Park far exceeds those previously recorded for this and other species of pitcher plants (Folkerts 1999, Juniper et al. 1989, Wray and Brimley 1943). Not counting spider eggs, we found 685 adult and juvenile spiders inhabiting seed heads of S. purpurea; by contrast, Wray and Brimley (1943) collected 226 (218 unidentified) spiders from the pitchers of Sarracenia flava L. (Yellow Pitcher Plant) in North Carolina. Without question, the spider abundances we observed associated with seed heads of S. purpurea in Maine were due chiefly to nesting female salticids and their progeny, and especially those of Tutelina similis. The observed variability in developmental-stage abundances (i.e., juveniles vs. adults) can be attributed to reproductive-period differences of individual species, e.g., those of Sitticus palustris, T. similis, and Phidippus clarus (Table 1). The preponderance of seed heads with multiple-spider occupants, compared to those with single-spider occupants, supports these conclusions. Spider-nesting habitats The reproductive behaviors of spiders vary widely among families and among species within families (Foelix 1996, Gertsch 1979, Nentwig and Heimer 1987). After depositing their eggs, all spiders cover their eggs with varying amounts of silk. Such coverings are called egg sacs or egg cocoons, and allegedly provide protection against desiccation, predators, and parasitoids (Foelix 1996; Hieber 1992a, 1992b). Maternal egg and brood care varies widely among spiders; hunting spiders, female salticids, clubionids, and some thomisids (Thomisidae) and gnaphosids (Gnaphosidae) usually deposit their egg sacs in silken retreats where the female remains until egg hatch and dispersal of the young spiderlings. Such egg retreats or nests are usually spun in protected shelters: for example, in rolled or folded leaves; under loose bark of stumps, logs, and tree boles; under rocks and in ground litter; and in, on, or under man-made structures. Unfortunately, few detailed studies have been made of the specific microhabitats and range of microhabitats selected by hunting spiders for construction of their maternal egg retreats or nests. Most information is 2008 D.T. Jennings, B. Cutler, and B. Connery 535 anecdotal; however, for salticids, see Edwards 2004, Jackson 1979, Jackson and Griswold 1979, Richman and Jackson 1992, and Tessler 1979. Of the 11 species of spiders associated with seed heads of Sarracenia purpurea at Acadia National Park, only four salticid species had spun nesting retreats; i.e., Eris militaris, Phidippus clarus, Sitticus palustris, and Tutelina similis. Previously recorded information about the nesting habitats of these four species is indeed limited. Although E. militaris is common in New England and southern Canada, its nesting habitats remain elusive. In Connecticut, Kaston (1981) noted that a female of Paraphidippus marginatus (Walckenaer) (now Eris militaris) guarded an egg sac fastened to the underside of a rolled leaf, but gave no information about leaf identity. Tessler (1979) found a nesting retreat of E. marginata (Walckenaer) (now E. militaris) within the confines of a eumenid wasp nest in Indiana. Our observations of retreats spun within the seed heads of S. purpurea represent a previously unknown nesting habitat for this salticid. Besides seed heads of the Northern Pitcher Plant, nests of Phidippus clarus were found on the apices of Hypericum perforatum L. (Common St. Johnswort) in southern Maine (D.T. Jennings, unpubl. data). In Indiana, Tessler (1979) observed nesting retreats of P. clarus in the umbels of Daucus carota L. (Queen Anne’s Lace), and in the tops of Rumex crispus L. (Curly Dock). In Kansas, Johnson (1995) noted that P. clarus reoccupied nests previously spun by Hibana gracilis (Hentz) (Garden Ghost Spider) (Family Anyphaenidae) in the expanded leaves of Asclepias sp. (milkweed). Edwards (2004) noted that P. clarus makes large white egg retreats in the tops of plants in old fields and weedy areas of open woodland, but gave no indication of plant identities. The nests we observed in Maine were similar to those described by Edwards (2004) for this salticid, but in a previously unknown microhabitat. In addition to seed heads of S. purpurea, the nests of Sitticus palustris in Maine also are found on dried inflorescences of Spiraea alba Du Roi var. latifolia (Ait.) Dippel (Meadowsweet) and on fallen curled leaves of Acer rubrum L. (Red Maple) (Jennings and Graham 2007). In Connecticut, Kaston (1981) observed females of S. palustris guarding egg sacs, but gave no details. Females of the related European Sitticus f. floricola (C.L. Koch) spin egg retreats on Juncus sp. in Poland (Prószyński 1980), and on seed heads of Eriophorum angustifolium Honckeny (Cotton Grass) in Great Britain (Wallace and Wallace 1991). We suspect that seed heads of Cotton Grass serve as a nesting microhabitat for S. palustris in Maine. Before our study, the nesting habitats of Tutelina similis were virtually unknown. Kaston (1981) noted that a female of Icius similis Banks (now T. similis) guarded an egg sac spun beneath loose bark of a tree in Connecticut. Unfortunately, the species of tree was not given. Our observations in Maine provide a previously unknown nesting habitat for T. similis. The behavioral mechanisms that govern the selection of microhabitats 536 Northeastern Naturalist Vol. 15, No. 4 for nesting are largely unknown for most species of spiders. It has been widely assumed that such selection is largely opportunistic; i.e., if a suitable habitat is encountered by a gravid female, then it (the habitat) is used. Our data suggest otherwise. Seed head-spider association frequencies The relatively high percentage of pitcher plant seed heads with spider webbing or retreats (63.8%, n = 301) indicates that spiders are common visitors, but not necessarily residents, of these microhabitats. The commonality of such spider-plant associations was due chiefly to species of Salticidae that utilized these protective shelters for egg laying and rearing of young. Spider residency in seed heads of S. purpurea, however, appears to be temporally mediated and influenced by life-stage development. We suspect that the relatively low percentage of seed heads with spiders (34.6%, n = 301) can be attributed to: 1) population densities of females seeking shelters; 2) juveniles abandoning moulting retreats; 3) parasitoids attacking eggs before egg hatch; 4) juveniles dispersing after egg hatch; 5) predators attacking eggs, juveniles, or adults; and 6) adult females making temporary foraging forays away from their nest. Our results indicate that seed heads sampled during mid-July and early August most likely will yield multiple-conspecific associations of spiders, i.e., if sufficient numbers of seed heads are collected. In Maine, spiders generally reproduce during mid- to late summer, especially E. militaris, P. clarus, S. palustris, and T. similis. Multiple-heterospecific associations of spiders in seed heads appear to be less common, perhaps due to pre-occupation of suitable nesting sites, competition, or araneophagy. The presence of juveniles and cast exuviae of juveniles on nests of T. similis indicates that the silk of pre-existing retreats may provide an anchoring platform for moulting by conspecifics and by heterospecifics. Jackson and Griswold (1979) noted several organisms, including spiders and cast exuviae of spiders, were found in the nests of Phidippus johnsoni (Peckham & Peckham) (Johnson Jumper) in California and Wyoming. Spider parasites, parasitoids, and other associates Our findings provide new host records for parasitic mites infesting females of Tutelina similis and Idris parasitoids emerging from eggs of E. militaris. We suspect that the parasitic mites are a species of Leptus (Erythraeidae) or Trombidium (Trombidiidae); both species parasitize Enoplognatha ovata (Clerck) (Cobweb Spider), a theridiid spider commonly found in diverse habitats along coastal Maine (Reillo 1989). Spider eggs, including those enclosed in egg sacs and nesting retreats, are not immune to attack by predators and parasitic insects. These natural enemies of spider eggs include flies (Diptera), wasps (Hymenoptera), and mantispids (Neuroptera) (Austin 1985, Eason et al. 1967, Hieber 1992b, Redborg 1983). The egg parasitoids we found within nests of E. militaris are a species of Idris (Hymenoptera: Scelionidae), a genus currently under 2008 D.T. Jennings, B. Cutler, and B. Connery 537 revision. Austin (1985) provides a list of spider families and species that serve as hosts for Idris in Australia; Eason et al. (1967) provides notes on the life history and behaviors of a species of Idris recovered from eggs of the lycosid Pardosa lapidicina Emerton (Lycosidae) (Wolf Spider) in Arkansas. Virtually nothing is known about the life history and habits of Idris species attacking spider eggs in Maine. Spider-pitcher plant interactions Clearly, senescent flowers of S. purpurea are used by salticid spiders for moulting, prenuptial mate-guarding, nesting, and rearing of young; the spiders spin silken cocoons, nests, and retreats among the floral components of senescing flowers and seed heads. These silken structures allegedly provide spiders some protection against abiotic and biotic factors (Austin 1985, Hieber 1992b, Jackson 1979, Richman and Jackson 1992). We suspect that the curled, closed or partially closed, umbrella bracts of pitcher plant seed heads also provide spiders some protection against the elements, and may act as impediments to potential predators, parasites, and parasitoids. However, we did not detect a preference on the part of spiders for any particular seed-head condition. How spiders encounter and select these specific microhabitats is largely unknown. Such encounters and selections may be purely accidental and opportunistic; obviously, choice experiments involving plants and their floral components may help to elucidate the behavioral mechanisms and processes involved (e.g., see Morse 1985). Our data suggest that spider-pitcher plant associations are not strictly accidental. The observed frequencies of spider webbing and retreats in seed heads of S. purpurea either met or exceeded the null-hypothesized expected frequencies. However, spiders are not always present in these microhabitats; some abandon their nests or retreats after moulting, egg-laying, and rearing of young, and during temporary foraging forays. Although statistically appropriate, our null-hypothesized expected frequency of equal proportions (i.e., 50% of sampled seed heads inhabited by spiders) may be biologically inappropriate or unrealistic. Without measuring spider population densities, spider nesting-moulting behaviors, and densities of pitcher plants and other potential nesting sites (e.g., curled leaves, dry inflorescences) in these plant communities, the expected frequency of spiders inhabiting seed heads of the Northern Pitcher Plant remains speculative. Our observations of pitcher plant seeds adhering to silk of some spider retreats, and especially retreats of Eris and Phidippus, possibly indicates that seed dispersal might be impeded. Such entanglements may have resulted during the collection, transport, and dissection of fruits; adherence of pitcher plant seeds to spider silk was not observed and measured in situ at Acadia National Park. Sarracenia purpurea also reproduces asexually by rhizomes (McDaniel 1971). In some plants, alteration of flowers by spiders has a minimal effect on seed production (Ott et al. 1998) and enhances seed production in others, that is to say by defense (Ruhren and Handel 1999). 538 Northeastern Naturalist Vol. 15, No. 4 Spiders that prey on or disrupt the feeding activities of herbivorous and frugivorous insects (e.g., larvae of Exyra and Endothenia, see Folkerts 1999) may be beneficial to S. purpurea reproduction. Although we did not observe spiders preying on insect inhabitants of seed heads, the presence of insect cadavers indicates possible predation by spiders. We frequently encountered spider-inhabited seed heads previously damaged by lepidopterous larvae, but the larvae were absent. We suspect that resident spiders may have disrupted the feeding activities of these frugivorous insects, or possibly fed on the larvae and afterwards discarded their cadavers. The presence of dead ants in some sampled seed heads may be attributable to predation by T. similis. In Utah, T. similis consistently stalks and preys on ants, including seed-gathering harvester ants associated with Artemisia tridentata Nutt. (Big Sagebrush) (Wing 1983). The microcosm of S. purpurea seed heads warrants further investigation by ecologists and araneologists. Acknowledgments We gratefully acknowledge the enthusiastic assistance of Elvira Flores and Paul Wilson, Biological Technicians, Acadia National Park. Arlene Banks and James Bird, University of Maine, Fogler Library, provided much-needed assistance with literature searches, acquisitions, and retrievals. David Manski, Chief Biologist, Acadia National Park, kindly issued a collecting permit. Joni Harper Dunn provided close-up photos of the salticid-nesting retreat; her expertise is greatly appreciated. Portions of this research were supported by: the USDI, Acadia National Park, Schoodic Education and Research Center; the USDA, Forest Service, Northern Research Station; and the Maine Entomological Society. We thank Jerry R. Longcore and two anonymous reviewers for their constructive comments on an earlier draft. Literature Cited Austin, A.D. 1985. The function of spider egg sacs in relation to parasitoids and predators, with special reference to the Australian fauna. Journal of Natural History 19:359–376. Bristowe, W.S. 1939. The Comity of Spiders, Volume I. Ray Society, London UK. Reprinted 1968. Johnson Reprint Corporation, New York, NY. 228 pp. Cresswell, J.E. 1991. Capture rates and composition of insect prey of the pitcher plant Sarracenia purpurea. The American Midland Naturalist 125:1–9. Cresswell, J.E. 1993. The morphological correlates of prey capture and resource parasitism in pitchers of the carnivorous plant Sarracenia purpurea. The American Midland Naturalist 129:35–41. Eason, R.R., W.B. Peck, and W.H. Whitcomb. 1967. Notes on spider parasites, including a reference list. Journal Kansas Entomological Society 40:422–434. Edwards, B.J. 2004. Revision of the Jumping Spiders of the Genus Phidippus (Araneae: Salticidae). Occasional Papers of the Florida State Collection of Arthropods. Volume 11. Florida Department of Agriculture and Consumer Services, Division of Plant Industry and The Center for Systematic Entomology, Gainesville, fl. 156 pp. 2008 D.T. Jennings, B. Cutler, and B. Connery 539 Ellison, A.M. 2005. Turning the tables: Plants bite back. Pp. 25–30, In C. Howell- Walte and M. Shepherd (Eds.). Wings. Publication of the Xerces Society, Portland, OR. Fall 2005. Foelix, R.R. 1996. Biology of Spiders, 2nd Edition. Oxford University Press, Inc. New York, NY. 330 pp. Folkerts, D. 1999. Pitcher plant wetlands of the southeastern United States. Pp. 247–275, In D.P. Batzer, R.B. Radar, and S.A. Wissinger (Eds.). Invertebrates in Freshwater Wetlands of North America: Ecology and Management. John Wiley and Sons, Inc., New York, NY. 1100 pp. Gertsch, W.J. 1979. American Spiders, 2nd Edition. Van Nostrand Reinhold, New York, NY. 272 pp. Hieber, C.S. 1992a. The role of spider cocoons in controlling desiccation. Oecologia 89:442–448. Hieber, C.S. 1992b. Spider cocoons and their suspension systems as barriers to generalist and specialist predators. Oecologia 91:530–535. Hilton, D.F.J. 1982. The biology of Endothenia daeckeana (Lepidoptera: Olethreutidae), an inhabitant of the ovaries of the Northern Pitcher Plant, Sarracenia p. purpurea (Sarraceniaceae). Canadian Entomologist 114:269–274. Jackson, R.R. 1979. Nests of Phidippus johnsoni (Araneae, Salticidae): Characteristics, pattern of occupation, and function. Journal of Arachnology 7:47–58. Jackson, R.R., and C.E. Griswold. 1979. Nest associates of Phidippus johnsoni (Araneae, Salticidae). Journal of Arachnology 7:59–67. Jennings, D.T., and F. Graham, Jr. 2007. Spiders (Arachnida: Araneae) of Milbridge, Washington County, Maine. USDA, Forest Service, Northern Research Station, Newtown Square, PA. GTR-16. 204 pp. Johnson, S.R. 1995. Nests of Hibana gracilis are reused by Phidippus clarus in wetlands of northeastern Kansas. Journal of Arachnology 23:44–45. Judd, W.W. 1959. Studies of the Byron Bog in southwestern Ontario. X. Inquilines and victims of the pitcher-plant, Sarracenia purpurea L. Canadian Entomologist 91:171–180. Juniper, B.E., R.J. Robins, and D.M. Joel. 1989. The Carnivorous Plants. Academic Press Inc., San Diego, CA. 353 pp. Kaston, B.J. 1981. Spiders of Connecticut, Revised Edition. State Geological and Natural History Survey of Connecticut, Hartford, CT. Bulletin 70. 1020 pp. McDaniel, S. 1971. The genus Sarracenia (Sarraceniaceae). Tall Timbers Research Station. Bulletin 9:1–36. Morse, D.H. 1985. Nests and nest-site selection of the crab spider Misumena vatia (Araneae: Thomisidae) on milkweed. Journal of Arachnology 13:383–390. Nentwig, W., and S. Heimer. 1987. Ecological aspects of spider webs. Pp. 211–225, In W. Nentwig (Ed.). Ecophysiology of Spiders. Springer-Verlag, Berlin, Germany. 448 pp. Nyoka, S.E., and C. Ferguson. 1999. Pollinators of Darlingtonia californica Torr., the California Pitcher Plant. Natural Areas Journal 19:386–391. Ott, J.R., J.A. Nelson, and T. Caillouet. 1998. The effect of spider-mediated flower alteration on seed production in Golden-eye Phlox. The Southwestern Naturalist 43:430–436. Platnick, N.I. 2008. The World Spider Catalog. Version 8.5. The American Museum of Natural History. Available online at http://research.amnh.org/entomology/ spiders/catalog/index.html. Accessed 12 January 2008. 540 Northeastern Naturalist Vol. 15, No. 4 Pocock, R.I. 1898. Spider and pitcher-plant. Nature (London) 58:274–275. Pollard, S.D. 2006. Fishing crab spiders in the hanging stomachs of Borneo. Pp. 32, In Abstracts, American Arachnological Society, 30th Annual Meeting, June 2006, Baltimore, MD. 73 pp. Procter, W. 1946. Biological Survey of the Mount Desert Region. The Wistar Institute of Anatomy and Biology. Philadelphia, PA. 566 pp. Prószyński, J. 1980. Revision of the spider genus Sitticus Simon, 1901 (Aranei, Salticidae). IV. Sitticus floricola (C.L. Koch) group. Annales Zoologici (Poland) 36(1):1–35. Redborg, K.E. 1983. A mantispid larva can preserve its spider egg prey: Evidence for an aggressive allomone. Oecologia 58:230–232. Reillo, P.R. 1989. Mite parasitism of the polymorphic spider, Enoplognatha ovata (Araneae, Theridiidae), from coastal Maine. Journal of Arachnology 17:246–249. Richman, D.B., and R.R. Jackson. 1992. A review of the ethology of jumping spiders (Araneae, Salticidae). Bulletin of the British Arachnological Society 9:33–37. Ruhren, S., and S.N. Handel. 1999. Jumping spiders (Salticidae) enhance the seed production of a plant with extrafloral nectaries. Oecologia 119:227–230. Slack, A. 1980. Carnivorous Plants. The Massachusetts Institute of Technology Press, Cambridge, MA. 240 pp. Smith, R.B., and T.P. Mommsen. 1984. Pollen feeding in an orb-weaving spider. Science 226:1330–1332. Sokal, R.R., and F.J. Rohlf. 1981. Biometry. The Principles and Practice of Statistics in Biological Research, 2nd Edition. W.H. Freeman and Company, New York, NY. 859 pp. Swales, D.E. 1969. Sarracenia purpurea L. as host and carnivore at Lac Carré, Terrebonne Co., QC. Naturaliste Canada 96:759–763. Taylor, R.M., and W.A. Foster. 1996. Spider nectarivory. American Entomologist 42:82–86. Tessler, S. 1979. A study of the retreat-sites of jumping spiders (Araneae: Salticidae) nesting on Queen Anne’s Lace (Daucus carota carota L.). M.Sc. Thesis. Purdue University, West Lafayette, IN. 121 pp. Ubick, D., P. Paquin, P.E. Cushing, and V. Roth (Eds.). 2005. Spiders of North America: An Identification Manual. American Arachnological Society, Keene, NH. 377 pp. Wallace, B., and I. Wallace. 1991. Some records and observations on Sitticus floricola (C.L. Koch). Newsletter of the British Arachnological Society (March 1991) 60:3–4. Wing, K. 1983. Tutelina similis (Araneae: Salticidae): An ant mimic that feeds on ants. Journal of the Kansas Entomological Society 56:55–58. Wray, D.L., and C.S. Brimley. 1943. The insect inquilines and victims of pitcher plants in North Carolina. Annals of the Entomological Society of America 36:128–137.